边缘区域约束引导的运动模糊图像复原  

Restoring Motion Blurred Images Based on Edge Region Constraints

作　　者：段江永[1] 孟高峰[1] 向世明[1] 潘春洪[1] 

机构地区：[1]中国科学院自动化研究所模式识别国家重点实验室,北京100190

出　　处：《计算机辅助设计与图形学学报》2012年第8期1038-1046,共9页Journal of Computer-Aided Design & Computer Graphics

基　　金：国家自然科学基金(60975037;61005036;61175025)

摘　　要：相机在拍摄物体时,物体运动或相机抖动会导致图像模糊.针对此问题,提出一种边缘区域约束引导的运动模糊图像复原算法.通过分析图像中阶跃边缘区域与运动模糊核之间的内在关系,建立了一种新的边缘区域约束;对该约束进一步结合RANSAC算法,可有效地引导运动模糊核的估计.大量对比实验结果表明,该算法简单、估计精度高,能够得到优于传统算法的高质量图像模糊恢复结果.Object motion or camera shake during exposure often leads to image blur. This paper proposes an algorithm for removing motion blur in images. Based on the observation that clear images often have step edges, we develop a new type of edge region constraints by analyzing the relationship between the step-edge region and the motion kernel. With this type of constraints, the motion kernel is efficiently estimated by using the RANSAC algorithm. Extensive comparative experiments are conducted, and verify that our algorithm is simple, highly accurate, and is capable of restoring higher quality images than traditional methods.

关 键 词：运动模糊 边缘区域约束 核估计 反卷积 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]