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：2005
卷号：XXXVI-3/W24
出版社：Copernicus Publications
摘要：Laser scanning or LiDAR data are increasingly used in forestry applications but also e.g. in urban environments or for building reconstructions. Huge point clouds are usually converted to a grid or are pre-processed in specific software packages. In this paper we present a methodology to extract and delineate single trees from small footprint, high intensity laser scanning point data in a GIS environment. Additional image data are only used for visualisation purposes and for accuracy assessment. The objective was to demonstrate the potential of a fully GIS-based workflow. After various pre-processing steps within the GIS, we developed a local maxima algorithm to identify tree tops. Secondly, we developed a region growing algorithm to delineating the respective tree crowns. It utilizes the original laser point data and not a derived raster data set such as a DSM. The algorithm was tested for six test plots located within the National Park Bavarian Forest (Germany) which is considered a natural or near-natural forest. For these plots, the results of extensive field surveys were available. Dominant trees could be detected with an accuracy of 72.2% but the overall tree detection rate was 51%. Suboptimal scan sampling distribution hinders perfect tree crown delineation. Our main goal - to develop and demonstrate a complete GIS-based workflow from Laser data pre-processing, algorithm development, analysis, to visualisation etc. – was reached. However, locating and counting trees within the LiDAR point cloud, particularly in multi-tiered deciduous plots and juvenile stands, requires the assistance of field-validation data and some subjective interpretation
关键词：LiDAR; Laser scanning; region growing algorithm; GIS; identification; crown delineation
