Local fractal dimension-based color quantization for error diffusion techniques

Abstract

Reconstruction of an image with a limited number of colors (color palette) leads to highly visible degradations in image quality known as false contours. A way to overcome this problem is to perform dithering techniques. In this paper we propose a color quantization method for use with color dithering techniques. It makes use of local fractal dimensions to allocate larger weight to pixels in low activity regions where false contours in an image are most likely to occur. The results show that our method significantly removes false contours and color impulses as well as preserves textures that are commonly lost in high activity regions when applying dithering techniques to color quantized images. © 2013 Springer.