Contents of Final Program

Welcome from the Organizers 1

Conference Organization 2

ICIC2009 Paper Reviewers 5

Conference Sponsor 7

General Information 8

How to Get to Ulsan 9

Hotel Information 10

Social Events 11

Industrial Tour 11

Ulsan City Tour 11

Floor Map of the Hotel 13

ICIC2009 Program at a Glance 14

Keynote Speakers 15

Technical Program Information 18

Technical Program Details

·Thursday, Sept 17 19

·Friday, Sept 18 27

Poster Paper 33

Abstract of Paper

·Thursday, Sept 17 40

·Friday, Sept 18 56

Author Index 67

- 1 -

Welcome from the Organizers

The International Conference on Intelligent Computing (ICIC) was formed to provide an annual forumdedicated to the emerging and challenging topics in artificial intelligence, machine learning, bioinformatics,and computational biology, etc. It aims to bring together researchers and practitioners from both academiaand industry to share ideas, problems and solutions related to the multifaceted aspects of intelligentcomputing.

ICIC 2009, held in Ulsan, Korea, September 16-19, 2009, constituted the Fifth International Conference onIntelligent Computing. It built upon the success of ICIC2008, ICIC2007, ICIC 2006 and ICIC 2005 held inShanghai, Qingdao, Kunming and Hefei, China, 2008, 2007, 2006 and 2005, respectively.

This year, the conference concentrated mainly on the theories and methodologies as well as the emergingapplications of intelligent computing. Its aim was to unify the picture of contemporary intelligentcomputing techniques as an integral concept that highlights the trends in advanced computationalintelligence and bridges theoretical research with applications. Therefore, the theme for this conference was“Emerging Intelligent Computing Technology and Applications”. Papers focusing on this theme weresolicited, addressing theories, methodologies, and applications in science and technology.

ICIC 2009 received 1082 submissions from 34 countries and regions. All papers went through a rigorouspeer review procedure and each paper received at least three review reports. Based on the review reports,the Program Committee finally selected 257 high-quality papers for presentation at ICIC 2009, of which214 papers have been included in two volumes of proceedings published by Springer: one volume ofLecture Notes in Computer Science (LNCS) and one volume of Lecture Notes in Artificial Intelligence(LNAI). The other 22 papers will be included in two international journals.

The organizers of ICIC 2009, including Ulsan University of Korea and the Institute of Intelligent Machinesof the Chinese Academy of Science, made an enormous effort to ensure the success of ICIC 2009. Wehereby would like to thank the members of the ICIC 2009 Advisory Committee for their guidance andadvice, the members of the Program Committee and the referees for their collective effort in reviewing andsoliciting the papers. We would like to thank Alfred Hofmann, executive editor from Springer, for his frankand helpful advice and guidance throughout and for his support in publishing the proceedings. In particular,we would like to thank all the authors for contributing their papers. Without the high-quality submissionsfrom the authors, the success of the conference would not have been possible. Finally, we are especiallygrateful to the IEEE Computational Intelligence Society, the International Neural Network Society and theNational Science Foundation of China for their sponsorship.

September 2009

De-Shuang HuangKang-Hyun JoHong-Hee LeeHee-Jun Kang

Vitoantonio Bevilacqua

Conference Organization

General Chairs:De-Shuang Huang, Institute of Intelligent Machines, Chinese Academy of Sciences, China Honghee Lee, University of Ulsan, KoreaFrank L Lewis, University of Texas at Arlington, USA

Program Committee Chairs: Kanghyun Jo, University of Ulsan, Korea Vitoantonio Bevilacqua, Polytechnic of Bari, Italy

Organizing Committee Chairs: Kang-Hyun Jo, University of Ulsan, KoreaIn-Soo Koo, University of Ulsan, KoreaYoungsoo Suh, University of Ulsan, KoreaNaoyuki Tsuruta, Fukuoka University, JapanChun-Hou Zheng, Institute of Intelligent Machines, Chinese Academy of Sciences, China

Organizing Committee Members: Myeong-Jae Yi, University of Ulsan, KoreaMyung-Kyun Kim, University of Ulsan, KoreaByeong-Ryong Lee, University of Ulsan, KoreaWon-Ho Choi, University of Ulsan, KoreaSang-Bock Cho, University of Ulsan, KoreaMunho Jeong, Korea Institute of Science and Technology, KoreaJongeun Ha, Seoul National University of Technology, KoreaDong-Joong Kang, Pusan National University, KoreaJong-Bae Lee, KETI, KoreaSang-Moo Lee, KITECH, Korea

Award Committee Chair: Daniel S. Levine, University of Texas at Arlington, USA

Publication Chair: Heejun Kang, University of Ulsan, Korea

Special Session Chairs: Prashan Premaratne, University of Wollongong, Australia Tokuro Matsuo, Yamagata University, Japan Vasily Gubarev, Novosibirsk State Technical University, Russia

Tutorial Chair: Laurent Heutte, The University of Rouen, France

International Liaison Chair: Frank Neumann, Max-Plank Institute for Informatics, Germany

Publicity Chairs: Kyungsook Han, Inha University, Korea Vladimir Filaretov, Russian Academy of Sciences, Russia Zhongming Zhao, Virginia Commonwealth University, USA

- 2 -

Maolin Tang, Queensland University of Technology, AustraliaMuhammad Khurram Khan, King Saud University, Saudi ArabiaValeriya Gribova, Far Eastern Branch of Russian Academy of Sciences, Russia

Exhibition Chairs: Young-Soo Suh, University of Ulsan, Korea In-Soo Koo, University of Ulsan, Korea Jin Hur, University of Ulsan, Korea

Program Committee Members:

- 3 -

- Andrea Francesco Abate, University of Salerno,Italy

- Shafayat Abrar, The University of Liverpool,UK

- Peter Andras, Newcastle University, UK- Sabri Arik, Istanbul University, Turkey- Vasily Aristarkhov, Nizhniy Novgorod State

Technical University, Russian Federation- Costin Badica, University of Craiova, Romania- Vitoantonio Bevilacqua, Polytechnic of Bari,

Italy- David B. Bracewell, General Electric Global

Research, USA- Uday K. Chakraborty, University of Missouri,

USA- Shih-Hsin Chen, NanHua University, Taiwan,

China - Wen-Sheng Chen, Shenzhen University, China- Xiyuan Chen, Southeast University, China- Yang Chen, Southeast University, China- Yuehui Chen, Jinan University, China- Sang-Bock Cho, University of Ulsan, Korea- Won-Ho Choi, University of Ulsan, Korea- Michal Choras, University of Technology & Life

Sciences, Poland- Tommy Chow, City University of Hong Kong,

Hong Kong, China- Jose Alfredo F. Costa, Federal University, Brazil - Angelo Ciaramella, University of Naples

“Parthenope”, Italy- Kevin Curran, University of Ulster, UK- Mingcong Deng, Okayama University, Japan- Eng. Salvatore Distefano, University of Messina,

Italy- Karim Faez, Amirkabir University of

Technology, Iran- Jianbo Fan, Ningbo University of Technology,

China- Minrui Fei, Shanghai University, China- Wai-Keung Fung, University of Manitoba,

Canada

- Liang Gao, Huazhong University of Science &Technology, China

- Qing-Wei Gao, Anhui University, China - Xiao-Zhi Gao, Helsinki University of

Technology, Finland- Chandan Giri, Bengal Engineering & Science

University, India- Dunwei Gong, China University of Mining and

Technology, China- Valeriya Gribova, Far Eastern Branch of Russian

Academy of Sciences, Russia - Kayhan Gulez, Yildiz Technical University,

Turkey - Ping Guo, Beijing Institute of Technology, China- Jongeun Ha, Seoul National University of

Technology, Korea- Aili Han, Shandong University, China- Fei Han, Jiangsu University, China- Kyungsook Han, Inha University, Korea- Haibo He, Stevens Institute of Technology, USA- Laurent Heutte, Université de Rouen, France- Wei-Chiang Hong, Oriental Institute of

Technology, Taiwan, China- Yuexian Hou, Tianjing University, China- Peter Hung, National University of Ireland,

Ireland- Chuleerat Jaruskulchai, Kasetsart University,

Thailand - Munho Jeong, Korea Institute of Science and

Technology, Korea- Li Jia, Shanghai University, China- Zhenran Jiang, East China Normal University,

China- Jih-Gau Juang, National Taiwan Ocean

University, Taiwan, China- Dah-Jing Jwo, National Taiwan Ocean

University, Taiwan, China- Dong-Joong Kang, Pusan National University,

Korea- Sanggil Kang, Inha University, Korea - Uzay Kaymak, Erasmus University Rotterdam,

- 4 -

Netherlands- Muhammad Khurram Khan, King Saud

University, Saudi Arabia- Myung-Kyun Kim, University of Ulsan, Korea- Sungshin Kim, Pusan National University, Korea- In-Soo Koo, University of Ulsan, Korea - Donald H. Kraft, Louisiana State University,

USA- Harshit Kumar, National University of Ireland,

Ireland- Yoshinori Kuno, Saitama University, Japan- Takashi Kuremoto, Yamaguchi University, Japan- Wen-Chung Kuo, National Formosa University,

Taiwan, China- Hak-Keung Lam, King’s College London, UK- Byeong-Ryong Lee, University of Ulsan, Korea- Jong-Bae Lee, KETI, Korea- Sang-Moo Lee, KITECH, Korea- Vincent C S Lee, Monash University, Australia- Guo-Zheng Li, Tongji University, China- Kang Li, Queen’s University Belfast, UK- Li Li, Shenzhen University, China- Peihua Li, Hei Long Jiang University, China- Hualou Liang, Drexel University, USA- Chunmei Liu, Howard University, USA- Ju Liu, Shandong University, China- Van-Tsai Liu, National Formosa University,

Taiwan, China- Marco Loog, University of Copenhagen,

Denmark - Ahmad Lotfi, Nottingham Trent University, UK- Jinwen Ma, Peking University, China- Shiwei Ma, Shanghai University, China- Vishnu Vardhan Makkapati, Philips Research

Asia, India - Cheolhong Moon, Gwangju University, Korea- Tarik Veli Mumcu, Technical University of

Kaiserslautern, Germany - Roman Neruda, Academy of Sciences of the

Czech Republic, Czech Republic- Frank Neumann, Max-Plank Institute for

Informatics, Germany- Minh Nhut Nguyen, Nanyang Technological

University, Singapore - Ben Niu, Shenzhen University, China- Sim-Heng Ong, National University of

Singapore, Singapore- Francesco Pappalardo, University of Catania,

Italy- Caroline Petitjean, Université de Rouen, France- Prashan Premaratne, University of Wollongong,

Australia

- Shaoqi Rao, Sun Yat-Sen University, China- Seeja K R, Hamdard University, India - Angel Sappa, Autonomous University of

Barcelona, Spain- Aamir Shahzad, Beihang University, China- Li Shang, Suzhou Vocational University, China - Jiatao Song, Ningbo University of Technology,

China- Nuanwan Soonthornphisaj, Kasetsart University,

Thailand- Joao Miguel Sousa, Technical University of

Lisbon, Portugal- Min Su, Purdue University, USA- Zhan-Li Sun, National University of Singapore,

Singapore- Maolin Tang, Queensland University of

Technology, Australia- Antonios Tsourdos, Cranfield University, UK- Naoyuki Tsuruta, Fukuoka University, Japan - Sergio Vitulano, Cagliari University, Italy- Anhua Wan, Sun Yat-Sen University, China- Chao-Xue Wang, Xi’an University of

Technology, China- Hong-Qiang Wang, University of Georgia, USA - Jinlian Wang, Capital Medical University, China- Ling Wang, Tsinghua University, China- Xueqin Wang, Sun Yat-Sen University, China- Xuesong Wang, China University of Mining and

Technology, China- Yong Wang, Academy of Mathematics and

System Science, CAS, China- Ling-Yun Wu, Chinese Academy of Sciences,

China- Shunren Xia, Zhejiang University, China- Yu Xue, University of Science and Technology

of China, China- Ching-Nung Yang, National Dong Hwa

University, Taiwan, China- Jun-Heng Yeh, Tatung University, Taiwan, China- Myeong-Jae Yi, University of Ulsan, Korea- Zhi-Gang Zeng, Huazhong University of Science

& Technology, China- Jun Zhang, Sun Yat-sen University, China- Yong Zhang, Jinan University, China- Xing-Ming Zhao, Shanghai University, China- Zhongming Zhao, Virginia Commonwealth

University, USA- Bo-Jin Zheng, South-Central University for

Nationalities, China- Fengfeng Zhou, University of Georgia, USA- Huiyu Zhou, Brunel University, UK

ICIC2009 Paper Reviewers

- 5 -

Adam GhandarAdriao DuarteAimin ZhouAizhong MiAjiboye OsunlekeAkira YanouAl SavvarisAlaa SagheerAlan RitzAldayr AraujoAlessandro CincottiAlessandro IbbaAlessandro PerfettoAlessia AlbaneseAlessia D’IntronoAlessio FeroneAlexander HogenboomAlexander KleschevAlexander PonomarenkoAmar BallaAmar KhoukhiAmelia BadicaAnbumani SubramanianAndreas KonstantinidisAndreas SchmidtAndrey LarionovAndrey LogvinovAngelo CiaramellaAngelo RiccioAnne CanutoAntonino StaianoAntonio CelestiAnvita BajpaiAravindan ChandraboseArdelio GallettiArun D. MahindrakarAsaduzzamanAsharaf SAsit DasAsunción MochónAtsushi ShimadaAtsushi YamashitaB V BabuBae-guen KwonBanghua YangBanu DiriBao Vo-NguyenBarbara PizzileoBeijing ChenBeilu ShaoBekir KarlikBen NiuBijaya Ketan PanigrahiBin LiBin QianBing XueBo ChenBo LiuBo-Chao ChengBogdan RaducanuBo-Hyeun WangBunyarit UyyanonvaraC.-H. YangCarme JuliaCarmelo RagusaCaroline PetitjeanChang Wook AhnChangan JiangChangho YunChangyin SunChao WangChao WuChee-Meng Chew

Jiayin ZhouJie GuiJie LeiJie SunJiguang WangJih-Gau JuangJi-Hun BaeJimmy LinJin ZhouJindong LiuJinn-Shing ChengJinpeng QiJinsoo KimJiping SHIJoaquin Torres-SospedraJohn EconomouJohn KleinJong Hyun ParkJong Min LeeJoo Seop YunJoongjae LeeJose Alfredo F. CostaJose AlvarezJosep M. Mirats TurJoydeb MukherjeeJuan Jose Gonzalez de la RosaJun DuJun LiuJun QinJun ZhangJunaid AhmedJunfeng XiaJungkyu RhoJunhao HuJunlin ChangJyh-Ching JuangKai JiangKang LiKanghee KimKarina ShakhgendyanKaushik RoyKazunori OnoguchiKazuyuki MatsumotoKe TangKok-Leong OngKrishna ChandramouliKrishnanand Kaipa NarasimhaKuei-Hsiang ChaoKunikazu KobayashiLaks RaghupathiLalit GuptaLance C FungLaurent HeutteLe DongLeh LuohLei LiuLei YangLei ZhangLi Ding Li NieLi QingfengLi ShangLiang ZhaoLiangxu LiuLihua JiangLi-Jen KaoLijuan XiaoLijun XuLin GaoLin LiLin WangLin ZhuLina Lu

Gouhei TanakaGuang JinGuang-Ming WuGuangwei ZhangGuilherme BarretoGuisheng ChenGuo WeidongGuohui ZhangGuowei ZhangGuoyu ZuoGurumurthy SwaminathanGwang-Hyun KimGwo-Ruey YuGyung-Jin HongH.K. Lamhai minHaibing GaoHaini QuHai-Tao ZhengHameed Ullah KhanHamid Abrishami MoghaddamHanif UllahHanlin HeHan-min ChienHasan JamilHassan TaheriHehua ZhangHerbert IuHoang-Yang LuHong FuHong-Bo LeiHongjun JiaHonorius GalmeanuHoushang DarabiHsiang-Yi LeeHualiang LiHuan XuHuang PingHuifang LiHuiran LiuHuiyu ZhouHyun-Deok KangHyun-Ju ParkHyun-Sik KimIbrahim ALISKANIbrahim Beklan KucukdemiralIkhyeon JangIndrajit BanerjeeIng-Chyuan WuIngo FeldmannInsoo KooIrene ArtemievaIrene ArtemievaIroshi AwasakiJackson SouzaJaehyung ParkJames CaiJames WaltonJames YehJanset DasdemirJavad HaddadniaJawid AziziJayasudha John SuseelaJeongHyun KimJiajun YanJian Xun PengJianbo FanJiande SunJiang jlJianhua CheJianli LiJianping QiaoJianting Cao

Chen AsiaChen HuiChen PengChen ShaoCheng SunChengjian WeiChenkun QiChenn-Jung HuangCheon Seong-PyoChi ZhouChia-Mei ChenChieh-yao ChangChien-Yuan LaiChih-Hung WangChi-Min LiChin-CHun ChangChin-Feng LinChing-Hung LeeChing-kun ChenChingti LiuChing-Ti LiuChin-yuan FanChiung-Hua HuangChonglun FangChuan ShiChuang MaChun ChenChungho ChoChungui XuChunhou ZhengChunlin ChenCuco CuistanaDajun DuDanfeng ZhuDao ZhouDario BruneoDavid BracewellDavid GeronimoDe XuDenis Orelderchian tsaihDingfei GEDmitry Serkindong yangDongsheng CheDunwei GongElena ShalfeevaElvira PopescuErkan ZergerogluErtan &OumlEung Nam KoEvgeni NurminskiEylem YucelFahad MuhayaFei GeFengfeng ZhouFilippo CastiglioneFrancesco CamastraFrancesco IorioFrancesco LongoFrancesco NapolitanoFrancesco TusaFrederik HogenboomFu YongguiFurong WangFu-Shiung HsiehGalip CanseverGang ChenGe LeiGerrit K. JanssensGinny WongGiuseppe AgrilloGiuseppe MAngioni

- 6 -

Ling WangLingFeng LiuLingling WangLing-po Li Linlin ShenLipo WangLiqing ZhangLitt Teen HiewLiu ZhipingLiya DingLouis WuLu XuLucia MorenoM LoogM Rafiq SwashMaciej HrebienManish SrivastavaMaqsood MahmudMarco CortellinoMargaret KnyasevaMargarita KnyazevaMario MarinelliMario RossiMark Halling-BrownMarkus KoskelaMarzio PennisiMasaru OkumuraMax PowerMaysam AbbodMeiling HouMengru TuMihnea ScafesMika SulkavaMilan LovricMin JiangMin WuMin ZhengMin-Chih ChenMine Tsunenoriming yuMingguang ShiMingxiao LiMingyu YouMinh-Tri PhamMi-ran YunMKM RahmanMohammad NarimaniMonalisa MazumdarMoussa HADDADMuhammad Khurram KhanMutsumi WatanabeNaeem RamzanNaiara AginakoNaoyuki Tsuruta Nataliya NikiforovaNelson MascarenhasNestor AranaNeyir OzcanNhan Nguyen-ThanhNi BuNidhi AroraNikolay MikhaylovNitthinun SuphasetthawitNuanwan SoonthornphisajOlesya KazakovaPallavi VajinepalliPandu DevarakotaPeilin JiaPeilin JiangPeixing LiPeng QiuPeng RenPeng ZhangPing WangPing ZhangPing-Min Hsu

Xinyu LiXiujun GongXiushan NieXuefen ZhuXuemei RenXue-qiang ZengYago SaezYan DongYan LiYan YangYang ChenYang ShiYang SongYangmin LiYanmin Liuyaou zhaoYaroslava KatuevaYasuhiro TaniguchiYasushi MaeYe BinYe XuYehu ShenYen Ming Chiu Yen-Wen WangYeonsik KangYF XuYifeng ZhangYihai ZhuYingke LeiYinglei SongYixiang LuYong WangYongcui WangYongsheng DongYoon-Seok NamYoshinori KobayashiYou OuyangYoungbae HwangYu WuYu XueYuan-Chang ChangYuanling HaoYu-Chen LinYu-Chiun ChiouYue WangYuhu ChengYun XuYu-Qing QiuYuxi LiuZanchao ZhangZhang LiZhang LiangshengZhao Yinggangzhaohui GanZhaohui SunZhenbing ZENGZhengyou WangZhi TengZhigang WangZhigang YanZhijun TanZhijun YangZhixia YangZhong JinZhongkun HeZhongming ZhaoZhongqiang WuZhongsheng WangZhu TengZikai WuZNERGIZZujun Hou

Tiantai GUOTian-Yu LiuTiefang HeTimo HonkelaTingxu YanTiong GohTolga EnsariTomasz AndrysiakTomasz RutkowskiTomoaki TsuruokaTomohiro HenmiToni ZgaljicToshiaki KondoToshihisa TanakaTsang-Long PaoTsung-Yi ChenU KaymakUttam RoyVanta DimitrovaVasily AristarkhovVenugopal ChakravarthyVictor JinVincenzo Daniele CunsoloVinod PathangayVitaliy SnytyukVito SantarcangeloVladimir BrusicWai-keung FungWang HailiWangheon LeeWan-Jui LeeWee Keong NgWei HuangWei JiaWei JingWei XiongWei YuWei-Chiang HongWei-Chih YangWeidong LiWeidong XuWeifeng LIWeitong HuangWeiWu WangWen ShengjunWen-Chung ChangWenjie LiWenkai LiWenyong DongWen-Yuan LiaoWenyun LiWing-Kuen LingWon-Kyu KimWoochang shinWoosung YangWorasait SuwannikWuchuan YangWudai LiaoXiangbin ZhuXiangrong Zhangxianjun shenXianwen RenXian-xia ZhangXiaochun CaoXiaoFeng WangXiaojian shaoXiaojing SongXiaojuan WangXiaolei XiaXiaomin LiuXiaoyan SunXiaoying WangXiao-Zhi GaoXin HaoXin ZouXinping Xie

Pradip GhantyPramod NCPramuditha SuraweeraPrashan PremaratneQi LiuQi WangQi YuQiao WANGQingwei GaoQuande QinQuang NguyenQuanke PanQun NiuRafal KozikRaffaele MontellaRandeep SinghRomina OlivaRong JinRong-xiang HuRosa de DuonniRoy PowerRuhul SarkerRui Jorge AlmeidaRyuzo OkadaS KimS. Jamal H ZaidiSaad BedrosSakashi MaedaSaleh AlySanggil KangSanto MottaSarif NaikSawomir LasotaSeeja K.R, H KSeokjin SungSeokjoo ShinSeong-Joe LimSeref Naci EnginShafayat AbrarShahid HussainShanwen ZhangShao jjShaohui LiuShaojing FanShaoli WangShaomin ZhangSheng ChenShengping ZhangSherif SherifShih-Ting YangShi-Jay Chen Shinya TakahashiShiuh-Jeng WangShripad KondraShuanghe ZhuShuhui BiSibel Senansoon-min HwangSriparna SahaSteve LingSteven GuanSulan ZhangSun Cheol BaeSungon LeeSungshin KimSupriya RaoSusana VieiraSuwon LeeSyed Ismail ShahTakashi KuremotoTariq ChatthaThanh Tho QuanThanhVu NguyenThomas O'DanielThomas TawiahThuc Kieu Xuan

- 7 -

Conference Sponsor

■Sponsored by

Institute of Control, Robotics and Systems, Korea

■Organized by

University of Ulsan, Korea

Chinese Academy of Sciences, China

■Technically co-sponsored by

Springer-Verlag

IEEE Computational Intelligence Society

International Neural Network Society

■Financially Contributed by

National Research Foundation of Korea

National Natural Science Foundation of China

Ulsan Metropolitan City

University of Ulsan - Brain Korea 21

Network-based Automation research Center (NARC)

e-vehicle Fusion Education Center

Korea Tourism Organization

- 8 -

General Information

Conference LocationICIC 2009 ville be held at Lotte Hotel Ulsan. TheHotel is located in Ulsan downtown. With state-of-the-art facilities and impeccablepersonalized services, Lotte Hotel has served someof the most prestigious guests in the world,establishing its reputation as one of the best hotels inthe world. The conference meeting rooms includeCrystal Ballroom on the 2nd floor and CharlotteRoom on the 3rd floor of the hotel.

Lotte Hotel Ulsan1480-1 Samsan-dong, Nam-gu, Ulsan, Korea(Zip code: 680-020)TEL 82-52-960-1000FAX 82-52-960-4334-5www.lottehotelulsan.com

Registration DeskThe registration desk will be open in the period ofthe following times;17:00 - 20:00 Wednesday, 16 September, 1F08:00 - 19:00 Thursday, 17 September, 2F08:00 - 16:30 Friday, 18 September, 2F

Information /Message BoardThe information/ message board will be located nearthe Registration Desk. Messages will be posted inthis area throughout the conference.

Name BadgeAll attendees must wear their name badges at alltimes to gain admission to all conference sessions,welcome reception, conference banquet andindustrial tour.

Conference ProceedingsAccepted 214 papers are included in two volumes ofproceedings published by Springer: one volume ofLecture Notes in Computer Science (LNCS) and onevolume of Lecture Notes in Artificial Intelligence(LNAI). Each book will be distributed to only authorcontributed to the Springer proceedings at the timeof conference. The other 22 papers will be publishedin two international journals. When the journals areissued out, the Chinese secretariat will post a copyof journal to author by airmail. Additional copy ofproceedings can be purchased at KRW 150,000 eachon-site.

Official LanguageThe official language of the conference is Englishand will be used for all presentations and printedmaterials.

Currency and Credit CardsThe unit of Korean currency is the Won. Foreigncurrency and traveler’s checks can be exchanged

into Korean Won at foreign exchange banks andother authorized money exchangers. Credit cards,including VISA, American Express, Diners Club,Master Card and JCB, are accepted at major hotels,department stores, and large restaurants. Theexchange rate is subject to fluctuation.

ElectricityIn Korea, outlets for 220 volts 60 Herz are dominant.Always check the power supply before using yourequipment.

Refreshment BreakCoffee or tea with some cookies will be servedaround the break twice or three times a day.

Registration KitAll registrants will be given a conference bag. Thebag includes final program, a copy of SpringerProceedings (LNCS or LNAI), name badge, receipt,participant list, tour guidebook, welcome reception,industrial / city tours, conference banquet andrefreshment breaks. The accompany person feeincludes name badge, a copy of SpringerProceedings (LNCS or LNAI), welcome reception,industrial / city tours, conference banquet andrefreshment breaks.

ClimateThe climate of Ulsan is relatively mild climateaffected by seasonal wind. Ulsan temperature isexpected to range 25 to 28℃(70~80℉) during theconference. It is sunny, windy and scattered showerin Ulsan. Relative humidity is 55~60%. The weatheris subject to change. Ulsan has the sunny weather inKorea, allowing tourists to enjoy their visit all yearlong.

RestaurantThere are a few Korean restaurants outside the LotteHotel Ulsan. Also, you can find the LotteDepartment Store right next to the hotel. There are alot of restaurants in this department store.

Conference SecretariatIf you have any inquiry on the ICIC2009, pleasecontact;

ICIC 2009 Korea Secretariat C-Agency Co. Meeting Management #805 Sungji Building 538 Dohwa-dong, Mapo-gu Seoul 121-040, KOREA Tel : 82-2-717-3280 Fax : 82-2-706-4879 Email : info@icic2009.org Website : http://icic2009.org

- 9 -

How to Get to Ulsan

··Ulsan Metropolitan City is a representative port city located in southeastern region of Korea, and the distance from Seoul to Ulsan is approximately 400 Km.

- 10 -

Hotel Information

··Lotte Hotel Ulsan - Conference VenueAddress: 1480-1 Samsan-dong, Nam-gu, Ulsan, Korea (Zip code: 680-020) Tel: +82-52-960-1000, Fax: +82-52-960-4334~5Location: 20 min by taxi from the Ulsan Airport. Taxi fare costs KRW 7,000. Website: www.lottehotelulsan.com

··Olympia HotelAddress: 1128-1 Shinjeoung 2- dong, Nam-gu, UlsanTel: +82-52-271-8401, Fax: +82-52-271-8410Location: 20-30 min by taxi from the Ulsan Airport. Taxi fare costs KRW 10,000.

10 min by taxi or bus from the Lotte Hotel 10 min by taxi or bus from Ulsan station or Ulsan Intercity Bus terminal

Website: http://www.olympiaulsan.co.kr/

··Cozy (Koji) MotelAddress: 1564-11 Samsan-dong, Nam-gu, UlsanTel: +82-52-268-0058, Fax: +82-52-245-0058Location: 20 min by taxi from the Ulsan Airport. Taxi fare costs KRW 7,000. It is near by Lotte Hotel Ulsan.

··Motel CherevilleAddress: 1564-16 Samsan-dong, Nam-gu, UlsanTel: +82-52-271-9069, Fax: +82-52-271-9046Location: 20 min by taxi from the Ulsan Airport. Taxi fare costs KRW 7,000.

It is near by Lotte Hotel Ulsan.

▼▼ Around the conference venue

- 11 -

Social Events

Welcome ReceptionThursday, September 17th at 6:30 pm - 8:30 pm

At Crystal Ballroom, 2nd floor of Lotte Hotel Ulsan

ICIC2009 Organizer invites all conference attendees to enjoy an eveningwith colleagues and friends. Various kinds of cuisines and drinks will beserved for all of the participants to welcome and relax.

Conference BanquetFriday, September 18th at 7 pm - 9 pm

At Crystal Ballroom, 2nd floor of Lotte Hotel Ulsan

Banquet will open with vedio presentation on Ulsan Metropolitan City whosupports financially this conference. There will be congratulatory addresses ofGeneral Chairs and Vice Mayor of Ulsan Metropolitan City. Program chair willreport you of technical program and paper statistics. Best paper award will bepresented by the Award Commitee. Korean food is served in full course at thebanquet. Korean traditional music instrument performance, SookmyungGayageum Orchestra will be performed after the dinner. Additional banquetcoupon is available at KRW 75,000.

