Search

This manuscript reports the feasibility of a sequential convolutional neural network (CNN) machine-learning model that correctly identifies 11 North American softwood species from 14× magnified macroscopic end-grain images. The convolutional network contained a large kernel size, max pooling layers, and leaky rectified linear units to accelerate training. To reduce overfitting of training data, we employed L2 regularization, custom initialization, and stratified 5-fold cross-validation techniques. The database consisted of 1789 wood end-grain images. The training data set consisted of 1431 images, whereas the validation set had approximately 358 images. In both sets, the input image size was 227 pixels × 227 pixels. Data augmentation was performed on-the-fly by flipping, rotating, and zooming the images. We tested the performance of the CNN against precision, sensitivity, specificity, F1 score, and adjusted accuracy. The adjusted accuracy for the entire model was 94.0%. Confusion matrices indicated the lowest performance was in correctly classifying ponderosa pine (Pinus ponderosa Douglas ex P. Lawson & C. Lawson) and eastern spruce (Picea spp. A. Dietr.) group with an average sensitivity of 89.0% for each. Even though high validation accuracy (>94.0%) was achieved, we concluded that a much larger data set is needed for wood identification to obtain industrially accurate identification of softwoods, mainly due to their visual and macroscopic similarities.

Ground support provided by crane mats ensures the safety of workers and integrity of heavy-duty machines. Analytical evaluation of timber mats is necessary to advance the development of engineering standards associated with wood-based as well as other types of mats. The goal of this study was to mechanically evaluate bending stiffness and strength of timber 8-inch and 12-inch deep mixed oak bolt-laminated timber mats. The 8-inch-deep specimen dimensions were 24 inches wide and 12 feet long, while 12-inch-deep specimens were 24 inches wide and 18 feet long. Twenty-eight specimens of each depth were flexurally tested over a 17:1 span to depth ratio. The modulus of elasticity along with the parametric and non-parametric design strength values are presented herein. The characterization of timber crane mats by analyzing their mechanical properties provides a starting point for producers and users to understand the performance and applications of their product. Further research will allow investigation of alternative materials, sizes, and structural designs.

In this study, an artificial neural network (ANN) model was designed to predict color change based on visual assessment of coated cross laminated timber (CLT) exposed outdoors. Coatings and stains were investigated based on ASTM protocols to assess wood surface visual rating, against checking, flaking, erosion, and mildew growth in the State of Mississippi (USA) during one year (2019–2020). It was hypothesized that accurate ratings would promote precise color prediction by the ANN model. Visual assessment inputs were used to develop the model for predicting total color change (ΔE). The training and validation splits of the network were based on a 10-fold cross-validation technique, and the ANN model performance was assessed on the validation set using mean squared error (MSE), mean average precision (MAE), and coefficient of determination (R 2) after permutation feature importance analysis (PFI). Results indicated that coating was the most important feature in color change model. Erosion, checking and flaking achieved similar importance with an approximate difference of 6%. The ANN model was able to effectively predict color change values based on visual ratings with overall accuracy of 95% on truly unseen data. These findings revealed that coating properties, visual appearance, time of exposure, are associated with discoloration. Accurate visual assessment and a well-trained ANN can successfully provide the desired values of ΔE with a smaller number of complex test procedures.

This study reports the feasibility of using deep convolutional neural networks (CNN), for automatically detecting knots on the surface of wood with high speed and accuracy. A limited dataset of 921 images were photographed in different contexts and divided into 80:20 ratio for training and validation, respectively. The “You only look once” (YoloV3) CNN-based architecture was adopted for training the neural network. The Adam gradient descent optimizer algorithm was used to iteratively minimize the generalized intersection-over-union loss function. Knots on the surface of wood were manually annotated. Images and annotations were analyzed by a stack of convolutional and fully connected layers with skipped connections. After training, model checkpoint was created and inferences on the validation set were made. The quality of results was assessed by several metrics: precision, recall, F1-score, average precision, and precision x recall curve. Results indicated that YoloV3 provided knot detection time of approximately 0.0102 s per knot with a relatively low false positive and false negative ratios. Precision, recall, f1-score metrics reached 0.77, 0.79, and 0.78, respectively. The average precision was 80%. With an adequate number of images, it is possible to improve this tool for use within sawmills in the forms of both workstation and mobile device applications.

Cross-laminated timber (CLT) is increasingly used in building construction worldwide. Durability of CLT against fungal attack has yet to be fully explored. Water intrusion in mass timber can yield dimensional changes and microbial growth. This study evaluated the performance of CLT coated with various water- and solvent-based stains commercially available in the United States. Twelve coatings were tested for moisture excluding effectiveness, water repellency effectiveness, volumetric swelling, and anti-swelling efficiency. Only five coatings repelled water, limiting dimensional changes. A modified version of AWPA E10-16 (2016) was performed to evaluate decay of the coated CLT samples. Weight losses were recorded after 18 weeks’ exposure to the brown-rot decay fungus Gloeophyllum trabeum. In accelerated mold testing, coated CLT samples were grown in chambers containing spores of Aspergillus sp., Rhizopus sp., and Penicillium sp. for 29 d and assessed visually for mold growth. In both tests, coating C (transparent, water-based, alkyd/acrylic resin) performed the best among the tested coatings. Mold growth was completely prevented, and weight loss caused by G. trabeum was approximately 1.33%. Although coating C prevented decay for 18 weeks, coatings are not intended to protect against decay fungi. However, they may offer short-term protection during transport, storage, and construction.

