摘要
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。
In this article, We design a combining physical scattering mechanism and statistical charac- teristic of polarization SAR unsupervised classification algorithm. Because the method based on physical scattering characteristics unable to distinguish between fine category differences and keep the resolu- tion. This algorithm is unsupervised classification combying H/A/α with wishart, this method is based on the physical mechanism of scattering and statistical characteristic, the California Camp Roberts region of JPL AIRSAR data as an example, selecting 900×900 pixels size of the study area, classification results show that the algorithm is availably in keeping the resolution and distinguish the subtle differences in the category of the works.
出处
《测绘科学》
CSCD
北大核心
2017年第9期94-97,120,共5页
Science of Surveying and Mapping
基金
2014年滁州学院规划项目(2014GH02)
2016年虚拟地理环境教育部重点实验室开放基金资助项目(2015VGE01)
2015年博士科研启动基金项目(2015qd06)
