Texture Feature Extraction and Segmentation of SAR Image Based on Over-Complete Wavelet

摘要：

Texture feature extraction of SAR image based on wavelet transform cannot guarantee translation invariant, which often leads to instability of the extracted feature. The disadvantage greatly affects the results of image segmentation. Compared with wavelet decomposition, the over-complete wavelet analysis doesnt down-sample the output of the filters. So it realizes the translation invariance by using redundancy information in the image. On the basis of texture extraction of SAR image, fuzzy clustering is adopted to the segmentation of SAR image according to its own characteristic, which is the diversity of grey-levels in SAR image. Experiments have proved the superiority of the method in SAR image segmentation.

作者: Rong Hong Lizhong Xu Min Tang Fengchen Huang

作者单位: College of Computer and Information Engineering Hohai University Nanjing, China, 210098

会议类型: 国际会议

会议名称: Fourth International Conference on Impulsive and Hybrid Dynamical Systems(ICIHDS 2007)(第四届国际脉冲和混合动力系统学术会议)

会议地点: 南宁

会议语种:英文

在线出版日期: 2007-07-20（万方平台首次上网日期，不代表论文的发表时间）