Image Restoration Using Group-based Sparse Representation Technique
Balwinder Kaur, Ashok Kumar Bathla · International Journal of Advanced Engineering Research and Science · 2016
This paper presents an efficient algorithm for solving restoration problem in the frame-based image restoration. In image restoration, the patch-based approach has been used for better results without degrading the image. Our proposed procedure for solving the stable optimal problem is based on a patched group strategy method. In this paper GSR (Group Based Sparse Representation) algorithm is proposed which is based on concept that group of patches are constructed which maintains the relationship among various patches of images and it is implemented for three image restoration problems i.e. DE noising, DE blurring and Image enhancement. Various parameters are taken into account like PSNR, SSIM, speed and time. The simulation experiments have been conducted in MATLAB and the results have been compared with existing scheme i.e. patch based sparse representation. Both standard and real time images have been included in simulation. It has been observed that after the reconstruction of an image values of parameters are increased which are more than existing scheme.