Classifying tree and nontree echoes from airborne laser scanning in the forest-tundra ecotone

被引:0
作者
Stumberg, Nadja [1 ]
Orka, Hans Ole [1 ]
Bollandsas, Ole Martin [1 ]
Gobakken, Terje [1 ]
Naesset, Erik [1 ]
机构
[1] Norwegian Univ Life Sci, Dept Ecol & Nat Resource Management, N-1432 As, Norway
关键词
INDIVIDUAL TREES; LIDAR DATA; CLASSIFICATION; INTENSITY; MODEL; AGREEMENT; MIGRATION; DISTANCE; COVER; STAND;
D O I
暂无
中图分类号
TP7 [遥感技术];
学科分类号
081102 ; 0816 ; 081602 ; 083002 ; 1404 ;
摘要
Temperature-sensitive ecosystems such as the forest-tundra ecotone are expected to be particularly affected by the changing climate. A large proportion of the total land area in Norway is represented by the forest-tundra ecotone, and effective monitoring techniques for these areas are required. Airborne laser scanning (ALS) has been proposed for detection of small pioneer trees and its height and intensity data may hold potential for monitoring tasks. The main objective of the present study was to assess the capability of high-density ALS data (e. g., >2 m(-2)) to classify tree and nontree echoes directly from the laser point cloud. For this purpose, the laser height and intensity, a geospatial variable represented by the area of Voronoi polygons, and the terrain variables of aspect and slope were used to distinguish between tree and nontree laser echoes along a 1000 km transect stretching from northern Norway (66 degrees 19' N) to the southern part of the country (58 degrees 3' N). Generalised linear models (GLM) and support vector machines (SVM) were employed for the classification using different combinations of the aforementioned variables. Total accuracy and the Cohen's kappa coefficient were used for performance assessment for the different models. A total accuracy of at least 93% was found irrespective of classification method or model, and Cohen's kappa coefficients indicated moderate fits for all models using both classification methods. Comparisons of Cohen's kappa coefficients revealed equivalent performances for the GLM and SVM classification methods for models consisting of different combinations of the laser height, intensity, geospatial variable, and aspect. However, SVM was superior when laser height and intensity were used together with slope. In summary, the capability of high-density ALS data for the classification of tree and nontree echoes directly from the laser point cloud could be verified irrespective of the classification method.
引用
收藏
页码:655 / 666
页数:12
相关论文
共 50 条
[41]   Characterizing riparian vegetation and classifying riparian extent using airborne laser scanning data [J].
Stackhouse, Leanna A. ;
Coops, Nicholas C. ;
White, Joanne C. ;
Tompalski, Piotr ;
Hamilton, Jeffery ;
Davis, Donald J. .
ECOLOGICAL INDICATORS, 2023, 152
[42]   Individual tree detection from airborne laser scanning data based on supervoxels and local convexity [J].
Ramiya, Anandakumar M. ;
Nidamanuri, Rama Rao ;
Krishnan, Ramakrishnan .
REMOTE SENSING APPLICATIONS-SOCIETY AND ENVIRONMENT, 2019, 15
[43]   Individual tree detection using template matching of multiple rasters derived from multispectral airborne laser scanning data [J].
Huo, Langning ;
Lindberg, Eva .
INTERNATIONAL JOURNAL OF REMOTE SENSING, 2020, 41 (24) :9525-9544
[44]   Bayesian Approach to Tree Detection Based on Airborne Laser Scanning Data [J].
Lahivaara, Timo ;
Seppanen, Aku ;
Kaipio, Jari P. ;
Vauhkonen, Jari ;
Korhonen, Lauri ;
Tokola, Timo ;
Maltamo, Matti .
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2014, 52 (05) :2690-2699
[45]   Mapping LAI in a Norway spruce forest using airborne laser scanning [J].
Solberg, Svein ;
Brunner, Andreas ;
Hanssen, Kjersti Holt ;
Lange, Holger ;
Naesset, Erik ;
Rautiainen, Miina ;
Stenberg, Pauline .
REMOTE SENSING OF ENVIRONMENT, 2009, 113 (11) :2317-2327
[46]   Airborne laser scanning performance analysis in the forest stand map production [J].
Sefercik, Umut Gunes ;
Atesoglu, Ayhan ;
Atalay, Can .
GEOMATIK, 2021, 6 (03) :179-188
[47]   Assessment of Low Density Full-Waveform Airborne Laser Scanning for Individual Tree Detection and Tree Species Classification [J].
Yu, Xiaowei ;
Litkey, Paula ;
Hyyppa, Juha ;
Holopainen, Markus ;
Vastaranta, Mikko .
FORESTS, 2014, 5 (05) :1011-1031
[48]   Forest inventory based on canopy height model derived from airborne laser scanning data [J].
Sackov, Ivan .
CENTRAL EUROPEAN FORESTRY JOURNAL, 2022, 68 (04) :224-231
[49]   A forest structure habitat index based on airborne laser scanning data [J].
Coops, Nicholas C. ;
Tompaski, Piotr ;
Nijland, Wiebe ;
Rickbeil, Gregory J. M. ;
Nielsen, Scott E. ;
Bater, Christopher W. ;
Stadt, J. John .
ECOLOGICAL INDICATORS, 2016, 67 :346-357
[50]   THE OPPORTUNITIES AND CHALLENGES OF USING AIRBORNE LASER SCANNING FOR FOREST INVENTORIES IN LITHUANIA [J].
Mozgeris, Gintautas ;
Bikuviene, Ina ;
Jonikavicius, Donatas .
8TH INTERNATIONAL SCIENTIFIC CONFERENCE RURAL DEVELOPMENT 2017: BIOECONOMY CHALLENGES, 2017, :698-702