Improved convex set projection algorithm for Image sequence reconstruction

The convex set projection (POCS) algorithm based on set theory has the advantages of flexible observation model, simple implementation principle and easy introduction of prior information of images. It has become an important method to solve the problem of super-resolution reconstruction of sequence images. In this paper, the projection reconstruction algorithm of convex set is studied in detail. In view of the shortcomings of traditional POCS algorithm, a series of improvements are made: the weight function is constructed at the edge pixels of the image to eliminate the edge ringing effect in POCS reconstruction image; the iterative step size of relaxation projection operator is adaptively selected to improve the image restoration quality and improve the computational efficiency. The experimental results show that the improved POCS algorithm can obtain better reconstruction effect.