Description

Color blindness is the state of decreased ability to identify differences between colors. Inorder to help the persons having color blindness the colors in the images can be changed so that the persons can be able to recognize the colors. The process of recoloring of images is proposed which changes the colors in the images without affecting the originality of the image. The proposed approach identifies the colors in the images and then clusters the colors and finally color transformation is applied and the similarities were measured. The clusters were extracted from Fuzzy-C means clustering algorithm. The performance of the process is measured with the help of the naturalness and perceptibility. In the color identification step based on the color blindness of the person the color is identified from the RGB image. If the person contains red deficiency the red color regions from the images were alone identified. Similarly for green and blue color deficiency the color regions were extracted from the images. The color clustering process groups image pixels that has similar colors in the images. For clustering Fuzzy c means algorithm is employed. In fuzzy clustering, every point has a degree of belonging to clusters, as in fuzzy logic, rather than belonging completely to just one cluster. Thus, points on the edge of a cluster, may be in the cluster to a lesser degree than points in the center of cluster. An overview and comparison of different fuzzy clustering algorithms is available. Any point in the image has a set of coefficients giving the degree of being in the kth cluster. With fuzzy c-means, the centroid of a cluster is the mean of all points, weighted by their degree of belonging to the cluster. Each clustered region represents regions of different intensity. Color transformation process is employed based on the application of transformation matric to the input images in the needed color portions. The transformation matrix converts the original pixels from the images into different color based on the color blindness of the person. The performance of the process is measured based on the naturalness and perceptibility of the resulting color transformed image. The values were calculated based on the difference between the original RGB image and the transformed image.