{"title":"A New Image Registration Algorithm Based on Mutual Information of Gradient and Improved PSO","authors":"Laiming Xu, Jun Liu","doi":"10.1109/ISDEA.2012.389","DOIUrl":null,"url":null,"abstract":"Because of the larger difference between the two images to be registered, such as exposure degree or noise pollution, the common registration algorithm that based on mutual information of original gray, is easy to fall into local extreme value, thus the registration accuracy is unsatisfactory. In this case, a new method of image registration algorithm that based on mutual information of gradient and improved PSO is developed in this paper. The experiment result shows the method works well, and comparing with traditional registration algorithm, the registration accuracy is significantly improved while the efficiency is also guaranteed.","PeriodicalId":267532,"journal":{"name":"2012 Second International Conference on Intelligent System Design and Engineering Application","volume":"20 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2012-01-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2012 Second International Conference on Intelligent System Design and Engineering Application","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ISDEA.2012.389","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 1
Abstract
Because of the larger difference between the two images to be registered, such as exposure degree or noise pollution, the common registration algorithm that based on mutual information of original gray, is easy to fall into local extreme value, thus the registration accuracy is unsatisfactory. In this case, a new method of image registration algorithm that based on mutual information of gradient and improved PSO is developed in this paper. The experiment result shows the method works well, and comparing with traditional registration algorithm, the registration accuracy is significantly improved while the efficiency is also guaranteed.