首页 | 本学科首页   官方微博 | 高级检索  
     检索      

超分辨率图象恢复
引用本文:余英林,黎冠昌,吴立忠,贾青.超分辨率图象恢复[J].华南理工大学学报(自然科学版),1987(1).
作者姓名:余英林  黎冠昌  吴立忠  贾青
作者单位:华南工学院无线电与自控研究所 (余英林,黎冠昌,吴立忠),华南工学院无线电与自控研究所(贾青)
摘    要:本文系统论述超分辨率图象恢复的两大类算法,极大熵法和凸集投影法.我们用可变空间域窗法改善了GP算法,并求得最大熵法的一种情况的非线性方程组的近似解.最后给出我们的实验结果.我们还在一维情况下当收到信号的分辨率比原信号低几倍时进行极大熵恢复也得列满意的结果.

关 键 词:图象恢复    凸集  准则  投影

IMAGE RESTORATION BY SUPER-RESOLUTION METHODS
Yu Yinglin Li Guanchang,Wu Lizhong and Jia Qing Radio & Autom. Contr. Res. Inst. South China Inst. of Tech..IMAGE RESTORATION BY SUPER-RESOLUTION METHODS[J].Journal of South China University of Technology(Natural Science Edition),1987(1).
Authors:Yu Yinglin Li Guanchang  Wu Lizhong and Jia Qing Radio & Autom Contr Res Inst South China Inst of Tech
Abstract:This paper is to present in detail the study of image restoration by superresolution methods with the special emphasis on super-resolution methods based on the maximum entropy criterion and alternating projection of special convex sets. By successfully decreasing the window width in spatial domain, the conventional GP algorithm is improved, and at the same time the approroximate solution to the nonlinear equation system, which is derived from the maximum entropy criterion, is obtained. In 1-D case, a satisfactory restoration is also obtained with incomplete data when the sampling rate of the incoming data is lower than that of the test data by 1-5 times, Finally the experiment results are reported.
Keywords:image restoration  entropy  convex sets  criterion  projection
本文献已被 CNKI 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号