Industrial Tour

Friday, September 18th at 3:00 pm - 6:30 pm

Ulsan is the home base for Hyundai companies and SK Corporation, which can be ranked in the fivebiggest companies in Korea. We are now ready to welcome and accept you as a companion based on a richindustrial foundation. The industrial tour will be scheduled at 3 pm, Friday, September 18, 2009. The tourprogram consists of Hyundai Motors and Daewang Rock. Those who wish to join the tour should sign upat the registration desk until 12 pm, Thursday, September 17. The tour is a complimentary program.

Ulsan City Tour

Saturday, September 19th at 8:30 am - 12:30 pm

Ulsan is one of the wealthy cities as a symbol of industrial and economic growth in Korea. Ulsan is blessedwith a mild climate and beautiful nature. The Taehwa River flows across Ulsan City adding beauty to thenature of Ulsan. If you wish to join the city tour, please sign up at the registration desk until 12 pm,Friday, September 18th. The city tour is free of charge.

BangudaeAt 8 kilometers to the northeast from Eonyang, a cliff of curious rocks soars onone range of Mt. Yunho. It is called Bangudae because it looks like a turtlelying flat on the ground. In front of it there is a projected rock, and it can beseen as a head of a turtle. But Sayun dam constructed at the interjection of thestream from the valley of stone carvings of Chonjonri and the tail of TaehwaRiver makes the whole area a lakeside. This dam is used for supplying water tothe Ulsan industrial complex. Bangudae has been famous for its beautifulscenery from the beginning, and Mookgaek, a poet, visited it often. The cliff of

Bangudae is over 30 meters high and has three layers. With the rock resembling a lying turtle, the sight isvery wonderful, and so people named it the second Mt. Kumgang.

Chun-jun-ri carved stonesThis valley is near Chun-jun-ri has carved stones, distinguished scenery, andabundant water that is clear and pure. Because there is national treasure #147,Ul-ju Chun-jun-ri carved stone, this is one of the two national treasures that Ul-san city owns, which many people visit constantly. Also there is Chi-san lecturehall and a historic site of Park Jae Sang, who is a meritorious retainer in Shillaage. The stone on which a faithful wife stood waiting for her husband until sheperished, Eun-ul-am, has each legend. This place is for family’s summervacation and education for children including a gigantic tree aged about 500

years, 223 meters in height, 12 meters in girth and #64 natural monument gingko tree at Du-seo-myun.

Jangsaengpo Whale MuseumThe only domestic Jangsaengpo Whale Museum is constructed in Jangsaengpowhere is the battlefield base of whaling in old days and it would like to collects,conserves and displays the whaling possessive remains that is appearing from1986 when prohibits the whaling and provides the various information relatedwhale and then it will also would like to use them as marine sightseeingresources with offering the spaces of marine ecology and educational studyingexperience.

- 12 -

- 13 -

Floor Map of the Hotel

■2nd floor of the Hotel

■3rd floor of the Hotel

ICIC2009 Program at a Glance

- 14 -

- 15 -

Keynote Speakers

Keynote Speech 1 09:15 am - 10:00 am, Thursday, September 17th

·Topic: Robot Town Project - A Platform for Service Robot in Daily Human Life Environment

·Speaker: Professor Tsutomu Hasegawa, Kyushu University, Japan·Summary: Ordinary environment of human daily life is a dynamically changing 3D real

space: there exist human beings walking and working around, and the layout of furnitureand obstacles may change so frequently that the map becomes obsolete quickly. Althoughit may be a simple and well-ordered environment for humans, it is too complex torecognize for a conventional self-contained autonomous robot equipped with as manysensors as possible within its limited body. Considering the state-of-the-art, we will not be

able to expect a capable robot executing various tasks independently in our daily life environment in near future.We propose an alternative approach to an intelligent robot working in our dairy life environment: the environmentstructured in informative way through the sensor network. Distributed sensors and RFID tags are embedded in theenvironment and are connected to a network to support the robot to recognize its surrounding situation. Thus theeffort required for robotic tasks execution is much reduced. Extending the idea to a larger area, we are implementing an information based structured environment platform,“Robot Town” Vision cameras are distributed and set up in a block of a town to observe and measure movingobjects including robots. RFID tags are attached to objects in the environment so that the robot easily knowsexisting objects in its surroundings. RFID tags are also distributed and attached to the fixed structure such as floor,wall, gate and even outdoor pedestrian area so that the robot approximately localizes itself. Sensory data management and interaction with robots are performed by a system called Town ManagementSystem (TMS). The system integrates the data from sensors into online database (DB) of the environment andprovides a robot with real time information of the dynamically changing situation of its surroundings. Three-dimensional map and the distribution of the RFID tags are also stored in the TMS.Several experiments of robotic service have been successfully performed in the platform. The Robot Town Projecthas been financially supported in the Coordination Program of Science and Technology Projects, the Council forScience and Technology Policy of the Japanese Government. ··Bio-Sketch: Tsutomu Hasegawa received the B.E. degree in 1973 in electronic engineering and the Ph.D.degree in 1987, both from the Tokyo Institute of Technology, Tokyo, Japan.He was associated with the Electrotechnical Laboratory of the Japanese Government from 1973 to 1992 where heperformed research on robotics. From 1981 to 1982, he was a Visiting Researcher at the Laboratoired’Automatique et d’Analyse des Systemes (LAAS/CNRS), Toulouse, France. He joined Kyushu University,Fukuoka, Japan, in 1992 and is currently a Professor with the Department of Intelligent Systems, Graduate Schoolof Information Science and Electrical Engineering, Kyushu University. His research interests are in manipulatorcontrol, geometric modeling and reasoning, motion planning, man-machine interaction, and ambient intelligence. Dr. Hasegawa is a recipient of Franklin V. Taylor Memorial Award from IEEE Systems, Man, and CyberneticsSociety in 1999. He is a member of the Institute of Electrical Engineers of Japan, the Society of Instrumental andControl Engineers in Japan, the Robotics Society of Japan, and Fellow of the Japanese Society of MechanicalEngineers.

Keynote Speech 2 1:15 pm - 2:00 pm, Thursday, September 17th

·Topic: Rough Classification: Algorithms and Applications·Speaker: Professor Ngoc Thanh Nguyen, Wroclaw University of Technology, Poland ·Summary: Rough classification methods, in general, serve to determining a set of attributeswhich generate an approximate classification referring to a given classification. In the Pawlak’s

concept of rough classification for a given classification C of set U of objects a roughclassification is the approximation of C. Assume that classification C is generated by set B ofattributes, then the approximation of C is based on determining a proper subset B’ of B suchthat the classification generated by B’differs “a little” from C. The small difference betweenthese classifications is illustrated by the difference of their accuracy measures which should

not be larger than some threshold. In our approach we consider other problem of rough classification: For a given

classification of set U which is generated by set A of attributes, one should determine such minimal set B ofattributes from A that the distance between the classification generated by attributes from B and the givenclassification is minimal. In this talk an approach for using rough classification methods to performrecommendation processes in intelligent e-learning systems is presented. Rough classification in this case is relatedto inconsistency aspect of knowledge of the system. The inconsistency here appears in two aspects: In the firstaspect inconsistency refers to difference of the passed scenarios of similar learners (belonging to the same class ofthe classification). In this case to determine an opening scenario for a new learner it is needed to calculate theconsensus of the passed scenarios of the members of the class. The second aspect of inconsistency refers to the factthat assumed to be similar learners (belonging to the same class of the classification) may have very miscellaneouspassed scenarios. This, in turn, may cause a lack of efficiency of the procedure proposed for the first aspect. Herewe propose to use a rough classification based method to redefine the criterion for classification. Apart from theapplication in e-learning systems, another application of the rough classification methods will be also presentedreferring to designing adaptive user interfaces.··Bio-Sketch: Professor Ngoc Thanh Nguyen (Ph.D., D.Sc.) works at Wroclaw University of Technology, Poland,where he is the head of Knowledge Management Systems Department in the Faculty of Computer Science andManagement. His scientific interests consist of knowledge integration methods, collective intelligence, intelligenttechnologies for conflict resolution, inconsistent knowledge processing, multi-agent systems, and E-learningmethods. He has edited 20 special issues in international journals, 6 books and 6 conference proceedings. He is theauthor of 4 monographs and about 180 other publications. His latest monograph entitled “Advanced Methods forInconsistent Knowledge Management” has been published by Springer last year. Prof. Nguyen serves as Editor-in-Chief of International Journal of Intelligent Information and Database Systems; Editor-in-Chief of two book series:Advances in Applied Intelligence Technologies and Computational Intelligence and its Applications for IGI GlobalPublishers (USA); Associate Editor of Neurocomputing, International Journal of Innovative Computing &Information Control, Journal of Information Knowledge System Management and KES Journal; and a Member ofEditorial Review Boards of several other prestigious international journals. He serves also as an expert for Ministryof Science and Higher Education and Ministry of Regional Development of Poland in evaluating R&D projects. Heis the Chair of KES Symposium Series on Agent and Multi-agent Systems. He has been General Chair or ProgramChair of more than 10 international conferences. Prof. Nguyen has been selected as the Vice-President ofInternational Society of Applied Intelligence (ISAI); Senior Member of IEEE and ACM, the largest computerscience societies in the world. He is also an expert of European Commission in evaluation research projects and anexpert of Polish Ministry of Science and Higher Education and Slovakia Research Agency. He is also the AssociateChair of KES International and many other functions in international societies like IFIP, WIC etc. In 2008 for hisactivities the President of Poland has rewarded Prof. Nguyen the Bronze Medal for Education. He has obtained alsoawards from the Polish Ministry of Science and Higher Education and the Rector of Wroclaw University ofTechnology.

Keynote Speech 3 09:00 am - 09:45 am, Friday, September 18th

·Topic: Robust Computer Vision Techniques and Applications·Speaker: Professor In-So Kweon, KAIST, Korea·Summary: Research in KAIST Robotics and Computer Vision (RCV) Lab. has been

focused on developing robust methods concerning important computer vision problems: 3Dstructure recovery, image processing and object recognition. In this talk, we first presentrobust methods for finding feature correspondences from an image pair with significantdeformation. We then introduce a new theory to model the sensor noise of CCD cameras for low-levelimage processing, such as edge and corner detection. The robustness against illumination

variations will be demonstrated by extensive experiments. Finally, we will present a graphical model based objectrecognition framework for recognizing objects under strong cluttered backgrounds. The framework is designed toresemble the characteristics of the human vision system. Experimental results using the standard DBs and realimages show the feasibility of the proposed method for real-world applications, such as intelligent service robots.··Bio-Sketch: In-So Kweon is Professor of School of Electrical Engineering and Computer Science, KAIST. InJun. 2009 at IEEE-CVPR’2009, he achieved Best Student Paper-Runner Up (with O. Duchenne), in Sep. 2008 atInternational Conference on Control, Automation and Systems (ICCAS 2008), he achieved Student Paper Award(with Jungho Kim). In Oct. 2008 at International Conference on Ubiquitous Robots and Ambient Intelligence 2008

- 16 -

- 17 -

(URAI 2008), he achieved Outstanding Poster Award (with K. Sung). In 2006 at Annual Summer Workshop onRobotics, Korea Robotics Society, he achieved Best Paper Award (with S. Kim). In 2001 at SPIE PhotonicsConference, he achieved Poster Paper Award, Boston, USA (with K. Yoon). In 2001 he achieved KAIST ResearchAward.

Keynote Speech 4 1:15 pm - 2:00 pm, Friday, September 18th

·Topic: Intelligent Pattern Recognition and Applications to Biometrics in InteractiveLearning Environment

·Speaker: Professor Patrick Wang, Northeastern University, USA·Summary: This talk deals with some fundamental aspects of biometrics and its applications.

It basically includes the following: Overview of Biometric Technology and Applications,Importance of Security: A Scenario of Terrorists Attack, What are Biometric Technologies?Biometrics: Analysis vs Synthesis, Analysis: Interactive Pattern Recognition Concept,Importance of Measurement and Ambiguity, How it works: Fingerprint Extraction andMatching, Iris, and Facial Analysis, Authentication Applications, Thermal Imaging:

Emotion Recognition. Synthesis in biometrics, Modeling and Simulation, and more Examples and Applications ofBiomedical Imaging in Interactive Fuzzy Learning Environment. Finally, some future research directions arediscussed.··Bio-Sketch: Prof. Patrick S.P. Wang, PhD. IAPR Fellow and IEEE Outstanding Achievement Awardee, and isTenured Full Professor, Northeastern University, USA, iCORE (Informatics Circle of Research Excellence)Visiting Professor, University of Calgary, Canada, Otto-Von-Guericke Distinguished Guest Professor, MagdeburgUniversity, Germany, Zijiang Visiting Chair, ECNU, Shanghai, China, as well as honorary advisory professor ofseveral key universities in China, including Sichuan University, Xiamen University, East China Normal University,Shanghai, and Guangxi Normal University, Guilin. Prof. Wang received his BSEE from National Chiao Tung University (Jiaotong University), MSEE from NationalTaiwan University, MSICS from Georgia Institute of Technology, and PhD, Computer Science from Oregon StateUniversity. Dr. Wang has published over 23 books, 130 technical papers, 3 USA/European Patents, inPR/AI/TV/Cybernetics/Imaging, and is currently founding Editor-in-Chief of IJPRAI (International Journal ofPattern Recognition and Artificial Intelligence), and Book Series of MPAI, WSP. In addition to his technicalinterests, Dr. Wang also published a prose book, “Harvard Meditation Melody” and many articles and poemsregarding Du Fu and Li Bai’s poems, Beethoven, Brahms, Mozart and Tchaikovsky’s symphonies, and Bizet,Verdi, Puccini and Rossini’s operas.

- 18 -

Technical Program Information

The technical program consists of 4 keynote sessions and 5 parallel oral sessions of accepted papers.

Keynote SpeechThe keynote speech will be given for 45 minuteseach. It will take place at 9 am and 1:15 pm atCrystal Ballroom during the conference.

Technical SessionsAll the technical sessions consist of oralpresentations. They will be held at CrystalBallroom on the 2nd floor and Charlotte Room onthe 3rd floor of hotel on 17 - 18 September.

Guide to Understanding Session NumberingEach session in the technical program is assigneda unique number which clearly indicates whenand where the paper is presented. A typicalnumber is shown below;

Typical Session Number: T1A

The first letter (i.e. T) indicates the day of theconference.T = Thursday F = Friday

The second number (i.e. 1) shows the time of theday1 = Morning2 = Afternoon3 = Late Afternoon

The third letter (i.e. A) indicates the location ofthe presentation which is one of the five sessionrooms. A = Crystal Ballroom #1B = Crystal Ballroom #2C = Crystal Ballroom #3D = Charlotte Room #1E = Charlotte Room #2

- 19 -

TTeecchhnniiccaall PPrrooggrraamm DDeettaaiillss

09:00-09:15 Opening Ceremony RM: Crystal Ballroom #2 09:15-10:00 Keynote Speech 1 RM: Crystal Ballroom #2 Robot Town Project - A Platform for Service Robot in Daily Human Life Environment Tsutomu Hasegawa, Kyushu University, Japan Chair: De-Shuang Huang, Chinese Academy of Sciences, China 10:00-10:15 Break 10:15-12:15 <T1A> Neural Networks / Swarm Intelligence and Optimization RM : Crystal Ballroom #1 Chairs: Kazuo Ohmi, Osaka Sangyo University, Japan Enrique Dominguez, University of Malaga, Spain 10:15 - 10:35 A SOM Based Stereo Pair Matching Algorithm for 3-D Particle Tracking Velocimetry

Kazuo Ohmi, Basanta Joshi, Sanjeeb Prasad Panday Osaka Sangyo University, Japan 10:35 - 10:55 Spiking Neural Network Performs Discrete Cosine Transform for Visual Images

QingXiang Wu, T. M. McGinnity, Liam Maguire, Arfan Ghani, Joan Condell University of Ulster at Magee Campus, UK 10:55 - 11:15 Spam Detection based on a Hierarchical Self-Organizing Map

E.J. Palomo, E. Domnguez, R.M. Luque, and J. Muñoz University of Malaga, Spain 11:15 - 11:35 An Ensemble of Neural Networks for Stock Trading Decision Making

Pei-Chann Chang1, Chen-Hao Liu3, Chin-Yuan Fan2, Jun-Lin Lin1, Chih-Ming Lai1 1Yuan Ze University, Taiwan, 2Ming Dao University, Taiwan, 3Kainan University, Taiwan 11:35 - 11:55 An Improved PSO Algorithm Encoding a Priori Information for Nonlinear Approximation

Tong-Yue Gu, Shi-Guang Ju, Fei Han Jiangsu University, China 11:55 - 12:15 Multi-objective Oriented Search Algorithm for Multi Objective Reactive Power Optimization

Xuexia Zhang, Weirong Chen Southwest Jiaotong University, China 10:15-12:15 <T1B> Evolutionary Learning & Genetic Algorithms RM : Crystal Ballroom #2 Chair: Juan Figueroa, Universidad Distrital Francisco Jose de Caldas, Colombia

T1A.6

T1A.5

T1A.4

T1A.3

T1A.2

T1A.1

Thursday, September 17

- 20 -

10:15 - 10:35 Interactive Genetic Algorithms with Individual's Fuzzy Fitness

Dun-wei Gong, Jie Yuan and Xiao-yan Sun China University of Mining and Technology, China 10:35 - 10:55 Interactive Genetic Algorithms with Variational Population Size

Jie Ren, Dun-wei Gong, Xiao-yan Sun, Jie Yuan and Ming Li China University of Mining and Technology, China 10:55 - 11:15 Two Step Template Matching Method with Correlation Coefficient and Genetic Algorithm

Gyeongdong Baek and Sungshin Kim Pusan National University, Korea 11:15 - 11:35 Missing Data Imputation in Multivariate Data by Evolutionary Algorithms

Juan C. Figueroa Garcìa1, Dusko Kalenatic2 and Cesar Amilcar Lopez Bello1 1Universidad Distrital Francisco José de Caldas, Bogotá – Colombia, 2Universidad de la Sabana, Chia – Colombia 11:35 - 11:55 Some Distributed Algorithms for Quantized Consensus Problem

Jianping He, Wenhai Chen, Lixin Gao Wenzhou University, China 11:55 - 12:15 On the Robustness of Type-1 and Type-2 Fuzzy Tests vs. ANOVA Tests on Means

Juan C. Figueroa Garcìa1, Dusko Kalenatic2 and Cesar Amilcar Lopez Bello1 1Universidad Distrital Francisco José de Caldas, Bogotá – Colombia, 2Universidad de la Sabana, Chia - Colombia 10:15-12:15 <T1C> Fuzzy Systems and Soft Computing / Kernel Methods and Supporting Vector Machines RM : Crystal Ballroom #3 Chairs: Mian Muhammad Awais, Lahore University of Management Sciences, Pakistan Keun-Chang Kwak, Chosun University, Korea 10:15 - 10:35 An FIS for Early Detection of Defect Prone Modules

Zeeshan Ali Rana, Mian Muhammad Awais, Shafay Shamail Lahore University of Management Sciences (LUMS), Pakistan 10:35 - 10:55 Variable Precision Concepts and Its Applications for Query Expansion

Fei Hao1, Shengtong Zhong2 1Korea Advanced Institute of Science and Technology, Korea, 2Norwegian University of Science and Technology, Norway 10:55 - 11:15 Combining Global model and Local Adaptive Neuro-Fuzzy Network

Yun-Hee Han, Keun-Chang Kwak Chosun University, Korea 11:15 - 11:35 Application of a Case Base Reasoning Based Support Vector Machine for Financial Time Series

Data Forecasting Pei-Chann Chang1, Chi-Yang Tsai1, Chiung-Hua Huang1,2, Chin-Yuan Fan3 1Yuan Ze University, Taiwan, 2Ta Hwa Institute of Technology, Taiwan, 3Ming Dao University, Taiwan 11:35 - 11:55 Cost-Sensitive Supported Vector Learning to Rank Imbalanced Data Set

Xiao Chang1,2, Qinghua Zheng1,2 and Peng Lin1,2

1Xi'an Jiaotong University, China, 2Shaanxi Key Lab. of Satellite and Computer Network, China 11:55 - 12:15 An Ensemble Classifier Based on Kernel Method for Multi-situation DNA Microarray Data

Xuesong Wang, Yangyang Gu, Yuhu Cheng and Ruhai Lei China University of Mining and Technology, China

T1C.6

T1C.5

T1C.4

T1C.3

T1C.2

T1C.1

T1B.6

T1B.5

T1B.4

T1B.3

T1B.2

T1B.1

- 21 -

10:15-12:15 <T1D> Combinatorial & Numerical Optimization / Systems & Computational Biology / Machine Learning Theory and Methods RM : Charlotte Room #1 Chairs: Naoyuki Tsuruta, Fukuoka University, Japan Alessandro Cincotti, Japan Advanced Institute of Science and Technology, Japan 10:15 - 10:35 An Effective Hybrid Algorithm Based on Simplex Search and Differential Evolution for Global

Optimization Ye Xu, Ling Wang, Lingpo Li Tsinghua University, China 10:35 - 10:55 Differential Evolution with Level Comparison for Constrained Optimization

Ling-po Li, Ling Wang, Ye Xu Tsinghua University, China 10:55 - 11:15 Agent based modeling of atherosclerosis: a concrete help in personalized treatments

Francesco Pappalardo1,2, Alessandro Cincotti3, Alfredo Motta4, and Marzio Pennisi2 1National Research Council(CNR), Italy, 2University of Catania, Catania, Italy, 3Japan Advanced Institute of Science and Technology, Japan, 4Politecnico di Milano, Italy 11:15 - 11:35 Construction of the ensemble of logical models in cluster analysis

Vladimir Berikov Sobolev Institute of mathematics, Russia 11:35 - 11:55 Ordinal Regression with Sparse Bayesian

Xiao Chang1,2, Qinghua Zheng1,2 and Peng Lin1,2

1Xi'an Jiaotong University, China, 2Shaanxi Key Lab. of Satellite and Computer Network, China 11:55 - 12:15 A Support System for Making Archive of Bi-Directional Remote Lecture –

Photometric Calibration Naoyuki Tsuruta, Mari Matsumura and Sakashi Maeda Fukuoka University, Japan 10:15-12:15 <T1E> Supervised Learning / Neural & Nature Inspired Computing and Optimization / Communication and Networks RM : Charlotte Room #2 Chairs: Taeho Jo, Inha University, Korea Zhang liming, Sun Yat-sen University, China 10:15 - 10:35 Profile based Algorithm to Topic Spotting in Reuter21578

Taeho Jo Inha University, Korea 10:35 - 10:55 A New Method of Morphological Associative Memories

Naiqin Feng, Xizheng Cao, Sujuan Li, Lianhui Ao and Shuangxi Wang Henan Normal University, China 10:55 - 11:15 Emergency resources scheduling on continuous consumption system based on adaptively mutate

genetic algorithm Zhang Liming, Lin Yuhua, Yang Guofeng, Chang Huiyou Sun Yat-sen University, China 11:15 - 11:35 Cloud@Home: Bridging the Gap between Volunteer and Cloud Computing

Vincenzo D. Cunsolo, Salvatore Distefano, Antonio Puliafito and Marco Scarpa University of Messina, Italy 11:35 - 11:55 An Intelligent Prediction Model for Generating LGD Trigger of IEEE 802.21 MIH

M. Yousaf, Sohail Bhatti, Maaz Rehan, A. Qayyum and S. A. Malik Center of Research in Networks and Telecommunications (CoReNeT), Pakistan T1E.5

T1E.4

T1E.3

T1E.2

T1E.1

T1D.6

T1D.5

T1D.4

T1D.3

T1D.2

T1D.1

- 22 -

11:55 - 12:15 GLRT Based Fault Detection in Sensor Drift Monitoring System

In-Yong Seo1, Ho-Cheol Shin1, Moon-Ghu Park1 and Seong-Jun Kim2 1Korea Electric Power Research Institute, Korea, 2Kangnung National University, Korea 12:15-13:15 Lunch 13:15-14:00 Keynote Speech 2 RM: Crystal Ballroom #2 Rough Classification: Algorithms and Applications Ngoc Thanh Nguyen, Wroclaw University of Technology, Poland Chair: Kanghyun Jo, University of Ulsan, Korea 14:00-14:15 Break 14:15-16:15 <T2A> Knowledge-Based Systems and Intelligent Computing In Medical Imaging RM : Crystal Ballroom #1 Chair: Laurent HEUTTE, Universite de Rouen, France 14:15 - 14:35 Characterization of endomicroscopic images of the distal lung for computer-aided diagnosis

Aurélien Saint-Réquier1, Benoît Lelandais1, Caroline Petitjean1, Chesner Désir1, Laurent Heutte1, Mathieu Salaün2 and Luc Thiberville2 1Université de Rouen, France, 2CHU de Rouen, France 14:35 - 14:55 Image processing framework for virtual colonoscopy

Vitoantonio Bevilacqua1,2, Marco Cortellino1,2, Michele Piccinni1, Antonio Scarpa1, Diego Taurino1, Giuseppe Mastronardi1,2, Marco Moschetta3, Giuseppe Angelelli3 1Politecnico di Bari, Italy, 2Spin-Off of Polytechnic of Bari, Italy, 3Policlinico Universitario di Bari, Italy 14:55 - 15:15 Combined use of densitometry and morphological analysis to detect flat polyps

Vitoantonio Bevilacqua1,2, Marco Cortellino1,2, Giuseppe Mastronardi1,2, Antonio Scarpa1, Diego Taurino1 1Politecnico di Bari, Italy, 2Spin-Off of Polytechnic of Bari, Italy 15:15 - 15:35 Relevant Measurements for polyps in 3D Virtual Colonoscopy

Vitoantonio Bevilacqua1,2, Marianna Notarnicola1, Marco Cortellino1,2, Antonio Scarpa1, Diego Taurino1, Giuseppe Mastronardi1,2 1Politecnico di Bari, Italy, 2Spin-Off of Polytechnic of Bari, Italy 15:35 - 15:55 Retinal Vessel Extraction by a Combined Neural Network–Wavelet Enhancement Method

Leonarda Carnimeo1, Vitoantonio Bevilacqua1,2, Lucia Cariello1,2, Giuseppe Mastronardi1,2 1Polytechnic of Bari, Italy, 2Spin-Off of Polytechnic of Bari, Italy 14:15-16:15 <T2B> Dynamic Spectrum Sharing Systems RM : Crystal Ballroom #2 Chair: Insoo Koo, University of Ulsan, Korea 14:15 - 14:35 Cooperative Spectrum Sensing Using Enhanced Dempster-Shafer Theory of Evidence in Cognitive Radio

Nhan Nguyen Thanh and Insoo Koo University of Ulsan, Korea 14:35 - 14:55 A secure distributed spectrum sensing scheme in cognitive radio

Nhan Nguyen-Thanh and Koo Insoo University of Ulsan, Korea T2B.2

T2B.1

T2A.5

T2A.4

T2A.3

T2A.2

T2A.1

T1E.6

- 23 -

14:55 - 15:15 An Optimal Data Fusion Rule in Cluster-Based Cooperative Spectrum Sensing

Hiep-Vu Van, Hyungseo Kang and Insoo Koo University of Ulsan, Korea 15:15 - 15:35 Exact Bit Error Probability of Multi-hop Decode-and-Forward Relaying with Selection Combining

Bao Quoc Vo-Nguyen and Hyung Yun Kong University of Ulsan, Korea 15:35 - 15:55 A Cooperative Transmission Scheme for Cluster based Wireless Sensor Networks

Asaduzzaman and Hyung Yun Kong University of Ulsan, Korea 15:55 - 16:15 A packet scheduling algorithm for IEEE 802.22 WRAN systems and calculation reduction

method thereof Young-du Lee, Tae-joon Yun and Insoo Koo University of Ulsan, Korea 14:15-16:15 <T2C> Knowledge Discovery and Data Mining / Intelligent Computing Algorithms in Banking and Finance RM : Crystal Ballroom #3 Chairs: Ou Liu, The Hong Kong Polytechnic University, Hongkong Tetsuya Takaishi, Hiroshima University of Economics, Japan 14:15 - 14:35 A Semantic Lexicon-based Approach for Sense Disambiguation and Its WWW Application

Vincenzo Di Lecce1, Marco Calabrese1, Domenico Soldo2 1Polytechnic of Bari, Italy, 2myHermes S.r.l., Italy 14:35 - 14:55 The establishment of verb logic and its application in universal emergency response

information system design Jian Tan, Xiang Tao Fan Laboratory of Digital Earth Sciences, China 14:55 - 15:15 Using Intelligent System for Reservoir Properties Estimation

Fariba Salehi1 and Arnoosh Salehi2 1Islamic Azad University, Iran, 2National Iranian Oil Company, Iran 15:15 - 15:35 On an Ant Colony-based Approach for Business Fraud Detection1

Ou Liu1, Jian Ma2, Pak-Lok Poon1 and Jun Zhang3 1The Hong Kong Polytechnic University, China, 2City University of Hong Kong, China, 3Sun Yat-sen University, China 15:35 - 15:55 Markov Chain Monte Carlo on Asymmetric GARCH Model Using the Adaptive Construction

Scheme Tetsuya Takaishi Hiroshima University of Economics, Japan 15:55 - 16:15 An Intelligent Computing Algorithm to Analyze Bank Stock Returns

Vincenzo Pacelli University of Foggia, Italy 14:15-16:15 <T2D> Advances in Intelligent Information Processing RM : Charlotte Room #1 Chair: Fengwen Cao, Suzhou Vocational University, China 14:15 - 14:35 Design of a Single-Phase Grid-Connected Photovoltaic Systems based on Fuzzy-PID Controller

