Automatic Change Detection of High Resolution Remote Sensing Images Based on Level Set Evolution and Support Vector Machine Classification

We propose a method for change detection in highresolution remote sensing images by means of level set evolution and Support Vector Machine (SVM) classification, which combined both pixellevel method and objectlevel method Both pixelbased change features and objectbased ones are extracted to improve...

Full description

Saved in:
Bibliographic Details
Main Authors: YAN Ming, CAO Guo, XIA Meng
Format: Article
Language:zho
Published: Harbin University of Science and Technology Publications 2019-02-01
Series:Journal of Harbin University of Science and Technology
Subjects:
Online Access:https://hlgxb.hrbust.edu.cn/#/digest?ArticleID=1640
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:We propose a method for change detection in highresolution remote sensing images by means of level set evolution and Support Vector Machine (SVM) classification, which combined both pixellevel method and objectlevel method Both pixelbased change features and objectbased ones are extracted to improve the discriminability between the changed class and the unchanged classAt the pixellevel, the change detection problem is formulated as a segmentation issue using level set evolution in the difference image At the objectlevel, potential training samples are selectedfrom the segmentation results without manual intervention into SVM classifier Thereafter, the final changes are obtained by combining the pixelbased changes and the objectbased changes A chief advantage of our approach is being able to select appropriate samples for SVM classifier training Furthermore, our proposed method helps improving the accuracy and the degree of automation We systematically evaluated it with a variety of SPOT5 images and aerial images Experimental results demonstrated the accuracy of our proposed method
ISSN:1007-2683