SAR image change detection based on Gabor wavelets and convolutional wavelet neural networks

• Published in: Multimedia tools and applications Vol. 82; no. 20; pp. 30895 - 30908
• Main Authors: Yi, Wen, Wang, Shijie, Ji, Nannan, Wang, Changpeng, Xiao, Yuzhu, Song, Xueli
• Format: Journal Article
• Language: English
• Published: New York Springer US 01.08.2023; Springer Nature B.V
• Subjects: Accuracy; Algorithms; Change detection; Classification; Clustering; Computer Communication Networks; Computer Science; Data Structures and Information Theory; Methods; Morlet wavelet; Multimedia; Multimedia Information Systems; Neural networks; Radar detection; Radar imaging; Special Purpose and Application-Based Systems; Synthetic aperture radar; Wavelet transforms; Synthetic aperture radar images; Convolutional wavelet neural networks; Change detection; Gabor wavelets; Fuzzy C-means clustering
• ISSN: 1380-7501; 1573-7721
• DOI: 10.1007/s11042-023-15106-5

Synthetic aperture radar (SAR) image change detection technology is of great significance. In the existing convolutional wavelet neural networks (CWNN) based SAR image change detection methods, the precision of preclassification is not high. The precision of preclassification will affect the performance of the network, and thus affect the accuracy of image change detection. In order to further improve the accuracy of change detection, the method based on Gabor wavelets and convolutional wavelet neural networks (GWCWNN) is applied to SAR image change detection in this paper. This method combines Gabor wavelets and fuzzy C-means clustering algorithm to provide high precision training samples for the networks, so as to improve the accuracy of image change detection. The results on three real data sets respectively show that the proposed method is better than the existing four methods.