Improvement on Image Decomposition Based on Morphological Component Analysis

This paper proposes a new method for image decomposition based on MCA. The basic idea presented in this paper is the use of non sub sampled contour let transform to represent structure parts, and the double density wavelet transform for the texture parts, then the Besov semi-norm is added for restricting structure parts. Finally, the block-coordinate-relaxation method is used to solve the new model. Experimental results show that the algorithm can effectively separate the structure and texture parts from noisy image, and improve the denoising performance.