A Study On The Algorithms Of Color Inverse Halftoning Based On Human Visual System

The usual inverse halftoning method doesn't consider the characteristics of human visual system and color spaces. Based on this causes the two algorithms are proposed for color inverse halftoning. The first is designed by simulating human visual perception. It processes halftoning image in Opponent Space in which the light and the chrominance are separated. The distortion of the relativities among R, G and B channels is less. The experimental results show it is better than the others without considering the relativities. The second is based on the perceptual characteristics about detail and smooth areas. An image of details enhancement will be reconstructed from two images which are generated by the first method with different filters. The experimental result shows the inverse halftone images generated by the algorithm fit human visual perception well.