运动模糊图像复原R-L算法的优化  

Optimization of R-L algorithm for motion blurred image restoration

作　　者：程敏睿 时康 张尘研 叶睿哲 Chen Minrui;Shi Kang;Zhang Chenyan;Ye Ruizhe(College of Electrical Engineering,Anhui Polytechnic University,Anhui Wuhu 241000,China)

机构地区：[1]安徽工程大学电气工程学院,安徽芜湖241000

出　　处：《长江信息通信》2021年第2期22-24,共3页Changjiang Information & Communications

基　　金：国家级创新创业训练计划项目201910363016,202010363036;安徽工程大学大学生科研项目2020DZ23。

摘　　要：图像获取过程中如果产生拍摄对象与设备之间的相对运动,就会产生图像变得模糊,这种情况称之为运动模糊。文章从图像复原基础原理出发,主要介绍了Richardson-Lucy算法对运动图像的复原,比较自相关和频谱投影模糊长度估算方法。通过实验发现频谱投影在较短模糊长度是较为精确,自相关估算模糊长度在较长模糊长度较为精确。由此提出组合优化模糊长度估算,通过实验图像复原发现组合优化的模糊长度复原图像峰值信噪比大13%,有效提高了图像复原精度。In the process of image acquisition,if the relative motion between the subject and the device is generated,the image will become blurred,which is called motion bluring.Based on the basic principle of image restoration,this paper mainly introduces the restoration of moving image by Richardson Lucy algorithm,and compares autocorrelation and spectral projection fuzzy length estimation methods.It is found that the spectrum projection is more accurate in the shorter fuzzy length,and the autocorrelation estimation is more accurate in the longer fuzzy length.It is found that the peak signal-to-noise ratio of the reconstructed image is 13%,which improves the precision of image restoration effectively.

关 键 词：运动图像复原 运动模糊 点扩散函数 Richardson-Lucy算法 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]