Guangting Li, Xin Zhang, Shikang Nie, Yibo Chen, Chenchen Lin, Yifeng He
{"title":"Road Extraction in SAR Iimage Based on the Image Dynamics and Watershed Transformation","authors":"Guangting Li, Xin Zhang, Shikang Nie, Yibo Chen, Chenchen Lin, Yifeng He","doi":"10.1109/icisfall51598.2021.9627474","DOIUrl":null,"url":null,"abstract":"In this paper, basing on the region character presented by the width of the road in SAR image, a new road extraction frame, combing the image dynamics with watershed transformation, is constructed. Firstly, the image dynamics calculation is researched, and a fast calculation of ridge dynamics, which is performed by uniting the one dimension dynamics of different directions, is proposed. Then, the ridge dynamics is used to extract the road seeds. Finally, the regions, which belong to the results of watershed transformation and contain the road seeds, are merged for road extraction. Both the bright lines and the dark lines in SAR images are extracted and constitute the comparatively integrated road net, which illustrates the effectiveness of the proposed method.","PeriodicalId":240142,"journal":{"name":"2021 IEEE/ACIS 20th International Fall Conference on Computer and Information Science (ICIS Fall)","volume":"4 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2021-10-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2021 IEEE/ACIS 20th International Fall Conference on Computer and Information Science (ICIS Fall)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/icisfall51598.2021.9627474","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
In this paper, basing on the region character presented by the width of the road in SAR image, a new road extraction frame, combing the image dynamics with watershed transformation, is constructed. Firstly, the image dynamics calculation is researched, and a fast calculation of ridge dynamics, which is performed by uniting the one dimension dynamics of different directions, is proposed. Then, the ridge dynamics is used to extract the road seeds. Finally, the regions, which belong to the results of watershed transformation and contain the road seeds, are merged for road extraction. Both the bright lines and the dark lines in SAR images are extracted and constitute the comparatively integrated road net, which illustrates the effectiveness of the proposed method.