Joint POCS algorithm with sub-pixel motion estimation for super-resolution of diffraction-limited images

Abstract

This paper proposes a novel super-resolution restoration algorithm based on jointly projecting onto convex sets (POCS) and sub-pixel motion estimation, which differs from the super-resolution algorithms proposed in the previous studies. This scheme specially considers: (1) motion estimation and reconstruction interrelations; (2) sub-pixel relative motion between observation frames contributes novel information which provides constraints for the high-resolution images. Hence, our algorithm offers a greater flexibility for simultaneously estimating sub-pixel motion parameters within the reconstruction iteration. By numerical experiments, the approach achieves better reconstruction performance