Statistical Analysis of the Lms Algorithm Applied to Super-Resolution Video Reconstruction

Super-resolution reconstruction of image sequences is highly dependent on the quality of the motion estimation between successive frames. This work presents a statistical analysis of the least mean square (LMS) algorithm applied to super-resolution reconstruction of an image sequence. Deterministic recursions are derived for the mean and mean square behaviors of the reconstruction error as functions of the registration errors. The new model describes the behavior of the algorithm in realistic situations, and significantly improves the accuracy of a simple model available in the literature. Monte Carlo simulations show good agreement between actual and predicted behaviors