Tree top detection using local maxima filtering: a parameter sensitivity analysis

The sensitivity of a treetop detection algorithm is investigated by automated evaluation of detection performance for several parameter combinations. The algorithm consists in digital elevation models computation, morphological filtering, Gaussian smoothing and local maxima extraction and selection. The analysis is performed on three field plots located in the French Alps. One is a Norway spruce stand while the two others are dominated by Silver fir and European beech. Detection rates above 42.9% are achieved with less than 4.1% of false positives. Even though some similarities exist regarding resolution and morphological filtering, optimal settings determined on one plot performed uncertainly on the others. Besides, optimised parameters may depend on both the laser data mainly point density and on the forest structure and species.

[1]  Installation de placettes permanentes dans les forêts résineuses des Alpes du Nord : objectifs, méthode, résultats , 1996 .

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

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

[4]  R. Hill,et al.  Quantifying canopy height underestimation by laser pulse penetration in small-footprint airborne laser scanning data , 2003 .

[5]  Tomas Brandtberg.  Detection and analysis of individual leaf-off tree crowns in small footprint, high sampling density lidar data from the eastern deciduous forest in North America , 2003 .

[6]  Y. H. Shikoku,et al.  THE EFFECTS OF FOOTPRINT SIZE AND SAMPLING DENSITY IN AIRBORNE LASER SCANNING TO EXTRACT INDIVIDUAL TREES IN MOUNTAINOUS TERRAIN , 2004 .

[7]  S. Popescu,et al.  Seeing the Trees in the Forest: Using Lidar and Multispectral Data Fusion with Local Filtering and Variable Window Size for Estimating Tree Height , 2004 .

[8]  F. Rey,et al.  Guide des sylvicultures de montagne Alpes du Nord Françaises , 2006 .

[9]  F. Berger,et al.  Real-size experiments and 3-D simulation of rockfall on forested and non-forested slopes , 2006 .

[10]  P. Gong,et al.  Isolating individual trees in a savanna woodland using small footprint lidar data , 2006 .

[11]  P. Gong,et al.  Detection of individual trees and estimation of tree height using LiDAR data , 2007, Journal of Forest Research.

[12]  M. Heurich.  Automatic recognition and measurement of single trees based on data from airborne laser scanning over the richly structured natural forests of the Bavarian Forest National Park , 2008 .