Change detection in high resolution SAR images based on multi-scale texture features
This paper studied on change detection algorithm of high resolution (HR) Synthetic Aperture Radar (SAR) images based on multi-scale texture features. Firstly, preprocessed multi-temporal Terra-SAR images were decomposed by 2-D dual tree complex wavelet transform (DT-CWT), and multi-scale texture features were extracted from those images. Then, log-ratio operation was utilized to get difference images, and the Bayes minimum error theory was used to extract change information from difference images. Lastly, precision assessment was done. Meanwhile, we compared with the result of method based on texture features extracted from gray-level cooccurrence matrix (GLCM). We had a conclusion that, change detection algorithm based on multi-scale texture features has a great more improvement, which proves an effective method to change detect of high spatial resolution SAR images.
High resolution SAR multi-scale texture feature wavelet transform change detection
Caihuan. Wen Ziqiang. Gao
Hebei Institute of Surveying and Mapping of Geological Prospecting Bureau，Langfang 065000;
国际会议
桂林
英文
1-7
2011-11-01（万方平台首次上网日期，不代表论文的发表时间）