Saturation avoidance color correction for digital color images

Abstract

The qualities of color images captured by digital imaging devices are vulnerable to the scene illumination settings of a given environment. The colors of captured objects may not be accurately reproduced when the illumination settings are uncontrollable or not known a priori. This undesirable property can inevitably degrade the qualities of captured images and lead to difficulties in subsequent image-processing stages. Considering that the task of controlling scene illumination is nontrivial, color correction has emerged as a plausible post-processing procedure to efficiently restore the scene chromatics of a given image. In this study, a new color correction technique called the Saturation Avoidance Color Correction (SACC) algorithm is proposed to remove the undesirable effect of scene illuminants. Unlike most well-established color correction algorithms, the proposed SACC comprises a nonlinear pixel adjustment mechanism to avoid the saturation effect during the color manipulation process. A collection of color images including indoor, outdoor, and underwater images are used to verify the capability of SACC. Extensive experimental studies reveal that the proposed algorithm is preferable to some existing techniques because the former has a high capability to mitigate the color saturation issue and is able to produce corrected images with more pleasant visualization.