Segmentation image re-coloring based on k-means clustering algorithm as a tool for partial color-blind people

Abstract

Color blindness is one of the eye anomalies that can not distinguish one color from another. Therefore, the information contained in the image may be lost seen by color blindness people. To reduce the possibility of sufferers losing color information contained in the image, a system was built in the form of re-coloring on images based on the ability of color blindness sufferers to distinguish colors. This ability is measured based on RGB cluster owned by different people. This research using color based segmentation in the process of re-coloring. Image Segmentation using K-means Clustering was chosen because of short processing time with optimum result. A total of 15 images without the coloring process were tested on color blind subjects. The result is that only 20% of the information can be seen by the color blind subject. The output of this study is the image that has been re-colored. The result of this research is that 53% of the images can be seen by the color-blind subject by using color-based segmentation in each image, 60% of the image can be seen by the color-blind subject by using color-based segmentation on all images, and processing time is shorter than previous research.