A machine-learning approach for classifying defects on tree trunks using terrestrial LiDAR

• Published in: Computers and electronics in agriculture 2020-04, Vol.171, p.105332-12, Article 105332
• Main Authors: Nguyen, Van-Tho, Constant, Thiéry, Kerautret, Bertrand, Debled-Rennesson, Isabelle, Colin, Francis
• Format: Article
• Language: English
• Subjects: Abnormalities; Agricultural sciences; Bark; Beech; Computer Science; Computer Vision and Pattern Recognition; Defects; Forestry; Lidar; Life Sciences; Machine learning; Random forests; Roundwood quality; Scars; Silviculture, forestry; Standing tree grading
• ISSN: 0168-1699; 1872-7107
• DOI: 10.1016/j.compag.2020.105332

•Defects with a diameter from 0.5 cm on tree trunks could be detected in TLS data.•Good accuracy of the defects classification was achieved with an F1 score of 0.86.•The accuracy was similar between tested species with different bark roughness. Three-dimensional data are increasingly prevalent in forestry thanks to terrestrial LiDAR. This work assesses the feasibility for an automated recognition of the type of local defects present on the bark surface. These singularities are frequently external markers of inner defects affecting wood quality, and their type, size, and frequency are major components of grading rules. The proposed approach assigns previously detected abnormalities in the bark roughness to one of the defect types: branches, branch scars, epicormic shoots, burls, and smaller defects. Our machine learning approach is based on random forests using potential defects shape descriptors, including Hu invariant moments, dimensions, and species. The results of our experiments involving different French commercial species, oak, beech, fir, and pine showed that most defects were well classified with an average F1 score of 0.86.