首页> 外文期刊>Forest Science >Identification of Scandinavian Commercial Species of Individual Trees from Airborne Laser Scanning Data Using Alpha Shape Metrics
【24h】

Identification of Scandinavian Commercial Species of Individual Trees from Airborne Laser Scanning Data Using Alpha Shape Metrics

机译:使用Alpha形状度量从机载激光扫描数据中识别单个树木的斯堪的纳维亚商业物种

获取原文
获取原文并翻译 | 示例
       

摘要

Airborne laser scanning (ALS) data are not usually considered to be very informative with respect to tree species, and this information is often obtained by combining such data with spectral image material. The aim here was to test the ability of variables derived solely from ALS data to describe the crown shape and structure characteristics required for tree species discrimination. For that purpose, we constructed tree crown approximations from the three-dimensional return data by applying a computational geometry approach, the alpha shape concept, and developed metrics for describing them. We examined the ability of these metrics to classify Scandinavian commercial species (pine, spruce, and deciduous trees) by means of linear discriminant analysis and compared the alpha shape metrics with groups of ALS-based height, density, intensity, and two-dimensional texture variables. For evaluating the classification accuracy, we used a test data set composed of 92 dominant or codominant trees detected and delineated manually from ALS data with a density of approximately 40 returns m^sup -2^. The alpha shape metrics proved capable of discriminating between all three species classes evaluated, and several height distribution and texture variables were found to discriminate between the coniferous tree species. An overall accuracy of approximately 95% and a κ coefficient of 0.90 was achieved using a combination of the variables. Because this initial application of the alpha shape metrics was carried out using a test data set on only such trees that are clearly detectable from remotely sensed data, further research is required to apply the approach presented here within stands with a continuous canopy. Furthermore, as the tree observations considered here were highly biased toward mature coniferous trees, experiments using more representative data sets are needed to generalize the result obtained. [PUBLICATION ABSTRACT]
机译:机载激光扫描(ALS)数据通常不被认为对树种很有帮助,而该信息通常是通过将此类数据与光谱图像材料结合而获得的。此处的目的是测试仅从ALS数据得出的变量描述树种区分所需的树冠形状和结构特征的能力。为此,我们通过应用计算几何方法,alpha形状概念并根据描述这些度量的度量标准,从三维返回数据构造树冠近似值。我们检查了这些指标通过线性判别分析对斯堪的纳维亚商业物种(松树,云杉和落叶树)进行分类的能力,并将alpha形状指标与基于ALS的高度,密度,强度和二维纹理的组进行了比较变量。为了评估分类的准确性,我们使用了由ALS数据手动检出并描绘出的92棵优势树或共轭树组成的测试数据集,密度约为40个返回m ^ sup -2 ^。事实证明,alpha形状度量能够区分所评估的所有三个物种类别,并且发现了几个高度分布和纹理变量来区分针叶树物种。使用这些变量的组合可获得大约95%的整体精度和0.9的κ系数。由于最初仅使用可以从遥感数据清楚地检测到的树木上的测试数据集来进行alpha形状度量的这种初始应用,因此需要进一步研究以在具有连续天篷的看台上应用此处介绍的方法。此外,由于此处考虑的树木观测结果高度偏向成熟的针叶树,因此需要使用更具代表性的数据集进行实验以概括得出的结果。 [出版物摘要]

著录项

  • 来源
    《Forest Science》 |2009年第1期|p.37-47|共11页
  • 作者单位

    Jari Vauhkonen, University of Joensuu, Faculty of Forest Sciences, P.O. Box 111, Joensuu 80101, Finland - Phone: 358132514519, jari.vauhkonen@joensuu.fi. Timo Tokola, University of Joensuu, Faculty of Forest Sciences- timo.tokola@joensuu.fi. Petteri Packalén, University of Joensuu, Faculty of Forest Sciences-petteri.packalen@joensuu.fi. Marti Maltamo, University of Joensuu, Faculty of Forest Sciences-matti.maltamo@joensuu.fi.Acknowledgments: We thank Academy of Finland (project 12193: High Resolution Remote Sensing Potential to Measure Single Trees and Site Quality) for the financial support. Destia Ltd. (Mr. Tauno Suominen) for providing the airborne laser scanning data, Dr. Ilkka Korpela for the advice with the field measurements, and Dr. Lauri Mehtätalo and three anonymous reviewers for their comments on the manuscript.Manuscript received November 28, 2007, accepted October 8, 2008 Copyright © 2009 by the Society of American Foresters,;

  • 收录信息 美国《科学引文索引》(SCI);美国《工程索引》(EI);
  • 原文格式 PDF
  • 正文语种 eng
  • 中图分类
  • 关键词

  • 入库时间 2022-08-17 13:45:59

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号