Editor/Reviewer login
Frequency : 12 issues per year
Subject : Computer Applications and Technology
ISSN : 2319–8656 (Online)
Home Archives Volume 10 Issue 4 ijcatr10041002

IJCATR Volume 10 Issue 4

SAR Image Change Detection Based on Multi-scale Feature Extraction

Xiaoqian Yuan, Chao Chen, Shan Tian, Jiandan Zhong
10.7753/IJCATR1004.1002
keywords : SAR image change detection; fuzzy c-means clustering; multi-scale feature

PDF
In order to improve the contrast of the difference image and reduce the interference of the speckle noise in the synthetic aperture radar (SAR) image, this paper proposes a SAR image change detection algorithm based on multi-scale feature extraction. In this paper, a kernel matrix with weights is used to extract features of two original images, and then the logarithmic ratio method is used to obtain the difference images of two images, and the change area of the images are extracted. Then, the different sizes of kernel matrix are used to extract the abstract features of different scales of the difference image. This operation can make the difference image have a higher contrast. Finally, the cumulative weighted average is obtained to obtain the final difference image, which can further suppress the speckle noise in the image.
@artical{x1042021ijcatr10041002,
Title = "SAR Image Change Detection Based on Multi-scale Feature Extraction",
Journal ="International Journal of Computer Applications Technology and Research(IJCATR)",
Volume = "10",
Issue ="4",
Pages ="77 - 81",
Year = "2021",
Authors ="Xiaoqian Yuan, Chao Chen, Shan Tian, Jiandan Zhong"}
  • The paper proposes a multi-scale feature extraction based method to generate difference image
  • Modified FCM algorithm is used to obtain classification results
  • The influence of window size on the detection results is studied
  • Two data sets are used to verify the effectiveness of the proposed algorithm.
Copyright © 2012-2024 International Journal of Computer Applications Technology and Research All Rights Reserved.