Perceptual Uniformity-Aware Image Recoloring Method for Red-Green Anomalous Trichromacy

Color vision deficiency (CVD) is one kind of common disability fundamentally caused by genetic disorders, which lowers individuals' capability to distinguish colors and results in a loss of chromatic contrast. Although the state-of-the-art method for CVD compensation has been proposed, which can evenly deal with varying degrees of CVD, it suffers from the limitation of the color space while performs the recoloring, which lacks perceptual uniformity. More-over, its high computational time as well as complex parameter settings hinder its practical use. In this paper, we put forward an image recoloring algorithm for contrast enhancement for CVD compensation in the CIE L*a*b* color space, which can be adapted to the personal degree of CVD. In addition, the proposed algorithm maintains the naturalness of the original image by imposing a constraint on the compensation range. Furthermore, a discrete solver has been introduced to tackle the optimization problem effectively. To verify the effectiveness of the proposed algorithm, quantitative and subjective experiments have been conducted, and the experimental results show that our method achieved the best scores for contrast enhancement.