A Improved Super-resolution Image Reconstruction Algorithm

Super-resolution image reconstruction (SRR) refers to a resolution enhancement technology that an image sequence of the same scene should be collect first in the same position by the imaging system with a camera or video camera, and then we can get a higher resolution image by using signal processing technology which containing more details that had been extracted from these lower resolution images. The existing low-resolution imaging systems can be still utilized, so it is a viable economic solution. SRR has a wide range of applications and has become one of the hottest image restoration research topics in the world. The high-precision image registration algorithm, fast and efficient reconstruction algorithm are the focuses and difficulties of Super-resolution reconstruction.In this paper, spatial super-resolution reconstruction algorithm is the main context, and briefly introduced some common image registration algorithms, focused on the SR reconstruction algorithm based on MAP theory. The selection of the cost-function and regularization term has been deeply analyzed and compared by experiment. After deeply study of the cost function and comparing a lot of main reconstruction algorithms, we found that there are differences among the low resolution images, such as the additional noise and registration. Because of the existence of these differences, the contribution of each low-resolution image to reconstruction result are very different, but the recent reconstruction algorithm does not considerate these, so we proposed an adaptive weighted coefficients of fidelity item method. But in the processing of adaptive weighted, it will fuzz edge in reconstruction image, so we further propose an adaptive weighted reconstruction algorithm based on edge preservation. Experimental results show that the proposed method can effectively improve the effect of reconstruction.