Fengwen Cao, Yiwang Wang Suzhou Vocational University, China

T2D.1

T2C.6

T2C.5

T2C.4

T2C.3

T2C.2

T2C.1

T2B.6

T2B.5

T2B.4

T2B.3

- 24 -

14:35 - 14:55 A Constrained Approximation Algorithm by Encoding Second-order Derivative Information

into Feedforward Neural Networks Qing-Hua Ling, Fei Han Jiangsu University of Science and Technology, China 14:55 - 15:15 Image Reconstruction Using NMF with Sparse Constraints Based on Kurtosis Measurement

Criterion Li Shang1,2, Jinfeng Zhang2, Wenjun Huai2, Jie Chen2 and Jixiang Du3,4,5 1JiangSu Province Support Software Engineering R&D Center for Modern Information Technology Application in Enterprise, China, 2Suzhou Vocational University, China, 3Huaqiao University, China, 4University of Science and Technology of China, China, 5Chinese Academy of Sciences, China 15:15 - 15:35 A Cyanobacteria Romate Monitoring System

Zhiqiang Zhao1,2,3, Yiming Wang3 1JiangSu Province Support Software Engineering R&D Center for Modern Information Technology Application in Enterprise, China 2Suzhou Vocational University, China, 3Suzhou University, China 15:35 - 15:55 Image Segmentation of Level Set Based on Maximization of Between-Class Variance and

Distance Constraint Function Chang-xiong Zhou1,2, Zhi-feng Hu1,2, Shu-fen Liu1,2, Ming Cui1,2 and Rong-qing Xu3 1JiangSu Province Support Software Engineering R&D Center for Modern Information Technology Application in Enterprise, China, 2Suzhou Vocational University, China, 3Nanjing University of Posts and Telecommunications, China 15:55 - 16:15 Active MMW Focal Plane Imaging System

Pingang Su1,2, Zongxin Wang2, Zhengyu Xu2 1Suzhou Vocational University, China, 2Southeast University, China 14:15-16:15 <T2E> Network-based Intelligent Technologies RM : Charlotte Room #2 Chairs: Youngsoo Suh, University of Ulsan, Korea HONG-HEE LEE, University of Ulsan, Korea 14:15 - 14:35 Handling Multi-Channel Hidden Terminals Using a Single Interface in Cognitive Radio Networks

Liang Shan, Myung Kyun Kim University of Ulsan, Korea 14:35 - 14:55 Network Construction using IEC 61400-25 Protocol in Wind Power Plants

Tae O Kim, Jung Woo Kim and Hong Hee Lee University of Ulsan, Korea 14:55 - 15:15 Stability and Stabilization of Nonuniform Sampling Systems using a Matrix Bound of a Matrix

Exponential Young Soo Suh University of Ulsan, Korea 15:15 - 15:35 Robot Visual Servo through Trajectory Estimation of a Moving Object using Kalman Filter

Min-Soo Kim, Ji-Hoon Koh, Ho Quoc Phuong Nguyen, Hee-Jun Kang University of Ulsan, Korea 15:35 - 15:55 Implementation of Induction Motor Control System Using Matrix Converter based on CAN

Network and Dual-Port RAM Hong-Hee Lee, Hoang M. Nguyen University of Ulsan, Korea 15:55 - 16:15 Device Integration Approach to OPC UA-based Process Automation Systems with FDT/DTM

and EDDL Vu Van Tan, Dae-Seung Yoo and Myeong-Jae Yi University of Ulsan, Korea

T2E.6

T2E.5

T2E.4

T2E.3

T2E.2

T2E.1

T2D.6

T2D.5

T2D.4

T2D.3

T2D.2

- 25 -

16:15-16:30 Break 16:30-18:30 <T3A> Intelligent Computing in Bioinformatics RM : Crystal Ballroom #1 Chairs: Kyungsook Han, Inha University, Korea Sebastian Handrich, OvG-University Magdeburg, Germany 16:30 - 16:50 Dynamic Identification and Visualization of Gene Regulatory Networks from Time-Series Gene

Expression Profiles Yu Chen and Kyungsook Han Inha University, Korea 16:50 - 17:10 Predicting RNA-Binding Sites in Proteins Using the Interaction Propensity of Amino Acid Triplets

Mi-Ran Yun, Yanga Byun and Kyungsook Han Inha University, Korea 17:10 - 17:30 On-line Signature Verification Based on Spatio-Temporal Correlation

Hao-Ran Deng and Yun-Hong Wang Beihang University, China 17:30 - 17:50 Contribution Degree’s Application in the Research of Elements of TCM Syndromes

Rongyao Zheng, Guangcheng Xi and Jing Chen Chinese Academy of Sciences, China 17:50 - 18:10 Gender Recognition from Gait using Radon Transform and Relevant Component Analysis

Lei Chen1, Yunhong Wang1, Yiding Wang2, De Zhang1 1Beihang University, China, 2North China University of Technology, China 18:10 - 18:30 A Biologically Plausible Winner-Takes-All Architecture

Sebastian Handrich, Andreas Herzog, Andreas Wolf, and Christoph S. Herrmann Otto-von-Guericke-University Magdeburg, Germany 16:30-18:30 <T3C> Applications of Intelligent Computing in Information Assurance & Security / Intelligent Agent and Web Applications RM : Crystal Ballroom #3 Chair: Heejun Kang, University of Ulsan, Korea 16:30 - 16:50 Modified AES using Chaotic Key Generator for Satellite Imagery Encryption

Fahad Bin Muhaya, Muhammad Usama, Muhammad Khurram Khan King Saud University, Kingdom of Saudi Arabia

16:50 - 17:10 Experimental Comparison among 3D Innovative Face Recognition Frameworks Vitoantonio Bevilacqua1,2, Giuseppe Mastronardi1,2, Raffaele Piarulli1, Vito Santarcangelo1,2, Rocco Scaramuzzi1,

Pasquale Zaccaglino1 1Polytechnic of Bari, Italy, 2Spin-Off of Polytechnic of Bari, Italy

17:10 - 17:30 Buyer Coalitions with Bundles of Items by Using Genetic Algorithm Laor Boongasame and Anon Sukstrienwong

Bangkok University, Thailand

17:30 - 17:50 A Representation Methodology for Performance Specifications in UML Domain S. Distefano, A. Puliafito and M. Scarpa

Universitµa di Messina, Italy

17:50 - 18:10 Web-based Unified-Directory Service for Social Networking Services and Ubiquitous Sensor Network Services

Yung Bok Kim Sejong University, Korea

T3C.5

T3C.4

T3C.3

T3C.2

T3C.1

T3A.6

T3A.5

T3A.4

T3A.3

T3A.2

T3A.1

- 26 -

18:10 - 18:30 An Approach to Automated User Interest Matching in Online Classified Advertising Systems

Valeriya Gribova, Pavel Kachanov Institute of Automation and Control Processes, Russia 16:30-18:30 <T3D> New Particle Swarm Optimization and its Applications / Advances in Intelligent Information Processing RM : Charlotte Room #1 Chairs: Ben Niu, ShenZhen University, China Li Li ,ShenZhen University, China 16:30 - 16:50 Symbiotic Multi-swarm PSO for Portfolio Optimization

Ben Niu1, Bing Xue1, Li Li1 and Yujuan Chai2 1Shenzhen University, China, 2McMaster University, Canada 16:50 - 17:10 A Novel Particle Swarm Optimization with Non-linear Inertia Weight Based on Tangent

Function Li Li1, Bing Xue1, Ben Niu1, Lijing Tan2 and Jixian Wang3 1Shenzhen University, China, 2Measurement Specialties Inc, China, 3Anhui Agricultural University, China 17:10 - 17:30 An Improved Two-Stage Camera Calibration Method Based on Particle Swarm Optimization

Hongwei Gao1, Ben Niu2, Yang Yu1, Liang Chen1 1Shenyang Ligong University, China, 2Shenzhen University, China 17:30 - 17:50 Study on Multi-Depots Vehicle Scheduling Problem and Its Two-Phase Particle Swarm

Optimization Suxin Wang1,1, Leizhen Wang1, Huilin Yuan1, Meng Ge1, Ben Niu2, Weihong Pang1, Yuchuan Liu1 1Northeastern University at Qinhuangdao, China, 2Shenzhen University, China 17:50 - 18:10 Application of RBF Network Based on Immune Algorithm in Human Speaker Recognition

Yan Zhou1,2 and Xinming Yu1,2 1Jiangsu Research & Development Centre for Modern Enterprise Information Software Engineering, China, 2Suzhou Vocational University, China 18:10 - 18:30 Adaptive Immune Response Network Model

Tao Liu1,2, Li Zhang2 and Binbin Shi2 1JiangSu Province Support Software Engineering R&D Center for Modern Information Technology Application in Enterprise, China, 2Suzhou Vocational University, China 16:30-18:30 <T3E> Network-based Intelligent Technologies / Biology and Drug Design / Signal Processing for Interactive Brain-Machine-Interfacing RM : Charlotte Room #2 Chairs: Tomasz Rutkowski, RIKEN Brain Science Institute, Japan Julio Facelli, University of Utah, United States 16:30 - 16:50 A SOA-based Framework for Building Monitoring and Control Software Systems

Vu Van Tan, Dae-Seung Yoo and Myeong-Jae Yi University of Ulsan, Korea 16:50 - 17:10 Parallel genetic algorithms for crystal structure prediction: Successes and failures in predicting

bicalutamide polymorphs Marta B. Ferraro 1, Anita M. Orendt2 and Julio C. Facelli2 1Universidad de Buenos Aires, Argentina, 2University of Utah, USA 17:10 - 17:30 EMD Based Power Spectral Pattern Analysis for Quasi-Brain-Death EEG

Qiwei Shi1, Juhong Yang1, Jianting Cao1,3,4, Toshihisa Tanaka2,3, Tomasz M. Rutkowski3, Rubin Wang4 and Huili Zhu5 1Saitama Institute of Technology, Japan, 2Tokyo University of Agriculture and Technology, Japan, 3Brain Science Institute, Japan 4East China University of Science and Technology, China, 5Huadong Hospital Affiliated to Fudan University, China

T3E.3

T3E.2

T3E.1

T3D.6

T3D.5

T3D.4

T3D.3

T3D.2

T3D.1

T3C.6

- 27 -

17:30 - 17:50 Proposal of Ride Comfort Evaluaiton Method Using the EEG

Hironobu Fukai1, Yohei Tomita1, Yasue Mitsukura1, Hirokazu Watai2, Katsumi Tashiro2 and Kazutomo Murakami2 1Tokyo University of Agriculture and Technology, Japan, 2Bridgestone Corporation, Japan 17:50 - 18:10 Interactive Components Extraction from fEEG and fNIRS for Affective Brain Machine

Interfacing Paradigms Tomasz M. Rutkowski1, Toshihisa Tanaka1,2, Andrzej Cichocki1,Donna Erickson3 and Danilo P. Mandic4 1RIKEN Brain Science Institute, Japan, 2Tokyo University of Agriculture and Technology, Japan, 3Showa Music University, Japan, 4Imperial College London, UK 18:30-20:30 Welcome Reception RM: Crystal Ballroom #2

09:00-09:45 Keynote Speech 3 RM: Crystal Ballroom #2 Robust Computer Vision Techniques and Applications In-So Kweon, KAIST, Korea Chair: Laurent Heutte, The University of Rouen, France 09:45-10:00 Break 10:00-12:00 <F1A> Intelligent Computing in Computer Vision RM : Crystal Ballroom #1 Chairs: Hoang-Hon Trinh, University of Ulsan, Korea Hamid reza Pourreza, Ferdowsi University of Mashhad, Iran 10:00 - 10:20 Vehicle Detection Algorithm Using Hypothesis Generation and Verification

Quoc Bao Truong, Byung Ryong Lee University of Ulsan, Korea 10:20 - 10:40 A Novel Method using Contourlet to Extract Features For Iris Recognition System

Amir Azizi1, Hamid Reza Pourreza2 1Islamic Azad University Mashhad Branch, Iran 2Ferdowsi University of Mashhad, Iran 10:40 - 11:00 Vehicle License Plate Detection Algorithm Based on Color Space and Geometrical Properties

Kaushik Deb1, Vasily V. Gubarevz1 and Kang-Hyun Jo2 1Novosibirsk State Technical University, Russia, 2University of Ulsan, Korea 11:00 - 11:20 Spatial Relation Model for Object Recognition in Human-Robot Interaction

Lu Cao, Yoshinori Kobayashi, and Yoshinori Kuno Saitama University, Japan 11:20 - 11:40 Window Extraction Using Geometrical Characteristics of Building Surface

Hoang-Hon Trinh, Dae-Nyeon Kim, Suk-Ju Kang, Kang-Hyun Jo University of Ulsan, Korea 11:40 - 12:00 Auto-Surveillance for Object to Bring in/out Using Multiple Camera

Taeho Kim1, Dong-Wook Seo2, Hyun-Uk Chae1 and Kang-Hyun Jo1 1University of Ulsan, Korea, 2MOTORWEL Corporation, Korea

F1A.6

F1A.5

F1A.4

F1A.3

F1A.2

F1A.1

T3E.5

T3E.4

Friday, September 18

- 28 -

10:00-12:00 <F1B> Intelligent Computing in Signal Processing RM : Crystal Ballroom #2 Chairs: Jong-Myon Kim, University of Ulsan, Korea Reyes-García, National Institute of Astrophysics, Optics and Electronics, Mexico 10:00 - 10:20 Automatic Music Transcription Based on Wavelet Transform

Amir Azizi1, Karim Faez2, Amin Rezaeian Delui3, Saeid Rahati1 1Islamic Azad University Mashhad Branch, Iran, 2Amir Kabir University of Technology, Iran, 3Toos Institute of Higher Education, Iran 10:20 - 10:40 Minimum Sum-of-Squares Clustering by DC programming and DCA

Le Thi Hoai An1, Pham Dinh Tao2 1University of Paul Verlaine, France, 2National Institute for Applied Sciences, France 10:40 - 11:00 Synthesis of Bowhead Whale Sound using Modified Spectral Modeling

Pranab Kumar Dhar, Sangjin Cho, Jong-Myon Kim University of Ulsan, Korea 11:00 - 11:20 Automatic Emphasis Labeling for Emotional Speech by Measuring Prosody Generation Error1

Jun Xu, Lianhong Cai Tsinghua National Laboratory for Information Science and Technology Tsinghua University, China 11:20 - 11:40 Type-2 Fuzzy Sets Applied to Pattern Matching for the Classification of Cries of Infants under

Neurological Risk Karen Santiago-Sánchez, Carlos A. Reyes-García, Pilar Gómez-Gil National Institute of Astrophysics, México 11:40 - 12:00 Real-Time Sound Synthesis of Plucked String Instruments using a Data Parallel Architecture

Huynh Van Luong, Sangjin Cho, Jong Myon Kim and Uipil Chong University of Ulsan, Korea 10:00-12:00 <F1C> Evolutionary Learning & Computational Genomics and Proteomics / Image Processing & Document Retrievals / Data Fusion RM : Crystal Ballroom #3 Chair: Vincenzo Di Lecce, Politecnico di Bari, Italy 10:00 - 10:20 CAPS Genomic Subtyping on Orthomyxoviridae

Sheng-Lung Peng1, Yu-Wei Tsay1, Chich-Sheng Lin2, and Chuan Yi Tang3 1National Dong Hwa University, Taiwan, 2National Chiao Tung University, Taiwan, 3National Tsing Hua University, Taiwan 10:20 - 10:40 Verification of Pathotyping by Quasispecies Model

Sheng-Lung Peng and Yu-Wei Tsay National Dong Hwa University, Taiwan 10:40 - 11:00 A Fuzzy Logic Based Approach to Feedback Reinforcement in Image Retrieval

Vincenzo Di Lecce, Alberto Amato Politecnico di Bari, Italy 11:00 - 11:20 Ontology-based Decision Support for Security Management in Heterogeneous Networks

Michal Choras1,2, Rafal Kozik2, Adam Flizikowski1,2, Rafal Renk1,3 and Witold Ho lubowicz1,3 1ITTI Ltd., Poland, 2Institute of Telecommunications, Poland, 3Adam Mickiewicz University, Poland 11:20 - 11:40 Recovering Facial Intrinsic Images from a Single Input

Ming Shao, Yun-Hong Wang Beihang University, China 11:40 - 12:00 DepthLimited Crossover in GP for Classifier Evolution

Hajira Jabeen, Abdul Rauf Baig National University of Computer and Emerging Sciences, Pakistan F1C.6

F1C.5

F1C.4

F1C.3

F1C.2

F1C.1

F1B.5

F1B.5

F1B.4

F1B.3

F1B.2

F1B.1

- 29 -

10:00-12:00 <F1D> Intelligent Control and Automation RM : Charlotte Room #1 Chairs: Cheol-Hong Moon, Gwangju University, Korea Cheolkeun Ha, University of Ulsan, Korea 10:00 - 10:20 Multiobjective Permutation Flow Shop Scheduling using a Memetic Algorithm with an

NEH-based Local Search Tsung-Che Chiang, Hsueh-Chien Cheng and Li-Chen Fu National Taiwan University, Taiwan 10:20 - 10:40 Intelligent Nonlinear Friction Compensation Using Friction Observer and Backstepping Control

Seong Ik Han1, Chan Se Jeong2, Sung Hee Park2, Young Man Jeong2, Chang Don Lee2, Soon Yong Yang2 1Suncheon First College, Korea, 2Ulsan University, Korea 10:40 - 11:00 Adaptive Control using Neural Network for Command Following of Tilt-Rotor Airplane in –Tilt

Angle Mode Jae Hyoung Im, Cheolkeun Ha University of Ulsan, Korea 11:00 - 11:20 Implementation of LED Array Color Temperature Controlled Lighting System using RISC IP Core

Cheol-Hong Moon, Woo-Chun Jang Gwangju University, Korea 11:20 - 11:40 A Universal Data Access Server for Distributed Data Acquisition and Monitoring Systems

Dae-Seung Yoo, Vu Van Tan and Myeong-Jae Yi University of Ulsan, Korea 11:40 - 12:00 INS/GPS Integration System with DCM based Orientation Measurement

Ho Quoc Phuong Nguyen, Hee-Jun Kang, Young-Soo Suh, Young-Shick Ro University of Ulsan, Korea 10:00-12:00 <F1E> Intelligent Computing in Robotics RM : Charlotte Room #2 Chairs: Kyoung Kwan Ahn, University of Ulsan, Korea Mun Ho Jeong, Korea Institute of Science and Technology, Korea 10:00 - 10:20 A Robot Visual/Inertial Servoing to an Object with Inertial Sensors

Ho Quoc Phuong Nguyen, Hee-Jun Kang, Young-Soo Suh, Young-Shick Ro University of Ulsan, Korea 10:20 - 10:40 A Service Framework of Humanoid in Daily Life

KangGeon Kim, Ji-Yong Lee, Seungsu Kim, Joongjae Lee, Mun-Ho Jeong, ChangHwan Kim and Bum-Jae You Korea Institute of Science and Technology, Korea 10:40 - 11:00 Self-stabilizing Human-like Motion Control Framework for Humanoids Using Neural Oscillators

Woosung Yang1, Nak Young Chong2, Syungkwon Ra1, Ji-Hun Bae1, Bum Jae You1 1Korea Institute of Science and Technology, Korea, 2Japan Advanced Institute of Science and Technology, Japan 11:00 - 11:20 Pseudorandom RFID Tag Arrangement for Improved Mobile Robot Localization

Sungbok Kim Hankuk University of Foreign Studies, Korea 11:20 - 11:40 Dynamic Model Identification of 2-Axes PAM Robot Arm Using Neural MIMO NARX Model

Kyoung Kwan Ahn1, Ho Pham Huy Anh2 1University of Ulsan, Korea, 2Ho Chi Minh City University of Technology, Viet Nam 12:00-13:15 Lunch

F1E.5

F1E.4

F1E.3

F1E.2

F1E.1

F1D.6

F1D.5

F1D.4

F1D.3

F1D.2

F1D.1

- 30 -

13:15-14:00 Keynote Speech 4 RM: Crystal Ballroom #2 Intelligent Pattern Recognition and Applications to Biometrics in Interactive Learning Environment Patrick Wang, Northeastern University, USA Chair: Young Soo Suh, University of Ulsan, Korea 14:00-14:15 Break 14:15-16:15 <F2A> Intelligent Computing in Computer Vision RM : Crystal Ballroom #1 Chairs: Yoshinori Kuno, Saitama University, Japan IBRAHIM INCE, Kyungsung University, Korea 14:15 - 14:35 Object Analysis for Outdoor Environment Perception using Multiple Features

Dae-Nyeon Kim, Hoang-Hon Trinh and Kang-Hyun Jo University of Ulsan, Korea 14:35 - 14:55 Building-based Structural Data for Core Functions of Outdoor Robot

Hoang-Hon Trinh, Dae-Nyeon Kim, Suk-Ju Kang, Kang-Hyun Jo University of Ulsan, Korea 14:55 - 15:15 Appearance Feature based Human Correspondence under Non-overlapping Views

Hyun-Uk Chae and Kang-Hyun Jo University of Ulsan Korea 15:15 - 15:35 A New Low-Cost Eye Tracking and Blink Detection Approach: Extracting Eye Features with

Blob Extraction Ibrahim Furkan Ince, Tae-Cheon Yang Kyungsung University, Korea 15:35 - 15:55 Assisted-Care Robot Based on Sociological Interaction Analysis

Wenxing Quan, Naoto Ishikawa, Yoshinori Kobayashi and Yoshinori Kuno Saitama University, Japan 15:55 - 16:15 A Novel User Created Message Application Service Design for Bidirectional TPEG

Sang-Hee Lee and Kang-Hyun Jo University of Ulsan, Korea 14:15-16:15 <F2B> Open Oral Session RM : Crystal Ballroom #2 14:15-16:15 <F2C> Intelligent Computing in Image Processing RM : Crystal Ballroom #3 Chairs: Nobuo Funabiki, Okayama University, Japan Yang Chen, Southeast University, China 14:15 - 14:35 Recent Progress of the Quasientropy Approach to Signal and Image Processing

Yang Chen and Zhimin Zeng Southeast University, China 14:35 - 14:55 HDR Image Generation based on Intensity Clustering and Local Feature Analysis

Andrey Vavilin and Kang-Hyun Jo University of Ulsan, Korea 14:55 - 15:15 A Framework for Recognition Books on Bookshelves

Nguyen-Huu Quoc and Won-Ho Choi University of Ulsan, Korea F2C.3

F2C.2

F2C.1

F2A.6

F2A.5

F2A.4

F2A.3

F2A.2

F2A.1

- 31 -

15:15 - 15:35 A Heuristic Optimization Algorithm for Panoramic Image Generation Problem from Multiple Cameras

Megumi Isogai, Nobuo Funabiki, Toru Nakanishi Okayama University, Japan 15:35 - 15:55 A New Method for Iris Recognition Based on Contourlet Transform and Non Linear

Approximation Coefficients Amir Azizi1, Hamid Reza Pourreza2 1Islamic Azad University Mashhad Branch, Iran, 2Ferdowsi University of Mashhad, Iran 15:55 - 16:15 A Unified Direct Approach to Image Registration and Object Recognition with a Hybrid

Evolutionary Algorithm Igor V. Maslov1, Izidor Gertner2 1Polytechnical University, Russia, 2The City College of New York, USA 14:15-16:15 <F2D> Intelligent Computing in Pattern Recognition / Intelligent Sensor Networks RM : Charlotte Room #1 Chair: Yongzhao Zhan, Jiangsu University, China 14:15 - 14:35 Two-Dimensional Heteroscedastic Discriminant Analysis for Facial Gender Classification

Jun-Ying Gan, Si-Bin He, Zi-Lu Ying, Lin-Bo Cai Wuyi University, China 14:35 - 14:55 Shot Retrival Based on Fuzzy Evolutionary aiNet and Hybrid Features

Xianhui Li, Yongzhao Zhan, Jia Ke Jiangsu University, China 14:55 - 15:15 Facial Expression Recognition with Local Binary Pattern and Laplacian Eigenmaps

Zilu Ying1,2, Linbo Cai1, Junying Gan1, Sibin He1 1Wuyi University, China, 2Beihang University, China 15:15 - 15:35 A Coverage and Energy Aware Cluster-Head Selection Algorithm in Wireless Sensor Networks

Thao P. Nghiem, Jong Hyun Kim, Sun Ho Lee, and Tae Ho Cho Sungkyunkwan University, Korea 15:35 - 15:55 u-Healthcare Service based on a USN Middleware Platform and Medical Device

Communication Framework Yung Bok Kim Sejong University, Korea 15:55 - 16:15 Energy Efficient MAC length Determination Method for Statistical En-Route Filtering using

Fuzzy Logic Hyeon Myeong Choi, Tae Ho Cho Sungkyunkwan University, Korea 14:15-16:15 <F2E> Natural Language Processing and Expert Systems / Ensemble Methods RM : Charlotte Room #2 Chair: Laurent HEUTTE, Universite de Rouen, France 14:15 - 14:35 Developing the KMKE Knowledge Management System Based on Design Patterns and Parallel

Processing Lien-Fu Lai1, Chao-Chin Wu1, Liang-Tsung Huang2 and Ya-Chin Chang1 1National Changhua University of Education, Taiwan, 2MingDao University, Taiwan 14:35 - 14:55 Analysis of Shipbuilding Fabrication Process with Enterprise Ontology

Ji-Hyun Park1, Kyung-Hoon Kim2, Jae-Hak J. Bae1 1University of Ulsan, Korea, 2Hyundai Heavy Industries Co., Ltd., Korea F2E.2

F2E.1

F2D.6

F2D.5

F2D.4

F2D.3

F2D.2

F2D.1

F2C.6

F2C.5

F2C.4

- 32 -

14:55 - 15:15 Estimate Word Emotions Based on Part of Speech and Positional Information

Kazuyuki Matsumoto1 and Fuji Ren1,2 1The University of Tokushima, Japan 2Beijing University of Posts and Telecommunications, China 15:15 - 15:35 Towards a Better Understanding of Random Forests through the Study of Strength and

Correlation Simon Bernard, Laurent Heutte and Sébastien Adam Université de Rouen, France 15:35 - 15:55 An Empirical Study of the Convergence of RegionBoost

Xinzhu Yang, Bo Yuan, Wenhuang Liu Tsinghua University, China 15:55 - 16:15 Binary Sequences with Good Aperiodic Autocorrelations Using Cross-Entropy Method

Shaowei Wang1, Jian Wang1, Xiaoyong Ji1, Yuhao Wang2 1Nanjing University, China 2Nanchang University, China 15:00-18:30 Industrial Tour / Poster Session 19:00-21:00 Conference Banquet RM: Crystal Ballroom #2

08:30-12:30 Ulsan City Tour

F2E.6

F2E.5

F2E.4

F2E.3

Saturday, September 19

- 33 -

Poster Session An Improved Harmony Search Algorithm for the Location of Critical Slip Surfaces in Slope Stability Analysis Liang Li1, Guang-Ming Yu1, Shi-Bao Lu1, Guo-Yan Wang1,2, Xue-Song Chu1 1Qingdao Technological University,.China, 2Liao Ning Technical University, China Improvement and Light Weight of Twin Seat Underframe in Multiple Unit Train RenLiang Wang, Hu Huang, XinTian Liu, LiHui Zhao Shanghai University of Engineering Science, China A Quantum Particle Swarm Optimization Used for Spatial Clustering with Obstacles Constraints Xueping Zhang1,2, Jiayao Wang1,3, Haohua Du4,Tengfei Yang1 and Yawei Liu1 1Henan University of Technology, China, 2Fuzhou University, China, 3PLA Information Engineering University, China, 4Beihang University, China. A Hybrid Ant Colony Algorithm for the Grain Distribution Centers Location Le Xiao1,3 and Qiuwen Zhang2 1Huazhong University of Science and Technology, China, 2China Airborne Missile Academy, China, 3 Henan University of Technology, China Learning Hereditary and Reductive Prolog Programs from Entailment Shahid Hussain1 and M.R.K. Krishna Rao2 1Bahria University, Pakistan, 2King Fahd University of Petroleum and Minerals,Saudi Arabia Computing Parallel Speeded-Up Robust Features (P-SURF) via POSIX Threads Nan Zhang Xi’an Jiaotong-Liverpool University, China Study on Fault Diagnosis of Rolling Mill Main Transmission System Based on EMD-AR Model and Correlation Dimension Gui-Ping DAI1,2, Man-Hua WU2 1JiangSu Province Support Software Engineering R&D Center for Modern Information Technology Application in Enterprise, China 2Suzhou Vocational University, China Analysis of Mixed Inflammable Gases Based on Single Sensor and RBF Neural Network Yu ZHANG1,2,Mei-Xing QI3 and Cai-Dong GU1,2 1JiangSu Province Support Software Engineering R&D Center for Modern Information Technology Application in Enterprise, China 2Suzhou Vocational University, China, 3Suzhou Industrial Park Institute of Vocational Technology, China Normalizing Human Ear in Proportion to Size and Rotation Ali Pour Yazdanpanah1, Karim Faez2 1Islamic Azad University of Najaf Abad, Iran, 2Amirkabir University of Technology, Iran Adaptive Immune Response Network Model Tao Liu1,2, Li Zhang2 and Binbin Shi2 1JiangSu Province Support Software Engineering R&D Center for Modern Information Technology Application in Enterprise, China 2Suzhou Vocational University, China Fault Diagnosis of Steam Turbine-Generator Sets Using CMAC Neural Network Approach and Portable Diagnosis Apparatus Implementation Chin-Pao Hung, Wei-Ging Liu and Hong-Zhe Su National Chin-Yi University of Technology, Taiwan Speech Emotion Recognition Research Based on Wavelet Neural Network for Robot Pet Yongming Huang, Guobao Zhang, Xiaoli Xu Southeast University, China

