文献类型：期刊文献

中文题名：基于高分二号多光谱数据的树种识别方法

英文题名：Tree Species Identification Method Based on GF-2 Images

作者：尹凌宇[1] 覃先林[1] 孙桂芬[1] 祖笑锋[1] 陈小中[2]

第一作者：尹凌宇

机构：[1]中国林业科学研究院资源信息所;[2]四川省林业信息中心

年份：2016

卷号：0

期号：4

起止页码：121-127

中文期刊名：林业资源管理

外文期刊名：Forest Resources Management

收录：北大核心:【北大核心2014】；

基金：国防科工局重大专项项目(21-Y30B05-9001-13/15);民用航天预研项目"基于多源空间数据的森林火灾综合监测技术与应用示范"

语种：中文

中文关键词：树种识别;高分二号;最大似然法;支持向量机法;GF;2

外文关键词：identification of tree species;maximum likelihood method;support vector machine

分类号：TP79;S757.2

摘要：树种识别一直是困扰遥感研究的一个难点,而国产高分二号识别地物和树种具有巨大潜力。选取四川省甘孜州道孚县为研究区,利用高分二号4m多光谱遥感影像,并结合该县的森林资源二类调查结果数据,分别采用最大似然法和支持向量机方法,对利用高分二号数据在树种识别应用中的可能性进行探讨。研究结果表明:所采用的两种方法识别出研究区域主要树种的精度都高于80%,其中:采用最大似然法分类精度为81.79%,支持向量机方法分类精度为86.75%。在先验知识的支持下,利用高分二号多光谱影像也可用于树种识别研究中。
Identification of tree species has always challenged the remote sensing research. However,GF-2 images be used for identifying ground objects and classifying tree species have the great research potential. Based on the data of GF-2 4-meter multispectral images of Daofu County of Ganzi Prefecture in Sichuan Province were combined with Forest Resource Inventory Data. The maximum likelihood classification and support vector machine （SVM）method were used for classification of trees. The possibility of using the GF-2 data for species identification applications was explored. The results show that the two methods for identifying the main tree species are better than the accuracy of 80%in the study area. The maximum like-lihood classification accuracy is 81. 79%,SVM classification accuracy 86. 75%. With the support of prior knowledge,GF-2 multispectral images can also be used to study the species identification.