| 引用本文: | 张潇予,赵凤军,李宁. 基于多通道特征的SAR图像变化检测[J]. 雷达科学与技术, 2017, 15(5): 509-518.[点击复制] |
| ZHANG Xiaoyu, ZHAO Fengjun, LI Ning. Change Detection in SAR Images Based on Multi-Channel Features[J]. Radar Science and Technology, 2017, 15(5): 509-518.[点击复制] |
|
| 摘要: |
| 为充分利用SAR图像的细节信息，提高SAR图像变化检测的检测精度及抗噪性能，提出一种基于多通道特征的SAR图像变化检测方法。该方法提出了一种适用于SAR图像的变化检测一体化框架，首先，为了在抑制相干斑噪声的同时尽可能多地保留SAR图像的边缘及局部信息，引入引导图像滤波方法；其次，提取8个通道特征，充分利用了图像的细节信息，获得了性能良好的差异图；最后，利用主成分分析(PCA)和K-means聚类进行差异图分析，得到最终的变化信息。实验结果表明，该方法有效提高了检测精度，并且具有良好的抗噪性能。 |
| 关键词: 合成孔径雷达 变化检测 引导图像滤波 多通道特征 |
| DOI： |
| 分类号: |
| 基金项目:国家自然科学基金青年基金(No.61601437) |
|
| Change Detection in SAR Images Based on Multi-Channel Features |
|
ZHANG Xiaoyu, ZHAO Fengjun, LI Ning
|
|
1.(1.Institute of Electronics, Chinese Academy of Sciences, Beijing 100190, China;2. 2.University of the Chinese Academy of Sciences, Beijing 100190, China)
|
| Abstract: |
| In order to make full use of the detail information of SAR images and improve the detection accuracy and anti-noise performance of change detection, a method based on multi-channel features for change detection in SAR images is proposed. The method presents an integrative framework which is suitable for change detection in SAR images. Firstly, the guided image filtering is introduced to keep the edge and local information of SAR images as much as possible while suppressing speckle noise. Secondly, the features of 8 channels are extracted to make full use of the images detail information and generate a difference image(DI) with good performance. Finally, the principal component analysis(PCA) and K-means clustering are used to analyze DI to get better change detection results. The experimental results show that the proposed method can improve the detection accuracy effectively and has good anti-noise performance. |
| Key words: synthetic aperture radar(SAR) change detection guided image filtering multi-channel features |