- 34 -

Fuzzy Data Fusion for Updating Information in Modeling Drivers’ Choice Behavior Mauro Dell’Orco and Mario Marinelli Technical University of Bari, Italy Bilateral Negotiation in a Multi-Agent Energy Market Fernando Lopes1, A. Q. Novais1 and Helder Coelho2 1LNEG − National Research Institute, Portugal, 2University of Lisbon, Portugal Conflict-free Incremental Learning Rong-Lei Sun Huazhong University of Science and Technology, China Fuzzy Failure Analysis of Automotive Warranty Claims Using Age and Mileage Rate SangHyun Lee, KyungIl Moon Honam University, Korea Training Neural Networks for Protein Secondary Structure Prediction: the Effects of Imbalanced Data Set Viviane Palodeto, Hernán Terenzi, Jefferson Luiz Brum Marques Federal University of Santa Catarina, Brazil An AIS-based E-mail Classification Method Jinjian Qing1, Ruilong Mao1, Rongfang Bie1 and Xiao-Zhi Gao2 1Beijing Normal University, China, 2Helsinki University of Technology, Finland A Fuzzy-GA Wrapper-based Constructive Induction Model Zohreh HajAbedi1, Mohammad Reza Kangavari2 1Islamic Azad University, Iran, 2Iran University of Scince and Technology, Iran A Petri Net-Based Ladder Logic Diagram Design Method for the Logic and Sequence Control of Photo Mask Transport Yun Liu Jimei University, China Quantum Quasi-cyclic Low-density Parity-check Codes Dazu Huang1,2, Zhigang Chen1, Xin Li1,2 and Ying Guo1 1Central South University, China, 2Hunan College of Finance and Economics, China DepthLimited Crossover in GP for Classifier Evolution Hajira Jabeen, Abdul Rauf Baig National University of Computer and Emerging Sciences, Pakistan Classification Rule Discovery Using a Correlation Based Ant Miner Waseem Shahzad, Abdul Rauf Baig National University of Computer & Emerging Sciences, Pakistan A Novel Face Detection Method Based on Contourlet Features Huan Yang, Yi Liu, Tao Sun, Yongmi Yang Shandong University, China Modeling of Micro-Piezoelectric Motion Platform for Compensation and Neuro-PID Controller Design Van-Tsai Liu, Ming-Jen Chen, Wei-Chih Yang National Formosa University, Taiwan The Fuzzy PIControl for the DSTATCOM Based on the Balance of Instantaneous Power Qun-feng ZHU, Lei HUANG, Zhan-bin HU, Jie TANG Shaoyang University, China

- 35 -

An Effective Edge-Adaptive Color Demosaicking Algorithm for Single Sensor Digital Camera Images Md. Foisal Hossain1, Mohammad Reza Alsharif1 and Katsumi Yamashita2 1Univeristy of the Ryukyus, Japan, 2Osaka Prefecture University, Japan Data Fusion Algorithm Based on Event-Driven and Minimum Delay Aggregation Path in Wireless Sensor Network Tianwei Xu1, Lingyun Yuan1, Ben Niu2 1Yunnan Normal University, China, 2College of Management Shenzhen University, China Image Segmentation to HSI Model Based on Improved Particle Swarm Optimization Bo Zhao1, Yajun Chen2, Wenhua Mao1, Xiaochao Zhang1 1The Institute of Electrical and Mechanical Technology, China, 2Xi’an University of Technology, China A Method for Modeling Gene Regulatory Network with Personal Computer Cluster Jinlian Wang, Jian Zhang, Lin Li Capital Medical University, China Study on the Agricultural Knowledge Representation Model Based on Fuzzy Production Rules Chunjiang Zhao1,2, Huarui Wu1,2 1National Engineering Research Center for Information Technology in Agriculture, China, 2the Ministry of Agriculture, China Using Non-extensive Entropy for Text Classification Lin Fu, Yuexian Hou Tianjin University, China Researches on Robust Fault-tolerant Control for Actuator Failures in Time-varying Delay System Dong Li Suzhou Institute of Trade and Ecommerce, China Signaling Pathway Reconstruction by Fusing Priori Knowledge Shanhong Zheng1,2, Chunguang Zhou1 and Guixia Liu1 1Jilin University, China, 2Changchun University of Technology,China Ship Classification by Superstructure Moment Invariants Prashan Premaratne and Farzad Safaei University of Wollongong, Australia Adaptive Chaotic Cultural Algorithm for Hyperparameters Selection of Support Vector Regression Jian Cheng1,2, Jiansheng Qian1 and Yi-nan Guo1 1China University of Mining and Technology, China, 2Tsinghua University, China Multi-View Ear Recognition Based on Moving Least Square Pose Interpolation Heng Liu1,2, David Zhang3, Zhiyuan Zhang4 1Southwest University of Science and Technology, China, 2Shanghai Jiao Tong University, China, 3The Hong Kong Polytechnic University, China 4Shenzhen Graduate School, China Emotional Particle Swarm Optimization Wei Wang, Zhiliang Wang, Xuejing Gu and Siyi Zheng University of Science and Technology Beijing, China A New Intrusion Detection Method Based on Antibody Concentration Jie Zeng, Tao Li, Guiyang Li, Haibo Li Sichuan University, China

- 36 -

Warning List Identification Based on Reliability Knowledge in Warranty Claims Information System SangHyun Lee1, ByungSu Jung2, KyungIl Moon1 1Honam University, Korea, 2Nambu University, Korea DDoS Attack Detection Method Based on Linear Prediction Model Jieren Cheng1,2, Jianping Yin1, Chengkun Wu1, Boyun Zhang3 and Yun Liu1 1National University of Defense Technology, China, 2Xiangnan University, China, 3Hunan Public Security College, China The Fault Diagnosis of Analog Circuits Based on Extension Theory Meng-Hui Wang, Yu-Kuo Chung, Wen-Tsai Sung National Chin-Yi University of Technology, Taiwan Ensemble Classifiers based on Kernel PCA for Cancer Data Classification Jin Zhou1, Yuqi Pan1, Yuehui Chen1 and Yang Liu2 1University of Jinan, China, 2Hong Kong Baptist University, Hong Kong A Numerical Simulation Study of the Dependence of Insulin Sensitivity Index on Parameters of Insulin Kinetics Lin Li, Wenxin Zheng Capital Medical University, China Locality Preserving Fisher Discriminant Analysis for Face Recognition Xu Zhao, Xiaoyan Tian Beijing University of Technology, China A Novel Time-Domain Structural Parametric Identification Methodology Based on the Equivalency of Neural Networks and ARMA Model Bin Xu1, Ansu Gong1, Jia He1 and Sami F. Masri1,2 1Hunan University, China, 2University of Southern, China Segmentation of Blood and Bone Marrow Cell Images via Learning by Sampling Chen Pan, Huijuan Lu, Feilong Cao China Jiliang University, China Using Bayesian Network and AIS to Perform Feature Subset Selection Boyun Zhang Hunan Public Security College, China Knowledge Representation and Consistency Checking in a Norm-Parameterized Fuzzy Description Logic Jidi Zhao1, Harold Boley2 and Weichang Du1 1University of New Brunswick, Canada, 2National Research Council of Canada, Canada Segment-Based Emotion Recognition from Continuous Mandarin Speech Jun-Heng Yeh, Ching-Yi Lin, Yao-Wei Tsai, Tsang-Long Pao and Yu-Te Chen Tatung University, Taiwan Study on Minimum Zone Evaluation of Flatness Errors Based on a Hybrid Chaos Optimization Algorithm Ke Zhang Shanghai Institute of Technology, China Research on Real-Time Software Sensors Based on Aspect Reconstruction and Reentrant Tao You, Cheng-lie Du and Yi-an Zhu Northwestern Polytechnical University, China A New Method of Color Map Segmentation Based on the Self-organizing Neural Network Zhenqing Xue and Chunpu Jia Shandong Business Institute, China

- 37 -

DCGene：A Novel Predicting Approach of the Disease Related Genes on Functional Annotation Fang Yuan1 and Wang Hui2 1Shenzhen Institute of Information Technology, China, 2Huazhong University of Science and Technology, China A Video-based Indoor Occupant Detection and Localization Algorithm for Smart Buildings Ling Chen, Feng Chen and Xiaohong Guan Tsinghua University, China Cluster Analysis and Fuzzy Query in Ship Maintenance and Design Jianhua Che1, Qinming He1, Yinggang Zhao2, Feng Qian1, Qi Chen1 1Zhejiang University, China, 2Henan Polytechnic University,China Rough Set Theory in Pavement Maintenance Decision Ching-Tsung Hung1, Jia-Ruey Chang2, Jyh-Dong Lin3, Gwo-Hshiung Tzeng4 1Kainan University, China, 2Minghsin University of Science and Technology, China, 3Assistant Feng Chia University, China 4Kainan University, China A Complex Fuzzy Controller for Reducing Torque Ripple of Brushless DC Motor Zhanyou Wang1, Shunyi Xie1, Zhirong Guo2 1Naval University of Engineering, China, 2Naval Bengbu Petty Officer Academy,China Exploiting Knowledge Ontology and Intelligent Agents for PPI Network Analysis Wei-Po Lee1, Wen-Shyong Tzou2 1National Sun Yat-sen University, Taiwan, 2National Taiwan Ocean University, Taiwan Combination of Gabor Wavelets and Circular Gabor Filter for Finger-Vein Extraction Jinfeng Yang, Jinli Yang, Yihua Shi Civil Aviation University of China, China Multi-UCAV Cooperative Path Planning Using Improved Coevolutionary Multi-Ant-Colony Algorithm Fei Su, Yuan Li, Hui Peng, Lincheng Shen National University of Defense Technology, China A method for Multiple Sequence Alignment based on Particle Swarm Optimization Fasheng Xu, Yuehui Chen University of Jinan, China Analysis and Improvement of An ID-Based Anonymous Signcryption Model Mingwu Zhang1, Yusheng Zhong1, Bo Yang1 and Wenzheng Zhang2 1South China Agricultural University, China, 2National Laboratory for Modern Communications, China Particle Swarm Optimizer Based on Dynamic Neighborhood Topology Yanmin Liu1,2, Qingzhen Zhao1, Zengzhen Shao1, Zhaoxia Shang1, Changling Sui2 1Shandong Normal University, China, 2Zunyi Normal College, China Inference of Differential Equation Models by Muti Expression Programming for Gene Regulatory Networks Bin Yang, Yuehui Chen and Qingfang Meng University of Jinan, China Function Sequence Genetic Programming Shixian Wang, Yuehui Chen and Peng Wu University of Jinan, China The EDML Format to Exchange Energy Profiles of Protein Molecular Structures* Dariusz Mrozek1, Bo ena Małysiakż -Mrozek1, Stanisław Kozielski1, Sylwia Górczy skań -Kosiorz2 1Silesian University of Technology, Poland, 2Silesian Medical University, Poland

- 38 -

A Design and Research of Eye Gaze Tracking System Based on Stereovision Pengyi Zhang1, Zhiliang Wang1, Siyi Zheng1,2, Xuejing Gu1,3 1Beijing University of Science and Technology, China, 2Beijing Command College of Chinese People's Armed Police Force, China 3Hebei Polytechnic University, China Performance Enhancement of Sum of Absolute Difference (SAD) Computation in H.264/AVC using Saturation Arithmetic Trung Hieu Tran, Hyo-Moon Cho and Sang-Bock Cho University of Ulsan, Korea Solar Radiation Forecasting using ad-hoc Time Series Preprocessing and Neural Networks 1 Christophe Paoli1, Cyril Voyant1,2, Marc Muselli1, Marie-Laure Nivet1 1University of Corsica, France, 2Hospital of Castelluccio, Radiotherapy Unit, France The Analysis of the Energy Function of Chaotic Neural Network with White Noise Xu Yaoqun, Qin Feng Harbin University of Commerce, China Combined Discrete Particle Swarm Optimization and Simulated Annealing for Grid Computing Scheduling Problem Ruey-Maw Chen1, Der-Fang Shiau2, Shih-Tang Lo3 1National Chin-yi University of Technology, Taiwan, 2Fooyin University, Taiwan, 3Kun-Shan University, Taiwan Pseudo Invariant Line Moment to Detect the Target Region of Moving Vessels Jia Ke, Yongzhao Zhan, Xiaojun Chen, Manrong Wang Jiangsu University, China Jobs Run-time Scheduling in a Java Based Grid Architecture Cataldo Guaragnella, Andrea Guerriero, Ciriaco C. Pasquale, Francesco Ragni Politecnico di Bari, Italy Self-adaptation of learning rate in XCS working in noisy and dynamic environments Maciej Troc and Olgierd Unold Wroclaw University of Technology, Poland Fuzzy Support Vector Classification Based on Fuzzy Optimization Zhimin Yang, XiaoYang, Bingquan Zhang Zhejiang University of Technology, China Tactical Aircraft Pop-up Attack Planning using Collaborative Optimization Nan Wang, Lin Wang, Yanlong Bu, Guozhong Zhang, Lincheng Shen Mechatronics and Automation School of National University of Defense Technology, China MotifMiner: A Table Driven Greedy Algorithm For DNA Motif Mining Seeja K.R,Alam M.A, Jain S.K Hamdard University, India Personal Identification Based on Finger-Vein Features Jinfeng Yang, Jinli Yang, Yihua Shi Civil Aviation University of China, China The Classification of a Simulation Data of a Servo System Via Evolutionary Artificial Neural Networks Asil Alkaya, G.Miraç Bayhan Dokuz Eylul University , Turkey Design of the Performance Evaluation Library for Speech Recognition Systems Based on SystemC Jin-wei Liu, Si-jia Huo, Zhang-qin Huang, Yi-bin Hou, Jin-jia Wang Beijing University of Technology, China

- 39 -

A Semi-Automated Dynamic approach to Threat Evaluation and Optimal Defensive Resource Allocation Huma Naeem1,3, Asif Masood1,4, Mukhtar hussain1,5, Shoab A Khan2,6 1Military College of Signals, Pakistan, 2National university of Science and Tehcnology, Pakistan Natural Language Human-robot Interface using Evolvable Fuzzy Neural Networks for Mobile Technology Wojciech Kacalak and Maciej Majewski Koszalin University of Technology, Poland E-learning Systems with Artificial Intelligence in Engineering Wojciech Kacalak and Maciej Majewski Koszalin University of Technology, Poland A Plausible Model for Cellular Self-defense Mechanisms in Response to Continuous Ion Radiation(IR) under Radiotherapy Jinpeng Qi, Shihuang Shao and Yizhen Shen Donghua University, China Plausible Model of Feedback-Control for Cellular Response based on Gene Regulatory Networks under Radiotherapy Jinpeng Qi, Shihuang Shao and Yizhen Shen Donghua University, China Synchronization Behavior Analysis for Coupled Lorenz Chaos Dynamic Systems via Complex Networks Yuequan Yang1, Xinghuo Yu2, Tianping Zhang1 1Yangzhou University, China, 2RMIT University, Australia Weighted Small World Complex Networks: Smart Sliding Mode Control Yuequan Yang1, Xinghuo Yu2 1Yangzhou University, China, 2RMIT University, Australia A Framework on Rough Set-based Partitioning Attribute Selection Tutut Herawan, Mustafa Mat Deris Universiti Tun Hussein Onn Malaysia, Malaysia Fuzzy Modeling and Position Control of an Ultrasonic Motor Behnood Rasti and Hamed mojallali University of Guilan, Iran Robust Acoustic Source Localization With TDOA based RANSAC Algorithm Peihua Li and Xianzhe Ma Heilongjiang University, China A Parabolic Detection Algorithm Based on Kernel Density Estimation Xiaomin Liu, Qi Song and Peihua Li Hei Long Jiang University, China

- 40 -

AAbbssttrraacctt ooff PPaappeerr

▶ 10:15-12:15 <T1A> Neural Networks / Swarm Intelligence and Optimization RM : Crystal Ballroom #1 A SOM Based Stereo Pair Matching Algorithm for 3-D Particle Tracking Velocimetry

Kazuo Ohmi, Basanta Joshi, Sanjeeb Prasad Panday Osaka Sangyo University, Japan A self-organizing map (SOM) based algorithm has been developed for 3-D particle tracking velocimetry (3-D PTV) in stereoscopic particle pairing process. In this process every particle image in the left-camera frame should be paired with the most probably correct partner in the right-camera frame or vice versa for evaluating the exact coordinate. In the present work, the performance of the stereoscopic particle pairing is improved by applying proposed SOM optimization technique in comparison to a conventional epipolar line analysis. The algorithm is tested with the 3-D PIV standard image of the Visualization Society of Japan (VSJ) and the matching results show that the new algorithm is capable of increasing the recovery rate of correct particle pairs by a factor of 9 to 23 % compared to the conventional epipolar-line nearest-neighbor method.

Spiking Neural Network Performs Discrete Cosine Transform for Visual Images QingXiang Wu, T. M. McGinnity, Liam Maguire, Arfan Ghani, Joan Condell University of Ulster at Magee Campus, UK The human visual system demonstrates powerful image processing functionalities. Inspired by the principles from neuroscience, a spiking neural network is proposed to perform the discrete cosine transform for visual images. The structure and the properties of the network are detailed in this paper. Simulation results show that the network is able to perform the discrete cosine transform for visual images. Based on this mechanism, the key features can be extracted in ON/OFF neuron arrays. These key features can be used to reconstruct the visual images. The network can be used to explain how the spiking neuron-based system can perform key feature extraction. The differences between the discrete cosine transform and the spiking neural network transform are discussed.

Spam Detection based on a Hierarchical Self-Organizing Map E.J. Palomo, E. Domnguez, R.M. Luque, and J. Muñoz University of Malaga, Spain The GHSOM is an articial neural network that has been widely used for data clustering. The hierarchical architecture of the GH-SOM is more exible than a single SOM since it is adapted to input data, mirroring inherent hierarchical relations among them. The adaptation process of the GHSOM architecture is controlled by two parameters. However, these parameters have to be established in advance and this task is not always easy. In this paper, a new hierarchical self-organizing model that has just one parameter is proposed. The performance of this model has been evaluated by building a spam detector. Experimental results conrm the goodness of this approach.

An Ensemble of Neural Networks for Stock Trading Decision Making Pei-Chann Chang1, Chen-Hao Liu3, Chin-Yuan Fan2, Jun-Lin Lin1, Chih-Ming Lai1 1Yuan Ze University, Taiwan, 2Ming Dao University, Taiwan, 3Kainan University, Taiwan Stock turning signals detection are very interesting subject arising in numerous financial and economic planning problems. In this paper, Ensemble Neural Network system with Intelligent Piecewise Linear Representation for stock turning points detection is presented. The Intelligent piecewise linear representation method is able to generate numerous stocks turning signals from the historic data base, then Ensemble Neural Network system will be applied to train the pattern and retrieve similar stock price patterns from historic data for training. These turning signals represent short-term and long-term trading signals for selling or buying stocks from the market which are applied to forecast the future turning points from the set of test data. Experimental results demonstrate that the hybrid system can make a significant and constant amount of profit when compared with other approaches using stock data available in the market.

T1A.4

T1A.3

T1A.2

T1A.1

Thursday, September 17

- 41 -

An Improved PSO Algorithm Encoding a Priori Information for Nonlinear Approximation Tong-Yue Gu, Shi-Guang Ju, Fei Han Jiangsu University, China In this paper, an improved PSO algorithm for nonlinear approximation is proposed. The particle swarm optimization is easy to lose the diversity of the swarm and trap into the local minima. In order to resolve this problem, in the proposed algorithm, when the swarm loses its diversity, the current each particle and its historical optimium are interrupted by random function. Moreover, the a priori information obtained from the nonlinear approximation problem is encoded into the PSO. Hence, the proposed algorithm could not only improve the diversity of the swarm but also reduce the likelihood of the particles being trapped into local minima on the error surface. Finally, two real data in chemistry field are used to verify the efficiency and effectiveness of the proposed algorithm.

Multi-objective Oriented Search Algorithm for Multi Objective Reactive Power Optimization Xuexia Zhang, Weirong Chen Southwest Jiaotong University, China This paper presents a novel algorithm, multi-objective oriented search algorithm (MOOSA), to deal with the problem of multi-objective reactive power optimization in power system. The multi-objective oriented search algorithm has strong ability to search optimal solutions and well distributed solutions in Pareto front. The results show that the proposed algorithm is able to balance the multi objects in multi-objective reactive power optimization through the simulations on IEEE 30-bus testing system. The paper concludes that MOOSA is an effective tool to handle the problem of multi objective reactive power optimization.

▶ 10:15-12:15 <T1B> Evolutionary Learning & Genetic Algorithms RM : Crystal Ballroom #2 Interactive Genetic Algorithms with Individual's Fuzzy Fitness

Dun-wei Gong, Jie Yuan and Xiao-yan Sun China University of Mining and Technology, China Interactive genetic algorithms are effective methods to solve an optimization problem with implicit or fuzzy indices, and have been successfully applied to many real-world optimization problems in recent years. In traditional interactive genetic algorithms, many researchers adopt an accurate number to express an individual's fitness assigned by a user. But it is difficult for this expression to reasonably reflect a user's fuzzy and gradual cognitive to an individual. We present an interactive genetic algorithm with an individual's fuzzy fitness in this paper. Firstly, we adopt a fuzzy number described with a Gaussian membership function to express an individual's fitness. Then, in order to compare different individuals, we generate a fitness interval based on ∂-cut set, and obtain the probability of individual dominance by use of the probability of interval dominance. Finally, we determine the superior individual in tournament selection with size two based on the probability of individual dominance, and perform the subsequent evolutions. We apply the proposed algorithm to a fashion evolutionary design system, a typical optimization problem with an implicit index, and compare it with two interactive genetic algorithms, i.e., an interactive genetic algorithm with an individual's accurate fitness and an interactive genetic algorithm with an individual's interval fitness. The experimental results show that the proposed algorithm is advantageous in alleviating user fatigue and looking for user's satisfactory individuals.

Interactive Genetic Algorithms with Variational Population Size Jie Ren, Dun-wei Gong, Xiao-yan Sun, Jie Yuan and Ming Li China University of Mining and Technology, China Traditional interactive genetic algorithms often have a small population because of a limited human-computer interface and user fatigue, which restricts these algorithms’ performances to some degree. In order to improve these algorithms’ performances and alleviate user fatigue effectively, we propose an interactive genetic algorithm with variational population size in this paper. In the algorithm, the whole evolutionary process is divided into two phases, i.e. fluctuant phase and stable phase of the user’s cognition. In fluctuant phase, a large population is adopted and divided into several coarse clusters according to the similarity of individuals. The user only evaluates these clusters’ centers, and the other individuals’ fitness is estimated based on the acquired information. In stable phase, the similarity threshold changes along with the evolution, leading to refined clustering of the population. In addition, elitist individuals are reserved to extract building blocks. The offspring is generated based on these building blocks, leading to a reduced population. The proposed algorithm is applied to a fashion evolutionary design system, and the results validate its efficiency.

Two Step Template Matching Method with Correlation Coefficient and Genetic Algorithm Gyeongdong Baek and Sungshin Kim Pusan National University, Korea This paper presents a rotation invariant template matching method based on two step matching process, cross correlation and genetic algorithm. In order to improve the matching performance, the traditional normalized correlation coefficient method is combined with genetic algorithm. Normalized correlation coefficient method computes probable local position of the template in the scene image. And genetic algorithm computes global position and rotation of the template in the scene image. The experimental results show that this algorithm has good rotate invariance, and high precision property.

T1B.3

T1B.2

T1B.1

T1A.6

T1A.5

- 42 -

Missing Data Imputation in Multivariate Data by Evolutionary Algorithms Juan C. Figueroa Garcìa1, Dusko Kalenatic2, and Cesar Amilcar Lopez Bello3 1Universidad Distrital Francisco José de Caldas, Bogotá – Colombia, 2Universidad de la Sabana, Chia – Colombia, 3Universidad Distrital Francisco José de Caldas, Bogotá - Colombia This paper presents a proposal based on an Evolutionary algorithm to impute missing observations in Multivariate Data. A genetic algorithm based on the minimization of an error function derived from their covariance matrix and mean is presented. All methodological aspects of the genetic structure are presented. An extended explanation of the design of the Fitness Function is provided. An application example is solved by the proposed method.

Some Distributed Algorithms for Quantized Consensus Problem Jianping He, Wenhai Chen, Lixin Gao Wenzhou University, China In this paper, we propose some distributed algorithms for quantized consensus. These algorithms are used to study the distributed averaging problem on arbitrary connected graphs and arbitrary connected weighted graphs, with the additional constraint that the weight value at each node is an integer. These algorithms can guarantee the system achieve consensus with some moderate assumptions and can use to solve several application problems, such as averaging in a network with finite capacity channels and load balancing in a processor network, which can be modeled as distributed averaging problem.

On the Robustness of Type-1 and Type-2 Fuzzy Tests vs. ANOVA Tests on Means Juan C. Figueroa Garcìa1, Dusko Kalenatic2 and Cesar Amilcar Lopez Bello1 1Universidad Distrital Francisco José de Caldas, Bogotá – Colombia, 2Universidad de la Sabana, Chia - Colombia This paper presents a simulation study on fuzzy tests vs. ANOVA test on means. Type-1, Interval Type-2 and ANOVA classical tests are compared through a simulated experiment for contrasting the stability of those approaches in front to a small change on sample. We perform an experiment of comparing the means of three groups where the classical ANOVA test is very nearby to the rejection p-value and the fuzzy tests get more robust results. In this way, we use bootstrap concepts to simulate the change of a random value of the sample to view the behavior of each technique in front to these changes.

▶ 10:15-12:15 <T1C> Fuzzy Systems and Soft Computing / Kernel Methods and Supporting Vector Machines RM : Crystal Ballroom #3

An FIS for Early Detection of Defect Prone Modules Zeeshan Ali Rana, Mian Muhammad Awais, Shafay Shamail Lahore University of Management Sciences (LUMS), Pakistan Early prediction of defect prone modules helps in better resource planning, test planning and reducing the cost of defect correction in later stages of software lifecycle. Early prediction models based on design and code metrics are difficult to develop because precise values of the model inputs are not available. Conventional prediction techniques require exact inputs, therefore such models cannot always be used for early predictions. Innovative prediction methods that use imprecise inputs, however, can be applied to overcome the requirement of exact inputs. This paper presents a fuzzy inference system (FIS) that predicts defect proneness in software using vague inputs defened as fuzzy linguistic variables. The paper outlines the methodology for developing the FIS and applies the model to a real dataset. Performance analysis in terms of recall, accuracy, misclassiffication rate and a few other measures has been conducted resulting in useful insight to the FIS application. The FIS model predictions at an early stage have been compared with conventional prediction methods (i.e. classification trees, linear regression and neural networks) based on exact values. In case of the FIS model, the maximum and the minimum performance shortfalls were noticed for true negative rate (TNRate) and F measure respectively. Whereas for Recall, the FIS model performed better than the other models even with the imprecise inputs.

Variable Precision Concepts and Its Applications for Query Expansion Fei Hao1, Shengtong Zhong2 1Korea Advanced Institute of Science and Technology, Korea, 2Norwegian University of Science and Technology, Norway One of the most important tasks of search engines is presenting more additional relevant web pages and reducing those pages which are useless for users. Query expansion is an efficient method for dealing with this task. In this paper, variable precision concept(VPC) based on formal concept analysis(FCA) is firstly proposed and its properties are discussed. Then a new strategy of expanding query terms based on VPC is proposed. According to this new strategy, users can set the query precision in terms of their interests and obtain the additional relevance web pages. Finally, application results show the efficiency and effectiveness of this method.

Combining Global model and Local Adaptive Neuro-Fuzzy Network Yun-Hee Han, Keun-Chang Kwak Chosun University, Korea This paper is concerned with a method for combining global model with local adaptive neuro-fuzzy network. The underlying principle of this approach is to consider a two-step development. First, we construct a standard linear regression as global model which could be treated as a preliminary design capturing the linear part of the data. Next, all modeling discrepancies are compensated by a collection of rules that become attached to the regions of the input space in which the error becomes localized. The incremented

T1C.3

T1C.2

T1C.1

T1B.6

T1B.5

T1B.4

- 43 -

neuro-fuzzy network is constructed by building a collection of information granules through some specialized fuzzy clustering, called context-based fuzzy c-means that is guided by the distribution of error of the linear part of its development. The experimental results reveal that the proposed method shows a good approximation and generalization capability in comparison with the previous works.

Application of a Case Base Reasoning Based Support Vector Machine for Financial Time Series Data Forecasting Pei-Chann Chang1, Chi-Yang Tsai1, Chiung-Hua Huang1,2, Chin-Yuan Fan3 1Yuan Ze University, Taiwan, 2Ta Hwa Institute of Technology, Taiwan, 3Ming Dao University, Taiwan This paper establishes a novel financial time series-forecasting model, by clustering and evolving support vector machine for stocks on S&P 500 in the U.S. This forecasting model integrates a data clustering technique with Case Based Reasoning (CBR) weighted clustering and classification with Support Vector Machine (SVM) to construct a decision-making system based on historical data and technical indexes. The future price of the stock is predicted by this proposed model using technical indexes as input and the forecasting accuracy of the model can also be further improved by dividing the historic data into different clusters. Overall, the results support the new stock price predict model by showing that it can accurately react to the current tendency of the stock price movement from these smaller cases. The hit rate of CBR-SVM model is 93.85% the highest performance among others.

