Improving iterative back projection super resolution model via anisotropic diffusion edge enhancement

Abstract

This improving technique based on combining an Iterative Back Projection (IBP) super resolution method with Anisotropic Diffusion (AD) technique for overcoming IBP weaknesses. The IBP has specialty to remove a reconstruction error and blurry effect iteratively manner in image registration place. However, it has a weakness from avoiding result image from chessboard effect and lost high frequency information. For this reason, this super resolution approach requires an edge enhancement technique to complement the weakness. Anisotropic diffusion is edge enhancement techniques and it has benefited to estimate a piecewise smooth image from a noisy input image. This paper proposed to integrate Anisotropic Diffusion techniques in IBP registration stages with an improvement in the IBP flow process model. This improvement in IBP model produced a result image in better appearing output with preserves high frequency information and less number of iteration process reconstruction.