This works aims at enhancing images by using the colour appearance model CIECAM02 for the visually impaired to view digital displays to complement the existing image processing approaches with a reference to normal visions. Specifically, by studying the images perceived by low-vision users, the colour ranges of these perceived views are compared with those viewed by normal vision and then characterized and represented using CIECAM02 correlates, which include lightness, colourfulness, and hue. For low-vision users, the extents of these attributes are therefore obtained. Subsequently, for any input image, these CIECAM02 attributes are subsequently enhanced through histogram equalizer technique within their respective ranges for low-vision users. In comparison with the approach of RGB histogram equalizer, the preliminary result has shown that the proposed method appears to be better to enhance the contents depicted in an image. The evaluation experiment was carried out using an array of low-vision simulator glasses to be worn by a group of subjects with normal vision. The next stage of the work remains to invite real low-vision users to evaluate the proposed work.
Xiaohong W. Gao, Monica Loomes, "A new approach to image enhancement for the visually impaired" in Proc. IS&T Int’l. Symp. on Electronic Imaging: Color Imaging XXI: Displaying, Processing, Hardcopy, and Applications, 2016, https://doi.org/10.2352/ISSN.2470-1173.2016.20.COLOR-325