Automated Parking Lot Attendant SDP ’07 Team Frasier Tom Cleary Matt Regan Bill Ryan Adam Bailin.

27
Automated Parking Automated Parking Lot Attendant Lot Attendant SDP ’07 SDP ’07 Team Frasier Team Frasier Tom Cleary Matt Regan Bill Ryan Adam Bailin
  • date post

    19-Dec-2015
  • Category

    Documents

  • view

    218
  • download

    1

Transcript of Automated Parking Lot Attendant SDP ’07 Team Frasier Tom Cleary Matt Regan Bill Ryan Adam Bailin.

Page 1: Automated Parking Lot Attendant SDP ’07 Team Frasier Tom Cleary Matt Regan Bill Ryan Adam Bailin.

Automated Parking Lot Automated Parking Lot AttendantAttendant

SDP ’07SDP ’07

Team FrasierTeam Frasier

Tom ClearyMatt Regan

Bill RyanAdam Bailin

Page 2: Automated Parking Lot Attendant SDP ’07 Team Frasier Tom Cleary Matt Regan Bill Ryan Adam Bailin.

Current SystemCurrent System•Disorderly•Confusing •Antiquated

Page 3: Automated Parking Lot Attendant SDP ’07 Team Frasier Tom Cleary Matt Regan Bill Ryan Adam Bailin.

Large Parking LotsLarge Parking Lots•The larger the parking lot, the more difficult it is to find a parking space

Page 4: Automated Parking Lot Attendant SDP ’07 Team Frasier Tom Cleary Matt Regan Bill Ryan Adam Bailin.

Choosing a LotChoosing a Lot

This is a bad lot

•Many obstacles

•Parking spaces are obscured

•Hard to map

Page 5: Automated Parking Lot Attendant SDP ’07 Team Frasier Tom Cleary Matt Regan Bill Ryan Adam Bailin.

The lot we choseThe lot we chose

This is a good lot

•Fewer obstacles

•Parking spaces easily identifiable

•Easier to map

Page 6: Automated Parking Lot Attendant SDP ’07 Team Frasier Tom Cleary Matt Regan Bill Ryan Adam Bailin.

HoweverHowever► Still things we need to worry aboutStill things we need to worry about

Page 7: Automated Parking Lot Attendant SDP ’07 Team Frasier Tom Cleary Matt Regan Bill Ryan Adam Bailin.

The CameraThe Camera►Axis 210 Network CameraAxis 210 Network Camera►Set up on 2Set up on 2ndnd Floor KEB Floor KEB

Page 8: Automated Parking Lot Attendant SDP ’07 Team Frasier Tom Cleary Matt Regan Bill Ryan Adam Bailin.

System OverviewSystem Overview

►Project all about image processingProject all about image processing►Two main parts: control and processingTwo main parts: control and processing

Need a central way to control systemNeed a central way to control system

►Basic steps of control systemBasic steps of control system 1) Take picture1) Take picture 2) Send to Matlab2) Send to Matlab 3) Receive from Matlab3) Receive from Matlab 4) Display to user4) Display to user

Page 9: Automated Parking Lot Attendant SDP ’07 Team Frasier Tom Cleary Matt Regan Bill Ryan Adam Bailin.

Controlling the SystemControlling the System

►Block diagram for control systemBlock diagram for control system

Generate readable result

Wait for timer to expire

Query camera, grab snapshot

Save snapshot locally with unique filename

Send image data to Matlabfor image processing

Matlab returns processing results

Display to user

Initialize system

Page 10: Automated Parking Lot Attendant SDP ’07 Team Frasier Tom Cleary Matt Regan Bill Ryan Adam Bailin.

