Fusion of crown and trunk detections from airborne UAS based laser scanning for small area forest inventories

Abstract High density point clouds from unmanned airborne laser scanning (UALS) systems have great potential for small area forest inventories. We propose an UALS-based tree level inventory method that takes advantage of both the segmented crowns and segmented trunks: hybrid tree detection (HTD). The method is tested at twenty 30 m × 30 m validation plots of varying maturity, tree species distribution and stocking density. With the traditional individual tree crown delineation (ITC) approach, tree attributes are only predicted for crown segmented trees. Here, we assume that a segmented crown can contain more than one tree. Our idea is to identify segmented crowns that contain more than one segmented trunk. One of the trunks is linked to a segmented crown (upper most tree) and the remainders are treated as understory trunks. Heights of the crown segmented trees and the diameters of the understory trunks are used as predictor variables in nonlinear mixed-effects models of tree volume. The %RMSE and %MD values of volume predictions at the 30 m × 30 m validation plot level were 22.2%, −13.3% and 18.8%, −8.3% for ITC and HTD, respectively. We conclude that the proposed HTD approach improves the accuracy of ITC in managed boreal forests when using UALS data.

[1]  Ladislav Kulla,et al.  A Comparison of Two Tree Detection Methods for Estimation of Forest Stand and Ecological Variables from Airborne LiDAR Data in Central European Forests , 2019, Remote. Sens..

[2]  R. Dubayah,et al.  Rapid, High-Resolution Forest Structure and Terrain Mapping over Large Areas using Single Photon Lidar , 2016, Scientific Reports.

[3]  Johan Holmgren,et al.  Tree crown segmentation based on a tree crown density model derived from Airborne Laser Scanning , 2019 .

[4]  Richard Barnes,et al.  Priority-Flood: An Optimal Depression-Filling and Watershed-Labeling Algorithm for Digital Elevation Models , 2015, Comput. Geosci..

[5]  Johannes Breidenbach,et al.  Estimation of Forest Growing Stock Volume with UAV Laser Scanning Data: Can It Be Done without Field Data? , 2020, Remote. Sens..

[6]  J. Hyyppä,et al.  DETECTING AND ESTIMATING ATTRIBUTES FOR SINGLE TREES USING LASER SCANNER , 2006 .

[7]  Raul Queiroz Feitosa,et al.  Assessment of CNN-Based Methods for Individual Tree Detection on Images Captured by RGB Cameras Attached to UAVs , 2019, Sensors.

[8]  A. Lucieer,et al.  Individual tree detection and crown delineation from Unmanned Aircraft System (UAS) LiDAR in structurally complex mixed species eucalypt forests , 2021 .

[9]  P. Axelsson.  DEM Generation from Laser Scanner Data Using Adaptive TIN Models , 2000 .

[10]  Sergio Marconi,et al.  Individual Tree-Crown Detection in RGB Imagery Using Semi-Supervised Deep Learning Neural Networks , 2019, Remote. Sens..

[11]  Juha Hyyppä,et al.  Automated Stem Curve Measurement Using Terrestrial Laser Scanning , 2014, IEEE Transactions on Geoscience and Remote Sensing.

[12]  Jouko Laasasenaho.  Taper curve and volume functions for pine, spruce and birch [Pinus sylvestris, Picea abies, Betula pendula, Betula pubescens] , 1982 .

[13]  Peter Surový,et al.  Very High Density Point Clouds from UAV Laser Scanning for Automatic Tree Stem Detection and Direct Diameter Measurement , 2020, Remote. Sens..

[14]  Juha Hyyppä,et al.  Comparison of Area-Based and Individual Tree-Based Methods for Predicting Plot-Level Forest Attributes , 2010, Remote. Sens..

[15]  Juha Hyyppä,et al.  Under-canopy UAV laser scanning for accurate forest field measurements , 2020, ISPRS Journal of Photogrammetry and Remote Sensing.

