The application of artificial intelligence (AI) techniques has improved the accuracy of forest species identification, particularly in timber inventories conducted under Sustainable Forest Management (SFM). This study developed and evaluated machine learning models to recognize 16 Amazonian timber species using digital images of tree bark. Data were collected from three SFM units located in Nova Maringá, Feliz Natal, and Cotriguaçu, in the state of Mato Grosso, Brazil. High-resolution images were processed into sub-images (256 × 256 pixels), and two feature extraction methods were tested: Local Binary Patterns (LBP) and pre-trained Convolutional Neural Networks (ResNet50, VGG16, InceptionV3, MobileNetV2). Four classifiers—Support Vector Machine (SVM), Artificial Neural Networks (ANN), Random Forest (RF), and Linear Discriminant Analysis (LDA)—were used. The best result (95% accuracy) was achieved using ResNet50 with SVM, confirming the effectiveness of transfer learning for species recognition based on bark texture. These findings highlight the potential of AI-based tools to enhance accuracy in forest inventories and support decision-making in tropical forest management.

The Brazil nut tree is considered symbolic of the Brazilian Amazon in function of its great importance, being one of the most significant extractivist products and a subsistence practice of the Indigenous people in many municipalities in Pará state. One of the main problems in different communities is related to the marketing process since it is not possible to distinguish the origin of the nuts and this causes inconvenience. The study evaluated the potential of VIS/NIR spectroscopy to identify the origin of leaves and nuts from Brazil nut trees growing in two indigenous villages, in the Xipaya Indigenous Lands, Pará state. Analysis was performed based on CIEL*a*b* parameters and using VIS (360–740 nm) and near-infrared spectra (1000–2500 nm). The samples were differentiated according to means tests, principal component analysis (PCA), and classification analysis based on k-NN. Color parameters and spectra were similar in both communities. Classification models based on k-NN produced adequate results for the distinction of villages in all evaluated situations, with accuracy of 98.54% for leaves, 89% and 90.91% for nuts with and without shell, respectively. Near infrared can be applied in forests as a technique for previous provenance identification and contribute to the subsistence and sustainable practice of extraction.

The application of artificial intelligence (AI) techniques has improved the accuracy of forest species identification, particularly in timber inventories conducted under Sustainable Forest Management (SFM). This study developed and evaluated machine learning models to recognize 16 Amazonian timber species using digital images of tree bark. Data were collected from three SFM units located in Nova Maringá, Feliz Natal, and Cotriguaçu, in the state of Mato Grosso, Brazil. High-resolution images were processed into sub-images (256 × 256 pixels), and two feature extraction methods were tested: Local Binary Patterns (LBP) and pre-trained Convolutional Neural Networks (ResNet50, VGG16, InceptionV3, MobileNetV2). Four classifiers—Support Vector Machine (SVM), Artificial Neural Networks (ANN), Random Forest (RF), and Linear Discriminant Analysis (LDA)—were used. The best result (95% accuracy) was achieved using ResNet50 with SVM, confirming the effectiveness of transfer learning for species recognition based on bark texture. These findings highlight the potential of AI-based tools to enhance accuracy in forest inventories and support decision-making in tropical forest management.