文章基本信息

标题：A Fully Automated Procedure for Delineation and Classification of Forest and Non-Forest Vegetation Based on Full Waveform Laser Scanner Data
本地全文：下载
作者：C. Straub ; H. Weinacker ; B. Koch 等
期刊名称：ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
印刷版ISSN：2194-9042
电子版ISSN：2194-9050
出版年度：2008
卷号：XXXVII-B8
页码：1013-1020
出版社：Copernicus Publications
摘要：Detailed geo-referenced information on the distribution and occurrence of forest and non-forest vegetation is essential for many different disciplines e.g. forestry, nature conservation, agriculture, landscaping and urban planning. This article presents a digital image processing procedure for automated delineation and classification of forest and non-forest vegetation which is solely using full waveform laser scanner data as input. The delineation of regions covered by vegetation is based on the assumption that many laser reflections will be found inside of vegetation from different vegetation layers between the top of the canopy and the bare earth which particularly applies to multiple echoes from full waveform data. The vegetation regions are classified into forest and non- forest vegetation based on criteria which are generally used for vegetation mapping such as height of the vegetation, tree crown cover, size and width of vegetation objects. Non-Forest vegetation is further classified into single tree objects or connected groups of trees based on geometrical features. To verify the applicability for large areas the procedure was tested in a study site in the Southern Black Forest Nature Park, Germany with a total size of 7.68 km2. An accuracy assessment of the automated method is given with a comparison to a delineation and classification result done by a human operator based on RGB true-orthophotos and with a terrestrial survey. An error matrix was used to verify the classification result. An overall accuracy of 97.73% was reached. The capability and the limitations of the method are discussed
关键词：Laser scanning (LiDAR); Classification; Aerial Survey; Feature extraction; Environmental monitoring; Forestry