欢迎访问林草资源研究

林业资源管理 ›› 2019›› Issue (5): 44-51.doi: 10.13466/j.cnki.lyzygl.2019.05.009

• 科学研究 • 上一篇    下一篇

基于XGBoost的高分二号影像树种识别

蔡林菲(), 吴达胜(), 方陆明, 郑辛煜   

  1. 1. 林业感知技术与智能装备国家林业和草原局重点实验室,浙江 临安 311300
    2. 浙江省林业智能监测与信息技术研究重点实验室,浙江 临安 311300
    3. 浙江农林大学 信息工程学院,浙江 临安 311300
  • 收稿日期:2019-09-18 修回日期:2019-10-20 出版日期:2019-10-28 发布日期:2020-09-18
  • 通讯作者: 吴达胜
  • 作者简介:蔡林菲(1995-),女,浙江临海人,在读硕士,主要研究方向为资源与环境信息系统。Email: 1527641144@qq.com
  • 基金资助:
    浙江省科技重点研发计划资助项目(2018C02013)

Tree Species Identification Using XGBoost Based on GF-2 Images

CAI Linfei(), WU Dasheng(), FANG Luming, ZHENG Xinyu   

  1. 1. Key Laboratory of Forestry Perception Technology and Intelligent Equipment,State Forestry and Grassland Administration,Linan,Zhejiang 311300,China
    2. Key Laboratory of Forestry Intelligent Monitoring and Information Technology Research,Linan,Zhejiang 311300,China
    3. School of Information Engineering,Zhejiang A&F University,Linan,Zhejiang 311300,China
  • Received:2019-09-18 Revised:2019-10-20 Online:2019-10-28 Published:2020-09-18
  • Contact: WU Dasheng

摘要:

树种分布是森林资源监测的一个重要指标,也是遥感影像在森林资源监测应用中的难点之一。基于国产高分二号卫星影像数据、森林资源二类调查数据、DEM数据,结合光谱、纹理、指数及地形因子等多种特征,比较支持向量机、随机森林和XGBoost等3种分类算法,根据分类精度选择最优算法(即XGBoost)进行特征筛选,对龙泉市的阔叶树、马尾松、杉木和毛竹等4种主要优势树种进行分类。结果表明:采用XGBoost分类模型的分类总精度为83.88%,Kappa系数0.78,较支持向量机和随机森林分类方法有明显提高。经特征选择后,虽未明显提高树种分类精度,但可以减少特征的冗余,为小样本数据下特征的选取降维提供了一定的参考。

关键词: 高分二号, 特征选择, XGBoost, 树种分类, 遥感

Abstract:

Tree species distribution is an important indicator of forest resources monitoring,and it is also one of the difficulties in the application of remote sensing images in forest resources monitoring.Based on domestic GF-2 satellite imagery data,forest resources survey data and DEM data,combined with spectral,texture,index and topographic factors and other characteristics.Three classification algorithms,support vector machine,random forest and XGBoost,are compared,and the optimal algorithm (XGBoost) is selected according to classification accuracy.Four dominant tree species,broad-leaved trees,Pinus massoniana,Chinese fir and Phyllostachys pubescens,were classified by row feature selection.The results show that the total classification accuracy of XGBoost classification model is 83.58% and the Kappa coefficient is 0.77,which is significantly higher than that of support vector machine and random forest classification method.After feature selection,the classification accuracy of tree species has not been improved obviously,but the redundancy of features can be reduced,which provides a reference for feature selection and dimensionality reduction in small sample data.

Key words: GF-2, feature selection, XGBoost, tree species identification, remote sensing

中图分类号: