Share Email Print

Journal of Applied Remote Sensing

Performance analysis of freeware filtering algorithms for determining ground surface from airborne laser scanning data
Author(s): Kalev Julge; Artu Ellmann; Anti Gruno
Format Member Price Non-Member Price
PDF $20.00 $25.00

Paper Abstract

Numerous filtering algorithms have been developed in order to distinguish the ground surface from nonground points acquired by airborne laser scanning. These algorithms automatically attempt to determine the ground points using various features such as predefined parameters and statistical analysis. Their efficiency also depends on landscape characteristics. The aim of this contribution is to test the performance of six common filtering algorithms embedded in three freeware programs. The algorithms’ adaptive TIN, elevation threshold with expand window, maximum local slope, progressive morphology, multiscale curvature, and linear prediction were tested on four relatively large (4 to 8  km2) and diverse landscape areas, which included steep sloped hills, urban areas, ridge-like eskers, and a river valley. The results show that in diverse test areas each algorithm yields various commission and omission errors. It appears that adaptive TIN is suitable in urban areas while the multiscale curvature algorithm is best suited in wooded areas. The multiscale curvature algorithm yielded the overall best results with average root-mean-square error values of 0.35 m.

Paper Details

Date Published: 8 August 2014
PDF: 15 pages
J. Appl. Remote Sens. 8(1) 083573 doi: 10.1117/1.JRS.8.083573
Published in: Journal of Applied Remote Sensing Volume 8, Issue 1
Show Author Affiliations
Kalev Julge, Tallinn Univ. of Technology (Estonia)
Artu Ellmann, Tallinn Univ. of Technology (Estonia)
Anti Gruno, Tallinn Univ. of Technology (Estonia)
Estonian Land Board (Estonia)

© SPIE. Terms of Use
Back to Top