color Image Enhancement with a Human Visual System Based Adaptive Filter

15
Color Image Enhancement with a Human Visual System Based Adaptive Filter Presented by K. ANURAMA 12S11D6502

description

enhancement of color in digital images

Transcript of color Image Enhancement with a Human Visual System Based Adaptive Filter

Page 1: color Image Enhancement with a Human Visual System Based Adaptive Filter

Color Image Enhancement with a Human Visual System Based

Adaptive Filter

Presented by K. ANURAMA12S11D6502

Page 2: color Image Enhancement with a Human Visual System Based Adaptive Filter

Abstract

Proposed method New color image enhancement algorithm

- Based on human visual system adaptive filter• Dividing three major parts

– Adaptive adjustment

– Color restoration

– Unlike traditional color image enhancement algorithms

• Use of color space conversion – Much better visibility

• Better effectiveness in reducing halo and color distortion

Page 3: color Image Enhancement with a Human Visual System Based Adaptive Filter

Introduction

Image enhancement technology– Problem of many factors

• Limited dynamic range• Lighting influences

• Display device

– Not suitable to directly use gray image enhancement technologies for color images

Page 4: color Image Enhancement with a Human Visual System Based Adaptive Filter

Color image enhancement technologiesBased on human visual characteristics Retinex– Very good enhanced result

– Color distortion

– Complex of calculation

Li Tao and VijayanK. Asari– Robust color image enhancement algorithm

Use of Gaussian filter Inaccurate estimation

Page 5: color Image Enhancement with a Human Visual System Based Adaptive Filter

Color image enhancement technologies

– New bio-inspired color image enhancement algorithm Combination of retinex and Li Tao and VijayanK. Asari algorithm

Use of bilateral filter

• New algorithm– Consideration of color information

– Use of adaptive filter to get the background image

Page 6: color Image Enhancement with a Human Visual System Based Adaptive Filter

HALO EFFECT

Halo artifacts are due to the proximity of two areas of very different intensity.

For example, if a dim area is close to a bright window, the bright pixels can influence the processing of the dim area and can cause a black halo around the bright area. Moreover, local filtering tends to make pure black and pure white low contrast areas turn gray. These phenomena are illustrated in Fig. 1. The shadow on the face is a halo artifact due to the background window. The black t-shirt looks washed out due to the local filtering

Fig. 1. Example of halo artifacts and graying-out. The shadow on the face is a halo artifact due to the background window. The black t-shirt looks washed out due to the local filtering

Page 7: color Image Enhancement with a Human Visual System Based Adaptive Filter

Color Image Enhancement algorithm Proposed method

– Framework of color enhancement algorithm

Fig. 1. The Framework of the proposed color image enhancement algorithm.

Page 8: color Image Enhancement with a Human Visual System Based Adaptive Filter
Page 9: color Image Enhancement with a Human Visual System Based Adaptive Filter
Page 10: color Image Enhancement with a Human Visual System Based Adaptive Filter

Fig. 2. The Example of luminance image and background image: from up to down, the original image, luminance image and background image

Page 11: color Image Enhancement with a Human Visual System Based Adaptive Filter
Page 12: color Image Enhancement with a Human Visual System Based Adaptive Filter

Image enhancement processUsing proposed algorithm

Fig. 3. Enhanced results of using the proposed algorithm. The left are original, and the right are enhanced images

Page 13: color Image Enhancement with a Human Visual System Based Adaptive Filter

Comparison with other techniques

Compared to the traditional algorithms

Fig. 4. Illustrate the difference of using different algorithm: (a) the algorithm in [4]; (b) the algorithm in [5]; (c) Retinex; (d) proposed algorithm 11

Page 14: color Image Enhancement with a Human Visual System Based Adaptive Filter

Fig. 11. Top: Gamma-encoded image. Middle: Image treated with the adaptivefilter method. Bottom: Image treated with the fast bilateral filtering method.

Page 15: color Image Enhancement with a Human Visual System Based Adaptive Filter

Conclusion

Proposed method– New color image enhancement algorithms• Considering human visual properties• Improvement of visual qualities of enhanced images

– Adaptive filter considering color information– Utilizing color space conversion to get luminance

image