文献类型：期刊文献

中文题名：多时相双极化SAR影像林地类型分类方法

英文题名：Multi-Temporal and Dual-Polarization SAR for Forest Land Type Classification

作者：王馨爽[1,2] 陈尔学[1] 李增元[1] 姚顽强[3] 赵磊[1]

第一作者：王馨爽

机构：[1]中国林业科学研究院资源信息研究所;[2]国家测绘地理信息局陕西基础地理信息中心;[3]西安科技大学测绘科学与技术学院

年份：2014

卷号：50

期号：3

起止页码：83-91

中文期刊名：林业科学

外文期刊名：Scientia Silvae Sinicae

收录：CSTPCD;;Scopus;北大核心:【北大核心2011】；CSCD:【CSCD2013_2014】；

基金：863计划课题"高分辨率SAR遥感综合实验与应用示范"(2011AA120405);"面向对象SAR影像地物高可信解译技术"(2011AA120404)

语种：中文

中文关键词：多时相;极化;InSAR;林地类型;分类

外文关键词：muhi-temporal; polarization ; InSAR ; forest land type; classification

分类号：S757

摘要：森林在全球碳循环及大自然空气调节中发挥着重要作用,对森林分布的监测与制图意义重大。以黑龙江省逊克县两景星载ALOS PALSAR多时相数据覆盖区为研究区,利用不同时相极化SAR、干涉SAR对植被结构变化特征的敏感性,结合后向散射系数与干涉相干性的时变特征进行林地类型分类研究,发展基于多时相、多极化、干涉SAR的SVM林地类型分类识别方法。结果表明:多时相的平均干涉相干性对有林地、疏林地及灌木林地的识别十分有效;综合运用所选取的多时相干涉、极化比等有效维度信息能很好地突出地物的边缘与结构,更细致地区分不同林地类型。
Forest plays an important role on global carbon cycle and nature disturbance, so it is of great significance to monitor and map forest resources. Xunke County of Heilongjiang Province was selected as the test site and the coverage of two scenes of ALOS PALSAR images were applied. The sensitivity of multi-temporal PolSAR, InSAR with forest structure variation and time varying characteristics of backscattering coefficients and interferometrie coherences were applied for forest land type classification. We developed a forest land type classification method based on SVM using multi-temporal, dual-polarization and interferometrie SAR （InSAR） data. The result showed that the average InSAR coherence of multi- temporal could effectively identify forest land, sparse forest land and shrub land. The muhi-temporal InSAR coherence, polarization ratios and other effective dimension information selected could effectively highlight the features and structures of objects and classify forest land types in detail.