A Color Recovery Algorithm Based on Color Space Transformation

Abstract

The same object will show different colors under different illumination. An illuminant-independ color representation were proposed. Firstly, estimate the illumination color, and transform the RGB cube to a color space XLYLZL, then transform the XLYLZL back to RGB color space. To verify the effectiveness of color correction algorithm, a set of images taken under CIE A illuminant which are yellow-biased were processed to evaluate the performance. The original images were transformed to images under the CIE Standard Illuminant D65. Color difference of recovered image and standard image as well as original images and standard image were calculated to evaluate the performance, the experimental results show that with the color correction algorithm based on color space transformation, the accuracy rate of the recovered image improved by 68.98% compared with the original image.