A Fuzzy based naturalness preserving image color compensation method for colorblindness

Eyesight is one of the key senses possessed by human beings to perceive the world of colors. Color blindness is a form of Color Vision Deficiency (CVD) that affects about 0.5% of female and 8% of the male population [1]-[4]. The colorblind have difficulty perceiving specific colors called confusion colors. Hence, they face difficulty in handling routine activities [5]. With the increased use of the web, it is imperative to provide digital solutions to help the colorblind perceive colors. The authors propose a novel method using fuzzy based clustering to provide color compensation in images for red green color blindness. The method aims at optimizing the color enhancement by clustering the confusing colors and color compensating them. To achieve this, the authors have used the fuzzy C means algorithm and the delta E method to calculate the Euclidean distance between the original color and the compensated color. The delta E method ensures that the difference computed is within the perception of colors by humans. The efficiency of color compensation is measured by a) subjective evaluation by normal vision users b) evaluation by colorblind users and c) comparison with state-of-the-art method. The evaluation results affirm the efficiency. With the increased use of visualization and digitization of daily activities, the proposed method would aid colorblind users and help them to distinguish confusion colors in the images effectively.