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

Abstract

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 F 1 score of 0.86.

Phyllotaxis transition over the lifespan of a palm tree using Magnetic Resonance Imaging (MRI) and Terrestrial Laser Scanning (TLS): the case of Jubaea chilensis

“…This technology allows acquiring valuable information from the stem surface of palm trees, but it might be also used for other species. TLS stem surface scanning has been applied to study wood quality of other species [ 44 , 67 ]. In our case, the use of TLS data and the proposed algorithm allowed us to identify precisely scars and their boundaries, despite of the relatively smooth surface of the J. chilensis stem.…”

CNN-based Method for Segmenting Tree Bark Surface Singularites

“…On ten meshes, the authors obtain an F1 score of 0.71. In 2020, they successfully classified the segmented regions [10] using a random forest learning process. The singularities regions were classified into four classes: branch, branch scar, bush, small singularities.…”

“…They are in two directories: INRAE1a, INRAE1b. The data come from the same acquisition device and have been transformed into a 3D mesh using the process described in [10]. Each tree trunk is associated to three files: the mesh (with extension .off), the point indices corresponding to singularities (suffixed by -groundtruth-points.id), the face indices corresponding to singularities (suffixed by -groundtruth.id).…”

Review of the current state of application of wood defect recognition technology

“…The practical value of these defect detection techniques has been demonstrated in the assessment of the structural health of urban and forest standing trees (Feng et al 2014;Kasal 2014;Papandrea et al 2022). In recent studies, some academics have compared the efficiency of automated wood defect detection techniques with manual inspection, verifying the improvements in efficiency of automated detection techniques (Nguyen et al 2020(Nguyen et al , 2021. These research results have contributed to the development of automated defect detection technology for use in wood processing lines.…”