一种改进的迭代图像盲恢复算法  

A Modified Iterative Blind Image Restoration Algorithm

作　　者：马耀锋[1] 李红丽[1] 

机构地区：[1]中州大学信息工程学院,郑州450044

出　　处：《西南大学学报（自然科学版）》2013年第7期151-155,共5页Journal of Southwest University(Natural Science Edition)

基　　金：河南省教育厅自然科学基金资助项目(2010C520020)

摘　　要：针对原NAS-RIF算法敏感于噪声且支撑域为矩形等不足,提出了一种改进方案.通过引入高阶统计量去噪,解决了NAS-RIF算法敏感于噪声的特性;利用图像分割技术进行图像支撑域的自估计,改变了NAS-RIF算法中支撑域必须是矩形的不利限制.仿真结果表明改进的NAS-RIF算法具有很好的恢复效果.An improved method is presented to overcome the drawbacks of the original NAS-RIF algorithm,i.e.noise sensitivity and a rectangular support region.To deal with the noise sensitivity of NAS-RIF,higher order statistics denoising is introduced.Self-estimation on image support based on image segmentation is used to change the rectangular support limitation.At last,simulation is done to various PSNR images and backgrounds.The simulation results show that the improved algorithm solves the existing problems of the original algorithm,and observably improves the quality of recovery image.

关 键 词：图像盲恢复 NAS-RIF算法 高阶统计量 支撑域估计 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]