测绘通报 ›› 2021, Vol. 0 ›› Issue (7): 29-33.doi: 10.13474/j.cnki.11-2246.2021.0204

• 学术研究 • 上一篇    下一篇

无人机高分辨率影像的森林株数密度制图

郭伟1, 杨春宇2, 吴子若1, 季翔林1, 杨春洁1, 赵传武1, 张玉环3   

  1. 1. 中国矿业大学(北京)地球科学与测绘工程学院, 北京 100083;
    2. 北京市地质工程设计研究院, 北京 101500;
    3. 生态环境部卫星环境应用中心, 北京 100094
  • 收稿日期:2021-05-12 修回日期:2021-05-17 出版日期:2021-07-25 发布日期:2021-08-04
  • 通讯作者: 张玉环。E-mail:yuhuan_rs@163.com
  • 作者简介:郭伟(1984-),男,博士,讲师,主要研究方向为城市遥感、矿山生态。E-mail:guowei_rs@163.com
  • 基金资助:
    国家自然科学基金重点项目(41930650);中央高校基本科研业务费专项资金(2020XJDC03);国家对地观测科学数据中心开放基金(NODAOP2020016)

Forest density mapping based on UAV high-resolution image

GUO Wei1, YANG Chunyu2, WU Ziruo1, JI Xianglin1, YANG Chunjie1, ZHAO Chuanwu1, ZHANG Yuhuan3   

  1. 1. College of Geoscience and Surveying Engineering, China University of Mining and Technology-Beijing, Beijing 100083, China;
    2. Beijing Institute of Geological Engineering, Beijing 101500, China;
    3. Ministry of Ecology and Environment Center for Satellite Application on Ecology and Environment, Beijing 100094, China
  • Received:2021-05-12 Revised:2021-05-17 Online:2021-07-25 Published:2021-08-04

摘要: 本文以大兴安岭无人机遥感数据为基础进行森林密度制图,并提出了一种局部阈值算法,通过与传统的Otsu方法去除背景噪声相比,得出了该方法在中等或较低森林株数密度地区能很好地去除噪声背景。结合局部最大值法取得很好的单木提取精度,其查全率达到了100%。传统的Otsu去除背景方法在较高森林株数密度地区具有较好的识别效果,但对非林分的空地信息存在错提取的现象。通过对以上两种方法的对比研究,得到大兴安岭森林株数密度制图结果,该研究可为稀疏森林区域的株数密度制图提供参考。

关键词: 无人机遥感, 森林株数密度, 高斯滤波, 阈值分割, 局部最大值

Abstract: Based on the UAV remote sensing data of Daxing'anling Mountains, this paper proposes a local threshold algorithm for forest density mapping. Compared with the traditional Otsu method to remove the background noise, it is concluded that this method can remove the noise background well in the areas with medium or low forest density. Combined with the local maximum method, it has good single tree extraction accuracy, and its recall rate reaches 100%. The traditional Otsu background removal method has a good recognition effect in high forest density area, but its disadvantage is the wrong information extraction of non-forest open space. Through the comparative study of the above two methods, the mapping results of forest plant density in Daxing'an Mountains are obtained, which can provide a reference for the mapping of forest plant density in sparse forest areas.

Key words: UAV remote sensing, forest density, Gaussian filtering, threshold segmentation, local maximum

中图分类号: