Deep features from Pre-trained CNNs and a vision transformer for amazonian timber species recognition using classical machine learning and random image patches
Souza, Deivison Venicio, Santos, Joielan Xipaia, Gama, Natally Celestino, Silva, Washington Duarte Silva, Sousa, Genilson Santana, Vieira, Helena Cristina, Acosta, Tawani Lorena Naide, Cirilo, Alexia Juliana Camodeco, Silva, Gabriela Cristina Costa, Muñiz, Graciela Inés Bolzon, Nisgoski, Silvana Journal of the Indian Academy of Wood Science
· Springer Nature
· 2026-12-01
· DOI: 10.1007/s13196-026-00425-5
Abstract
Accurate identification of timber species is essential for biodiversity conservation, sustainable forest management, and timber supply chain control. This study evaluated four pre-trained convolutional neural networks—ResNet50, VGG16, InceptionV3, and MobileNetV2 and a Vision Transformer (ViT) as fixed feature extractors for the automated identification of Amazonian timber species. The resulting deep representations were classified using an Artificial Neural Network (ANN), Support Vector Classifier (SVC), Linear Discriminant Analysis (LDA), and Random Forest (RF). The dataset comprised 864 macroscopic images of wood cross-sections representing six species and five genera from the Brazilian Amazon. Two image-representation strategies were compared: original images and 20 random 224 × 224-pixel patches extracted from each image. Patch-level class probabilities were aggregated at the image level through soft voting, and predictive performance was evaluated using tree-level Leave-One-Group-Out Cross-Validation. The patch-based strategy improved classifier performance and produced more consistent recognition across species, particularly by increasing the recall of classes with the highest error rates when original images were used. Among the evaluated feature extractors, ViT representations yielded the best results when combined with ANN, SVC, and LDA. The ViT + SVC model achieved the highest predictive performance, with 98.29% accuracy and an MCC of 97.96%. ViT + LDA achieved comparable performance, with 97.70% accuracy and an MCC of 97.26%, while requiring substantially less time for hyperparameter optimization. These findings indicate that combining pre-trained deep feature extractors, conventional machine learning algorithms, random patch sampling, and soft-voting aggregation is a promising and computationally efficient strategy for the automated identification of Amazonian timber species.
Souza, Deivison Venicio, Santos, Joielan Xipaia, Gama, Natally Celestino et al. Deep features from Pre-trained CNNs and a vision transformer for amazonian timber species recognition using classical machine learning and random image patches. // Journal of the Indian Academy of Wood Science. – 2026. – DOI: 10.1007/s13196-026-00425-5. – URL: http://link.springer.com/openurl/fulltext?id=doi:10.1007/s13196-026-00425-5
Souza, Deivison Venicio, Santos, Joielan Xipaia, Gama, Natally Celestino, Silva, Washington Duarte Silva, Sousa, Genilson Santana, Vieira, Helena Cristina, Acosta, Tawani Lorena Naide, Cirilo, Alexia Juliana Camodeco, Silva, Gabriela Cristina Costa, Muñiz, Graciela Inés Bolzon, Nisgoski, & Silvana (2026). Deep features from Pre-trained CNNs and a vision transformer for amazonian timber species recognition using classical machine learning and random image patches. Journal of the Indian Academy of Wood Science. https://doi.org/10.1007/s13196-026-00425-5
Souza, et al. “Deep features from Pre-trained CNNs and a vision transformer for amazonian timber species recognition using classical machine learning and random image patches.” Journal of the Indian Academy of Wood Science, 2026. doi:10.1007/s13196-026-00425-5.
@article{Souza2026,
title = {Deep features from Pre-trained CNNs and a vision transformer for amazonian timber species recognition using classical machine learning and random image patches},
author = {Souza and Venicio, Deivison and Santos and Xipaia, Joielan and Gama and Celestino, Natally and Silva and Silva, Washington Duarte and Sousa and Santana, Genilson and Vieira and Cristina, Helena and Acosta and Naide, Tawani Lorena and Cirilo and Camodeco, Alexia Juliana and Silva and Costa, Gabriela Cristina and Muñiz and Bolzon, Graciela Inés and Nisgoski and Silvana},
journal = {Journal of the Indian Academy of Wood Science},
year = {2026},
doi = {10.1007/s13196-026-00425-5},
url = {http://link.springer.com/openurl/fulltext?id=doi:10.1007/s13196-026-00425-5}
}
The Calycophyllum spruceanum , known in Brazil, as “pau-mulato”, is a rapidly growing forest species native to the Amazon that produces dense wood used in floors, plywood, civil and naval construction and firewood, as it has high...
The growing demand for renewable energy sources has stimulated interest in biomass as a sustainable alternative, particularly the use of residues from the wood-processing industry. This study evaluated the energy potential of sawdust...
Fibrous lignocellulosic substrates such as wood used in the marine environment, eg., wooden docks, are highly susceptible to biofilm formation, which can progress to fouling, resulting in material degradation and increased maintenance...
Fiber-cement composites are used in lightweight construction due to their favorable balance of mechanical and physical properties. This study investigates the influence of rice husk-derived cellulose microfibers on the performance of...
This study details the fabrication and characterization of low-cost bio-based composite panels designed for sustainable construction and eco-friendly packaging. The non-structural boards utilize agricultural by-products, reinforcing a...