Cost-Sensitive Supported Vector Learning to Rank Imbalanced Data Set Xiao Chang1,2, Qinghua Zheng1,2

and Peng Lin1,2

1Xi'an Jiaotong University, China, 2Shaanxi Key Lab. of Satellite and Computer Network, China In recent years, the algorithms of learning to rank have been proposed by researchers. Most of these algorithms are pairwise approach. In many real world applications, instances of ranks are imbalanced. After the instances of ranks are composed to pairs, the pairs of ranks are imbalanced too. In this paper, a cost-sensitive risk minimum model of pairwise learning to rank imbalance data sets is proposed. Following this model, the algorithm of cost-sensitive supported vector learning to rank is investigated. In experiment, the convention Ranking SVM is used as baseline. The document retrieval data set is used in experiment. The experimental results show that the performance of cost-sensitive supported vector learning to rank is better than Ranking SVM on the document retrieval data set.

An Ensemble Classifier Based on Kernel Method for Multi-situation DNA Microarray Data Xuesong Wang, Yangyang Gu, Yuhu Cheng and Ruhai Lei China University of Mining and Technology, China In order to deal with the interaction between genes effectively, a kernel technology was adopted into a subspace method in our study. A linear subspace classifier was generalized to a nonlinear kernel subspace classifier by using a kernel principle component analysis method to constitute nonlinear feature subspaces. Because DNA microarray data have characteristics of high dimension, few samples and strong nonlinearity, three types of classifiers based on kernel machine learning method were designed, i.e., support vector machine (SVM), kernel subspace classifier (KSUB-C) and kernel partial least-squares discriminant analysis (KPLS-DA). But the performances of these classifiers lie on the optimum setting of kernel functions and parameters. Therefore, to avoid the difficulty of selecting optimal kernel functions and parameters and to further improve the accuracy and generalization property of the classifiers, an ensemble classifier based on kernel method for multi-situation DNA microarray data was proposed by adopting the idea of ensemble learning. The ensemble classifier combines the classification results of the SVM, KSUB-C and KPLS-DA classifiers. Experimental results involving three public DNA microarray datasets indicate that the proposed ensemble classifier has high classification accuracy and perfect generalization property.

▶ 10:15-12:15 <T1D> Combinatorial & Numerical Optimization / Systems & Computational Biology / Machine Learning Theory and Methods RM : Charlotte Room #1

An Effective Hybrid Algorithm Based on Simplex Search and Differential Evolution for Global Optimization Ye Xu, Ling Wang, Lingpo Li Tsinghua University, China In this paper, an effective hybrid NM-DE algorithm is proposed for global optimization by merging the searching mechanisms of Nelder-Mead (NM) simplex method and differential evolution (DE). First a reasonable framework is proposed to hybridize the NM simplex-based geometric search and the DE-based evolutionary search. Second, the NM simplex search is modified to further improve the quality of solutions obtained by DE. By interactively using these two searching approaches with different mechanisms, the searching behavior can be enriched and the exploration and exploitation abilities can be well balanced. Based on a set of benchmark functions, numerical simulation and statistical comparison are carried out. The comparative results show that the proposed hybrid algorithm outperforms some existing algorithms including hybrid DE and hybrid NM algorithms in terms of solution quality, convergence rate and robustness.

Differential Evolution with Level Comparison for Constrained Optimization Ling-po Li, Ling Wang, Ye Xu Tsinghua University, China The effectiveness of a constrained optimization algorithm depends on both the searching technique and the way to handle constraints. In this paper, a differential evolution (DE) with level comparison is put forward to solve the constrained optimization problems. In

T1D.2

T1D.1

T1C.6

T1C.5

T1C.4

- 44 -

particular, the (comparison level) constrained method is adopted to handle constraints, while the DE-based evolutionary search is used to find promising solutions in the search space. In addition, the scale factor of the DE mutation is set to be a random number to vary the searching scale, and a certain percentage of population is replaced with random individuals to enrich the diversity of population and to avoid being trapped at local minima. Moreover, we let the level increase exponentially along with the searching process to stress feasibility of solution at later searching stage. Experiments and comparisons based on the 13 well-known benchmarks demonstrate that the proposed algorithm outperforms or is competitive to some typical state-of-art algorithms in terms of the quality and efficiency.

Agent based modeling of atherosclerosis: a concrete help in personalized treatments Francesco Pappalardo1,2, Alessandro Cincotti3, Alfredo Motta4, and Marzio Pennisi2 1National Research Council(CNR), Italy, 2University of Catania, Catania, Italy, 3Japan Advanced Institute of Science and Technology, Japan 4Politecnico di Milano, Italy Atherosclerosis, a pathology affecting arterial blood vessels, is one of most common diseases of the developed countries. We present studies on the increased atherosclerosis risk using an agent based model of atherogenesis that has been previously validated using clinical data. It is well known that the major risk in atherosclerosis is the persistent high level of low density lipoprotein (LDL) concentration. However, it is not known if short period of high LDL concentration can cause irreversible damage and if reduction of the LDL concentration (either by life style or drug) can drastically or partially reduce the already acquired risk. We simulated four different clinical situations in a large set of virtual patients (200 per clinical scenario). In the first one the patients lifestyle maintains the concentration of LDL in a no risk range. This is the control case simulation. The second case is represented by patients having high level of LDL with a delay to apply appropriate treatments; The third scenario is characterized by patients with high LDL levels treated with specific drugs like statins. Finally we simulated patients that are characterized by several oxidative events (smoke, sedentary life style, assumption of alcoholic drinks and so on so forth) that effective increase the risk of LDL oxidation. Those preliminary results obviously need to be clinically investigated. It is clear, however, that SimAthero has the power to concretely help medical doctors and clinicians in choosing personalized treatments for the prevention of the atherosclerosis damages.

Construction of the ensemble of logical models in cluster analysis Vladimir Berikov Sobolev Institute of mathematics, Russia In this paper, the algorithm of cluster analysis based on the ensemble of tree-like logical models (decision trees) is proposed. During the construction of the ensemble, the algorithm takes into account distances between logical statements describing clusters. Besides, we consider some properties of the Bayes model of classification. These properties are used at the motivation of information-probabilistic criterion of clustering quality. The results of experimental studies demonstrate the effectiveness of the suggested algorithm.

Ordinal Regression with Sparse Bayesian Xiao Chang1,2, Qinghua Zheng1,2 and Peng Lin1,2

1Xi'an Jiaotong University, China, 2Shaanxi Key Lab. of Satellite and Computer Network, China In this paper, a probabilistic framework for ordinal prediction is proposed, which can be used in modeling ordinal regression. A sparse Bayesian treatment for ordinal regression is given by us, in which an automatic relevance determination prior over weights is used. The inference techniques based on Laplace approximation is adopted for model selection. By this approach accurate prediction models can be derived, which typically utilize dramatically fewer basis functions than the comparable supported vector based and Gaussian process based approaches while offering a number of additional advantages. Experimental results on the real-world data set show that the generalization performance competitive with support vector-based method and Gaussian process-based method.

A Support System for Making Archive of Bi-Directional Remote Lecture - Photometric Calibration Naoyuki Tsuruta, Mari Matsumura and Sakashi Maeda Fukuoka University, Japan We are developing a system that supports making lecture movie archive. This system enables us to combine CG or another movie with a lecture scene movie by using intelligent computer vision techniques, and supports us to generate an effective lecture archive. In this paper, we concentrate on a scenario to generate a movie that seems to be like a lecture done in one lecture room based on two movies of lectures done in different remote lecture rooms. Because the source movies are captured on the different camera work and the different illumination condition, not only geometric calibration but also photometric calibration are important to make the target movie realistic. To overcome this problem, we propose a photometric calibration technique based on “fast separation of direct and global components” method. By using this method, estimation of color of illumination on the scene becomes more stable than our conventional method.

T1D.6

T1D.5

T1D.4

T1D.3

- 45 -

▶ 10:15-12:15 <T1E> Supervised Learning / Neural & Nature Inspired Computing and Optimization / Communication and Networks RM : Charlotte Room #2

Profile based Algorithm to Topic Spotting in Reuter21578 Taeho Jo Inha University, Korea This research proposes an alternative approach to machine learning based ones for categorizing online news articles in Reuter21578. For using machine learning based approaches for any task of text mining or information retrieval, documents should be encoded into numerical vectors; two problems, huge dimensionality and sparse distribution, caused by encoding so. Although there are various tasks of text mining such as text categorization, text clustering, and text summarization, the scope of this research is restricted to text categorization. The idea of this research is to avoid the two problems by encoding a document or documents into a table, instead of numerical vectors. Therefore, the goal of this research is to improve the performance of text categorization by avoiding the two problems.

A New Method of Morphological Associative Memories Naiqin Feng, Xizheng Cao, Sujuan Li, Lianhui Ao and Shuangxi Wang Henan Normal University, China The morphological associative memories (MAM) have many attractive advantages. However, they can not give a guarantee that morphological hetero-associative memories are perfect, even if input patterns are perfect. In addition, the problem with the associative memory matrixes WXY and MXY is that WXY is incapable of handling dilative noise while MXY is incapable of effectively handling erosive noise. In this paper, the new methods of MAM, +WXY and +MXY are proposed. The certain qualifications that make +WXY and +MXY be perfect memories are analyzed and proved. As far as the hetero-associative memories are concerned, although +WXY and +MXY are not perfect, they are complements to original WXY and MXY. +WXY is capable of handling dilative noise while +MXY is capable of effectively handling erosive noise. Therefore they can be put together with original WXY and MXY to learn from others’ strong points to offset ones’ own weakness and to make the effect of hetero-associative memories and pattern recognition better. The calculation results demonstrate that both +WXY and +MXY are effectual

Emergency resources scheduling on continuous consumption system based on adaptively mutate genetic algorithm Zhang Liming, Lin Yuhua, Yang Guofeng, Chang Huiyou Sun Yat-sen University, China The emergency resources dispatch is a critical work in emergency relief, while it is quite difficult to achieve an optimized scheduling, adjusting to practical situation. In this paper, a large-scale emergency resources scheduling model is built, which simulates the realistic problem, this model includes multiple suppliers with a variety of resources, a single accident site and some restrictions, all closed to the practical event. Then we apply an adaptively mutate genetic algorithm to figure out a superior solution, which adopts the Binary Space Partitioning tree for heuristic searching and adaptive mutation. As the experimental result shows, this novel method proposed in our work performs well.

Cloud@Home: Bridging the Gap between Volunteer and Cloud Computing Vincenzo D. Cunsolo, Salvatore Distefano, Antonio Puliafito and Marco Scarpa University of Messina, Italy The ideas of using geographically distributed resources in a secure way (Network-Internet/Grid computing), providing self-management capabilities (Autonomic computing), quantifying and billing computing costs (Utility computing), in order to perform specific modular applications (web services), have been grouped altogether into the concept of Cloud computing. Only commercial Cloud solutions have been implemented so far, offering computing resources and (web) services for renting. Some interesting projects, such as Nimbus, OpenNEbula, Reservoir, work on Cloud. One of their aims is to provide a Cloud infrastructure able to provide and share resources and services for scientific purposes. The encouraging results of Volunteer computing projects such as SETI@home and FOLDING@home and the great flexibility and power of the emergent Cloud technology, suggested us to address our research efforts towards a combined new computing paradigm we named Cloud@Home, merging the benefits and overcoming the weaknesses of both the original computing paradigms. In this paper we present the Cloud@Home paradigm, describing its contribution to the actual state of the art on the topic of distributed and Cloud computing. We thus detail the functional architecture and the core structure implementing such a new paradigm, demonstrating how it is really possible to build up a Cloud@Home infrastructure.

An Intelligent Prediction Model for Generating LGD Trigger of IEEE 802.21 MIH M. Yousaf, Sohail Bhatti, Maaz Rehan, A. Qayyum and S. A. Malik Center of Research in Networks and Telecommunications (CoReNeT), Pakistan IEEE recently standardized 802.21-2008 Media Independent Handover (MIH) standard. MIH is a key milestone toward the evolution of integrated heterogeneous 4G wireless networks. MIH provides number of link layer events in a unified way that facilitate upper layer protocols in making handover decisions. One such event is Link Going Down (LGD) trigger. LGD is a predictive event that is generated when link conditions are expected to degrade in near future. Traditionally such link quality degradations and connectivity losses are predicted on the basis of a single parameter only i.e. received signal strength. However, in varying wireless conditions, simple predictions relying on single link layer parameter may generate false LGD triggers. This false triggering may initiate

T1E.5

T1E.4

T1E.3

T1E.2

T1E.1

- 46 -

unnecessary handovers that rather than facilitating upper layer mobility management protocols, may cause overhead and may degrade the overall network performance. In this paper, we present an intelligent model for generating MIH LGD trigger reliably. In our implementation, we used 'Time Delay Neural Networks (TDNN)' approach using multiple link layer parameters for LGD predictions. We also analyzed the prediction accuracy and the feasibility of using such intelligent technique for mobile devices.

GLRT Based Fault Detection in Sensor Drift Monitoring System In-Yong Seo1, Ho-Cheol Shin1, Moon-Ghu Park1 and Seong-Jun Kim2 1Korea Electric Power Research Institute, Korea, 2Kangnung National University, Korea In a nuclear power plant (NPP), periodic sensor calibrations are required to assure sensors are operating correctly. However, only a few faulty sensors are found to be calibrated. For the safe operation of an NPP and the reduction of unnecessary calibration, on-line calibration monitoring is needed. This paper presents an on-line sensor drift monitoring technique, based on a Generalized Likelihood Ratio Test (GLRT), for detecting and estimating mean shifts in sensor signal. Also, principal component-based Auto-Associative support vector regression (AASVR) is proposed for the sensor signal validation of the NPP. Response surface methodology (RSM) is employed to efficiently determine the optimal values of SVR hyperparameters. The proposed model was confirmed with actual plant data of Kori NPP Unit 3. The results show that the accuracy of the model and the fault detection performance of the GLRT are very competitive.

▶ 14:15-16:15 <T2A> Knowledge-Based Systems and Intelligent Computing In Medical Imaging RM : Crystal Ballroom #1

Characterization of endomicroscopic images of the distal lung for computer-aided diagnosis Aurélien Saint-Réquier1, Benoît Lelandais1, Caroline Petitjean1, Chesner Désir1, Laurent Heutte1, Mathieu Salaün2 and Luc Thiberville2 1Université de Rouen, France, 2CHU de Rouen, France This paper presents a new approach for the classification of pathological vs. healthy endomicroscopic images of the alveoli. These images, never seen before, require an adequate description. We investigate two types of feature vector for discrimination: a high-level feature vector based on visual analysis of the images, and a pixel-based, generic feature vector, based on Local Binary Patterns (LBP). Both feature sets are evaluated on state-of-the-art classifiers and an intensive study of the LBP parameters is conducted. Indeed best results are obtained with the LBP-based approach, with correct classification rates reaching up to 91.73% and 97.85% for non-smoking and smoking groups, respectively. Even though tests on extended databases are needed, first results are very encouraging for this difficult task of classifying endomicroscopic images of the distal lung.

Image processing framework for virtual colonoscopy Vitoantonio Bevilacqua1,2, Marco Cortellino1,2, Michele Piccinni1, Antonio Scarpa1, Diego Taurino1, Giuseppe Mastronardi1,2, Marco Moschetta3, Giuseppe Angelelli3 1Politecnico di Bari, Italy, 2Spin-Off of Polytechnic of Bari, Italy, 3Policlinico Universitario di Bari, Italy This paper describes a complete image processing framework for Virtual Colonscopy. The developed algorithms cover the entire process that allows a virtual navigation inside the colon lumen, starting from a dataset of axial CT slices. The implemented modules are: electronic colon cleansing, lumen segmentation, skeletonization, rendering and navigation. In particular for the centerline problem two different techniques are proposed and evaluated.

Combined use of densitometry and morphological analysis to detect flat polyps Vitoantonio Bevilacqua1,2, Marco Cortellino1,2, Giuseppe Mastronardi1,2, Antonio Scarpa1, Diego Taurino1 1Politecnico di Bari, Italy, 2Spin-Off of Polytechnic of Bari, Italy This paper describes a CAD system to detect a particular type of colon cancer, flat polyps. The identification of suspicious regions is based on two type of analysis executed in succession: a densitometry analysis that researches contrast fluid on polyp surface and a morphological analysis that reduces number of false positives, calculating a curvature index of the surface.

Relevant Measurements for polyps in 3D Virtual Colonoscopy Vitoantonio Bevilacqua1,2, Marianna Notarnicola1, Marco Cortellino1,2, Antonio Scarpa1, Diego Taurino1, Giuseppe Mastronardi1,2 1Politecnico di Bari, Italy, 2Spin-Off of Polytechnic of Bari, Italy Virtual Colonoscopy is an innovative method to discover colon neoplasias created in order to alleviate patients aches generated by the standard colonoscopy. For the same reason, we have realized an automatic process finalized to find polyps into the lumen through the extraction of colon centerline and the calculation of polyps distance from anus. This paper contains the description of what is implemented. In particular, the developed algorithms build up following steps: colon lumen segmentation starting from a dataset of axial CT slices, 3D rendering, centerline extraction and evaluation of polyps distance from anus.

Retinal Vessel Extraction by a Combined Neural Network-Wavelet Enhancement Method Leonarda Carnimeo1, Vitoantonio Bevilacqua1,2, Lucia Cariello1,2, Giuseppe Mastronardi1,2 1Polytechnic of Bari, Italy, 2Spin-Off of Polytechnic of Bari, Italy This paper presents a combined approach to automatic extraction of blood vessels in retinal images. The proposed procedure is composed of two phases: a wavelet transform-based preprocessing phase and a NN-based one. Several neural net topologies and

T2A.5

T2A.4

T2A.3

T2A.2

T2A.1

T1E.6

- 47 -

training algorithms are considered with the aim of selecting an effective combined method. Human retinal fundus images, derived from the publicly available ophthalmic database DRIVE, are processed to detect retinal vessels. The approach is tested by considering performances in terms of sensitivity and specificity values obtained from vessel classification. The quality of vessel identifications is evaluated on obtained image by computing both sensitivity values and specificity ones and by relating them in ROC curves. A comparison of performances by ROC curve areas for various methods is reported

▶ 14:15-16:15 <T2B> Dynamic Spectrum Sharing Systems RM : Crystal Ballroom #2 Cooperative Spectrum Sensing Using Enhanced Dempster-Shafer Theory of Evidence in Cognitive Radio

Nhan Nguyen Thanh and Insoo Koo University of Ulsan, Korea Cooperative is an appropriate method for improving the performance of spectrum sensing when cognitive radio system is under the deep shadowing and fading environment. The Dempster-Shafer theory of evidence for fusion has similar reasoning logic with human. Thus an enhanced scheme for cooperative spectrum sensing based on an enhanced Dempster-Shafer Theory of Evidence is proposed in this paper. Our scheme utilizes the signal to noise ratios to evaluate the degree of reliability of each local spectrum sensing terminal on a distributed Cognitive Radio network to adjust the sensing data more accuratly before making fusion by Dempter-Shafer theory of evidence. Simulation results show that significant improvement of the cooperative spectrum sensing gain is achieved.

A secure distributed spectrum sensing scheme in cognitive radio Nhan Nguyen-Thanh and Koo Insoo University of Ulsan, Korea Distributed spectrum sensing provides an improvement for primary user detection but leads a new security threat into CR system. The spectrum sensing data falsification malicious users can decrease the cooperative sensing performance. In this paper, we propose a distributed scheme in which the presence and absence hypotheses distribution of primary signal is estimated based on past sensing received power data by robust statistics, and the data fusion are performed according to estimated parameters by Dempster-Shafer theory of evidence. Our scheme can achieve a powerful capability of malicious user elimination due to the abnormality of the distribution of malicious users compared with that of other legitimate users. In addition, the performance of our data fusion scheme is enhanced by supplemented nodes’ reliability weight.

An Optimal Data Fusion Rule in Cluster-Based Cooperative Spectrum Sensing Hiep-Vu Van, Hyungseo Kang and Insoo Koo University of Ulsan, Korea In this paper, we consider a cluster-based cooperative spectrum sensing approach to improve the sensing performance of cognitive radio (CR) network. In the cluster-based cooperative spectrum sensing, CR users with the similar location are grouped into a cluster. In each cluster, the most favorable user namely cluster header, will be chosen to collect data from all CR users and send the cluster decision to common receiver who makes a final decision on the presence of primary user. In the cluster-based cooperative spectrum sensing, data fusion rule in the cluster takes an important role to reduce the rate of reporting error. Subsequently we propose optimal fusion rule for each cluster header with which we can minimize the sum of probability of false alarm and probability of missed detection in each cluster header.

Exact Bit Error Probability of Multi-hop Decode-and-Forward Relaying with Selection Combining Bao Quoc Vo-Nguyen and Hyung Yun Kong University of Ulsan, Korea In this paper, an exact closed-form bit error rate expression for M-PSK is presented for multi-hop Decode-and-Forward Relaying (MDFR) scheme, in which selection combining technique is employed at each node. We have shown that the proposed protocol offers remarkable diversity advantage over direct transmission as well as the conventional decode-and-forward relaying (CDFR) scheme. Simulations are performed to confirm our theoretical analysis.

A Cooperative Transmission Scheme for Cluster based Wireless Sensor Networks Asaduzzaman and Hyung Yun Kong University of Ulsan, Korea In this paper, a cross layer approach is used to obtain spatial diversity in physical layer. We develop a low complexity cooperative diversity protocol for Low Energy Adaptive Clustering Hierarchy (LEACH) based wireless sensor networks. A simple modification in clustering algorithm of LEACH protocol is proposed to exploit virtual MIMO based user cooperation. Instead of selecting a single cluster-head at network layer, we proposed M cluster-heads in each cluster to obtain a diversity order of M in long distance communication. Due to the broadcast nature of wireless transmission cluster-heads are able to receive the data from sensor nodes at the same time. This fact ensures the synchronization required to implement a virtual MIMO based space time block code (STBC) in cluster-head to sink node transmission. Analysis and simulation results show that our proposed cooperative LEACH protocol can save a huge amount of energy over LEACH protocol with same data rate, bit error rate, delay and bandwidth requirements.

T2B.5

T2B.4

T2B.3

T2B.2

T2B.1

- 48 -

A packet scheduling algorithm for IEEE 802.22 WRAN systems and calculation reduction method thereof Young-du Lee, Tae-joon Yun and Insoo Koo University of Ulsan, Korea In the paper, a scheduling algorithm based on the estimation of packet loss amount is proposed for supporting real-time traffic in IEEE 802.22 WRAN systems. The proposed scheduling algorithm takes two steps for resource allocation. In the first step, it assigns resource to users having packets that are estimated to be dropped at next frame. In the second step, it assigns the remaining resource to the other users according to one of the existing scheduling algorithms. The simulation results show that the proposed scheduling algorithm provides much better performance than the PLFS and MLWDF algorithm. In addition, a simple calculation method is proposed for the proposed scheme. With the simple method, we can reduce the number of checking packets in the queue as much as about a half time without any performance degradation in the case of video traffic.

▶ 14:15-16:15 <T2C> Knowledge Discovery and Data Mining / Intelligent Computing Algorithms in Banking and Finance RM : Crystal Ballroom #3

A Semantic Lexicon-based Approach for Sense Disambiguation and Its WWW Application Vincenzo Di Lecce1, Marco Calabrese1, Domenico Soldo2 1Polytechnic of Bari, Italy, 2myHermes S.r.l., Italy This work proposes a basic framework for resolving sense disambiguation through the use of Semantic Lexicon, a machine readable dictionary managing both word senses and lexico-semantic relations. More specifically, polysemous ambiguity characterizing Web documents is discussed. The adopted Semantic Lexicon is WordNet, a lexical knowledge-base of English words widely adopted in many research studies referring to knowledge discovery. The proposed approach extends recent works on knowledge discovery by focusing on the sense disambiguation aspect. By exploiting the structure of WordNet database, lexico-semantic features are used to resolve the inherent sense ambiguity of written text with particular reference to HTML resources. The obtained results may be extended to generic hypertextual repositories as well. Experiments show that polysemy reduction can be used to hint about the meaning of specific senses in given contexts.

The establishment of verb logic and its application in universal emergency response information system design Jian Tan, Xiang Tao Fan Laboratory of Digital Earth Sciences, China It is always a challenge to build up a stable and high-level integrated system capable of different types of emergencies. The biggest obstacle is how to build a universal work flow mode for the different events. To solve the problem, our research adopts an unusual way based on the self-evident truth that full text description of phenomena is a whole map of it. Then the system analysis’ subject can be altered from the real emergency response to the text description of it. Therefore semantic annotation which uses the semantic labels in propbank can be employed in the analysis process. The annotation subjects are the documents that each of them described a full emergency response process of different emergency type. After classification and statistic, three linguistic rules are found out. First, every sentence have a predicate verb which indicate an executable action and it belongs to a fixed set, second, each verb coexists with semantic role Arg0(actor), third, all the complement roles of predicate verbs converge into a fixed subset of semantic roles ,these conclusions are named together as verb logic. It is a high abstract semantic model, for it not only contains domains but also tell the relations among domains. Based on verb logic, universal work flow mode is constructed, and a universal emergency response system can be built up. The design of the system is also stated in this paper.

Using Intelligent System for Reservoir Properties Estimation Fariba Salehi1 and Arnoosh Salehi2 1Islamic Azad University, Iran, 2National Iranian Oil Company, Iran Reservoir description for simulation studies requires good knowledge of the permeability. Reliable permeability is only available from laboratory tests on cores, which are usually taken from a small percentage of the wells. In an offshore gas field only three wells have core data and all wells have full set of conventional log data. By using concept of hydraulic flow unit, statistical methods and intelligent systems is made a model for estimation of reservoir properties. Graphical statistical methods are applied for classification of hydraulic flow units. The Sugeno-type of fuzzy inference system conjunction with backpropagation network and subtractive clustering is used for prediction of flow zone indicator, permeability is then calculated from mean flow zone indicator value and corresponding porosity.

On an Ant Colony-based Approach for Business Fraud Detection1 Ou Liu1, Jian Ma2, Pak-Lok Poon1 and Jun Zhang3 1The Hong Kong Polytechnic University, China, 2City University of Hong Kong, China, 3Sun Yat-sen University, China Nowadays we witness an increasing number of business frauds. To protect investors’ interest, a financial firm should possess an effective means to detect such frauds. In this regard, artificial neural networks (ANNs) are widely used for fraud detection. Traditional back-propagation-based algorithms used for training an ANN, however, exhibit the local optima problem, thus reducing the effectiveness of an ANN in detecting frauds. To alleviate the problem, this paper proposes an approach to training an ANN using an ant colony optimization technique, through which the local optima problem can be solved and the effectiveness of an ANN in fraud detection can be improved. Based on our approach, an associated prototype system is designed and implemented, and an exploratory study is performed. The results of the study are encouraging, showing the viability of our proposed approach.

T2C.4

T2C.3

T2C.2

T2C.1

T2B.6

- 49 -

Markov Chain Monte Carlo on Asymmetric GARCH Model Using the Adaptive Construction Scheme Tetsuya Takaishi Hiroshima University of Economics, Japan We perform Markov chain Monte Carlo simulations for a Bayesian inference of the GJR-GARCH model which is one of asymmetric GARCH models. The adaptive construction scheme is used for the construction of the proposal density in the Metropolis-Hastings algorithm and the parameters of the proposal density are determined adaptively by using the data sampled by the Markov chain Monte Carlo simulation. We study the performance of the scheme with the artificial GJRGARCH data. We find that the adaptive construction scheme samples GJR-GARCH parameters effectively and conclude that the Metropolis-Hastings algorithm with the adaptive construction scheme is an efficient method to the Bayesian inference of the GJR-GARCH model.

An Intelligent Computing Algorithm to Analyze Bank Stock Returns Vincenzo Pacelli University of Foggia, Italy The objective of this paper is to propose an intelligent computing algorithm, represented by an artificial neural network model, to analyze the dynamics of stock prices of banks. Through the empirical application of the model developed, it is expected to obtain indications about the ability of the artificial neural network model developed to generalize the phenomenon analyzed. So the research aims to provide empirical results about the use of non-linear methods of analysis for the study of the dynamics of banks’ stock prices, enriching the prospects for research in terms of methodological tools.

▶ 14:15-16:15 <T2D> Advances in Intelligent Information Processing RM : Charlotte Room #1 Design of a Single-Phase Grid-Connected Photovoltaic Systems based on Fuzzy-PID Controller

Fengwen Cao, Yiwang Wang Suzhou Vocational University, China The output power of photovoltaic(PV) module varies with module temperature, solar isolation and 1oads changes etc. In order to control the output power of single-phase grid-connected PV system according to the output power PV arrays. In this paper design a Fuzzy-PID controller for single-phase grid connected PV system, which include a DC/DC converter and a single-phase DC/AC inverter that connected to utility grid. Fuzzy-PID control technique is used to realize the system control. The matlab simulation experimental results show that, the proposed method has the good performance

A Constrained Approximation Algorithm by Encoding Second-order Derivative Information into Feedforward Neural Networks Qing-Hua Ling, Fei Han Jiangsu University of Science and Technology, China In this paper, a constrained learning algorithm is proposed for function approximation. The algorithm incorporates constraints into single hidden layered feedforward neural networks from the a priori information of the approximated function. The activation functions of the hidden neurons are specific polynomial functions based on Taylor series expansions, and the connection weight constraints are obtained from the second-order derivative information of the approximated function. The new algorithm has been shown by experimental results to have better generalization performance than other traditional learning ones.

