Research Of Blind Image Restoration Algorithms

With the rapid development of multimedia and computer technology, image processing becomes more and more widely used, and it has already far broken through the early application, widely applied to a lot of fields, such as scientific research, industrial and agricultural production, military technology, government department, health care, etc. Blind image restoration is a great part of image processing, which refers to the technology field that study how to restore the original image from the observed image without sufficient a prior knowledge about the degradation process. Blind image restoration has a strong background of mathematics, including estimation theory, ill-posed problem solution method, line algebra, stochastic process, numerical analysis, and so on. The general sequence of blind image restoration is: firstly, model the degradation processing in mathematical form, and secondly, solve this inverse problem to get an estimation of original image.In this paper, some commonly used classical image restoration algorithms are introduced. EM and PBD algorithms are analyzed, and NAS-RIF algorithm is the important issue. The complicated EM algorithm and PBD algorithms are validated with Matlab , the differences between them are compared and the weaknesses of these algorithm are putted out. In NAS-RIF algorithm, the improved algorithm of supporting and correcting the no uniform background image are proposed, these two algorithms are validated with Matlab and the differences between the original and improved algorithms are compared. It is clear that the improved algorithm improves the efficiency and quality of restoration image.In the end of this paper, several kinds of algorithms studied in this article are compared and the directions and emphases in future research are promised as well.