典型滨海湿地植被全极化SAR显著性特征分析与融合分类  

Analysis and Classification of Full-Polarization SAR Characteristics of Typical Vegetation in Coastal Wetland

作　　者：万剑华[1] 张乃心 任广波[2] 刘善伟[1] WAN Jianhua;ZHANG Naixin;REN Guangbo;LIU Shanwei(School of Geoscience,China University of Petroleum,Qingdao 266580,China;First Institute of Oceanography,State Oceanic Administration,Qingdao 266061,China)

机构地区：[1]中国石油大学(华东)地球科学与技术学院,山东青岛266580 [2]国家海洋局第一海洋研究所,山东青岛266061

出　　处：《测绘地理信息》2018年第5期4-8,共5页Journal of Geomatics

基　　金：国家重点研发计划资助项目(2017YFC405600);国家自然科学基金资助项目(41776182)

摘　　要：以黄河口湿地作为研究区域,选取2016年9月14日的Radarsat-2全极化SAR（synthetic aperture radar）数据,基于Freeman-Durden分解法,分别分析了互花米草、翅碱蓬、芦苇这三种典型湿地植被在不同极化分解特征通道上的信号响应差异,结果表明地物在不同分解通道中具有显著差异性,可用于增益地物分类。此外,还根据分析结果对极化分解特征进行融合,通过比较分析发现融合图像在总体地物可分性略高于同分辨率光学影像,在个别地物可分性上显著优于同分辨率光学影像。In this paper,we selected the Radarsat-2 full-polarization SAR of Yellow River estuary wetland on September 14,2016.We got three kinds of polarization components by Freeman-Durden polarimetric decomposition and analyzed signal response difference of Phragmites australis,Spartina alterniflora,and Suaeda salsa in different polarization components.The results showed that the characteristics in different decomposition channels have significant differences,which can be used to improve classification accuracy of ground objects.In addition,we integrated the significant characteristics of polarization decomposition according to the analytical results.Through comparative analysis,we found that the overall accuracy of integrated image is slightly higher than the same resolution optical images,the accuracy of Spartina alterniflora and Suaeda salsa of integrated image is significantly better than the optical image.

关 键 词：全极化SAR 滨海湿地 极化分解 特征分析 分类 

分 类 号：TP751[自动化与计算机技术—检测技术与自动化装置]