Herramientas de corte para optimizar parámetros de clasificación de especies maderables con redes neuronales convolucionales / Cutting tools to optimize classification parameters of timber species with convolutional neural networks

Introducción: La gran diversidad de especies maderables tropicales demanda el desarrollo de nuevas tecnologías de identificación con base en sus patrones o características anatómicas. La aplicación de redes neuronales convolucionales (CNN) para el reconocimiento de especies maderables tropicales se ha incrementado en los últimos años por sus resultados prometedores. Objetivo: Evaluamos la calidad de las imágenes macroscópicas con tres herramientas de corte para mejorar la visualización y distinción de las características anatómicas en el entrenamiento del modelo CNN. Métodos: Recolectamos las muestras entre el 2020 y 2021 en áreas de explotación forestal y aserraderos de Selva Central, Perú. Luego, las dimensionamos y, previo a la identificación botánica y anatómica, las cortamos en secciones transversales. Generamos una base de datos de imágenes macroscópicas de la sección transversal de la madera, a través del corte, con tres herramientas para ver su rendimiento en el laboratorio, campo y puesto de control. Resultados: Usamos tres herramientas de corte para obtener una alta calidad de imágenes transversales de la madera; obtuvimos 3 750 imágenes macroscópicas con un microscopio portátil que corresponden a 25 especies maderables. El cuchillo ''Tramontina'' es duradero, pero pierde el filo con facilidad y se necesita una herramienta para afilar, el cúter retráctil ''Pretul'' es adecuado para madera suave y dura en muestras pequeñas de laboratorio; el cuchillo ''Ubermann'' es apropiado para el campo, laboratorio y puesto de control, porque tiene una envoltura duradera y láminas intercambiables en caso de pérdida de filo. Conclusiones: La calidad de las imágenes es decisiva en la clasificación de especies maderables, porque permite una mejor visualización y distinción de las características anatómicas en el entrenamiento con los modelos de red neuronal convolucional EfficientNet B0 y Custom Vision, lo cual se evidenció en las métricas de precisión.

Introduction: The great diversity of tropical timber species demands the development of new technologies capable of identifying them based on their patterns or anatomical characteristics. The application of convolutional neural networks (CNN) for the recognition of tropical timber species has increased in recent years due to the promising results of CNNs. Objective: To evaluate the quality of macroscopic images with three cutting tools to improve the visualization and distinction of anatomical features in the CNN model training. Methods: Samples were collected from 2020 to 2021 in areas of logging and sawmills in the Central Jungle, Peru. They were later sized and, after botanical and anatomical identification, cut in cross sections. A database of macroscopic images of the cross-section of wood was generated through cutting with three different tools and observing its performance in the laboratory, field, and checkpoint. Results: Using three cutting tools, we obtained high quality images of the cross section of wood; 3 750 macroscopic images were obtained with a portable microscope and correspond to 25 timber species. We found the ''Tramontina'' knife to be durable, however, it loses its edge easily and requires a sharpening tool, the ''Pretul'' retractable cutter is suitable for cutting soft and hard wood in small laboratory samples and finally the ''Ubermann'' knife is suitable for use in the field, laboratory, and checkpoint, because it has a durable sheath and interchangeable blades in case of dullness. Conclusion: The quality of the images is decisive in the classification of timber species, because it allows a better visualization and distinction of the anatomical characteristics in training with the EfficientNet B0 and Custom Vision convolutional neural network models, which was evidenced in the precision metrics.

Trace elements distribution in tropical tree rings through high-resolution imaging using LA-ICP-MS analysis.

