Implementation Of Super-resolution Image Restoration Based On Genetic Algorithms

Recent years have seen growing interest in the problem of super-resolution restoration of still image or video sequences. Whereas in the traditional single image restoration problem only a single input image is available for processing, the task of reconstructing super-resolution images from multiple undersampled and degraded images can take advantage of the additional spatio-temporal data available in the image sequence. In particular, camera and scene motion lead to frames in the source image sequence containing similar, but not identical information. The additional information available in these frames makes possible reconstruction of visually superior frames at higher resolution than that of the original data.Super-resolution restoration from an image sequence is a well recognized example of an ill-posed inverse problem. Such problems may be approached using regularization methods which constrain the feasible solution space by employing a-priori knowledge, hi this paper, we regard the restoration problem as a regularized image interpolation problem, and mainly deal with image interpolation algorithms to enhance the image quality in the sense of resolution. The regularized image interpolation is to find the inverse solution defined by the image degradation model subject to a priori constraints. We propose a general framework based on interpolation between image frames. Restored high-resolution image with smoothness constraint can be obtained from low resolution image frames.Genetic algorithm is a search algorithm based on mechanics of natural selection and natural genetics. It uses probabilistic transition rules to guide itself toward an optimum solution. Compared to other search algorithms, genetic algorithm has an excellent parallelism and mathematic simplicity. Recent years, genetic algorithm has been widely used in many domains, including image processing, hi this paper, genetic algorithm is applied to implement both motion estimation and high-resolution restoration and the experiment results show its feasibility.