Avocado tree point clouds with class labels

Published: 01-02-2021| Version 1 | DOI: 10.17632/h49fpprg6c.1
Fredrik Westling


LiDAR scans of 24 avocado trees from several years. Each point cloud has been annotated with two labels: one label for leaf (0) vs trunk (1) matter, and another for which tree each point belongs to: ground (0), center tree (1), north tree (2), south tree (3), uncategorised (4). Each point cloud is stored as a binary file with the following format (ui = 4-byte unsigned integer, d = 8-byte double): 3d,2ui,d Each line in the file has the fields: x,y,z,matter_label,tree_label,height XYZ is in North-East-Down orientation. The data was created in, and is easy to view using, ACFR comma/snark open-source tools (https://github.com/acfr/comma/wiki) Also included in the dataset is a list of trunk points in simple CSV format with tree IDs ; there are other trees in that list as well, but all trees represented in this dataset are in there.


Steps to reproduce

The trees were all manually scanned using a handheld GEOSLAM ZEB-1 LiDAR, with a trajectory involving two circuits of the tree (one far away, one close). The output point clouds were then manually geolocated by alignment to previous scans taken using a mobile LiDAR platform with high-accuracy GPS. Each point cloud was cropped to include only the three central trees in the scan, to discard the less-scanned surrounding trees, and a height value was calculated by automatically extracting the ground and computing the distance from each point to the nearest ground point. Finally, each point cloud was manually labelled in 3D with the two labels described; this was primarily done using the ACFR Snark tool "label-points"