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 Part B3b
页码：273-276
出版社：Copernicus Publications
摘要：In forestry applications, the great challenge is to automatically extract as much information as possible on forest structure, the vertical and horizontal distribution of vegetation, the delineation of individual trees and identification of their species. Lidar data has great efforts on forest application because of its precisely 3d Geometry information. In the current papers, the methods mainly focus on individual tree's classification in urban areas or manage the forest area as a whole. In the original techniques, it is difficult to extract two or three trees which grow together. The new method proposed in this paper can solve the problem rightly, which is based on the gridding segmentation, which can both extract individual trees and classification. The method in this paper is also an automation process for vegetation further classification and it includes five steps: location, segmentation, statistic, analysis and further classification. In the test area, we should distinguish at the individual tree level between Conifer (pine trees) and Broad- leaved (poplars). While a clear distinction between these two species was not always visually obvious at the individual tree level, due to other extraneous sources of variation in the dataset, the observation was supported in general at the site level. Sites dominated by Conifer exhibited a lower proportion of singular returns compared to sites dominated by Broad-leaved, and the method can distinguish the sorts of the trees in the test area and each sort has a 70% correct classification
