3D modelling of individual trees using a handheld camera: Accuracy of height, diameter and volume estimates

Accurate tree measurement is necessary for applications such as resource appraisal, and biophysical and ecological modelling. This research tests the potential to use a low-cost hand-held camera alongside structure-from-motion with multi-view stereo-photogrammetry (SfM-MVS) to accurately measure trees. SfM-MVS is a computer vision technique, in which the geographic coordinates of objects are calculated from a series of photographs, resulting in a 3D point cloud model. This research tested the ability of SfM-MVS to reconstruct spatially accurate 3D models from which 2D (height, crown spread, crown depth, stem diameter) and 3D (volume) tree metrics could be estimated. Thirty small, potted trees were photographed and measured with traditional dendrometry to evaluate SfM-MVS derived tree size estimates. Tree volume was obtained via the water displacement approach xylometry. SfM-MVS estimates of 2D tree metrics had errors (RMSE - root mean squared error) as low as 3.74% (RMSEtree (height) = 3.74%, RMSEcrown depth = 11.93%, RMSEcrown (spread) = 14.76%, RMSEDBH = 9.6%). SfM-MVS estimates of 3D tree metrics were better for the main stem than for the slender branches (RMSEStem = 12.33%, RMSEBranches = 47.53%, RMSETotal Volume = 18.53%). Apart from height and crown depth, all modelled variables had negative bias, suggesting that SfM-MVS tends to underestimate the size of trees. The results show that SfM-MVS is capable of producing estimates of 2D and 3D metrics with accuracy comparable to that of laser scanning (i.e. LiDAR). Factors like the position of the tree relative to its surroundings, the background scene and the ambient lighting, appear to affect model success. SfM-MVS provides a low-cost alternative to remote sensing technologies currently used such as terrestrial laser scanning and, as no specialised equipment is required, it is able to be used by people with little expertise or training. Future research is required for exploring the suitability of SfM-MVS for specific applications requiring accurate dendrometry. (C) 2015 Elsevier GmbH. All rights reserved.