Image Reconstruction Using NMF with Sparse Constraints Based on Kurtosis Measurement Criterion Li Shang1,2, Jinfeng Zhang2, Wenjun Huai2, Jie Chen2 and Jixiang Du3,4,5 1JiangSu Province Support Software Engineering R&D Center for Modern Information Technology Application in Enterprise, China, 2Suzhou Vocational University, China, 3Huaqiao University, China, 4University of Science and Technology of China, China, 5Chinese Academy of Sciences, China A novel image reconstruction method using non-negative matrix factorization (NMF) with sparse constraints based on the kurtosis measurement is proposed by us. This NMF algorithm with sparse constraints exploited the Kurtosis as the maximizing sparse measure criterion of feature coefficients. The experimental results show that the natural images’ feature basis vectors can be successfully extracted by using our algorithm. Furthermore, compared with the standard NMF method, the simulation results show that our algorithm is indeed efficient and effective in performing image reconstruction task.

A Cyanobacteria Romate Monitoring System Zhiqiang Zhao1,2,3, Yiming Wang3 1JiangSu Province Support Software Engineering R&D Center for Modern Information Technology Application in Enterprise, China 2Suzhou Vocational University, China, 3Suzhou University, China This article analyzes the major factors and features of the Cyanophyta problem, and makes a research on some key attributs so as to build up a monitor system by analyzing all its steps, and utilizing the current theory and methodology. This article, in view of the traits of the Cyanophyta monitor techniques, fully focuses on the remote wireless monitoring of Cyanophyta in Tai Lake, and have realized software and hardware design of the end device, data excess point and remote monitoring platform. The application of WSNs has improved the Cyanophyta monitoring technique.

T2D.4

T2D.3

T2D.2

T2D.1

T2C.6

T2C.5

- 50 -

Image Segmentation of Level Set Based on Maximization of Between-Class Variance and Distance Constraint Function Chang-xiong Zhou1,2, Zhi-feng Hu1,2, Shu-fen Liu1,2, Ming Cui1,2 and Rong-qing Xu3 1JiangSu Province Support Software Engineering R&D Center for Modern Information Technology Application in Enterprise, China, 2Suzhou Vocational University, China, 3Nanjing University of Posts and Telecommunications, China In most existing level set models for image segmentation, it is necessary to constantly re-initialize the level set function, or to acquire the gradient flow information of the image to restrict the evolution of the curve. A novel image segmentation model of level set is proposed in the paper, which is based on the maximization of the between-class variance and the distance-based constraint function. In this model, the distance-based constraint function is introduced as the internal energy to ensure that the level set function is always the signed distance function (SDF), so that the constant re-initialization of the level set function during the evolution process is avoided. Meanwhile, the external energy function (between-class variance function) is constructed based on the weighted sum of square of the difference between the average grey levels of the target region and the overall region, the background and the overall region respectively. This function is maximized to ensure that the curve represented by zero level set converges towards the target boundary stably. Experimental results show that the constant re-initialization in traditional models has been eliminated in the proposed model. Furthermore, since region information has been incorporated into the energy function, the model renders good performance in the segmentation of both weak edges images and those with Gaussian noise or impulse noise.

Active MMW Focal Plane Imaging System Pingang Su1,2, Zongxin Wang2, Zhengyu Xu2 1Suzhou Vocational University, China, 2Southeast University, China Millimeter wave imaging technology has received a lot of attention in recent years. It has been widely applied in aircraft landing guidance system, dangerous substance inspection, plasma tests and human carry on safety inspection. It can be applied in many areas in the future. This paper studies an active millimeter wave focal plane imaging system. Among the theory and key techniques of millimeter wave imaging, the optical imaging system is designed and automatic scanning, automatic data collection and image processing are implemented. In the experiment system, the automatic focal plane scanning of the receiver antenna array, data collection around the imaging spots of the 94GHz millimeter wave, automatic imaging and image optimization are carried out at the computer. The experiment verifies the practicability of the design of the diffractive lens and the detection of hidden objects.

▶ 14:15-16:15 <T2E> Network-based Intelligent Technologies RM : Charlotte Room #2 Handling Multi-Channel Hidden Terminals Using a Single Interface in Cognitive Radio Networks

Liang Shan, Myung Kyun Kim University of Ulsan, Korea Cognitive networks enable efficient sharing of the radio spectrum. Multi-hop cognitive network is a cooperative network in which cognitive users take help of their neighbors to forward data to the destination. Control signals exchanged through a common control channel (CCC) to enable cooperation communication. But, using a common control channel introduces a new issue like channel saturation which degrades the overall performance of the network. Moreover, the multi-channel hidden terminal problem will be another important challenge in cognitive radio networks, in which the multi-channel hidden terminals can decrease the throughput, cause much overhead, and sometimes even make the whole network invalidated. In this paper, a novel MAC protocol to resolve the multi-channel hidden terminal problem using a single interface which avoid using the CCC.

Network Construction using IEC 61400-25 Protocol in Wind Power Plants Tae O Kim, Jung Woo Kim and Hong Hee Lee University of Ulsan, Korea In recent years, the wind power plants are widely developing as an alternative energy source. In order to provide a uniform communications basis for the monitoring and control of wind power plants, IEC 61400-25 has been developed. This paper describes a Web service based network construction using communication protocol stack which is included in IEC 61400-25-4. This system is necessary to implement remote control systems for wind power plants.

Stability and Stabilization of Nonuniform Sampling Systems using a Matrix Bound of a Matrix Exponential Young Soo Suh University of Ulsan, Korea This paper is concerned with stability and stabilization of networked control systems, where sampling intervals are time-varying. A nonuniform sampling system is modeled as a time-varying discrete time system. With the assumption that the sampling time variation bounds are known, the stability condition is derived in the form of linear matrix inequalities. Compared with previous results, a less conservative stability condition is derived using a matrix bound of a matrix exponential

Robot Visual Servo through Trajectory Estimation of a Moving Object using Kalman Filter Min-Soo Kim, Ji-Hoon Koh, Ho Quoc Phuong Nguyen, Hee-Jun Kang University of Ulsan, Korea

T2E.4

T2E.3

T2E.2

T2E.1

T2D.6

T2D.5

- 51 -

In this paper, a robot visual servo control algorithm is proposed by combining the conventional image based robot visual servoing algorithm with a trajectory estimation algorithm of a moving object using Kalman filter. The erroneous image information of a moving object due to the imprecise camera characteristics is compensated by applying Kalman filter to the process model of a moving object. The robot visual servo control algorithm is simulated, implemented and discussed with a Samsung FARA AT-2 robot and a MV50 Camera for its effectiveness, in both cases of with/without a trajectory estimation algorithm of a moving object using Kalman filter.

Implementation of Induction Motor Control System Using Matrix Converter based on CAN Network and Dual-Port RAM Hong-Hee Lee, Hoang M. Nguyen University of Ulsan, Korea This paper presents induction motor control system operation using matrix converter based on the controller area network (CAN). The hardware control system is designed with dual microcontrollers which communicate to each other by a dual-port RAM. The advantages of matrix converter are utilized with the CAN network on the field oriented control method of induction motor. The performances of the motor control fully guarantee the system stability and the successful data communication by network. The experimental results are given on 5Hp induction motor to verify the effectiveness and feasibility of the control system using network.

Device Integration Approach to OPC UA-based Process Automation Systems with FDT/DTM and EDDL Vu Van Tan, Dae-Seung Yoo and Myeong-Jae Yi University of Ulsan, Korea Manufacturers and distributors alike are seeking better ways to more efficiently manage the assets of their operators. Advances in communication technologies and standards are now making them easier and more cost justifiable to deliver information from measurement instrumentation as well as manage these instrumentation assets more efficiently. Today's technologies such as Electronic Device Description Language (EDDL) and Field Device Tool (FDT) are available for device integration. This paper introduces a flexible device integration approach to achieving the advantages of both such technologies for the device management to new OPC Unified Architecture (UA)-based process automation systems. This approach is suitable not only for the process automation industry, but also for the factory automation industry. Visibility of and access to field device information through the OPC standards, FDT, and EDDL contribute to efforts to improve the life cycle management of process plants and associated distribution operations.

▶ 16:30-18:30 <T3A> Intelligent Computing in Bioinformatics RM : Crystal Ballroom #1 Dynamic Identification and Visualization of Gene Regulatory Networks from Time-Series Gene

Expression Profiles Yu Chen and Kyungsook Han Inha University, Korea Recent improvements in high-throughput proteomics technology have produced a large amount of time-series gene expression data. The data provide a good resource to uncover causal gene-gene or gene-phenotype relationships and to characterize the dynamic properties of the underlying molecular networks for various biological processes. Several methods have been developed for identifying the molecular mechanisms of regulation of genes from the data, but many of the methods consider static gene expression profiles only. This paper presents a new method for identifying gene regulations from the time-series gene expression data and for visualizing the gene regulations as dynamic gene regulatory networks. The method has been implemented as a program called DRN Builder (Dynamic Regulatory Network Builder; http://wilab.inha.ac.kr/drnbuilder/) and successfully tested on actual gene expression profiles. DRN Builder will be useful for generating potential gene regulatory networks from a large amount of time-series gene expression data and for analyzing the identified networks.

Predicting RNA-Binding Sites in Proteins Using the Interaction Propensity of Amino Acid Triplets Mi-Ran Yun, Yanga Byun, and Kyungsook Han Inha University, Korea In protein-RNA interactions, amino acids often exhibit different preferences for its RNA partners with different neighbor amino acids. Hence, the interaction propensity of an amino acid can be better assessed by considering neighbors of the amino acid than examining the amino acid alone. In this study, we computed the interaction propensity of three consecutive amino acids (called amino acid triplet or triple amino acids) instead of individual amino acids from the rigorous analysis of the recent structure data of protein-RNA complexes. Most amino acid triplets have no interaction, and only 1.1% of them interact with a nucleotide. We used the interaction propensity to predict RNA-binding sites in protein sequences with a support vector machine (SVM) classifier, and observed that the interaction propensities of amino acid triplets are more effective than other biochemical properties of amino acids for predicting RNA-binding sites in proteins. Experimental results with nonredundant 134 protein sequences showed that the SVM classifier achieved a sensitivity of 77%, specificity of 76%, net prediction of 77%, and correlation coefficient of 0.535. Comparison of the SVM classifier with RNABindR and BindN demonstrated that it outperforms the other two methods in the net prediction and correlation coefficient. Our SVM classifier can also be used to predict protein-binding nucleotides in RNA sequences.

T3A.2

T3A.1

T2E.6

T2E.5

- 52 -

On-line Signature Verification Based on Spatio-Temporal Correlation Hao-Ran Deng and Yun-Hong Wang Beihang University, China In this paper, a novel signature verification algorithm based on spatio-temporal correlation and an improved template selection method are proposed. The proposed algorithm computes the spatio-temporal matrices of signatures and considers them as images, then applies image processing methods to obtain features. The correlation matrices will enlarge the variations of the signatures, so an improved template selection method considering the intra-class variation of enrollment signatures is also adopted. The algorithms are validated on the SVC 2004 database and inspiring results are obtained.

Contribution Degree’s Application in the Research of Elements of TCM Syndromes Rongyao Zheng, Guangcheng Xi and Jing Chen Chinese Academy of Sciences, China Using unsupervised algorithms to cluster for diagnosis information data is a mainstream and difficult area of TCM clinical research, and the optimal symptoms’ number of the syndrome is even more difficult to gain. However, there is no relevant and effective research on it yet. An unsupervised clustering algorithm is proposed based on the concepts of complex system entropy and contribution degree in this work. The algorithm is based on the familiar unsupervised complex system entropy cluster algorithm, simultaneously, it introduces contribution degree to self-adaptively select the symptoms’ number. This work carried out three clinical epidemiology surveys about depression, chronic renal failure and chronic hepatitis b, and obtained 1787 cases, each of which has measurements for 76 symptoms. The algorithm discovers 9 patterns, and 6 of them fit the syndrome in clinic. Therefore, we conclude that the algorithm provides an effective solution to discover syndrome from symptoms.

Gender Recognition from Gait using Radon Transform and Relevant Component Analysis Lei Chen1, Yunhong Wang1, Yiding Wang2, De Zhang1 1Beihang University, China, 2North China University of Technology, China In this paper, a new method for gender recognition via gait silhouettes is proposed. In the feature extraction process, Radon transform on all the 180 angle degrees is applied to every silhouette to construct gait templates and the initial phase of each silhouette in an entire gait cycle is also associated to the templates representing dynamic information of walking. Then the Relevant Component Analysis (RCA) algorithm is employed on the radontransformed templates to get a maximum likelihood estimation of the within class covariance matrix. At last, the Mahalanobis distances are calculated to measure gender dissimilarity in recognition. The Nearest Neighbor (NN) classifier is adopted to determine whether a sample in the Probe Set is male or female. Experimental results in comparison to state-of-the-art methods show considerable improvement in recognition performance of our proposed algorithm.

A Biologically Plausible Winner-Takes-All Architecture Sebastian Handrich, Andreas Herzog, Andreas Wolf, and Christoph S. Herrmann Otto-von-Guericke-University Magdeburg, Germany Winner-takes-all (WTA) is an important mechanism in artificial and biological neural networks. We present a biologically plausible two layer WTA architecture with biologically plausible spiking neuron model and conductance based synapses. The excitatory neurons in the WTA layer receive spiking signals from an input layer and can inhibit other excitatory WTA neurons via related inhibitory neurons. The connections from the input layer to WTA layer can be trained by Spike-Time-Dependent Plasticity to discriminate between different classes of input patters. The overall input of the WTA neurons are controlled by synaptic scaling.

▶ 16:30-18:30 <T3C> Applications of Intelligent Computing in Information Assurance & Security / Intelligent Agent and Web Applications RM : Crystal Ballroom #3

Modified AES using Chaotic Key Generator for Satellite Imagery Encryption Fahad Bin Muhaya, Muhammad Usama, Muhammad Khurram Khan King Saud University, Kingdom of Saudi Arabia In this paper, we propose a new modified version of Advanced Encryption Standard (AES) using chaotic key generator for satellite imagery security. We analyze and examine the Modified AES and chaotic key generator to enhance the key space and sensitivity, performance, and security level for reducing the risk of different attacks. The chaotic key generator utilizes multiple chaotic maps named as Logistic, Henon, Tent, Cubic, Sine and Chebyshev. The proposed algorithm presents numerous interesting and attractive features, such as a high level of security, large enough key-space with improved key sensitivity, pixel distributing uniformity and an acceptable encryption and decryption speed. The presented algorithm is ideal for real-time applications to deal with redundant, bulky, complex and stubborn satellite imagery.

Experimental Comparison among 3D Innovative Face Recognition Frameworks Vitoantonio Bevilacqua1,2, Giuseppe Mastronardi1,2, Raffaele Piarulli1, Vito Santarcangelo1,2, Rocco Scaramuzzi1, Pasquale Zaccaglino1 1Polytechnic of Bari, Italy, 2Spin-Off of Polytechnic of Bari, Italy In this paper, starting to the previous work on 3D face recognition, is presented an optimization of the search of the points ALS and ALD of the nose and a new graph approach for the recognition base on several new points. Experiments are performed on a dataset (44 3D faces) acquired by a 3D laser camera at eBIS lab with pose and expression variations. The face recognition performance on the

T3C.2

T3C.1

T3A.6

T3A.5

T3A.4

T3A.3

- 53 -

44 faces considered reach the 100% percentage.

Buyer Coalitions with Bundles of Items by Using Genetic Algorithm Laor Boongasame and Anon Sukstrienwong Bangkok University, Thailand There are several existing buyer coalition schemes. These schemes do not consider forming a buyer coalition with bundles of items. There is only one scheme that forms a buyer coalition with bundles of items. Nevertheless, the scheme suffers from computational complexity. In this paper, we have applied genetic algorithms (GA) to form buyer coalitions with bundles of items, called the GroupPackageString scheme. The fitness function is defined by total discount of the buyer coalitions over the GA to measures the GroupPackageString scheme. The coalition results show that the total discount of any coalition in this scheme is higher than those in the GroupBuyPackage scheme.

A Representation Methodology for Performance Specifications in UML Domain S. Distefano, A. Puliafito and M. Scarpa Universitµa di Messina, Italy Performance related problems play a key role in the Software Development Process (SDP). In particular an early integration of performance specifications in the SDP has been recognized during last years as an effective approach to speed up the production of high quality and reliable software. Some time ago, we defined and implemented a methodology for automatically evaluating performance aspects in a UML Software Architecture (SA) model. To pursue this goal it is needed to map the starting UML model into a performance domain afterwards analyzed. The performance indices are inserted in the UML model exploiting the OMG Profile for Schedulability, Performance and Time Specification standard. However, to really automate the process, it becomes mandatory to organize and formalize the representation of the UML SA project by fixing semantic rules. The goal of this paper is the formalization of the SDP representation, characterizing the syntax and the semantics through which specifying performance requirements and behaviors into UML models.

Web-based Unified-Directory Service for Social Networking Services and Ubiquitous Sensor Network Services Yung Bok Kim Sejong University, Korea For integrated social networking and sensor networking services, a unified approach using a unified directory service based on web-based directory was studied. As a convenient and usable mobile web service for unified social/sensor networking service, the multi-lingual single-character domain names as mobile user interface for accessing the metadata of social/sensor information in unified directory are convenient, efficient and consistent. For searching for social/sensor information as well as registering metadata of sensor/social information, we introduce the web-based unified-directory service with the requirements, performance metrics for QoS, resource utilization and real-time estimation of the performance metrics.

An Approach to Automated User Interest Matching in Online Classified Advertising Systems Valeriya Gribova, Pavel Kachanov Institute of Automation and Control Processes, Russia The paper presents an approach to automated user interest matching in online classified advertising systems which is based on the analysis of the structure and semantics of classified ads. A classified advertisement structure and classified ad types along with the examples are described in the paper.

▶ 16:30-18:30 <T3D> New Particle Swarm Optimization and its Applications / Advances in Intelligent Information Processing RM : Charlotte Room #1

Symbiotic Multi-swarm PSO for Portfolio Optimization Ben Niu1, Bing Xue1, Li Li1 and Yujuan Chai2 1Shenzhen University, China, 2McMaster University, Canada This paper presents a novel symbiotic multi-swarm particle swarm optimization (SMPSO) based on our previous proposed multi-swarm cooperative particle swarm optimization. In SMPSO, the population is divided into several identical sub-swarms and a center communication strategy is used to transfer the information among all the sub-swarms. The information sharing among all the sub-swarms can help the proposed algorithm avoid be trapped into local minima as well as improve its convergence rate. SMPSO is then applied to portfolio optimization problem. To demonstrate the efficiency of the proposed SMPSO algorithm, an improved Markowitz portfolio optimization model including two of the most important limitations are adopted. Experimental results show that SMPSO is promising for this class of problems

A Novel Particle Swarm Optimization with Non-linear Inertia Weight Based on Tangent Function Li Li1, Bing Xue1, Ben Niu1, Lijing Tan2 and Jixian Wang3 1Shenzhen University, China, 2Measurement Specialties Inc, China, 3Anhui Agricultural University, China Inertia weight is a most important parameter of particle swarm optimization (PSO), which can keep a right balance between the

T3D.2

T3D.1

T3C.6

T3C.5

T3C.4

T3C.3

- 54 -

global search and local search. In this paper, a novel PSO with non-linear inertia weight based on the tangent function is provided. The paper also presents the method of determining a control parameter in our proposed method, saving the user from a tedious trial and error based approach to determine it for each specific problem. The performance of the proposed PSO model is amply demonst-rated by applying it for four benchmark problems and comparing it with other three PSO algorithms. From experimental results, it can be concluded that using a non-linear dynamic inertia weight makes the rapidity of convergence rate with higher precision.

An Improved Two-Stage Camera Calibration Method Based on Particle Swarm Optimization Hongwei Gao1, Ben Niu2, Yang Yu1, Liang Chen1 1Shenyang Ligong University, China, 2Shenzhen University, China According to the calibration of binocular vision, an improved twostage camera calibration method involved with multi-distortion coefficients is introduced in this paper. At the first stage, the 3D points’ coordinate are calculated by the imitated direct linear transformation (DLT) triangulation based on distortion compensation. And at the second stage, particle swarm optimization (PSO) is selected to determine two cameras’ parameters. In this way the parameters of the two cameras can be tuned simultaneously. In order to assist estimating the performance of the proposed method, a new cost function is designed. Simulation and experiment are made under the same calibration data sets. The performance of PSO used to tune the parameters is also compared to that of GA. The experiment results show that the strategy of taking the 3D reconstruction errors as object function is feasible and PSO is the best choice for camera parameters’ optimization.

Study on Multi-Depots Vehicle Scheduling Problem and Its Two-Phase Particle Swarm Optimization Suxin Wang1,1, Leizhen Wang1, Huilin Yuan1, Meng Ge1, Ben Niu2, Weihong Pang1, Yuchuan Liu1 1Northeastern University at Qinhuangdao, China, 2Shenzhen University, China To get global solution in multi-depots vehicle scheduling problem (MDVSP), MDVSP models are established. Two-phase particle swarm optimization (TPPSO) is established to solve MDVSP. The optimization course are as follow: first phase, set up goods number dimension particle position vector, vector’s every column corresponds to goods, vector elements are random vehicle serial number, thus we can assign goods to vehicles. Second phase, particle position matrix is set up, matrix’s column number equal to vehicle freight goods number, every column corresponds to a goods, and matrix has two row, the first row correspond to goods start depot, second row correspond to end depot, matrix elements are random number between 0 and 1, matrix elements are sort ascending according to sort rules, we can get single vehicle route. Then evaluate and filtrate particles by optimization aim, circulate until meet terminate qualification. TPPSO can assign all freights to all vehicles and easy to get optimized solution.

Application of RBF Network Based on Immune Algorithm in Human Speaker Recognition Yan Zhou1,2 and Xinming Yu1,2 1Jiangsu Research & Development Centre for Modern Enterprise Information Software Engineering, China, 2Suzhou Vocational University, China When using traditional clustering algorithms based on Radial Basis Function(RBF)network to recognize human speakers, it is hard to decide the numbers and locations of the cluster centers. To overcome these shortcomings, this paper proposes an RBF network based on artificial immune mechanism for human speaker recognition. The artificial immune mechanism can adaptively compute the number and initial locations of the centers in the hidden layer of the RBF network based on the audio sample data set. Experimental tests show that the system has a fast learning speed for network weights. The system is very good at searching for global optimum. It has a high recognition rate and is a new practical method for human speaker recognition.

Adaptive Immune Response Network Model Tao Liu1,2, Li Zhang2 and Binbin Shi2 1JiangSu Province Support Software Engineering R&D Center for Modern Information Technology Application in Enterprise, China, 2Suzhou Vocational University, China Artificial immune system (AIS) and its applications have become a search hotspot in recent years. According to the theory of immune response and immune network, the Adaptive Immune Response Network (AIRN) is presented in this paper. In AIRN, the expression and procedure are given. The AIRN is applied to the clustering analysis, and this cluster method can receive data modes more quickly. The testing results show that the AIRN has better performance in data partition and pattern recognition than the clustering algorithm based on GA and the others.

▶ 16:30-18:30 <T3E> Network-based Intelligent Technologies / Biology and Drug Design / Signal Processing for Interactive Brain-Machine-Interfacing RM : Charlotte Room #2

A SOA-based Framework for Building Monitoring and Control Software Systems Vu Van Tan, Dae-Seung Yoo and Myeong-Jae Yi University of Ulsan, Korea This paper proposes a SOA-based framework for building complex monitoring and control software systems used in modern process and factory automation today where production processes will span over all types of systems. This framework is developed with utilization of the OPC Unified Architecture (UA) specifications and Object-Oriented Design (OOD). It provides generic components upon which sophisticated production processes can be modeled. Solutions to security of remote invocations are implemented to make this framework capable and reliable. The preliminary experiment results are provided, and the comparison

T3E.1

T3D.6

T3D.5

T3D.4

T3D.3

- 55 -

with existing approaches and the discussion are also presented. They demonstrate that the proposed framework is feasible for applying to web service-based monitoring and control system applications.

Parallel genetic algorithms for crystal structure prediction: Successes and failures in predicting bicalutamide polymorphs Marta B. Ferraro 1, Anita M. Orendt2 and Julio C. Facelli2 1Universidad de Buenos Aires, Argentina, 2University of Utah, USA This paper describes the application of our distributed computing framework for crystal structure prediction, Modified Genetic Algori-thms for Crystal and Cluster Prediction (MGAC), to predict the crystal structure of the two known polymorphs of bicalutamide. The paper describes our success in finding the lower energy polymorph and the difficulties encountered in finding the second one. The results show that genetic algorithms are very effective in finding low energy crystal conformations, but unfortunately many of them are not plausible due to spurious effects introduced by the energy potential function used in the selection process. We propose to solve this by using a multi objective optimization GA approach, adding the unit cell volume as a second optimization target.

EMD Based Power Spectral Pattern Analysis for Quasi-Brain-Death EEG Qiwei Shi1, Juhong Yang1, Jianting Cao1,3,4, Toshihisa Tanaka2,3, Tomasz M. Rutkowski3, Rubin Wang4 and Huili Zhu5 1Saitama Institute of Technology, Japan, 2Tokyo University of Agriculture and Technology, Japan, 3Brain Science Institute, Japan 4East China University of Science and Technology, China, 5Huadong Hospital Affiliated to Fudan University, China Evaluating the significance differences between the group of comatose patients and the group of brain death is important in the determination of brain death. This paper presents the power spectral pattern analysis for Quasi-Brain-Death EEG based on Empirical Mode Decomposition (EMD).We first decompose a single-channel recorded EEG data into a number of components with different frequencies. We then focus on the components which are related to the brain activities. Since the power of spontaneous activities in the brain is usually higher than that of non-activity components. Therefore, we can evaluate the power spectral patterns between comatose patients and quasi-brain-deaths. Our experimental results illustrate the effectiveness of proposed method.

Proposal of Ride Comfort Evaluaiton Method Using the EEG Hironobu Fukai1, Yohei Tomita1, Yasue Mitsukura1, Hirokazu Watai2, Katsumi Tashiro2 and Kazutomo Murakami2 1Tokyo University of Agriculture and Technology, Japan, 2Bridgestone Corporation, Japan In this study, we propose the ride comfort evaluation method by using the electroencephalography (EEG). Recently, the subjective evaluation method that is questionnaire survey etc. is used for introducing the human sensibility. However, it is not established because of the difficulty of obtaining the human sensibility. Moreover, the objective evaluation method is hoped because the subjective evaluation method has ambiguous criterion by individual, and difference of sensitivity. Therefore, we propose the evaluation method by using the EEG that objective evaluation is possible. In this study, we investigate a ride comfort of car driving. We use the general car and we investigate the ride comfort according to the difference of the tire. The EEG is measured in driving condition. Moreover, the ride comfort subjective evaluation is surveyed by semantic differential method (SD method). The feature of the EEG during the driving and feature of the subjective evaluation is extracted by the factor analysis (FA). From the result, the EEG feature and subjective evaluation feature has correlation. Thus, the effectiveness of the proposed method as an objective evaluation method was shown.

Interactive Components Extraction from fEEG and fNIRS for Affective Brain Machine Interfacing Paradigms Tomasz M. Rutkowski1, Toshihisa Tanaka1,2, Andrzej Cichocki1,Donna Erickson3 and Danilo P. Mandic4 1RIKEN Brain Science Institute, Japan, 2Tokyo University of Agriculture and Technology, Japan, 3Showa Music University, Japan, 4Imperial College London, UK The paper presents an approach to combine human interactive communication modeling principles in application to a novel interaction paradigm designed for brain-computer/machine-interfacing technologies. As a test platform for such an intelligent human-computer communication application an emotional stimuli paradigm was chosen of moving faces and speech. From information processing point of view several challenges with multimodal signal conditioning and stimuli dynamic response extraction in time frequency domain are addressed. Emotions play an important role in human daily life and human-to-human communication. This is why involvement of affective stimuli principles to human-machine-communication utilizing multichannel neurophysiological and periphery physiological signals is discussed.

T3E.5

T3E.4

T3E.3

T3E.2

- 56 -

▶ 10:00-12:00 <F1A> Intelligent Computing in Computer Vision RM : Crystal Ballroom #1 Vehicle Detection Algorithm Using Hypothesis Generation and Verification

Quoc Bao Truong, Byung Ryong Lee University of Ulsan, Korea In this paper, we present a two-stage vision-based approach to detect front and rear vehicle views in road scene images using eigenspace and a support vector machine for classification. The first stage is hypothesis generation (HG), in which potential vehicles are hypothesized. During the hypothesis generation step, we use a vertical, horizontal edge map to create potential regions where vehicles may be present. In the second stage verification (HV) step, all hypotheses are verified by using a Principle Component Analysis (PCA) for feature extraction and a Support Vector Machine (SVM) for classification, which is robust for both front and rear vehicle view detection problems. Our methods have been tested on different real road images and show very good performance.

A Novel Method using Contourlet to Extract Features For Iris Recognition System Amir Azizi1, Hamid Reza Pourreza2 1Islamic Azad University Mashhad Branch, Iran 2Ferdowsi University of Mashhad, Iran In different areas of Biometrics, recognition by iris images in nowadays has been taken into consideration by researchers as one of the common methods of identification like passwords, credit cards or keys. Iris recognition a novel biometric technology has great advantages such as variability, stability and security. Although the area of the iris is small it has enormous pattern variability which makes it unique for every one and hence leads to high reliability. In this paper we propose a new feature extraction method for iris recognition based on contourlet transform. Contourlet transform captures the intrinsic geometrical structures of iris image. It decomposes the iris image into a set of directional sub-bands with texture details captured in different orientations at various scales so for reducing the feature vector dimensions we use the method for extract only significant bit and information from normalized iris images. In this method we ignore fragile bits. At last, the feature vector is created by using Cooccurrence matrix properties. For analyzing the desired performance of our proposed method, we use the CASIA dataset, which is comprised of 108 classes with 7 images in each class and each class represented a person. And finally we use SVM and KNN classifier for approximating the amount of people identification in our proposed system. Experimental results show that the proposed increase the classification accuracy and also the iris feature vector length is much smaller versus the other methods.

