Robust super-resolution reconstruction algorithm based on total variation

A super-resolution algorithm based on L1 norm and total variation(TV) regularization is proposed.The L1 norm is used to constrain the fidelity of the reconstructed image,so the reconstruction error caused by the inaccuracy of the model estimation in the traditional algorithm is reduced.The total variation regularization is implemented to overcome the ill-poseness of the problem,the edge of the image is preserved effectively.A sequence of simulant low resolution images are reconstructed by the proposed algorithm,and the result is compared with which is constructed by the algorithm based on L2 norm and Tikhonov regularization in two aspects,visual effect and PSNR,the experimental results confirm that the proposed algorithm is more robust and edge-preserving.