Super-Resolution Image Reconstruction From Low-Resolution Image Sequences

Super-resolution image reconstruction technique refers to restoring a high-resolution and high-quality image from multiple low-resolution observations (or a video sequence) degraded by warping, blurring, noise and aliasing. Super-resolution (SR) restoration is one of the most important research subjects in the field image processing and it becomes more and more attractive in this field during recent years. It has a researching significance in theoretically as well as a broad application in practically. Super-resolution restoration is an economic image processing method that extracting higher resolution images containing more details from an image sequence of lower resolution, through such methods as motion estimation, image fusion, de-blurring and de-noising.In this paper, based on the method of Image super-resolution reconstruction, we study on the algorithm of POCS, according to the theory of POCS, we design a simulation-correction iteration method to reconstruct high-resolution images.A sequence images super-resolution reconstruction algorithm based on NEDI (new-edge-directed interpolation) is proposed. Before interpolation, a frequency domain algorithm by Vandewalle is employed to register a sequence of aliased low-resolution images precisely.Based on the algorithm of NEDI, the combination of the algorithm of NEDI and wavelet transform ,we reconstruct high-resolution images by wavelet inverse transformation- NEDI-wavelet transform and simulation results of Contourlet transform to provide a flexible multi-resolution, local and directional expansion for low-resolution image prove which is effective and feasible.The algorithms are tested in simulations, and compared to other similar SR methods.