Vehicle License Plate Detection Algorithm Based on Color Space and Geometrical Properties Kaushik Deb1, Vasily V. Gubarevz1 and Kang-Hyun Jo2 1Novosibirsk State Technical University, Russia, 2University of Ulsan, Korea In this paper, an algorithm for vehicle license plate detection (VLPD) is proposed, to select automatically statistical threshold value in HSI color space. The proposed VLPD algorithm consists of two main stages. Initially, HSI color space is adopted for detecting candidate regions. According to different colored LP, these candidate regions may include LP regions; geometrical properties of LP are then used for classification. The proposed method is able to deal with candidate regions under independent orientation and scale of the plate. Finally, the decomposition of candidate regions contains predetermined LP alphanumeric characters by using position in the histogram to verify and detect vehicle license plate (VLP) region. In experiment more than 150 images were used, they were taken from the variety of conditions such as complex scenes, illumination changing, distances and varied weather etc. Under these conditions, success of LP detection has reached to more than 94%.

Spatial Relation Model for Object Recognition in Human-Robot Interaction Lu Cao, Yoshinori Kobayashi, and Yoshinori Kuno Saitama University, Japan Carrying out user commands entails target object detection for service robots. When the robot system suffers from a limited object detection capability, effective communication between the user and the robot facilitates the reference resolution. We aim to develop a service robot, assisting handicapped and elderly people, where most of the user requests are directly or indirectly linked to some objects in the scene. Objects can be described using features like color, shape, size etc. For simple objects on simple backgrounds, theses attributes can be determined with satisfactory results. For complex scenes, position of an object and spatial relation with other objects in the scene, facilitate target object detection. This paper proposes a spatial relation model for the robot to interpret user’s spatial relation descriptions. The robot can detect a target object by asking the user the spatial relationship of the object and some known objects automatically recognized.

Window Extraction Using Geometrical Characteristics of Building Surface Hoang-Hon Trinh, Dae-Nyeon Kim, Suk-Ju Kang, Kang-Hyun Jo University of Ulsan, Korea This paper describes an approach to extract windows by analyzing geometrical characteristics of building surface. Firstly, building surfaces are detected and then wall region is extracted by using hue color of pixel; this step was well described in our previous works.

F1A.5

F1A.4

F1A.3

F1A.2

F1A.1

Friday, September 18

- 57 -

The non-wall regions are considered as candidates of other components of building such as windows, doors, columns and so on. To extract the windows, the image of candidates is recovered in rectangular shape. Then the ambiguous candidates which have irregular shape, for example, long and thin or very small are coarsely rejected. The geometrical characteristics such as the center coordinates, area, aspect ratio and the aligned coexistence are used for extracting the windows. The proposed approach has been experimented for a database with 150 building surfaces comprising 1607 windows. We obtained 93:34% extraction rate.

Auto-Surveillance for Object to Bring in/out using Multiple Camera Taeho Kim1, Dong-Wook Seo2, Hyun-Uk Chae1 and Kang-Hyun Jo1 1University of Ulsan, Korea, 2MOTORWEL Corporation, Korea This paper describes an auto-surveillance system which tracks a person who comes in/out an office using multiple camera system. Furthermore it automatically recognize whether the person bring an object in/out. For this purpose, we set three steps. The first step is detecting a person using MBM(Multiple Background Model) and TMB(Temporal Median Background). The second step is calculation of correspondence between persons detected by different view-point cameras in the multiple camera system. We simply calculate the correspondence based on the principal axis and homography. The last step is generating global color model, which includes every local color model organized by GMM(Gaussian Mixture Model) from each camera, of the person. The global color model represented by GMM checks the temporally varied error and detects the object to bring in or out objects. In the experiment, we show the detected human silhouette by background subtraction and the tracking result by correspondence of multiple views. We also show the color segmentation using GMM and the recognition result for detecting objects brought in/out by the tracked person.

▶ 10:00-12:00 <F1B> Intelligent Computing in Signal Processing RM : Crystal Ballroom #2 Automatic Music Transcription Based on Wavelet Transform

Amir Azizi1, Karim Faez2, Amin Rezaeian Delui3, Saeid Rahati1 1Islamic Azad University Mashhad Branch, Iran 2Amir Kabir University of Technology, Iran 3Toos Institute of Higher Education, Iran In this paper, we introduce a method which uses a note model and signal post processing for a musical instrument to make a piece of music .one of the important issues in note transcription is extraction of multiple pitches. Most of the examined methods face error in joint harmonics and frequencies. A good model for note of a specified musical instrument can help us identify a note better. The presented method is based on wavelet transform, onset detection, note model and conformity reduction error algorithm or regression and postprocessing for improved result. The results obtained show that detecting musical notes in a piece played on the guitar is, in comparison with similar methods, of higher detection accuracy and even in the case of noisy sound signals, the results are more acceptable.

Minimum Sum-of-Squares Clustering by DC programming and DCA Le Thi Hoai An1, Pham Dinh Tao2 1University of Paul Verlaine, France, 2National Institute for Applied Sciences, France In this paper, we propose a new approach based on DC (Difference of Convex functions) programming and DCA (DC Algorithm) to perform clustering via minimum sum-ofsquares Euclidean distance. The so called Minimum Sum-of-Squares Clustering (MSSC in short) is first formulated in the form of a hard combinatorial optimization problem. It is afterwards recast as a (continuous) DC program with the help of exact penalty in DC programming. A DCA scheme is then investigated. The related DCA is original and very inexpensive because it amounts to computing, at each iteration, the projection of points onto a simplex and/or onto a ball, that all are given in the explicit form. Numerical results on real word data sets show the efficiency of DCA and its great superiority with respect to K-means, a standard method of clustering.

Synthesis of Bowhead Whale Sound using Modified Spectral Modeling Pranab Kumar Dhar, Sangjin Cho, Jong-Myon Kim University of Ulsan, Korea Spectral modeling synthesis (SMS) considers a sound as a combination of a deterministic plus a stochastic component that makes possible for a synthesized sound to attain all the perceptual characteristics of the original sound. However, sometimes considerable phase variations occur in the deterministic component by using SMS since the addition of different frequency sinusoids in the overlap region causes amplitude distortion. As a result, subtraction between original and deterministic signal in time domain do not provide a good approximation of the residual signal. To overcome this problem, we propose a modified SMS that provides good approximation of the residual signal by calculating the complex residual spectrum in frequency domain. Analysis and simulation results for synthesizing bowhead whale sounds suggest that the proposed method is comparable to the SMS in both time and frequency domain. However, the proposed method outperforms the SMS in better spectrum matching because of the use of original phase information to synthesize the deterministic component as well as good approximation of the residual signal by subtracting the deterministic spectrum from the original spectrum and then utilizing spectral fitting.

Automatic Emphasis Labeling for Emotional Speech by Measuring Prosody Generation Error1 Jun Xu, Lianhong Cai Tsinghua National Laboratory for Information Science and Technology Tsinghua University, China Emotion helps human to express their feelings and intentions clearly. And the emphasis labels of speeches are the key of speech emotion analysis and synthesis. In order to label the emotion emphasis of speech samples from a corpus with only phonetic and

F1B.4

F1B.3

F1B.2

F1B.1

F1A.6

- 58 -

prosodic information, this paper introduces an automatic labeling algorithm by measuring the prosody generation error (PGE) of the result from a statistical synthesizer. Classification and Regression Tree (CART) and Maximum Entropy (ME) modeling are adopted for automatically labeling. Experiment shows that both models are helpful for labeling.

Type-2 Fuzzy Sets Applied to Pattern Matching for the Classification of Cries of Infants under Neurological Risk Karen Santiago-Sánchez, Carlos A. Reyes-García, Pilar Gómez-Gil National Institute of Astrophysics, México Crying is an acoustic event that contains information about the functioning of the central nervous system, and the analysis of the infant´s crying can be a support in the distinguishing diagnosis in cases like asphyxia and hyperbilirrubinemia. The classification of baby cry has been intended by the use of different types of neural networks and other recognition approaches. In this work we present a pattern classification algorithm based on fuzzy logic Type 2 with which the classification of infant cry is realized. Experiments as well as results are also shown.

Real-Time Sound Synthesis of Plucked String Instruments using a Data Parallel Architecture Huynh Van Luong, Sangjin Cho, Jong Myon Kim and Uipil Chong University of Ulsan, Korea Recent advances in physics-based sound synthesis have offered huge potential possibilities for the creation of new musical instruments. Despite that research on physics-based sound synthesis is going on for almost three decades, its higher computational complexity has limited its use in real-time applications. Conventional serial computation is inadequate for handling the physics-based sound synthesis of most instruments. To yield computation time compatible with real-time performance, we introduce a parallel approach to the physics-based sound synthesis. In this paper, with a parallel processing engine we implemented the physical modeling for one of traditional Korean plucked string instruments, called Gayageum, which has 12 silk strings. Analysis and simulation results suggest that our parallel approach has the potential to support the real-time sound synthesis of the Gayageum instrument. Moreover, our parallel approach outperforms today’s DSPs in terms of performance and energy efficiency.

▶ 10:00-12:00 <F1C> Evolutionary Learning & Computational Genomics and Proteomics / Image Processing & Document Retrievals / Data Fusion RM : Crystal Ballroom #3

CAPS Genomic Subtyping on Orthomyxoviridae Sheng-Lung Peng1, Yu-Wei Tsay1, Chich-Sheng Lin2, and Chuan Yi Tang3 1National Dong Hwa University, Taiwan, 2National Chiao Tung University, Taiwan, 3National Tsing Hua University, Taiwan The Orthomyxoviridae is a family of single strained RNA viruses including five genera: Influenza virus A, Influenza virus B, Influenza virus C, Thogotovirus, and Isavirus. Usually, Influenza viruses are identified by antigenic differences in their nucleoprotein and matrix protein. In this paper, we propose an algorithm to determine a set of suitable restriction enzymes for producing recognizable restriction maps on Orthomyxoviridae. Our method is applied to viral strains of highly pathogenic avian influenza (HPAI), containing potentially homozygous, heterozygous, and various genetic variations. In the analysis of CAPS (Cleaved Amplified Polymorphic Sequence) subtyping, our method outperforms the RNA coding of representative and epidemiologically significant human wild-type viruses, including H3N8, H5N1, H5N9, H7N1, H7N7, and H9N2. These isolates are analyzed by CAPS with enzymes AgeI, EciI, KpnI, and XbaI. The HPAI strains show a different RFLP (Restriction Fragment Length Polymorphism) profile by comparing with other low pathogenic avian influenza (LPAI) strains. We provide a rapid, specific, and reproducible identification of the genotypes on Orthomyxoviridae. It permits us to quickly confirm subtypes of Orthomyxoviridae.

Verification of Pathotyping by Quasispecies Model Sheng-Lung Peng and Yu-Wei Tsay National Dong Hwa University, Taiwan Discrimination using genetic diversity provides a significant support in genetic research and applications. Mostly, DNA markers indicate a process of determining the genotype presented at specific locations along the DNA molecule. Some developed DNA marker methods are RFLP, RAPD, AP-PCR, DAF, and AFLP. For these systems, enzymes play an important role. In this paper, we propose a mechanism to verify the enzyme efficacy for pathotyping. A procedure is given to inspect the validation on cleavage pattern by restriction enzymes, adapting the concept of genetic algorithm to quasispecies model { a genetic evolutionary processes of self-replicating macromolecules. The proposed mechanism is applied to viral strains of HPV (Papillomaviridae), including mutated strains from quasispecies model of homozygous, heterozygous, and various genetic variations. In the analysis of full length DNA strain PCR-RFLP subtyping, results showed that if digested patterns of HPV can be discriminated by specific enzyme set from non-high-risk and other papillomavirus, then it is also can be discriminated by the same enzyme set, under the condition of mutated simulation with quasispecies model. In addition, a measure of genetic diversity also evaluates the utility for PCR-RFLP markers in pathotyping, depending on the degree of digestion variation. We provide a specific and valid mechanism of examination on PCR-based pathotyping. Our approach offers a practical and verifiable direction for genomic pathotyping.

A Fuzzy Logic Based Approach to Feedback Reinforcement in Image Retrieval Vincenzo Di Lecce, Alberto Amato Politecnico di Bari, Italy

F1C.3

F1C.2

F1C.1

F1B.6

F1B.5

- 59 -

Nowadays, due to the spread of digital imaging technologies, the design of effective content based image retrieval (CBIR) systems is perceived by the research community as a primary problem. Various techniques such as clustering and relevance feedback were proposed to obtain a certain level of knowledge about a given image database. Often clustering techniques were used to obtain a first level characterization of the image database used to speed up the successive stage of queries. In this work the authors use the knowledge obtained using a fuzzy clustering algorithm to reinforce the user feedback. The system was tested on the Columbia Coil-20 image database and the obtained results seem to be encouraging.

Ontology-based Decision Support for Security Management in Heterogeneous Networks Michal Choras1,2, Rafal Kozik2, Adam Flizikowski1,2, Rafal Renk1,3 and Witold Ho lubowicz1,3 1ITTI Ltd., Poland, 2Institute of Telecommunications, Poland, 3Adam Mickiewicz University, Poland In this paper our original methodology of applying ontologybased logic into decision support system for security management in heterogeneous networks is presented. Such decision support approach is used by the off-network layer of security and resiliency mechanisms developed in the INTERSECTION Project. Decision support application uses knowledge about networks vulnerabilities to support off-network operator to manage and control in-networks components such as probes, intrusion detection systems, Complex Event Processor, Reaction and Remediation. Hereby, both IV O (Intersection Vulnerability Ontology) as well as PIV OT - decision support system based on the vulnerability ontology are presented.

Recovering Facial Intrinsic Images from a Single Input Ming Shao, Yun-Hong Wang Beihang University, China According to Barrow and Tenenbaum’s theory, an image can be decomposed into two images: a reflectance image and an illumination image. This midlevel description of images attracts more and more attentions recently owing to its application in computer vision, i.e. facial image processing and face recognition. However, due to its ill-posed characteristics, this decomposition remains difficult. In this paper, we concentrate on a slightly easier problem: given a simple frontal facial image and a learned near infrared image, could we recover its reflectance image? Experiments show that it is feasible and promising. Based on extensive study on hyperspectral images, skin color model and Quotient Image, we proposed a method to derive reflectance images through division operations. That is to divide visual frontal face images by learned near infrared images which are generated by super-resolution in tensor space. With the operation on grey distribution of frontal facial images, the results after division can represent the reflectance of skin, rarely bearing any illumination information. Experimental results show that our method is reasonable and promising in image synthesis, processing and face recognition.

DepthLimited Crossover in GP for Classifier Evolution Hajira Jabeen, Abdul Rauf Baig National University of Computer and Emerging Sciences, Pakistan Genetic Programming (GP) provides a novel way of classification having key features like transparency, flexibility and versatility. Presence of these features makes GP a powerful tool for classifier evolution. GP suffers from code bloat, which is highly undesirable in case of classifier evolution. In this paper, we have proposed a technique called “DepthLimited crossover”. This technique does not let trees increase in complexity while still maintaining diversity and efficient search during evolution. We have compared performance of traditional GP with DepthLimited crossover on data classification problem and found that DepthLimited crossover technique provides compatible results without expanding the search space beyond initial limits. The proposed technique is efficient in terms of classification accuracy, complexity of population and simplicity of evolved classifiers.

▶ 10:00-12:00 <F1D> Intelligent Control and Automation RM : Charlotte Room #1 Multiobjective Permutation Flow Shop Scheduling using a Memetic Algorithm with an NEH-based Local Search

Tsung-Che Chiang, Hsueh-Chien Cheng and Li-Chen Fu National Taiwan University, Taiwan In this paper we address scheduling of the permutation flow shop with minimization of makespan and total flow time as the objectives. We propose a memetic algorithm (MA) to search for the set of non-dominated solutions (the Pareto optimal solutions). The proposed MA adopts the permutation-based encoding and the fitness assignment mechanism of NSGA-II. The main feature is the introduction of an NEH-based neighborhood function into the local search procedure. We also adjust the size of the neighborhood dynamically during the execution of the MA to strike a balance between exploration and exploitation. Forty public benchmark problem instances are used to compare the performance of our MA with that of twenty-seven existing algorithms. Our MA provides close performance for small-scale instances and much better performance for large-scale instances. It also updates more than 90% of the net set of non-dominated solutions for the large-scale instances.

Intelligent Nonlinear Friction Compensation Using Friction Observer and Backstepping Control Seong Ik Han1, Chan Se Jeong2, Sung Hee Park2, Young Man Jeong2, Chang Don Lee2, Soon Yong Yang2 1Suncheon First College, Korea, 2Ulsan University, Korea In this article, a robust nonlinear friction control strategy is developed using friction observer and recurrent fuzzy neural network. The adaptive dynamic friction observer based on the LuGre friction model is proposed to estimates the friction parameters and a

F1D.2

F1D.1

F1C.6

F1C.5

F1C.4

- 60 -

directly immeasurable friction state variable. A RFNN approximator and reconstructed error compensator is also designed to give additional robustness to the control system due to the presence of the friction model uncertainty. A proposed composite control scheme with basic basckstepping controller is applied to the position tracking control of the servo mechanical system.

Adaptive Control using Neural Network for Command Following of Tilt-Rotor Airplane in -Tilt Angle Mode Jae Hyoung Im, Cheolkeun Ha University of Ulsan, Korea This paper deals with an autonomous flight algorithm design problem for the tilt-rotor airplane under development by Korea Aerospace Research Institute for simulation study. The objective of this paper is to design a guidance and control algorithm to follow the given command precisely. The approach to this objective is that model-based inversion is applied to the highly nonlinear tilt-rotor dynamics at fixed-wing mode (nacelle angle=0 deg), and then the classical controller is designed to satisfy overall system stabilization and precise command following performance. Especially, model uncertainties due to the tilt-rotor model itself and inversion process are adaptively compensated for in a simple neural network for performance robustness. The designed algorithm is evaluated from the nonlinear airplane simulation in fixed-wing mode to analyze the command following performance for given trajectory. The simulation results show that the command following performance is satisfactory and control responses are within control limits without saturation.

Implementation of LED Array Color Temperature Controlled Lighting System using RISC IP Core Cheol-Hong Moon, Woo-Chun Jang Gwangju University, Korea In this article, an LED Array Color Temperature Controlled Lighting System has been implemented using an 8 bit RISC IP Core for the lighting control system, as well as a Color Temperature Controlling IP and Delta-Sigma DAC IP designed to control the system. The light sources are made of an LED Array, and the LEDs are configured to have 10 stick bars, such as 3 chips of white (30EA), daylight (30EA), red (30EA), green (30EA), and blue (30EA) 0.1W SMD. The time information is acquired through a Real Time Clock, and bio rhythm compatible presentation is made through the LED Array Color Temperature control. The temperature control IP and Delta-Sigma DAC IP are interfaced by accessing the SFR address of the 8 bit RISC IP Core. The system is configured so that the Delta-Sigma DAC IP would produce 0V~3.3V analogue signals through a low bandwidth passing filter and control the lighting system through the serial communications with a PC using the serial port.

A Universal Data Access Server for Distributed Data Acquisition and Monitoring Systems Dae-Seung Yoo, Vu Van Tan and Myeong-Jae Yi University of Ulsan, Korea This paper introduces a universal data access (DA) server for modern distributed data acquisition and monitoring systems used in process and factory automation. This system is proposed with utilization of the OPC (Openness, Productivity, and Collaboration) technology and XML to achieving interoperability and platform independence. It allows to easily aggregate a large number of existing OPC DA servers and new OPC XML-DA servers into a unified and °exible system that supports exchange of data among these servers. By using binary data encoding to the SOAP messages, the proposed system has a sufficient good performance. The security consideration is discussed to provide more information to technical-level readers. The comparison of the proposed system with the existing approaches is also presented.

INS/GPS Integration System with DCM based Orientation Measurement Ho Quoc Phuong Nguyen, Hee-Jun Kang, Young-Soo Suh, Young-Shick Ro University of Ulsan, Korea This paper works toward the development and implementation of a INS/GPS integration system for the land vehicle application. A developed INS system is introduced to keep measuring the position/orientation of the vehicle when the vehicle is passed through GPS signal shading area. A new orientation scheme is studied to full fill the measurement states of the integration system. Roll/pitch estimation compensating external acceleration is performed with inertial sensors and yaw angle is obtained with GPS information. And then, the orientation information is supplied to the linearized Kalman filter of error model. This process is shown to improve the performance of the integration system. The field test was performed along a non-flat contour with some dismissals of GPS on it.

▶ 10:00-12:00 <F1E> Intelligent Computing in Robotics RM : Charlotte Room #2 A Robot Visual/Inertial Servoing to an Object with Inertial Sensors

Ho Quoc Phuong Nguyen, Hee-Jun Kang, Young-Soo Suh, Young-Shick Ro University of Ulsan, Korea The paper introduces a robot visual/inertial servoing algorithm when the robot needs to track an object with inertial sensors inside. In this situation, first, inertial Jacobian is newly defined to show the relationship between an angle set velocity vector and angular velocity vector of the robot tip. That is combined with the conventional image Jacobian for the proposed robot servoing algorithm. While four landmarks have been used in the conventional visual servoing algorithm, the proposed algorithm requires only two landmarks with help of the IMU to track a moving object. Simulation and Implementation have been done to verify the feasibility of the proposed methodology.

F1E.1

F1D.6

F1D.5

F1D.4

F1D.3

- 61 -

A Service Framework of Humanoid in Daily Life KangGeon Kim, Ji-Yong Lee, Seungsu Kim, Joongjae Lee, Mun-Ho Jeong, ChangHwan Kim and Bum-Jae You Korea Institute of Science and Technology, Korea This paper presents a service framework of a humanoid robot for the coordinated task execution. To execute given tasks, various sub-systems of the robot need to be coordinated effectively. The goal of our paper is to develop the service framework which makes it possible to execute various tasks in daily life environments. A script is used as a tool for describing tasks to easily regulate actions of the sub-systems while the robot is performing the task. The performance of the presented framework is experimentally demonstrated as follows: A humanoid robot, as the platform of the task execution, recognizes the designated object. The object pose is calculated by performing model-based object tracking using a particle filter with back projection-based sampling. An approach proposed by Kim et al. [1] is used to solve a human-like arm inverse kinematics and then the control system generates smooth trajectories for each joint of the humanoid robot. The results of our implementation show the robot can execute the task efficiently in human workspaces, such as an office or home.

Self-stabilizing Human-like Motion Control Framework for Humanoids Using Neural Oscillators Woosung Yang1, Nak Young Chong2, Syungkwon Ra1, Ji-Hun Bae1, Bum Jae You1 1Korea Institute of Science and Technology, Korea, 2Japan Advanced Institute of Science and Technology, Japan We propose an efficient and powerful alternative for adaptation of human motions to humanoid robots keeping the bipedal stability. For achieving a stable and robust whole body motion of humanoid robots, we design a biologically inspired control framework based on neural oscillators. Entrainments of neural oscillators play a key role to adapt the nervous system to the natural frequency of the interacted environments, which show superior features when coupled with virtual components. The coupled system allows an unstable system to stably move according to environmental changes. Hence the feature of the coupled system can be exploited for sustaining the bipedal stability of humanoid robots. Also based on this, a marionette-type motion conversion method to adapt captured motions to a humanoid robot is developed owing that there are the differences in capabilities of dynamics and kinematics between a robot and a human. Then this paper discuss on how to stably show human motions with a humanoid robot. We verify that a real humanoid robot can successfully sustain the bipedal stability exhibiting captured whole body motions from various simulations and experiments.

Pseudorandom RFID Tag Arrangement for Improved Mobile Robot Localization Sungbok Kim Hankuk University of Foreign Studies, Korea In passive RFID environment, this paper presents a pseudorandom tag arrangement for improved performance of RFID based mobile robot localization. It is assumed that a mobile robot travels along a series of linear segments, each at a constant velocity, and the number of tags sensed at one time is at most one. First, using spatial and temporal information during tag traversing, a simple but effective mobile robot localization method is developed. Second, four repetitive tag arrangements, including square, parallelogram, tilted square, and equilateral triangle, are examined. For each tag arrangement, the difficulty in tag installation and the problem of tag invisibility are discussed and compared. Third, inspired from the Sudoku puzzle, a pseudorandom tag arrangement is proposed, which is superior to conventional deterministic tag arrangement in terms of tag invisibility and tag installation.

Dynamic Model Identification of 2-Axes PAM Robot Arm Using Neural MIMO NARX Model Kyoung Kwan Ahn1, Ho Pham Huy Anh2 1University of Ulsan, Korea, 2Ho Chi Minh City University of Technology, Viet Nam In this paper, a novel Forward Dynamic MIMO Neural NARX model is used for simultaneously modeling and identifying both joints of the 2-axes PAM robot arm’s forward dynamic model. The contact force variation and highly nonlinear cross effect of both links of the 2-axes PAM robot arm are modeled thoroughly through a Forward Neural MIMO NARX Model-based identification process using experiment input-output training data. The results show that the novel Forward Dynamic Neural MIMO NARX Model trained by Back Propagation learning algorithm yields outstanding performance and perfect accuracy.

▶ 14:15-16:15 <F2A> Intelligent Computing in Computer Vision RM : Crystal Ballroom #1 Object Analysis for Outdoor Environment Perception using Multiple Features

Dae-Nyeon Kim, Hoang-Hon Trinh and Kang-Hyun Jo University of Ulsan, Korea This paper describes the method to know objects for autonomous robot navigation in an unknown outdoor environment. The method segments the objects from an image taken by moving robot on outdoor environment. In the beginning object segmentation, this uses multiple features to obtain the objects of segmented region. Multiple features are color, context information, line segments, edge, Hue Cooccurrence Matrix (HCM), Principal Components (PCs) and Vanishing Points (VPs). The model of the objects for outdoor environment defines their characteristics individually. We segment the region as mixture using the proposed features and methods. Next the stage classifies the object into natural and artificial ones. We detect sky and trees of natural object and detect building of artificial object using the combination of appearance and context information. Then we estimate the dimensions of building. Extensive experiments with the object segmentation and analysis on outdoor environment confirm the validity of the approach.

F2A.1

F1E.5

F1E.4

F1E.3

F1E.2

- 62 -

Building-based Structural Data for Core Functions of Outdoor Robot Hoang-Hon Trinh, Dae-Nyeon Kim, Suk-Ju Kang, Kang-Hyun Jo University of Ulsan, Korea The most important things to realize such an intelligent robot are core functions such as landmark detection, recognition and reconstruction. Since where we have core functions, the robot can propagate other procedures like navigation, mapping, localization, etc. Thus, this paper describes an approach to construct a structural data for core functions of outdoor robot by using geometrical structure of building. Firstly, line segments are detected. Then several processes such as rejecting noises, calculating dominant vanishing points, filtering the edges of building are used to detect the building surfaces. The criteria are created for decision of building detection function. Secondly, for each surface, a generative model including area, wall histogram and a list of local features are computed for the recognition function. Finally, the geometrical features as windows, doors, floors or rooms are estimated for reconstructing the building. The proposed method has been performed with large databases and sound results of all functions.

Appearance Feature based Human Correspondence under Non-overlapping Views Hyun-Uk Chae and Kang-Hyun Jo University of Ulsan Korea In this paper, a method is proposed, to solve correspondence problem under structured space which is installed multiple cameras. The correspondence between different cameras is an important task to use the multiple camera system. For solving this problem, the proposed method is consists of three steps which are detection of moving object, feature extraction and correspondence among different cameras. First step is to detect moving people by background subtraction from multiple background model. The temporal difference is used jointly to remove noise occurred from temporary change. The detected regions are divided using labeling as individual person. The second step is to segment the each person by a criterion with appearance and context information. The segmented regions in a person are estimated as Gaussian mixture model (GMM) for correspondence. The final step is process of correspondence between different cameras. A GMM from a camera is matched with another GMM from other cameras. A ratio of those GMMs is used as a criteria to identify same person. The experiment was performed with the specific scenarios in quantitative results.

A New Low-Cost Eye Tracking and Blink Detection Approach: Extracting Eye Features with Blob Extraction Ibrahim Furkan Ince, Tae-Cheon Yang Kyungsung University, Korea The systems let user track their eye gaze information have been technologically possible for several decades. However, they are still very expensive. They have limited use of eye tracking and blink detection infra-structure. The purpose of this paper is to evaluate cost effects in the sector and explain our new approach in detail which reduces high costs of current systems apparently. This paper introduces an algorithm for fast and sub-pixel precise detection of eye blobs for extracting eye features. The algorithm is based on differential geometry and still exists in OpenCV library as a class. Hence, blobs of arbitrary size that means eye size can be extracted by just adjusting the scale parameter in the class function. In addition, center point and boundary of an eye blob, also are extracted. These describe the specific eye location in the face boundary to run several algorithms to find the eye-ball location with its central coordinates. Several examples on real simple web-cam images illustrate the performance of the proposed algorithm and yield an efficient result on the idea of low-cost eye tracking, blink detection and drowsiness detection system.

Assisted-Care Robot Based on Sociological Interaction Analysis Wenxing Quan, Naoto Ishikawa, Yoshinori Kobayashi and Yoshinori Kuno Saitama University, Japan This paper presents on-going work in developing service robots that provide assisted-care to the elderly in multi-party settings. In typical Japanese day-care facilities, multiple caregivers and visitors are co-present in the same room and any caregiver may provide assistance to any visitor. In order to effectively work in such settings, a robot should behave in such a way that a person who has a request can easily initiate communication with the robot. Based on findings from observations at several day-care facilities, we have developed a robot system that displays availability to multiple persons and then displays recipiency to an individual person who initiates interaction. This paper details this robot system and its experimental evaluation.

