摘要
图像序列的超分辨率恢复是目前国际图像复原界的一个研究热点 .本文提出了一种基于分层机制的图像序列超分辨率恢复方法 .在将问题采用正则化方法变为适定问题后 ,提出了一种级联模板的快速运算方法 .
The super-resolution technique from image sequence is an active research issue in the field of image restoration.For this problem, a novel method based on delaminating strategy is described in this paper.The delaminating strategy divides the original problem into several sub-problems.Each of these sub-problems is only connected to one degrading factor.In addition, it is suggested that the ally-template algorithm,which is implemented by a serious mask operator,can resolve each sub-problem.To show the effectiveness of the presented algorithm,the simulative results carried out on an aerial image sequence are compared with those of conventional methods such as the maximum a posteriori probability (MAP) method and the projections onto convex sets (POCS) method.Furthermore,the complexity of the new method is discussed.The experimental result prove this method to be efficient.
出处
《电子学报》
EI
CAS
CSCD
北大核心
2004年第12期2074-2077,共4页
Acta Electronica Sinica
基金
国家自然科学基金 (No .60 2 72 0 99)
国家 863计划项目 (No.2 0 0 2AA40 1 0 0 1 4B)
测绘遥感信息工程国家重点实验室开放基金 (No .WKL(0 2 ) 0 1 0 3)
