Integration of Wood Anatomy and Artificial Intelligence: A Technological Framework Based on the UTN Xylotheque for Forensic Identification and Forest Governance in Ecuador
Preprint
- 15 June 2026
- preprint
- Published by MDPI AG in Preprints.org
Abstract
Accurate identification of timber species is fundamental for sustainable forest management and mitigating illegal logging in Ecuador. This study characterizes the Wood Anatomy Laboratory and Xylotheque (LAMX) at the Universidad Técnica del Norte, a strategic collection housing 510 cataloged samples (127 species, 56 families) and 2,267 histological preparations. Based on an analysis of records from the Forest Management System (SAF), a prioritized list of 129 commercial species was developed, identifying critical gaps in the collection’s representation. To facilitate on-site digital monitoring, the mobile application “Wood Identifier UTN” was developed, based on lightweight Convolutional Neural Networks (CNNs) for the automated macroscopic recognition of five high-value pilot Neotropical species. Validation of the model through field tests in timber yards in Ibarra demonstrated a high capacity for taxonomic discrimination, achieving an overall accuracy of 94.04% and an F1-score of 0.976 with the MobileNetV2 architecture. This hybrid approach establishes a scalable, transparent, and low-cost technological framework that strengthens regulatory compliance and enhances forest governance in Neotropical regions.Keywords
All Related Versions
- Version 1, 2026-06-15, Preprints
- Published version: Forests, 17 (7), 781.
This publication has 23 references indexed in Scilit:
- Deep learning methodology for the identification of wood species using high-resolution macroscopic images and patch-votingApplied Intelligence, 2026
- WoodGLNet: a multi-scale network integrating global and local information for real-time classification of wood imagesJournal of Real-Time Image Processing, 2024
- Dynamic Confidence-Aware Multi-Modal Emotion RecognitionIEEE Transactions on Affective Computing, 2023
- Anchor tuning in Faster R-CNN for measuring corn silage physical characteristicsComputers and Electronics in Agriculture, 2021
- Comparison of two forensic wood identification technologies for ten Meliaceae woods: computer vision versus mass spectrometryWood Science and Technology, 2020
- Grad-CAM: Visual Explanations from Deep Networks via Gradient-Based LocalizationInternational Journal of Computer Vision, 2019
- Tree sleuths are using DNA tests and machine vision to crack timber crimesNature, 2019
- MobileNetV2: Inverted Residuals and Linear Bottlenecks2018
- Deep Residual Learning for Image Recognition2016
- Forensic timber identification: It's time to integrate disciplines to combat illegal loggingBiological Conservation, 2015