Extraction of tree parameters using terrestrial 3D LIDAR data

Abstract

Availability of sufficient green spaces becomes one of the mandatory requirements nowadays
for any cities of the world as the growing population and urban sprawl leads to reduction to
green cover. Popularly called as urban green spaces (UGS), they not only help to reduce the
heat islands in a city but also ensure healthy well-being of the city residents. UGS extraction
from satellite images as attempted in many studies has many limitations. For example, one can
see only two-dimensional spread of UGS and it may not be possible to extract the tree
parameters such as height, diameter, etc. as one can measure only area of UGS that too not at
individual tree level if medium and low-resolution satellite images are used. To overcome all
these drawbacks with satellite images, LIDAR technology can be used by laser scanning of
trees and the collected LIDAR data can then be used to built the 3D models of trees. The
present study did the same by carrying out the LIDAR survey for a sample tree located on a
highway side in Vellore, India. The LIDAR scans were then registered at an accuracy of 6 mm
in order to get a single point cloud which is free from noises and was then used to extract the
required parameters, namely, the height of tree, diameter at breast height and crown base
height. The results revealed that terrestrial LIDAR survey is one of the potential data sources
for extracting the tree parameters precisely.