文献类型：期刊文献

中文题名：基于高分二号影像的森林变化快速检测方法研究

英文题名：Study on Rapid Forest Change Detection Method Based on GF-2 Images

作者：冯林艳[1] 谭炳香[1] 王晓慧[1] 郝家田[2] 侯晓巍[2]

第一作者：冯林艳

机构：[1]中国林业科学研究院资源信息研究所,北京100091;[2]国家林业和草原局西北调查规划设计院,陕西西安710048

年份：2019

卷号：32

期号：5

起止页码：142-148

中文期刊名：林业科学研究

外文期刊名：Forest Research

收录：CSTPCD;;Scopus;北大核心:【北大核心2017】；CSCD:【CSCD2019_2020】；

基金：中央级公益性科研院所基本科研业务费专项资金(CAFYBB2017MB012);中国科学院A类先导专项(GrantNo.XDA19030500)

语种：中文

中文关键词：森林;GF-2;变化检测;NDVI;影像差值

外文关键词：Forest;GF-2;change detection;NDVI;image difference

分类号：S758.4

摘要：[目的]探讨用于快速更新森林资源数据库的森林变化检测方法,监测短时期内森林采伐与更新的动态变化。[方法]以变化频繁快速,高度集约经营的广西上思县人工林作为研究区,以两个时相的高分二号遥感影像作为数据源,分别利用红波段、近红外波段和NDVI 3种特征的影像差值,并基于分布函数确定阈值,对研究区进行快速的变化检测,并提取变化区域和变化类型。[结果]表明,3种特征差值的检测精度排序为:NDVI差值法最优,红波段差值法次之,近红外波段差值法最差。其中NDVI的总体精度为87.12%,Kappa系数为0.76,[结论]该方法在实现快速检测变化的目的下,可用于森林资源数据库的更新。
[Objective] To explore the forest change detection method used to update the forest resource database rapidly and to monitor the dynamic changes in forest harvesting and renewal in a short period.[Method] The research area was set in the plantation of Shangsi County, Guangxi Zhuang Autonomous Region, where the plantation area changed frequently and rapidly and was under integrated management. The GF-2 remote sensing images of two phases were used as data sources. The image difference values of the red band, near-infrared band and NDVI were also exploited respectively. The threshold was determined based on the distribution function. The changes in the research area were detected rapidly. The changing area and the change type were extracted based on the detection.[Result] The detection accuracy of the three feature differences are as follows: NDVI difference method showed the best, the red band difference method was the second, and the near-infrared band difference method was the worst. The overall accuracy of NDVI was 87.12%, and the Kappa coefficient was 0.76.[Conclusion] This method can be applied to quickly detecting the changes and can be used to update the forest resource database.