Automatic Delineation and Height Measurement of Regenerating Conifer Crowns under Leaf-Off Conditions Using UAV Imagery

Andrew J Chadwick,Barry White,Nicholas C Coops,Anne Hervieux,Tristan R H Goodbody,Dominik Röeser,Christopher W Bater,Lee A Martens

doi:10.3390/rs12244104

Andrew J Chadwick, Barry White + Show 6 more

Open Access

https://doi.org/10.3390/rs12244104

Copy DOI

Abstract

The increasing use of unmanned aerial vehicles (UAV) and high spatial resolution imagery from associated sensors necessitates the continued advancement of efficient means of image processing to ensure these tools are utilized effectively. This is exemplified in the field of forest management, where the extraction of individual tree crown information stands to benefit operational budgets. We explored training a region-based convolutional neural network (Mask R-CNN) to automatically delineate individual tree crown (ITC) polygons in regenerating forests (14 years after harvest) using true colour red-green-blue (RGB) imagery with an average ground sampling distance (GSD) of 3 cm. We predicted ITC polygons to extract height information using canopy height models generated from digital aerial photogrammetric (DAP) point clouds. Our approach yielded an average precision of 0.98, an average recall of 0.85, and an average F1 score of 0.91 for the delineation of ITC. Remote height measurements were strongly correlated with field height measurements (r2 = 0.93, RMSE = 0.34 m). The mean difference between DAP-derived and field-collected height measurements was −0.37 m and −0.24 m for white spruce (Picea glauca) and lodgepole pine (Pinus contorta), respectively. Our results show that accurate ITC delineation in young, regenerating stands is possible with fine-spatial resolution RGB imagery and that predicted ITC can be used in combination with DAP to estimate tree height.

Highlights

Successful reforestation following harvest is an integral component of sustainable forest management, as the degree to which stands become established will influence the entire lifecycle of a forest [1]
They were approximately 14 years in age with crowns much smaller than those found in mature forests
We have presented a workflow for the automatic delineation of individual conifer crowns in true colour orthomosaics using Mask R-Convolutional neural networks (CNN) and the subsequent extraction of remote height measurements

Summary

Introduction

Successful reforestation following harvest is an integral component of sustainable forest management, as the degree to which stands become established will influence the entire lifecycle of a forest [1]. Given that regenerating stands are harvested approximately 80 years after planting, the reforestation phase represents a significant financial and ecological investment, the success of which rests heavily upon the ability to monitor and manage these young, regenerating stands [1,2]. The pronounced size, abundance, and remoteness of Alberta’s forested land base relative to other jurisdictions increases the cost and difficulty of operating field-based assessments. These factors are confounded by the restricted availability of field crews as well as poor road access to many harvested sites.

Methods

Results

Discussion

Conclusion