►Take picture every 3 seconds using Take picture every 3 seconds using TimerTimer

►Run m-file from MatlabRun m-file from Matlab►Wait for Matlab to return resultsWait for Matlab to return results►Arrange results in human-readable Arrange results in human-readable

formform►Create image – layout of parking lot Create image – layout of parking lot

with indication as to which spots are with indication as to which spots are takentaken

Controlling the SystemControlling the System

Page 11: Automated Parking Lot Attendant SDP ’07 Team Frasier Tom Cleary Matt Regan Bill Ryan Adam Bailin.

SoftwareSoftware

►Using Microsoft’s .NET frameworkUsing Microsoft’s .NET framework►ClassesClasses

WebRequest() - request web resource WebRequest() - request web resource (image.jpg)(image.jpg)

HttpWebResponse() – returns jpg data HttpWebResponse() – returns jpg data streamstream

FileStream() – saves stream locallyFileStream() – saves stream locally Timer() - take pictures at intervalTimer() - take pictures at interval

Page 12: Automated Parking Lot Attendant SDP ’07 Team Frasier Tom Cleary Matt Regan Bill Ryan Adam Bailin.

User InterfaceUser Interface

► Will present user Will present user with computer with computer generated map of generated map of parking lotparking lot

Page 13: Automated Parking Lot Attendant SDP ’07 Team Frasier Tom Cleary Matt Regan Bill Ryan Adam Bailin.

Problems and SolutionsProblems and Solutions

►Learning curve for Visual Studio and Learning curve for Visual Studio and MATLABMATLAB

►Network congestion (wireless vs. wired)Network congestion (wireless vs. wired)►.jpg image size (640x480).jpg image size (640x480)►Delays to/from MatlabDelays to/from Matlab

Page 14: Automated Parking Lot Attendant SDP ’07 Team Frasier Tom Cleary Matt Regan Bill Ryan Adam Bailin.

We have our picture We have our picture on file, now what?on file, now what?

►Must read picture into MatlabMust read picture into Matlab““imread(‘c:\snapshot.jpg’)”imread(‘c:\snapshot.jpg’)”

►Image is 3-dimensional(red, green, Image is 3-dimensional(red, green, blue)blue)

Snapshot 480x640x3 uint8Snapshot 480x640x3 uint8

Page 15: Automated Parking Lot Attendant SDP ’07 Team Frasier Tom Cleary Matt Regan Bill Ryan Adam Bailin.

Our image ProcessingOur image Processing

► Basic idea: Image Differencing!Basic idea: Image Differencing! Is the new snapshot different from the base Is the new snapshot different from the base

snapshot?snapshot?► If so, something must have changedIf so, something must have changed

► Cut large snapshot into smaller piecesCut large snapshot into smaller pieces Each small piece is of one parking spotEach small piece is of one parking spot Pixels are manually mapped to each spotPixels are manually mapped to each spot

► All processing done on small pictures All processing done on small pictures individuallyindividually

Page 16: Automated Parking Lot Attendant SDP ’07 Team Frasier Tom Cleary Matt Regan Bill Ryan Adam Bailin.

Scaling ExampleScaling Example

► This is one example This is one example of pixel mappingof pixel mapping

► Most processing will Most processing will be done on these be done on these small picturessmall pictures

Page 17: Automated Parking Lot Attendant SDP ’07 Team Frasier Tom Cleary Matt Regan Bill Ryan Adam Bailin.

How Different?How Different?

► No two pictures are alikeNo two pictures are alike Glare, shadows, random ambiences.Glare, shadows, random ambiences.

► How different are two pictures?How different are two pictures? Correlation coefficient!Correlation coefficient!

► Variable which represents how different or alike two Variable which represents how different or alike two pictures arepictures are

► Between -1 and 1, 1 being two identical picturesBetween -1 and 1, 1 being two identical pictures A correlation coefficient below the threshold A correlation coefficient below the threshold

causes concern!causes concern!► State of parking spot is changedState of parking spot is changed► New snapshot becomes the baseNew snapshot becomes the base

Page 18: Automated Parking Lot Attendant SDP ’07 Team Frasier Tom Cleary Matt Regan Bill Ryan Adam Bailin.

