TITLE:
Validation of High-Density Airborne LiDAR-Based Feature Extraction Using Very High Resolution Optical Remote Sensing Data
AUTHORS:
Shridhar D. Jawak, Satej N. Panditrao, Alvarinho J. Luis
KEYWORDS:
LiDAR; WorldView-2; Feature Extraction
JOURNAL NAME:
Advances in Remote Sensing,
Vol.2 No.4,
December
16,
2013
ABSTRACT:
This work uses the canopy height
model (CHM) based workflow for individual tree crown delineation from LiDAR
point cloud data in an urban environment and evaluates its accuracy by using very
high-resolution PAN (spatial) and 8-band WorldView-2 imagery. LiDAR point cloud
data were used to detect tree features by classifying point elevation values.
The workflow includes resampling of LiDAR point cloud to generate a raster
surface or digital terrain model, generation of hill-shade image and intensity
image, extraction of digital surface model, generation of bare earth digital
elevation model and extraction of tree features. Scene dependent extraction
criteria were employed to improve the tree feature extraction. LiDAR-based
refining/filtering techniques used for bare earth layer extraction were crucial
for improving the subsequent tree feature extraction. The PAN-sharpened WV-2
image (with 0.5 m spatial resolution) used to assess the accuracy of
LiDAR-based tree features provided an accuracy of 98%. Based on these
inferences, we conclude that the LiDAR-based tree feature extraction is a potential
application which can be used for understanding vegetation characterization in
urban setup.