Color enhancement for red–green color deficiency using multispectral image

Previous studies have proposed methods to recolor images and help individuals with color vision deficiencies distinguish colors they typically confuse. These methods often simulate the appearance of images for dichromats using tristimulus values, which represent color within a narrow range. In this paper, we propose an image enhancement method tailored for individuals with red-green color vision deficiencies utilizing multispectral images. Principal component analysis (PCA) is applied to the multispectral data to reduce its dimensionality, allowing specific wavelength components corresponding to red-green color deficiencies to be enhanced. Using the recolored images, we demonstrate that this method improves color discrimination for red-green color-deficient individuals while preserving the naturalness of the images. Additionally, evaluations based on the number of discernible colors and image difference metrics quantitatively demonstrate that this method is more effective in enhancing color discrimination while preserving the naturalness of the image compared to conventional methods.