The Taub Faculty of Computer Science Events and Talks

Color information plays a major role in image processing and visual communication although presently most algorithms and tools are developed mainly for monochromatic images. Usually, the processing of color images is performed either in the RGB color space or in another color space chosen rather arbitrarily, such as YUV or YIQ. In this work we propose new frameworks for color image processing and coding based on an optimized approach. These frameworks along with rate-distortion analysis optimize the stages of coding and color processing for visual communication. The Mean Square Error (MSE( and the visual-oriented Weighted Mean Square Error (WMSE) are used to account for both quantitative and subjective visual fidelity. We exploit the high inter-color correlations of the RGB primaries to introduce a correlation-based coding approach rather than the ordinary decorrelation-based method, which is currently used in most algorithms. The two approaches are further generalized to provide a unified correlation/decorrelation-based framework.

The new color processing approach can be helpful in several fields of image processing, including image compression and image demosaicing. We show that for compression, the new framework outperforms presently available compression algorithms, including well established ones, such as JPEG 2000, while having comparable complexity. For demosaicing, instead of using the ordinary RGB color space, the correlation of primary colors can be better exploited in an optimized color space, resulting in demosaicing that is superior to presently available methods. Additional aspects of the new frameworks will be presented and discussed. Our conclusion is that by optimizing the use of color information, major operations in image processing and visual communication could be improved quantitatively and visually.