基于2-D公共因子提取的单幅图像盲复原研究(英文)  

Research on Blind Restoration of One Blurred Image Based on Extraction of 2-D GCD

作　　者：王睿[1] 方勇[1] 

机构地区：[1]上海大学通信与信息工程学院,上海200072

出　　处：《系统仿真技术》2008年第1期6-9,29,共5页System Simulation Technology

基　　金：国家自然基金(60472103);教育部博士点基金(20060280003)共同资助项目

摘　　要：利用2-D公共因子(GCD)提取可以有效地实现图像盲复原,但通常需要原始图像的两幅降晰图像,这在某些条件下并不能满足,极大限制了此方法的应用。发展了一种基于2-D公共因子的单幅图像盲复原方法,利用2-D公共因子提取从一幅降晰图像中估计降晰函数,再利用经典的图像反卷积技术,从而复原原始图像,仿真实验证明了该方法的有效性。The method of 2-D great common divisor （GCD） extraction can be used to restore the blurred image. However, two blurred images are required for this approach. This condition can usually not be satisfied, so this method is limited greatly. A new method is proposed for image restoration with one blurred image based on 2-D GCD extraction. Firstly, the blurring function is estimated by 2-D GCD extraction according to one blurred image. Then the classical deconvolution technique is exploited for image restoration. Simulation results show that the proposed algorithm is efficient for blurring image restoration from one blurred image restoration.

关 键 词：2-D公共因子提取 图像盲复原 反卷积 

分 类 号：TN911.73[电子电信—通信与信息系统]