Colorfulness Enhancement Using Image Classifier Based on Chroma-histogram

The paper Proposes a colorfulness enhancement of pictorial images using image classifier based on chroma histogram. This approach firstly estimates strength of colorfulness of images and their types. With such determined infomation, the algorithm automatical- ly adjusts image colorfulness for a better natural image look. With the help of an additional detection of skin colors and a pixel chroma adaptive local processing, the algodtlan produces more natural image look. The algorithm perfomance had been tested with an image quality judgment experiment of 20 persons. The experimental result indicates a better image preference.