Skip to main content
SLU publication database (SLUpub)

Conference paper2008Peer reviewed

A method for linking field-surveyed and aerial-detected single trees using cross correlation of position images and the optimization of weighted tree list graphs

Olofsson, Kenneth; Lindberg, Eva; Holmgren, Johan

Abstract

With the present number of forestry remote sensing and field plot survey methods several data sources can be combined to potentially achieve a higher accuracy than with a single data source. Some of these datasets contain single tree information and it would be of great value when designing survey techniques if one could automatically link data from different sources belonging to the same tree. In this paper such a method for linking field-surveyed and aerial detected trees is described and evaluated. A simulation study and experimental data show the accuracy of the algorithm at different settings. The method correctly links >90 % of the trees if the corresponding datasets have a position error standard deviation of 1 [m] and 10 % omission and commission errors

Keywords

LiDAR; digital aerial images; data fusion; field plots; single tree detection and linking

Published in


ISBN: 978-0-85538-774-7

Conference

SilviLaser 2008