[16]  Juha Hyyppä,et al.  Autonomous Collection of Forest Field Reference - The Outlook and a First Step with UAV Laser Scanning , 2017, Remote. Sens..

[17]  Ilkka Korpela,et al.  Mapping forest plots: an efficient method combining photogrammetry and field triangulation , 2007 .

[18]  Michele Dalponte,et al.  Delineation of Individual Tree Crowns from ALS and Hyperspectral data: a comparison among four methods , 2015 .

[19]  M. Maltamo,et al.  Imputation of single-tree attributes using airborne laser scanning-based height, intensity, and alpha shape metrics , 2010 .

[20]  Åsa Persson,et al.  Detecting and measuring individual trees using an airborne laser scanner , 2002 .

[21]  Jinqiang Cui,et al.  UAV LiDAR for below-canopy forest surveys , 2013 .

[22]  Arko Lucieer,et al.  Evaluating Tree Detection and Segmentation Routines on Very High Resolution UAV LiDAR Data , 2014, IEEE Transactions on Geoscience and Remote Sensing.

[23]  M. Maltamo,et al.  Comparison of multispectral airborne laser scanning and stereo matching of aerial images as a single sensor solution to forest inventories by tree species , 2019, Remote Sensing of Environment.

[24]  Albert R. Stage,et al.  Most Similar Neighbor: An Improved Sampling Inference Procedure for Natural Resource Planning , 1995, Forest Science.

[25]  Yu Oishi,et al.  The Impact of Different Support Vectors on GOSAT-2 CAI-2 L2 Cloud Discrimination , 2017, Remote. Sens..

[26]  Julia Armesto,et al.  Individual Tree Detection in a Eucalyptus Plantation Using Unmanned Aerial Vehicle (UAV)-LiDAR , 2020, Remote. Sens..

[27]  Petteri Packalen,et al.  Bayesian approach to single-tree detection in airborne laser scanning – use of training data for prior and likelihood modeling , 2018, Journal of Physics: Conference Series.

[28]  Juha Hyyppä,et al.  Combination of individual tree detection and area-based approach in imputation of forest variables using airborne laser data , 2012 .

[29]  Adrián Cardil,et al.  Measuring Individual Tree Diameter and Height Using GatorEye High-Density UAV-Lidar in an Integrated Crop-Livestock-Forest System , 2020, Remote. Sens..

[30]  E. Næsset,et al.  Prediction of species specific forest inventory attributes using a nonparametric semi-individual tree crown approach based on fused airborne laser scanning and multispectral data , 2010 .

[31]  D. Bates,et al.  Nonlinear mixed effects models for repeated measures data. , 1990, Biometrics.

[32]  Juha Hyyppä,et al.  In situ biomass estimation at tree and plot levels: What did data record and what did algorithms derive from terrestrial and aerial point clouds in boreal forest , 2019, Remote Sensing of Environment.

[33]  Lammert Kooistra,et al.  Non-destructive tree volume estimation through quantitative structure modelling: Comparing UAV laser scanning with terrestrial LIDAR , 2019, Remote Sensing of Environment.

[34]  Johan Holmgren,et al.  Forest Variable Estimation Using a High Altitude Single Photon Lidar System , 2018, Remote. Sens..

[35]  Eija Honkavaara,et al.  Forest in situ observations using unmanned aerial vehicle as an alternative of terrestrial measurements , 2019, Forest Ecosystems.

[36]  Joanne C. White,et al.  Remote Sensing Technologies for Enhancing Forest Inventories: A Review , 2016 .

[37]  Norbert Pfeifer,et al.  Rigorous Strip Adjustment of UAV-based Laserscanning Data Including Time-Dependent Correction of Trajectory Errors , 2016 .

[38]  M. Vastaranta,et al.  THE FUSION OF INDIVIDUAL TREE DETECTION AND VISUAL INTERPRETATION IN ASSESSMENT OF FOREST VARIABLES FROM LASER POINT CLOUDS , 2012 .