摘要
湿地植被制图是湿地自然资源管理过程中的一项重要任务。文中选取黄河口湿地为研究区,应用海岸带高光谱成像仪影像,提出了一种基于包络线去除的改进型光谱角匹配(Spectral Angle Mapper based on Continuum Removal,SAM_CR)方法,对芦苇、狄草、碱蓬和怪柳等湿地典型植被进行分类提取。结果表明,SAM_CR湿地典型植被种类识别的总体精度由传统SAM方法的74.87%提高到80.61%。
Vegetation mapping in wetland is an important task of wetland natural resource management.In this article,we selected the Yellow River Estuary Wetland as the study area,using Hyperspectral Imager for the Coastal Ocean image and we proposed a new method of Spectral Angle Matching(SAM_CR),which was based on the removal of envelope for typical wetland vegetation in Yellow River Estuary,including reeds,Spartina,Suaeda salsa and tamarisk.Studies have shown that the overall accuracy of SAM_CR compared with traditional SAM methods,increased from 74.87%to 80.61%.
作者
赵子飞
江涛
ZHAO Zifei;JIANG Tao(College of Geomatics,Shandong University of Science and Technology,Qingdao Shandong 266590,China)
出处
《北京测绘》
2018年第4期378-383,共6页
Beijing Surveying and Mapping
基金
国家自然科学基金项目(41471331
41601408)
关键词
湿地植被
黄河口
海岸带高光谱成像仪(HICO)
包络线去除
光谱角填图
wetland vegetation
Yellow River Estuary
Hyperspectral Imager for the Coastal Ocean(HICO)
continuum removal
spectral angle matching