A visual..A visual..

t0…… t5….. t10…

►A visual of how the program will run

Page 19: Automated Parking Lot Attendant SDP ’07 Team Frasier Tom Cleary Matt Regan Bill Ryan Adam Bailin.

Differencing IssuesDifferencing Issues

►Ambience's blocking camera positionAmbience's blocking camera position What if a truck blocks the view?What if a truck blocks the view?

►Solution! Timing bufferSolution! Timing buffer The base picture is only changed if the The base picture is only changed if the

new picture is different for a timenew picture is different for a time►Something that is blocking the camera will Something that is blocking the camera will

likely move awaylikely move away

Page 20: Automated Parking Lot Attendant SDP ’07 Team Frasier Tom Cleary Matt Regan Bill Ryan Adam Bailin.

More Issues…More Issues…

►Cars aren’t the only thing that can Cars aren’t the only thing that can cause a changecause a change Daylight gradually changes the new Daylight gradually changes the new

snapshot from the basesnapshot from the base

►Solution! Use full snapshotSolution! Use full snapshot A subtraction will show where the most A subtraction will show where the most

change took placechange took place

Page 21: Automated Parking Lot Attendant SDP ’07 Team Frasier Tom Cleary Matt Regan Bill Ryan Adam Bailin.

Determine Ambient Determine Ambient ConditionsConditions

►Look at area of Look at area of just pavementjust pavement

► If average of If average of pixels is pixels is similar, spot is similar, spot is probably probably emptyempty

Page 22: Automated Parking Lot Attendant SDP ’07 Team Frasier Tom Cleary Matt Regan Bill Ryan Adam Bailin.

MDR SpecificationsMDR Specifications

► Mount camera in good location overlooking a lot Mount camera in good location overlooking a lot near Knowles Engineering Building and connect to near Knowles Engineering Building and connect to network network

► Able to import an image into an image processing Able to import an image into an image processing programprogram

► Able to manipulate an image using basic image Able to manipulate an image using basic image processing techniquesprocessing techniques

Page 23: Automated Parking Lot Attendant SDP ’07 Team Frasier Tom Cleary Matt Regan Bill Ryan Adam Bailin.

Live view of cameraLive view of camera

http://abyss.ecs.umass.edu:8080

Page 24: Automated Parking Lot Attendant SDP ’07 Team Frasier Tom Cleary Matt Regan Bill Ryan Adam Bailin.

ImagesImages

RGB Grayscale

Edge Detection

Page 25: Automated Parking Lot Attendant SDP ’07 Team Frasier Tom Cleary Matt Regan Bill Ryan Adam Bailin.

ImagesImages

Picture 2

(Picture 1) – (Picture 2)

Picture 1

Page 26: Automated Parking Lot Attendant SDP ’07 Team Frasier Tom Cleary Matt Regan Bill Ryan Adam Bailin.

Looking ahead…Looking ahead…

► Need to explore the effects of weather conditions such as Need to explore the effects of weather conditions such as rain and snowrain and snow

► May need to consider alternate image processing solutions May need to consider alternate image processing solutions due to the following observations:due to the following observations: Pixel subtraction may not be accurate based on time of dayPixel subtraction may not be accurate based on time of day Obstructions (groups of people, cars driving through parking Obstructions (groups of people, cars driving through parking

lot)lot) Glare on window directly in front of camera – solved with box Glare on window directly in front of camera – solved with box

► Have many ways of determining spots – can average them, Have many ways of determining spots – can average them, have threshold for ‘spot taken’ eventhave threshold for ‘spot taken’ event

► We’re over the learning curveWe’re over the learning curve► Our demoOur demo

Page 27: Automated Parking Lot Attendant SDP ’07 Team Frasier Tom Cleary Matt Regan Bill Ryan Adam Bailin.

Questions?Questions?