Identifying crown areas in an undulating area planted with eucalyptus using unmanned aerial vehicle near-infrared imagery

Abstract

In this letter, we propose an identification method of tree crown areas for imagery captured by a near-infrared camera on board an unmanned aerial vehicle platform over an undulating Eucalyptus planting area in Guangdong Province, China. The method extracts crown areas by applying mathematical morphology, unsupervised segmentation based on J-value segmentation, local spatial statistics, and Iterative Self-Organizing Data Analysis Technique Algorithm. Two morphology filters and four segmentation scales were compared between densely and sparsely planted plots as well as sunlit and shaded plots. The opening operation by the window size of 9×9 pixel and segmentation by the seed area sized 65×65 pixel achieved the best performance with overall accuracy of 91%, 93%, 89% and 91% in densely sunlit, sparsely sunlit, densely shaded and sparsely shaded plots.

Mapping of Rubber Forest Growth Models Based on Point Cloud Data

“…At the same time, it can generate a 3D model of the forest structure, including tree distribution, density and spatial distribution, to deepen the understanding of the forest structure. In 3D reconstruction, the accurate detection of the target crop is a key step in building a fine map [15]. With the breakthrough progress of artificial intelligence, especially deep learning technology, the field of 3D target detection has made great progress.…”

Hypertemporal Imaging Capability of UAS Improves Photogrammetric Tree Canopy Models

“…In this instance, UAS LIDAR provided independent measurement showing photogrammetric occlusion of upper canopy branches and sub-canopy ground features [19]. Other studies of canopy structure in Eucalyptus examine forestry plantations where tree canopy is light-limited and conical in shape and UAV photogrammetry products are compared to independent methods at a single time [21,53,54]. Appropriate minimum specifications for sensor, environment, and processing to achieve reproducible Eucalyptus canopy models are currently uncertain.…”