Background: Multiple challenges are faced by industry and certification agencies when commercializing tropical species. Anatomical similarities of tropical hardwoods impair identification. Deep learning models can facilitate microscopic identification of wood by using sophisticated techniques such as deep convolutional networks (DCNN). Our objective was to microscopically identify 23 commercially available Brazilian wood species using a custom DCNN model. Results: Photographs from microscopic slides of each wood species were processed, and the final data set contained 2,448 images. We applied stratified k-fold cross-validation technique during training to increase model’s robustness and trustworthiness. Thus, the dataset was divided into approximately 80% training (1,958 images) and 20% validation (490 images) for each fold. A series of augmentations were performed only on training data to include variations in rotation, zoom, and perspective. Image augmentation was performed on-the-fly. The network consisted of convolutions, max pooling, global average pooling, and fully connected layers. We tested the performance of the DCNN against accuracy, precision, recall, and F1-score on the validation set for each fold. Conclusion: The custom machine learned model accuracy was higher than 0.90. The model’s worst performance was identified in distinguishing between Toona ciliate and Khaya ivorensis, which was due more to wood variability than to a machine learning deficiency. Future studies should focus on integration, verification/monitoring, and updating of current models for end user manipulation, trust, ethics, and security.

Attack by defoliator insects reduces wood production and may affect quality. The objective of this work was to evaluate the tree growth, chemical, physical and anatomical properties of 5-year-old Eucalyptus grandis × Eucalyptus urophylla wood, both sound and attacked by Gonipterus platensis (Coleoptera: Curculionidae) and Thyrinteina arnobia (Lepidoptera: Geometridae). Eucalypt pulpwood-class trees were collected in commercial areas attacked by both defoliator insects in Itarare and Lencois Paulistas, Sao Paulo, Brazil. Attacks by G. platensis reduced tree height growth, decreased the mean annual increment, increased pentosans content, but did not affect basic density and anatomical characteristics. Attacks by T. arnobia reduced the cell wall thickness, basic density, insoluble and total lignin, and ash content. The radial profile of trees attacked by T. arnobia expressed lower air-dry density when compared to sound ones. The intensity of the effects on wood depends on the season, length of time, and severity of the attack. The alterations on tree growth and wood properties due to defoliation may jeopardize the final use of the wood product. As the studied trees were destined for pulpwood and paper production, important variables like wood consumption, yield of pulp, and pulpwood properties would be compromised.

This manuscript confirms the feasibility of using a long short-term memory (LSTM) recurrent neural network (RNN) to forecast lumber stock prices during the great and Coronavirus disease 2019 (COVID-19) pandemic recessions in the USA. The database was composed of 5012 data entries divided into recession periods. We applied a timeseries cross-validation that divided the dataset into an 80:20 training/validation ratio. The network contained five LSTM layers with 50 units each followed by a dense output layer. We evaluated the performance of the network via mean squared error (MSE), root mean squared error (RMSE), and mean absolute error (MAE) for 30, 60, and 120 timesteps and the recession periods. The metrics results indicated that the network was able to capture the trend for both recession periods with a remarkably low degree of error. Timeseries forecasting may help the forest and forest product industries to manage their inventory, transportation costs, and response readiness to critical economic events.

This manuscript provides a brief review about chemical and biological agents used to bioremediate treated wood waste. Wood preservatives have been used to increase wood’s useful life, because any species is subject to decay. Studies indicate that the disposal of treated wood after its service has drawn concern and scrutiny. Practices have included disposal in landfills or construction sites as well as destruction by burning, so it is apparent that more environmentally friendly options are needed. To mitigate these problems, acidic agents, fungi, and bacteria can be used as alternatives to remove heavy metals. At optimum temperature and concentration, acids play a major role in the removal process. The process is enhanced when a bioremediation technique is used after chemical extraction. In fact, bioremediation has been shown to be a remarkable technique for recovering copper, arsenic, creosote, and other compounds. The major drawback is the extensive duration of fungal activity for release of heavy metals.

