MULTI-FRAME IMAGE SUPER-RESOLUTION BASE ON GRADIENT VECTOR FLOW

Multi-frame super-resolution reconstruction aims to fuse several low resolution images into one image with high optical resolution. In this paper, a new regularization-based multi-frame image super-resolution algorithm is presented. Utilizing the gradient directional information, the proposed algorithm uses the gradient vector flow to measure the gradient consistency between the high-resolution image and low-resolution images. Experiment results show that the proposed algorithm achieves better performance than other multi-frame image super-resolution methods