A Novel User Created Message Application Service Design for Bidirectional TPEG Sang-Hee Lee and Kang-Hyun Jo University of Ulsan, Korea The T-DMB (Terrestrial-Digital Multimedia Broadcasting) is the national service, currently successful in use in Korea since 2008. Among other services, TPEG (Transport Protocol Experts Group) service has been spotlighted in the aspects of creating earnings. At present, TPEG service is not so popular as it fails to satisfy the user’s demands on various aspects. Thus, the variety of services including bidirectional service is necessary in stage of DMB2.0. In this paper, the limitations of existing TPEG-POI (Point of Interest) application service using the wireless communication network are indicated. To overcome such limitations, we propose a business model for TPEG-UCM (User Created Message) application service which uses individual bidirectional media. The experiment shown in this paper proves the usability and operability of the proposed method, suggesting that the implementation of the proposed method would be overcome a lack of variety and unidirectional of existing TPEG application.

F2A.6

F2A.5

F2A.4

F2A.3

F2A.2

- 63 -

14:15▶ -16:15 <F2C> Intelligent Computing in Image Processing RM : Crystal Ballroom #3 Recent Progress of the Quasientropy Approach to Signal and Image Processing

Yang Chen and Zhimin Zeng Southeast University, China The quasientropy (QE) is a class of infinitely many functions of probabilities that is similar to the Shannon entropy. In this paper, we review the application of the QE approach to independent component analysis (ICA) and chaotic time series analysis. We also report the new progress of the QE approach to textural features extraction in image processing.

HDR Image Generation based on Intensity Clustering and Local Feature Analysis Andrey Vavilin and Kang-Hyun Jo University of Ulsan, Korea This paper describes a cluster-based method for combining differently exposed images to increase dynamic range. Initially an image is decomposed into a set of arbitrary shaped regions. For each region we compute utility function based on amount of presented information and entropy. This function is used to select most appropriate exposure for each region. After exposures are selected, bilateral filtering is applied in order to make interregional transitions smooth. As a result we obtain weighting coefficients for each exposure and pixel. Output image is combined from clusters of input images using weights. Each pixel of output image is calculated as a weighted sum of exposures. The proposed method allows recovering details from overexposed and underexposed parts of image without producing additional noise. Our experiments show effectiveness of the algorithm for the high dynamic range scenes. It requires no information about shutter speed or camera parameters. This method shows robust results even if the exposure difference between input images is 2-stops or higher.

A Framework for Recognition Books on Bookshelves Nguyen-Huu Quoc and Won-Ho Choi University of Ulsan, Korea In this paper, we present a framework to recognize books on bookshelves by reading its title on book spines. The framework consists of control and recognition module. Control module moves camera to suitable positions for image capturing while recognition one processes taken images to know which books are shelved on the shelves. Firstly, images are captured from random position. Secondly, we separate it into book and non-book regions. Then, books in book region are segmented by using line segment and MSAC based dominant vanishing point (DVP). After book verification stage, adaptive Otsu’s thresholding is employed to extract book titles and ready for recognition of next stage. In case recognizing unsuccessfully, we feedback control information to control module to adjust camera location and repeat the above procedure.

A Heuristic Optimization Algorithm for Panoramic Image Generation Problem from Multiple Cameras Megumi Isogai, Nobuo Funabiki, Toru Nakanishi Okayama University, Japan Recently, a panoramic image has been expected in various applications due to the advantage of expressing a wide range of scenes by one image. In this paper, we propose a heuristic optimization algorithm for the panoramic image generation problem from multiple cameras. Our three-stage algorithm composed of the approximate calibration, the detailed calibration, and the image synthesis, trans-forms the images of the side cameras to be fit to the image of the central camera as best as possible. The first stage transforms the coordinates of the side image to resolve the difference caused by view directions and locations of cameras, where the parameters are optimized to match the feature points in overlapped areas among two images. The second stage adjusts both the coordinates and color densities to match the color density of every pixel in overlapped areas. In both stages, the parameters are optimized by a local search method with a Tabu period as a typical heuristic optimization method. Through experiments, we show the effectiveness of our proposal.

A New Method for Iris Recognition Based on Contourlet Transform and Non Linear Approximation Coefficients Amir Azizi1, Hamid Reza Pourreza2 1Islamic Azad University Mashhad Branch, Iran, 2Ferdowsi University of Mashhad, Iran In different methods of Biometrics, recognition by iris images in recent years has been taken into consideration by researchers as one of the common methods of identification like passwords, credit cards or keys. Iris recognition a new biometric technology has great advantages such as variability, stability and security. In this paper we propose a new feature extraction method for iris recognition based on contourlet transform. Contourlet transform captures the intrinsic geometrical structures of iris image. It decomposes the iris image into a set of directional sub-bands with texture details captured in different orientations at various scales so for reducing the feature vector dimensions we use the method for extract only significant bit and information from normalized iris images. In this method we ignore fragile bits. At last the feature vector is approximated by non linear approximation coefficient .Experimental results show that the proposed method reduces processing time and increase the classification accuracy and outperforms the wavelet based method.

F2C.5

F2C.4

F2C.3

F2C.2

F2C.1

- 64 -

A Unified Direct Approach to Image Registration and Object Recognition with a Hybrid Evolutionary Algorithm Igor V. Maslov1, Izidor Gertner2 1Polytechnical University, Russia, 2The City College of New York, USA The paper proposes a unified direct approach to a number of problems arising in image processing. In particular, the areas of image registration, and object or pattern recognition are addressed when the images of interest display significant geometric distortion due to some physical or geometrical conditions. The proposed method performs a direct multi-objective search in image response space for an optimal piece-wise affine transformation of the images using a hybrid evolutionary algorithm. In its most general form, the entire algorithm works in two relatively independent passes. First, the global search attempts to find the optimal solution for the principal affine transformation. During the second pass, the correction procedure seeks for the optimal piece-wise approximation of the actual image transformation using the result of the first pass as the initial approximation.

▶ 14:15-16:15 <F2D> Intelligent Computing in Pattern Recognition / Intelligent Sensor Networks RM : Charlotte Room #1

Two-Dimensional Heteroscedastic Discriminant Analysis for Facial Gender Classification Jun-Ying Gan, Si-Bin He, Zi-Lu Ying, Lin-Bo Cai Wuyi University, China In this paper, a novel discriminant analysis named two-dimensional Heteroscedastic Discriminant Analysis (2DHDA) is presented, and used for gender classification. In 2DHDA, equal within-class covariance constraint is removed. Firstly, the criterion of 2DHDA is defined according to that of 2DLDA. Secondly, the criterion of 2DHDA, log and rearranging terms are taken, and then the optimal projection matrix is solved by gradient descent algorithm. Thirdly, face images are projected onto the optimal projection matrix, thus the 2DHDA features are extracted. Finally, Nearest Neighbor classifier is selected to perform gender classification. Experimental results show that higher recognition rate is obtained by way of 2DHDA compared with 2DLDA and HDA.

Shot Retrival Based on Fuzzy Evolutionary aiNet and Hybrid Features Xianhui Li, Yongzhao Zhan, Jia Ke Jiangsu University, China In this paper, a novel method for shot retrieval is proposed which based on fuzzy evolutionary aiNet and hybrid features. At first, fuzzy evolutionary aiNet algorithm is utilized to extract key-frame in video-sequence. Then, take the extracted key-frame in the same shot as an ensemble and mapped them to high dimension space by nonlinear mapping, supposed that the frame-ensemble at high dimension space obey Gaussian distribution, then we can get shot similarity by computing the probabilistic distance of frame-ensemble. At last, the similarity measures of two frame-ensemble are combined by weighting features mode. Be applied to the field of shot similarity measure, it realize the effective shot retrival. Theoretical analysis and experimental results show the validity of the proposed method.

acial Expression Recognition with Local Binary Pattern and Laplacian Eigenmaps Zilu Ying1,2, Linbo Cai1, Junying Gan1, Sibin He1 1Wuyi University, China, 2Beihang University, China A new approach to facial expression recognition is constructed by combining the Local Binary Pattern and Laplacian Eigenmaps. Firstly, each image is transformed by an LBP operator and then divided into 3×5 non-overlapping blocks. The features of facial expression images are formed by concatenating the LBP histogram of each block. Secondly, linear graph embedding framework is used as a platform, and then Laplacian Eigenmaps is developed under this framework and applied for feature dimensionality reduction. Finally, Support Vector Machine is used to classify the seven expressions (anger, disgust, fear, happiness, neutral, sadness and surprise) on JAFFE database. The maximum facial expression recognition rate of the proposed algorithm reaches to 70.48% for person-independent recognition, which is much better than that of LBP+PCA and LBP+LDA algorithms. The experiment results prove that the facial expression recognition with local binary pattern and Laplacian Eigenmaps is an effective and feasible algorithm.

A Coverage and Energy Aware Cluster-Head Selection Algorithm in Wireless Sensor Networks Thao P. Nghiem, Jong Hyun Kim, Sun Ho Lee, and Tae Ho Cho Sungkyunkwan University, Korea The issue of identifying appropriate cluster-heads has recently been the focus of extensive research and development in wireless sensor networks. Unfortunately, cluster-heads are generally chosen either in a random manner or mainly based on nodes' residual energy. Accordingly, there is no guarantee that network coverage is well-preserved while this QoS is vital in target tracking and surveillance applications. In order to enhance both coverage preservation and energy efficiency, we propose a Coverage and Energy Aware Cluster-Head Selection Algorithm which fully considers three critical factors: the node's energy, location and especially coverage cost metric. Simulation results demonstrate that our algorithm cannot only prolong the network lifetime over 11%, but also substantially enlarge network coverage, from the middle phase of the network lifetime, by over 20% compared to the traditional energy-based selection methods in LEACH and HYENAS system.

F2D.4

F2D.3

F2D.2

F2D.1

F2C.6

- 65 -

u-Healthcare Service based on a USN Middleware Platform and Medical Device Communication Framework Yung Bok Kim Sejong University, Korea We developed a middleware platform, i.e. COSMOS (Common System for Middleware of Sensor Network), for several types of sensor networks including the Zigbee wireless sensor network, the CDMA cellular phone-based network, the RFID reader-based network and the IP-USN based on 6LowPAN. Development has been focused on interfaces for application programs as well as on sensor network abstractions for various ubiquitous sensor networks (USN). Standard interfaces were defined between the USN middleware and USN networks as well as application services. We studied several USN services including u-Healthcare to examine important issues about middleware platform for integrating with other standardized communication framework, e.g. a medical device communication framework. We introduce application services and the real-time data analysis for QoS in the u-healthcare service.

Energy Efficient MAC length Determination Method for Statistical En-Route Filtering using Fuzzy Logic Hyeon Myeong Choi, Tae Ho Cho Sungkyunkwan University, Korea In wireless sensor networks (WSNs) individual sensor nodes are subject to security compromises. An adversary can use compromised sensor nodes to inject false reports into the WSN. If undetected, these false reports are forwarded to the base station. Such attacks by compromised sensor nodes can not only result in false alarms but also depletion of the limited amount of energy in battery powered sensor nodes. The statistical en-routing filtering (SEF) scheme can detect and drop false reports during the forwarding process. In SEF, the number of the message authentication codes (MAC length) is important for detecting false reports and saving energy in network. In this paper, we present a fuzzy-based MAC length determination method for SEF. If there are fewer nodes surrounding the occurred event in the field in the network than the MAC length, the node cannot generate a legitimate report in SEF. The fuzzy-based method can prevent this problem and provide energy savings. We evaluated the proposed method’s performance via simulation.

▶ 14:15-16:15 <F2E> Natural Language Processing and Expert Systems / Ensemble Methods RM : Charlotte Room #2

Developing the KMKE Knowledge Management System Based on Design Patterns and Parallel Processing Lien-Fu Lai1, Chao-Chin Wu1, Liang-Tsung Huang2 and Ya-Chin Chang1 1National Changhua University of Education, Taiwan, 2MingDao University, Taiwan KMKE provides a knowledge engineering approach to integrating knowledge management activities (such as knowledge modeling, knowledge verification, knowledge storage and knowledge querying) into a systematic framework. In this paper, we develop the KMKE knowledge management system based on design patterns and parallel processing. First, several design patterns are applied to develop the KMKE system for enhancing its flexibility and extensibility. Making the KMKE system flexible and extensible is useful to deal with continuous changes originated in knowledge. Second, JAVA programs and CLIPS programs are bound to offer the capability of knowledge inference for the KMKE system. Knowledge verification and knowledge querying can then be performed through the execution of CLIPS rules. Finally, we propose the Parallel CLIPS to shorten the execution time of the KMKE system. Since a large amount of knowledge may increase the execution time substantially, parallelizing the execution of CLIPS rules in cluster system could effectively reduce the search space of the CLIPS inference engine.

Analysis of Shipbuilding Fabrication Process with Enterprise Ontology Ji-Hyun Park1, Kyung-Hoon Kim2, Jae-Hak J. Bae1 1University of Ulsan, Korea, 2Hyundai Heavy Industries Co., Ltd., Korea This paper describes the analysis and evaluation of shipbuilding process based on an enterprise ontology. Shipbuilding process is composed of steel fabrication, assembly, erection, launching, sea trials, naming, and delivery. Among them, the fabrication process has been analyzed and evaluated in this study. An enterprise ontology is a cognitive model containing knowledge unique to the enterprise, and enables the representation and sharing of the enterprise’s process knowledge. We have built an enterprise ontology, and represented the shipbuilding process using plug-ins of Protégé. In addition, we have analyzed the current state of the process and dependency among the workflow elements using a Prolog inference engine, and evaluated the shipbuilding process.

Estimate Word Emotions Based on Part of Speech and Positional Information Kazuyuki Matsumoto1 and Fuji Ren1;2 1The University of Tokushima, Japan 2Beijing University of Posts and Telecommunications, China Recently, the studies on emotion recognition technology have been made in the fields of natural language processing, speech signal processing, image data processing and brain wave analysis, with the goal of letting the computer have some idea of ambiguous information such as emotion or sensibility. This paper statistically studies the features of emotional expressions of Japanese and English examining an emotion annotated parallel corpus and proposes a method to estimate emotion of the emotional expressions from utterance. The proposed method estimates the emotion category of the emotional expressions by focusing on the three kinds of

F2E.3

F2E.2

F2E.1

F2D.6

F2D.5

- 66 -

features: part of speech of emotional expression, position of emotional expression and part of speech of the previous/next morpheme of the target emotional expression. The evaluation experiment of the method based on part of the speech features resulted over 90.0% (joy, hate) of accuracy.

Towards a Better Understanding of Random Forests through the Study of Strength and Correlation Simon Bernard, Laurent Heutte and Sébastien Adam Université de Rouen, France In this paper we present a study on the Random Forest (RF) family of ensemble methods. From our point of view, a “classical” RF induction process presents two main drawbacks : (i) the number of trees has to be a priori fixed (ii) trees are independently, thus arbitrarily, added to the ensemble due to the randomization. Hence, this kind of process offers no guarantee that all the trees will well cooperate into the same committee. In this work we thus propose to study the RF mechanisms that explain this cooperation by analysing, for particular subsets of trees called sub-forests, the link between accuracy and properties such as Strength and Correlation. We show that these properties, through the Correlation/Strengh2 ratio, should be taken into account to explain the sub-forest performance.

An Empirical Study of the Convergence of RegionBoost Xinzhu Yang, Bo Yuan, Wenhuang Liu Tsinghua University, China RegionBoost is one of the classical examples of Boosting with dynamic weighting schemes. Apart from its demonstrated superior performance on a variety of classification problems, relatively little effort has been devoted to the detailed analysis of its convergence behavior. This paper presents some results from a preliminary attempt towards understanding the practical convergence behavior of RegionBoost. It is shown that, in some situations, the training error of RegionBoost may not be able to converge consistently as its counterpart AdaBoost and a deep understanding of this phenomenon may greatly contribute to the improvement of RegionBoost.

Binary Sequences with Good Aperiodic Autocorrelations Using Cross-Entropy Method Shaowei Wang1, Jian Wang1, Xiaoyong Ji1, Yuhao Wang2 1Nanjing University, China, 2Nanchang University, China Cross Entropy (CE) has been recently applied to combinatorial optimization problems with promising results. In this short paper a CE based algorithm is presented to search for binary sequences with good aperiodic autocorrelation properties. The algorithm proposed can explore and exploit the solution space efficiently. In most cases, it can frequently find out binary sequences with higher merit factor and lower peak sidelobe level very quickly.

F2E.6

F2E.5

F2E.4

- 67 -

AAuutthhoorr IInnddeexx A. Puliafito T3C.4 A. Qayyum T1E.4 Abdul Rauf Baig F1C.6 Adam Flizikowski F1C.4 Alberto Amato F1C.3 Alessandro Cincotti T1D.3 Alfredo Motta T1D.3 Amin Rezaeian Delui F1B.1 Amir Azizi F1A.2, F1B.1, F2C.5 Andreas Herzog T3A.6 Andreas Wolf T3A.6 Andrey Vavilin F2C.2 Andrzej Cichocki T3E.5 Anita M. Orendt T3E.2 Anon Sukstrienwong T3C.3 Antonio Puliafito T1E.3 Antonio Scarpa T2A.2, T2A.3, T2A.4 Arfan Ghani T1A.2 Arnoosh Salehi T2C.3 Asaduzzaman T2B.5 Aurélien Saint-Réquier T2A.1 Bao Quoc Vo-Nguyen T2B.4 Basanta Joshi T1A.1 Ben Niu T3D.1, T3D.2, T3D.3, T3D.4 Benoît Lelandais T2A.1 Binbin Shi T3D.6 Bing Xue T3D.1, T3D.2 Bo Yuan F2E.5 Bum Jae You F1E.2, F1E.3 Byung Ryong Lee F1A.1 Carlos A. Reyes-García F1B.5 Caroline Petitjean T2A.1 Cesar Amilcar Lopez Bello T1B.4, T1B.6 Chan Se Jeong F1D.2, F1D.3 Chang Don Lee F1D.2, F1D.3 Chang Huiyou T1E.2 ChangHwan Kim F1E.2 Chang-xiong Zhou T2D.5 Chao-Chin Wu F2E.1 Chen-Hao Liu T1A.4 Cheol-Hong Moon F1D.4 Chesner Désir T2A.1 Chich-Sheng Lin F1C.1 Chih-Ming Lai T1A.4 Chin-Yuan Fan T1A.4, T1C.4 Chiung-Hua Huang T1C.4 Chi-Yang Tsai T1C.4 Christoph S. Herrmann T3A.6 Chuan Yi Tang F1C.1 Dae-Nyeon Kim F1A.5, F2A.1, F2A.2 Dae-Seung Yoo F1D.5, T2E.5, T3E.1 Danilo P. Mandic T3E.5 De Zhang T3A.5 Diego Taurino T2A.2, T2A.3, T2A.4 Domenico Soldo T2C.1

Dong-Wook Seo F1A.6 Donna Erickson T3E.5 Dun-wei Gong T1B.1, T1B.2 Dusko Kalenatic T1B.4, T1B.6 E. Domnguez T1A.3 E.J. Palomo T1A.3 Fahad Bin Muhaya T3C.1 Fariba Salehi T2C.3 Fei Han T1A.5, T2D.2 Fei Hao T1C.2 Fengwen Cao T2D.1 Francesco Pappalardo T1D.3 Fuji Ren F2E.3 Giuseppe Angelelli T2A.2 Giuseppe Mastronardi T2A.2, T2A.3, T2A.4, T2A.5, T3C.2 Guangcheng Xi T3A.4 Gyeongdong Baek T1B.3 Hajira Jabeen F1C.6 Hamid Reza Pourreza F1A.2, F2C.5 Hao-Ran Deng T3A.3 Hee-Jun Kang F1E.1, T2E.4 Hiep-Vu Van T2B.3 Hirokazu Watai T3E.4 Hironobu Fukai T3E.4 Ho Pham Huy Anh F1E.5 Ho Quoc Phuong Nguyen F1E.1 Hoang M. Nguyen T2E.4 Hoang-Hon Trinh F1A.5, F2A.1, F2A.2 Ho-Cheol Shin T1E.5 Hong-Hee Lee T2E.2. T2E.5 Hongwei Gao T3D.3 Hsueh-Chien Cheng F1D.1 Huili Zhu T3E.3 Huilin Yuan T3D.4 Huynh Van Luong F1B.6 Hyeon Myeong Choi F2D.6 Hyung Yun Kong T2B.4, T2B.5 Hyungseo Kang T2B.3 Hyun-Uk Chae F1A.6, F2A.3 Ibrahim Furkan Ince F2A.4 Igor V. Maslov F2C.6 Insoo Koo T2B.1, T2B.2, T2B.3, T2B.6 In-Yong Seo T1E.5 Izidor Gertner F2C.6 J. Muñoz T1A.3 Jae-Hak J. Bae F2E.2 Jia Ke F2D.2 Jian Ma T2C.4 Jian Tan T2C.2 Jian Wang F2E.6 Jianping He T1B.5 Jianting Cao T3E.3 Jie Chen T2D.3

Jie Ren T1B.2 Jie Yuan T1B.1, T1B.2 Ji-Hun Bae F1E.3 Ji-Hyun Park F2E.2 Jinfeng Zhang T2D.3 Jing Chen T3A.4 Jixian Wang T3D.2 Jixiang Du T2D.3 Ji-Yong Lee F1E.2 Joan Condell T1A.2 Jong Hyun Kim F2D.4 Jong Myon Kim F1B.6, F1B.3 Joongjae Lee F1E.2 Juan C. Figueroa Garcìa T1B.4, T1B.6 Juhong Yang T3E.3 Julio C. Facelli T3E.2 Jun Xu F1B.4 Jun Zhang T2C.4 Jun-Lin Lin T1A.4 Jun-Ying Gan F2D.1 Junying Gan F2D.3 KangGeon Kim F1E.2 Kang-Hyun Jo F1A.3, F1A.5, F1A.6,

F2A.1, F2A.2, F2A.3, F2A.6, F2C.2 Karen Santiago-Sánchez F1B.5 Karim Faez F1B.1 Katsumi Tashiro T3E.4 Kaushik Deb F1A.3 Kazuo Ohmi T1A.1 Kazutomo Murakami T3E.4 Kazuyuki Matsumoto F2E.3 Keun-Chang Kwak T1C.3 Kyoung Kwan Ahn F1E.5 Kyung-Hoon Kim F2E.2 Kyungsook Han T3A.1, T3A.2 Laor Boongasame T3C.3 Laurent Heutte F2E.4, T2A.1 Le Thi Hoai An F1B.2 Lei Chen T3A.5 Leizhen Wang T3D.4 Leonarda Carnimeo T2A.5 Li Li T3D.1, T3D.2 Li Shang T2D.3 Li Zhang T3D.6 Liam Maguire T1A.2 Liang Chen T3D.3 Liang Shan T2E.1, T2E.2 Liang-Tsung Huang F2E.1 Lianhong Cai F1B.4 Li-Chen Fu F1D.1 Lien-Fu Lai F2E.1 Lijing Tan T3D.2 Lin Yuhua T1E.2 Lin-Bo Cai F2D.1, F2D.3 Ling Wang T1D.1, T1D.2 Lingpo Li T1D.1, T1D.2

- 68 -

Lixin Gao T1B.5 Lu Cao F1A.4 Luc Thiberville T2A.1 Lucia Cariello T2A.5 M. Scarpa T3C.4 M. Yousaf T1E.4 Maaz Rehan T1E.4 Marco Calabrese T2C.1 Marco Cortellino T2A.2, T2A.3, T2A.4 Marco Moschetta T2A.2 Marco Scarpa T1E.3 Mari Matsumura T1D.6 Marianna Notarnicola T2A.4 Marta B. Ferraro T3E.2 Marzio Pennisi T1D.3 Mathieu Salaün T2A.1 Megumi Isogai F2C.4 Meng Ge T3D.4 Mian Muhammad Awais T1C.1 Michal Choras F1C.4 Michele Piccinni T2A.2 Ming Cui T2D.5 Ming Li T1B.2 Ming Shao F1C.5 Mi-Ran Yun T3A.2 Moon-Ghu Park T1E.5 Muhammad Khurram Khan T3C.1 Muhammad Usama T3C.1 Mun-Ho Jeong F1E.2 Myeong-Jae Yi F1D.5, T2E.6, T3E.1 Myung Kyun Kim T2E.1 Nak Young Chong F1E.3 Naoto Ishikawa F2A.5 Naoyuki Tsuruta T1D.6 Nguyen-Huu Quoc F2C.3 Nhan Nguyen Thanh T2B.1, T2B.2 Nobuo Funabiki F2C.4 Ou Liu T2C.4 Pak-Lok Poon T2C.4 Pasquale Zaccaglino T3C.2 Pavel Kachanov T3C.6 Pei-Chann Chang T1A.4, T1C.4 Peng Lin T1C.5, T1D.5 Pham Dinh Tao F1B.2 Pilar Gómez-Gil F1B.5 Pingang Su T2D.6 Pranab Kumar Dhar F1B.3 Qing-Hua Ling T2D.2 Qinghua Zheng T1C.5, T1D.5 QingXiang Wu T1A.2 Qiwei Shi T3E.3 Quoc Bao Truong F1A.1 R.M. Luque T1A.3 Rafal Kozik F1C.4 Rafal Renk F1C.4 Raffaele Piarulli T3C.2 Rocco Scaramuzzi T3C.2 Rong-qing Xu T2D.5

Rongyao Zheng T3A.4 Rubin Wang T3E.3 Ruhai Lei T1C.6 S. A. Malik T1E.4 S. Distefano T3C.4 Saeid Rahati F1B.1 Sakashi Maeda T1D.6 Salvatore Distefano T1E.3 Sang-Hee Lee F2A.6 Sangjin Cho F1B.3, F1B.6 Sanjeeb Prasad Panday T1A.1 Sebastian Handrich T3A.6 Sébastien Adam F2E.4 Seong Ik Han F1D.2, F1D.3 Seong-Jun Kim T1E.5 Seungsu Kim F1E.2 Shafay Shamail T1C.1 Shaowei Wang F2E.6 Sheng-Lung Peng F1C.1, F1C.2 Shengtong Zhong T1C.2 Shi-Guang Ju T1A.5 Shu-fen Liu T2D.5 Si-Bin He F2D.1 Sibin He F2D.3 Simon Bernard F2E.4 Sohail Bhatti T1E.4 Soon Yong Yang F1D.2, F1D.3 Suk-Ju Kang F1A.5, F2A.2 Sun Ho Lee F2D.4 Sung Hee Park F1D.2, F1D.3 Sungbok Kim F1E.4 Sungshin Kim T1B.3 Suxin Wang T3D.4 Syungkwon Ra F1E.3 T. M. McGinnity T1A.2 Tae Ho Cho F2D.4, F2D.6 Tae-Cheon Yang F2A.4 Taeho Jo T1E.1 Taeho Kim F1A.6 Tae-joon Yun T2B.6 Tao Liu T3D.6 Tetsuya Takaishi T2C.5 Thao P. Nghiem F2D.4 Tomasz M. Rutkowski T3E.3, T3E.5 Tong-Yue Gu T1A.5 Toru Nakanishi F2C.4 Toshihisa Tanaka T3E.3, T3E.5 Tsung-Che Chiang F1D.1 Uipil Chong F1B.6 Valeriya Gribova T3C.6 Vasily V. Gubarevz F1A.3 Vincenzo D. Cunsolo T1E.3 Vincenzo Di Lecce F1C.3, T2C.1 Vincenzo Pacelli T2C.6 Vito Santarcangelo T3C.2 Vitoantonio Bevilacqua T2A.2, T2A.3, T2A.4, T2A.5, T3C.2 Vladimir Berikov T1D.4 Vu Van Tan F1D.5, T2E.5, T3E.1

Weihong Pang T3D.4 Weirong Chen T1A.6 Wenhai Chen T1B.5 Wenhuang Liu F2E.5 Wenjun Huai T2D.3 Wenxing Quan F2A.5 Witold Ho lubowicz F1C.4 Won-Ho Choi F2C.3 Woo-Chun Jang F1D.4 Woosung Yang F1E.3 Xiang Tao Fan T2C.2 Xianhui Li F2D.2 Xiao Chang T1C.5, T1D.5 Xiao-yan Sun T1B.1, T1B.2 Xiaoyong Ji F2E.6 Xinming Yu T3D.5 Xinzhu Yang F2E.5 Xuesong Wang T1C.6 Xuexia Zhang T1A.6 Ya-Chin Chang F2E.1 Yan Zhou T3D.5 Yang Chen F2C.1 Yang Guofeng T1E.2 Yang Yu T3D.3 Yanga Byun T3A.2 Yangyang Gu T1C.6 Yasue Mitsukura T3E.4 Ye Xu T1D.1, T1D.2 Yiding Wang T3A.5 Yiming Wang T2D.4 Yiwang Wang T2D.1 Yohei Tomita T3E.4 Yongzhao Zhan F2D.2 Yoshinori Kobayashi F1A.4, F2A.5 Yoshinori Kuno F1A.4, F2A.5 Young Man Jeong F1D.2, F1D.3 Young Soo Suh T2E.3 Young-du Lee T2B.6 Young-Shick Ro F1E.1 Young-Soo Suh F1E.1 Yu Chen T3A.1 Yuchuan Liu T3D.4 Yuhao Wang F2E.6 Yuhu Cheng T1C.6 Yujuan Chai T3D.1 Yung Bok Kim F2D.5, T3C.5 Yun-Hee Han T1C.3 Yun-Hong Wang T3A.3, T3A.5, F1C.5 Yu-Wei Tsay F1C.1, F1C.2 Zeeshan Ali Rana T1C.1 Zhang Liming T1E.2 Zhengyu Xu T2D.6 Zhi-feng Hu T2D.5 Zhimin Zeng F2C.1 Zhiqiang Zhao T2D.4 Zi-Lu Ying F2D.1, F2D.3 Zongxin Wang T2D.6