A Fast Algorithm for Vectorial TV-Based Image Restoration

Abstract

In this paper, we first extend a simple algorithm proposed by Jia et al. [16] to color/vectorial images, and then apply the vectorial algorithm to some variational models for image restoration problems including color image denoisings with the red-green-blue (RGB) and chromaticity-brightness (CB) color representations, CB based colorization and image inpainting. The variational models are all total variation (TV)-based. The proposed vectorial algorithm is simple and straightforward to implement. Some numerical experiments show that it is fast and efficient.