{"title":"Integrated Change Detection Method for Landslide Monitoring","authors":"S. Khairunniza-Bejo, Siti Rusaniza Jusoh","doi":"10.1109/ICSAP.2009.10","DOIUrl":null,"url":null,"abstract":"This paper presents change detection method forlandslide monitoring using image registration and changeconditional statement approach. It is based on the invocationof deformation operators which imitate the deformationsexpected to be observed when a landslide occurs. Thesimilarity between two images is measured by a similarityfunction which takes into consideration value of mutualinformation, geometric deformation and maximumoverlapping area between the two images. A map of landslideareas is created by using the amount of pixel movement duringa registration process. This map is then thresholded in order toproduce map of change and unchanged areas. The conditionallogic statement of change and unchanged conditions isimplemented in order to produce the same type of map. Bothmaps are then compared for its consistency. From the results,it has been shown that both maps give almost similar locationof change and unchanged location with less than 4% ofmistmatch locations.","PeriodicalId":176934,"journal":{"name":"2009 International Conference on Signal Acquisition and Processing","volume":"1 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2009-04-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2009 International Conference on Signal Acquisition and Processing","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICSAP.2009.10","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
This paper presents change detection method forlandslide monitoring using image registration and changeconditional statement approach. It is based on the invocationof deformation operators which imitate the deformationsexpected to be observed when a landslide occurs. Thesimilarity between two images is measured by a similarityfunction which takes into consideration value of mutualinformation, geometric deformation and maximumoverlapping area between the two images. A map of landslideareas is created by using the amount of pixel movement duringa registration process. This map is then thresholded in order toproduce map of change and unchanged areas. The conditionallogic statement of change and unchanged conditions isimplemented in order to produce the same type of map. Bothmaps are then compared for its consistency. From the results,it has been shown that both maps give almost similar locationof change and unchanged location with less than 4% ofmistmatch locations.