| 一种结合散射相似性和Wishart的极化SAR图像分类方法 |
| |
| 引用本文： | 贺婷婷,谭维贤,黄平平,徐伟,胡楚锋. 一种结合散射相似性和Wishart的极化SAR图像分类方法[J]. 信号处理, 2019, 35(5): 904-910. DOI: 10.16798/j.issn.1003-0530.2019.05.023 |
| |
| 作者姓名： | 贺婷婷 谭维贤 黄平平 徐伟 胡楚锋 |
| |
| 作者单位： | 内蒙古工业大学信息工程学院 |
| |
| 基金项目： | 国家自然科学基金重点项目(61631011)；国家自然科学地区基金项目(61661043)；内蒙古自然科学基金面上项目(2016MS0606)
|
| |
| 摘 要： | 针对基于H-a 平面的目标分类方法运算量偏大的缺点，本文提出了一种结合散射相似性和Wishart的极化合成孔径雷达图像分类方法，提升了数据处理效率。该方法使用参数替代，将目标相干矩阵进行归一化，得到和 H-a 相似的分类效果，且避免计算特征值和特征向量，从而大幅提高了分类效率；通过结合Wishart迭代分类器进行分类来提升分类精度。最后通过对Radarsat-2卫星获取的黄河冰凌区域进行地物分类实验验证了本文方法的有效性，且该方法具有更好的运算效率和地物分类效果。
|
| 关 键 词： | 极化合成孔径雷达 地物分类 散射相似性 Wishart |
| 收稿时间： | 2019-01-02 |
A Method of Polarimetric SAR Image Classification Combined Scattering Similarity with Wishart |
| |
| Affiliation: | College of Information Engineering, Inner Mongolia University of TechnologyInner Mongolia Key Laboratory of Radar Technology and Application |
| |
| Abstract: | Aiming at the deficiency of target classification method based on H-a, such as large computation , in this paper, an algorithm combined scattering similarity with wishart for the images classification of the polarimetric SAR, which greatly saves the time of data processing. A method of parameter substitution by normalizing target coherence matrix, a classification method similar to H-a is obtained, which avoids calculating eigenvalues and eigenvectors and improves the classification efficiency. In order to further improve the classification effect, Wishart iterated classifier is combined for classify. Finally, the Yellow River ice cream area obtained by the Radarsat-2 satellite is classified to verify the effectiveness of the method, the proposed method can improve computational efficiency and improve classification effect of ground features. |
| |
| Keywords: | |
|
| 点击此处可从《信号处理》浏览原始摘要信息 |
|
点击此处可从《信号处理》下载免费的PDF全文 |
