demosaic.zip_MATLAB demosaic_MSE2.m_demosaic MATL_image restorat

This package contains MATLAB implementation of the CFA demosaicing algorithm presented in the paper, X. Li, "Demosaicing by successive approximation" and two benchmark algorithms, namely 1. B. K. Gunturk, Y. Altunbasak and R. M. Mersereau, ``Color plane interpolation using alternating projections ,'' {\em IEEE Trans. on Image Proc.}, vol. 11, no. 9, pp. 997-1013, Sep. 2002. 2. Wenmain Lu and Yap-peng Tan, ``Color filter array demosaicing: new method and performance measures,'' {\em IEEE Trans. on Image Proc.}, vol. 12, no. 10, pp. 1194-1210, Oct. 2003. To test these algorithms, simply run the following functions under MATLAB > run_cfa('img',1,12); % test benchmark scheme 1 > run_cfa1('img',1,12); % test benchmark scheme 2 > run_cfa2('img',1,12); % test our own scheme Note: cfa_main and cfa_debug are the two main functions of our proposed iterative demosaicing algorithm. The difference between them is the stopping criterion: cfa_main adopts a universal threshold and cfa_debug uses a two-class adaptive threshold. Two text files (mse_result.txt and cielab_result.txt) includes the experiment results when running those test programs. If you have any question regarding the above work, please contact Xin Li at xinl@csee.wvu.edu.