The genera Eucalyptus and Corymbia are widely used in Brazil. Although they present remarkable applicability, they manifest substantial end-splitting and surface checks, which allows wood decay organisms to penetrate the wood. Thereby, the resistance of Eucalyptus grandis x Eucalyptus urophylla (EU) and Corymbia citriodora (CT) treated with chromated copper arsenate type-C (CCA-C) against fungi decay was evaluated. Seventy-two fence posts were assessed; for each species, there were 18 posts treated with CCA-C and 18 non-treated posts. The posts were 2.20 m long and classified into three classes of diameter. The 2% active ingredient was used with a vacuum-pressure cycle. On each fence, disks measuring 2.0 cm thick were cut at the outcrop zone. Two sets of depths were analyzed: the edge at 0 cm to 1.5 cm as well as the inner part at 1.5 cm to 3.0 cm. The samples were subjected to Postia placenta, Gloeophyllum trabeum, and Trametes versicolor attack. The treatment was effective against all fungi, but for CT, the diameter range of 8 cm to 12 cm was optimal. The treated wood from EU samples reached the lowest weight loss for all fungi. Heartwood-sapwood ratio played a major role. By comparing the non-treated woods, EU yielded the highest mass loss.

The productive sustainability of eucalyptus planting is threatened when both diseases and disorders of growth occur. These interferences can have abiotic origin as environmental conditions can negatively affect the wood quality. This study has the goal of evaluating influence of different levels of physiological disorders in dendrometric variables and wood quality from seven-year-old Eucalyptus grandis x Eucalyptus urophylla hybrids intended for wood pulping production. The trees from commercial plantations in State of Maranhão, Brazil were separated into three levels of physiological disorders by means of morphologic changes. Then, they were evaluated according to dendrometric variables, anatomical characteristics, chemical properties, density of wood, and kraft pulp process. The physiological disorder levels significantly influenced all evaluated properties. Moreover, the wood from trees with higher severity of symptoms resulted in higher average values of dendrometric variables, basic density, and fibers dimensions. In addition, apparent density exhibited a different standard in the pith region with increase on radial wood profile. Trees with the least severe disorders exhibited changes in dendrometric and anatomical variables, mainly in wood volume and fibers dimensions, respectively.

Heat treatment (HT) of a mixture of tannins and inorganic boron compounds showed effective results against wood decay organisms. Boron compounds play a critical role in the preservation of wood against wood decay organisms. The use of tannins and HT represents a relatively new environmentally friendly approach to the wood preservation industry. The aim of this study was to investigate the effect of tannin impregnation and HT on boron depletion, and termite and fungi resistance. Southern yellow pine (SYP) and yellow-poplar (YP) cube samples were used for this study. A mixture of condensed tannins from the Quebracho tree and disodium octaborate tetrahydrate (DOT) was injected into the specimens using a vacuum/pressure cycle, and the specimens were then heat-treated under N2 atmosphere for four hours at 190 &deg, C to investigate both the tannin&rsquo, s fixative ability to reduce boron leaching and the performance of the mixture against brown and white-rot fungi and termites. Tannins restricted boron leaching in 46% and 34% for SYP and YP, respectively, and also significantly increased the resistance against white-rot fungi for YP. Tannins and HT showed unpredictably good performance against termites. Tannins may be acting to denature proteins, in that case, fungal enzymes would be inhibited. This study revealed the importance of HT on a mixture of boron and tannins to decrease boron leaching and increase the durability of preservative systems.

Cross-laminated timber (CLT) market demand is on the rise in the United States. Adequate protective measures have not been extensively studied. The objective of this study was to investigate the weathering performance of exterior wood coatings. We evaluated coated CLT sample surfaces based on visual appearance, color change (CIE*L*a*b), gloss changes, and water intrusion. From the five exterior wood coatings evaluated, only two showed adequate performance after twelve months field exposure. Based on visual ratings following the ASTM procedures, coating failure occurs more quickly in Mississippi than in Wisconsin, due to its greater decay zone. Both location and coating type impacted the aging of the samples. Artificial weathering results were consistent with natural weathering indicating the two adequate coatings were the most resistant to failure, color, and gloss change. For future studies, new coatings designed for the protection of end-grain in CLT panels should be a target of research and development.

Resumen: El objetivo de la investigación fue evaluar el efecto del ataque de Leptocybe invasa en las propiedades dendrométricas de la madera de dos clones de eucalipto (Eucalyptus urophylla y Eucalyptus tereticornis), cultivados en los municipios de Imperatriz (Estado do Maranhão) y Darcinópolis (Estado de Tocantins), Brasil, destinados a la producción de pulpa kraft. Se analizaron 10 árboles de 3 años de edad atacados por Leptocybe invasa, para cada clon. El diagnóstico fue basado en la formación de grandes agallas en la copa, y con orificio de emergencia (cerrando el ciclo). Los resultados fueron constatados por la variación de las características de las especies evaluadas con relación a su altura total, diámetro, volumen de corteza, frecuencia de vasos, longitud de fibras, perfil radial de densidad y contenido de cenizas. Los árboles hospederos presentaron alteraciones en las variables dendrométricas, contenido de pentosanos y cenizas en la madera para ambos materiales genéticos, no siendo observadas variaciones en la densidad básica, dimensión de vasos y fibras, así como en los contenidos de lignina y extractivos de la madera.