An Optimized Algorithm for Face Image Super-resolution
Xiongli Sun · Radio Engineering · 2012
To mitigate the blocking effect and local distortion in traditional super-resolution algorithms,an improved algorithm is proposed based on Markov model.This algorithm can meet the constraints of image reconstruction,obtain the optimal block from the training set using nonlinear local searching technique,and achieve compatibility between matching blocks through horizontal compatibility checking.It also provides weighted processing based on Sigmoid function to improve matching accuracy.Experimental results demonstrate that the proposed algorithm can effectively prevent the blocking effect and local distortion in the process to obtain HR image,and improve the quality of the super-resolution image.Compared with the traditional algorithms,this algorithm also provides strong robustness.