前视红外图像复原及边界振铃效应消去研究.pdfVIP

32 15 武 汉 理 工 大 学 学 报 Vol. 32 No. 15 2010 8 JOURNAL OF WUHAN UNIVERSITY OF TECHNOLOGY Aug. 2010 DOI: 10. 3963/j. issn. 2010. 15.036 1 2 3 杨文霞 , 陈湘平 , 蔡 超 ( 1. , 430070; 2. , 430019; 3. , 430074) : 为提高前视红外图像复原的计算速度和复原效果, 首先提出一种加速的 andweber 迭代复原算法, 使松弛迭 代 子在迭代过程中,根据复原图像的梯度特征进行自适应选择, 使收敛所需迭代次数降低为传统 landweber 算法的 1/ 5~ 1/ 6,提高了算法的实时性同时,设计出改进的Neumann 边界条件, 对观测图像进行延拓, 将图像复原误差转移 到图像主值序列之外, 有效降低复原图像边界处的振铃效应 : 图像复原; 振铃效应; landweber 迭代; Neumann 边界延拓 : TP 391. 41 : A :2010) Research on Forward Looking In rared Image Restoration with Ringing Reduction 1 2 3 YAN G Wenx ia , CH EN X iangp ing , CAI Chao ( 1.School of Sciences, Wuhan University of Technology, Wuhan 430070, China; 2.Dept. of Information and Command Automation, Air Force Radar Academy, Wuhan 430019, China; 3. Institute of Pattern Recognition and Artificial Intelligence, Multi spectral Image Information Processing Key aboratory of National Defense, Huazhong University of Science and Technology, Wuhan 430074, China) Abstract: Aiming at the problems of improving the performance and computational efficiency of the Forward ooking In frared (F IR) image restoration procedure, an improved landw eber iterat ive accelerating technique is proposed first. Instead of being fixed, the relaxation factor is updated adaptively at each iteration considering the gradient character of the restoring im age. Experimental results demonstrate that the required iterative timesof the proposedmethod are redu