首页 | 本学科首页   官方微博 | 高级检索  
     

评价多传感器图像融合效果方法的比较
引用本文:王海晖,彭嘉雄,吴巍.评价多传感器图像融合效果方法的比较[J].红外与激光工程,2004,33(2):189-193.
作者姓名:王海晖  彭嘉雄  吴巍
作者单位:华中科技大学,图像识别与人工智能研究所,图像信息处理与智能控制教育部重点试验室,湖北,武汉,430074
基金项目:精确制导与自动目标识别国防科技重点实验室基金资助项目(51483040103JW0515)
摘    要:为了选择一种在复杂条件下能与主观评价图像融合方法取得一致效果的客观评价方法,对一些方法进行了比较。通过量度融合图像与源图像的交互信息,评价多传感器图像的融合效果具有简单、准确、抗干扰性好的特点。通过红外图像与可见光图像融合的实验,将交互信息量的评价方法与其他评价方法相比较,结果表明,交互信息量是一种很好的评价指标。

关 键 词:图像融合  评价准则  融合性能  交互信息量
文章编号:1007-2276(2004)02-0189-05
收稿时间:2003/5/14
修稿时间:2003年5月14日

Comparison of the methods of evaluating the performance of multi-sensor image fusion
WANG Hai-hui,PENG Jia-xiong,WU Wei Institute for Pattern Recognition and Artificial Intelligence,Huazhong University of Science & Technology,Wuhan ,China.Comparison of the methods of evaluating the performance of multi-sensor image fusion[J].Infrared and Laser Engineering,2004,33(2):189-193.
Authors:WANG Hai-hui  PENG Jia-xiong  WU Wei Institute for Pattern Recognition and Artificial Intelligence  Huazhong University of Science & Technology  Wuhan  China
Affiliation:WANG Hai-hui,PENG Jia-xiong,WU Wei Institute for Pattern Recognition and Artificial Intelligence,Huazhong University of Science & Technology,Wuhan 430074,China)
Abstract:In order to choose an objective evaluation criterion which is always consistent with the subject evaluation criteria in the complicated conditions, some criteria are analyzed and compared in this paper. Through calculating the mutual information between source images and fused images, the fusion performance are evaluated. This method has the characters of simple, precise and anti-jamming. Furthermore, infrared image and visible image are merged. Compared with other evaluation criteria methods, the experimental results show that mutual information is better.
Keywords:Image fusion  Evaluation criterion  Fusion performance  Mutual information
本文献已被 CNKI 维普 万方数据 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