文献类型：期刊文献

中文题名：基于遥感技术的森林树种识别研究进展

英文题名：Research Progress of Forest Tree Species Identification Based on Remote Sensing Technology

作者：李华玉[1,2] 陈永富[2] 陈巧[2] 王娟[1,2] 张超[1]

第一作者：李华玉

机构：[1]西南林业大学林学院,云南昆明650224;[2]中国林业科学研究院资源信息研究所,国家林业和草原局林业遥感与信息技术重点实验室,北京100091

年份：2021

卷号：36

期号：6

起止页码：220-229

中文期刊名：西北林学院学报

外文期刊名：Journal of Northwest Forestry University

收录：CSTPCD;;北大核心:【北大核心2020】；CSCD:【CSCD_E2021_2022】；

基金：中央级公益性科研院所基本科研业务费专项资金(CAFYBB2018SZ008)。

语种：中文

中文关键词：树种识别;遥感技术;无人机遥感;深度学习

外文关键词：tree species identification;remote sensing technology;UAV remote sensing;deep learning

分类号：S771.5

摘要：基于遥感手段的森林类型/树种(组)的精准识别是森林参数提取和计算的前提,是林业遥感领域的研究前沿,可为宏观尺度快速获取森林资源信息提供重要途径。对多源遥感数据在森林树种识别中的应用研究进行总结分析发现,当前基于遥感数据的树种识别已成为林业遥感的研究热点。利用卫星遥感数据和近地低空遥感数据结合随机森林、支持向量机等分类方法进行树种识别已相对成熟。近年来,无人机技术不断发展,以其灵活性强的特点在林业中有较强的应用潜力,在计算机技术、数字图像处理技术和机器学习领域不断发展成熟背景下,低空遥感数据结合深度学习技术应用于森林树种的精确识别是值得深入研究的科学问题。
The accurate identification of forest types/tree species(groups)based on remote sensing is the premise of forest parameter extraction and calculation,and is the research frontier in the field of forest remote sensing,which can provide an important way to obtain forest resource information at macro scale.By summarizing and analyzing the application of multi-source remote sensing data in forest tree species identification,it is found that tree species identification based on remote sensing data has become a hot research topic in forest remote sensing.Now,using satellite remote sensing data and near-Earth low-altitude remote sensing data combined with Random Forest,Support Vector Machine and other classification methods for tree species identification have been relatively mature.In recent years,unmanned aerial vehicle(UAV)technology has been well developed,presenting the strong application potential in forestry for its flexibility.Under the background of continuous development and maturity in the field of computer technology,digital image processing technology and machine learning,the application of low altitude remote sensing data combined with depth learning technology to the accurate identification of forest tree species is a scientific problem worthy of further study.