论文部分内容阅读
针对仅利用物理散射特征分类方法无法实现精细类别差异区分和分辨率保持的问题,该文设计了一种将物理散射机制和统计特征相结合的极化SAR非监督分类算法。该算法采用加利福尼亚州Camp Roberts地区的JPL AIRSAR数据,通过H/A/珔α和Wishart分类相结合的非监督分类方法,对900像素×900像素大小的研究区进行了分类,分类结果表明,该算法在保持分辨率及区分精细类别差异方面是有效的。
Aiming at the problem that the classification of fine categories can not be achieved by using only the classification method of the physical scattering features, the paper designs a non-supervised classification algorithm of polarimetric SAR combining physical scattering mechanism and statistical features. The algorithm uses JPL AIRSAR data from Camp Roberts, California, and classifies the study area of 900 pixels × 900 pixels by the unsupervised classification method combining H / A / 珔α and Wishart classification. The classification results show that the Algorithms are effective in maintaining resolution and differentiating fine classifications.