期刊
URBAN FORESTRY & URBAN GREENING
卷 13, 期 3, 页码 562-570出版社
ELSEVIER GMBH
DOI: 10.1016/j.ufug.2014.03.005
关键词
Individual tree detection; LiDAR; Random Forests; Tree registers; Urban forest
资金
- Finnish Cultural Foundation
This case study describes a method for utilizing leaf-off airborne laser scanning (ALS) data for mapping characteristics of urban trees. ALS data were utilized to detect and update all street trees in the tree inventory of the City of Helsinki, Finland. The inventory consists of roughly 20,000 street trees with mean diameter at breast height (DBH) of 24 cm and mean height of 10.6 m. The large number of trees makes the manual updating process very laborious. The automatic mapping procedure presented in this paper detected 88.8% of all trees in the inventory. Tree height was predicted with root mean square error (RMSE) of 1.27 meters and-tree DBH with RMSE of 6.9 cm. The presented method provides a practical and cost-effective tool for the mapping of urban tree characteristics. The cost-efficiency was further enhanced because the used ALS data were originally collected for other urban planning purposes. (C) 2014 Elsevier GmbH. All rights reserved.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据