BACKGROUND: The distribution of trace elements in tree rings although poorly known may be useful to better understand environmental changes, pollution trends, long-term droughts, forest dieback processes, and biology of trees. METHOD: Laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS) is used for imaging micronutrients and potentially toxic elements distribution, allowing the investigation of trace elements at high spatial resolution within the tree rings. To ensure a more efficient determination of micronutrients and potentially toxic elements, LA-ICP-MS instrumental conditions were optimized and carbon, a major element in wood, is used as an internal standard during analysis to correct for random fluctuations. RESULTS: Spatial distributions maps of Ba, Cu, Fe, Mn, Ni, and Pb in growth layers of six tropical tree species were built-up using the LA-iMageS software, namely: Amburana cearensis (Fabaceae), Cedrela fissilis (Meliaceae), Hymenaea courbaril (Fabaceae), Maclura tinctoria (Moraceae), Parapiptadenia zehntneri (Fabaceae), Peltogyne paniculata (Fabaceae). A correlation between the trace element composition and different cell types (parenchyma, fiber, and vessel) was distinctly observed. It was observed a general pattern of Ba, Cu, Ni, Mn, and Pb accumulation mainly in the axial parenchyma and vessels. But the elemental composition of xylem cells is strongly species dependent. The multivariate analysis also points to a distinct accumulation of minerals between heartwood and sapwood in the same species. CONCLUSIONS: Imaging both essential and deleterious element distributions in the tree rings may improve visualization and can effectively contribute to understanding the lifetime metabolism of trees and evaluating the effects of environmental changes related to climatic seasonality, pollution, and future paleoclimate reconstructions.

Brazilian cerrado species: wood characteristics / Espécies do cerrado brasileiro: características da madeira

The characterization of wood anatomy and proprieties can provide subsidies for rational use of tree species. Enables, oftentimes, the definition of technological potential of wood - and wood products - through the assessment of quality information and also allows it ́s correct identification, contributing to the timber production chain. The objective of this study was to contribute to the knowledge about wood properties of the Cerrado biome (Brazilian savanna) species trough the anatomical characterization and determination of density profile. Twenty trees species belonging to thirteen families occurring in Pirenópolis, Goiás and Brasília, Federal District, Brazil, was selected and radial samples were removed from the tree trunks at the DBH (1.30 m) in a non-destructive way using an incremental probe and a motorized extractor. The qualitative and quantitative anatomical parameters of vessels and fibers and the wood density were determined in the collected samples. Species presented fibers with mean values between 900.5 and 2052.9 µm for the length; 18.1 and 27.7 µm for the diameter; and 4.4 to 9.4 µm for the wall thickness. The species presented a variation in the vessel dimensions from 29.2 to 155.6 µm for the diameter; 17.7 to 32.5% for occupied area; and 2.6 to 165 to vessels.mm-2. The apparent density presented mean values between 0.36 and 1.21 g cm-3. The specie that presented the highest variation of wood density was A. fraxinifolium. According to the radial profiles two patterns of increasing and stable variation in the apparent density of the pith to bark were defined as a function of the wood anatomical characteristics.

A caracterização anatômica e das propriedades da madeira pode auxiliar na tomada de decisão quanto ao uso racional de espécies. Possibilita, muitas vezes, a definição do potencial tecnológico da madeira - e produtos - fornecendo acesso a informações sobre a sua qualidade, além de sua correta identificação, contribuído com a cadeia produtiva madeireira. O objetivo deste estudo foi investigar e caracterizar as propriedades da madeira de espécies arbóreas do Cerrado, através da descrição anatômica e determinação do perfil de densidade. Foram selecionadas 20 espécies pertencentes a treze famílias, na cidade de Pirenópolis, Goiás e Brasília, Distrito Federal, Brasil, e amostras radiais foram retiradas dos troncos das árvores no DAP (1,30 m) de forma não destrutiva, utilizando-se uma sonda de incremento e um extrator motorizado. Nas amostras coletadas, foram determinados os parâmetros anatômicos qualitativos e quantitativos dos vasos e das fibras e a densidade aparente da madeira. As espécies apresentaram fibras com valores médios entre 900,5 a 2052,9 µm para o comprimento; 18,1 a 27,7 µm para o diâmetro e 4,4 a 9,4 µm para a espessura da parede; e variação nas dimensões dos vasos de 29,2 a 155,6 µm para o diâmetro; 17,7 a 32,5% para área ocupada; e 2,6 a 165 vasos.mm-2 para a frequencia dos vasos. A densidade aparente apresentou valores médios entre 0,36 e 1,21 g cm-3. A espécie que apresentou a maior variação de densidade de madeira foi A. fraxinifolium. De acordo com os perfis radiais, dois padrões de variação crescente e estável na densidade aparente, da medula para casca, foram definidos como uma função das características anatômicas.