Search

Near-infrared (NIR) spectroscopy is widely used as a nondestructive evaluation (NDE) tool for predicting wood properties. When deploying NIR models, one faces challenges in ensuring representative training data, which large datasets can mitigate but often at a significant cost. Machine learning and deep learning NIR models are at an even greater disadvantage because they typically require higher sample sizes for training. In this study, NIR spectra were collected to predict the modulus of elasticity (MOE) of southern pine lumber (training set = 573 samples, testing set = 145 samples). To account for the limited size of the training data, this study employed a generative adversarial network (GAN) to generate synthetic NIR spectra. The training dataset was fed into a GAN to generate 313, 573, and 1000 synthetic spectra. The original and enhanced datasets were used to train artificial neural networks (ANNs), convolutional neural networks (CNNs), and light gradient boosting machines (LGBMs) for MOE prediction. Overall, results showed that data augmentation using GAN improved the coefficient of determination (R 2 ) by up to 7.02% and reduced the error of predictions by up to 4.29%. ANNs and CNNs benefited more from synthetic spectra than LGBMs, which only yielded slight improvement. All models showed optimal performance when 313 synthetic spectra were added to the original training data; further additions did not improve model performance because the quality of the datapoints generated by GAN beyond a certain threshold is poor, and one of the main reasons for this can be the size of the initial training data fed into the GAN. LGBMs showed superior performances than ANNs and CNNs on both the original and enhanced training datasets, which highlights the significance of selecting an appropriate machine learning or deep learning model for NIR spectral-data analysis. The results highlighted the positive impact of GAN on the predictive performance of models utilizing NIR spectroscopy as an NDE technique and monitoring tool for wood mechanical-property evaluation. Further studies should investigate the impact of the initial size of training data, the optimal number of generated synthetic spectra, and machine learning or deep learning models that could benefit more from data augmentation using GANs.

North American conifers exhibit three radial specific gravity (SG) patterns (Type 1, 2, and 3), which balance hydraulic and mechanical requirements. Type 1 and 2 patterns (Pinaceae) have low SG and high microfibril angle (MFA) corewood ensuring compliance, whereas in outerwood high SG/low MFA provide stiffness and strength resisting bending. Hydraulically, corewood, especially in Type 2 species, is resistant to embolism, whereas outerwood has higher specific conductivity. Cupressaceae (Type 3) have hydraulically very efficient, low SG outerwood, facilitating rapid growth. Corewood is flexible, whereas outerwood is mechanically weak and compensated for by more conical stems and durable heartwood (which prevents Brazier buckling). Radially earlywood (EW) decreases, and latewood (LW) increases for all types, whereas %latewood (%LW) increases (Type 1), decreases, then increases (Type 2) and decreases (Type 3). Ring SG increases when increasing LW SG and %LW are sufficient to counteract decreasing EW SG. Shade tolerance, crown recession, hormone gradients, and environmental variation affect patterns. Auxin concentration decreases with increasing distance from juvenile foliage slowing cell division, concomitantly gibberellin concentration (lignification) and carbohydrates (cell wall thickening) increase, producing higher %LW. Across a species range regions receiving relatively high summer rainfall have trees with higher %LW (by ring). [ABSTRACT FROM AUTHOR]

A chemistry-based explainable machine learning (ML) approach was used to predict wood properties using near infrared (NIR) spectral data collected from rough and smooth surfaces, and to provide better understanding of the role of important NIR wavelengths (features) in the performance of ML models. NIR spectra collected from western hemlock (Tsuga heterophylla) and coastal Douglas-fir (Pseudotsuga menziesii) boards with rough and smooth surfaces were fed into random forest and TreeNet; a gradient boosting machine algorithm, for predicting wood density, modulus of elasticity (MOE) and modulus of rupture (MOR). The TreeNet model could predict the MOE, MOR, and density with R2 of 0.66, 0.64, and 0.64 using spectra collected from rough surface and R2 of 0.54, 0.46, and 0.46 using spectra collected from smooth surface. TreeNet outperformed the random forest, and for both ML algorithms higher R2 and lower error were obtained using NIR data collected from rough surfaces. This suggested that for Douglass fir and western hemlock, NIR spectra could be collected on a sawn surface prior to surface planing. However, it is difficult to generalize the impact of surface roughness on the performance of predictive model as different factors (e.g. what constitutes a smooth or rough surface, variability of data set in terms of wood properties) impact the success of predictive models. NIR features having the greatest influence on TreeNet models were examined and consistently had wood chemistry specific band assignments. The most important features occurred in the O-H first overtone, and C–H second overtone regions and a narrow zone (approximately 2400–2500 nm) of the C–H stretch C–C stretch combination region. Important features also differed by property and with surface roughness. Explaining ML model performance using the relative importance of the NIR features showed the importance of wood chemistry related information when developing models, however MOE and MOR TreeNet models based on smooth surface NIR spectra showed an increased importance of water related features. Overall, the chemistry-based explainable machine learning model approach allows for identification of important NIR features, and regions, and aids in understanding how they contribute to the performance of NIR-based wood property predictive models. [ABSTRACT FROM AUTHOR]

The quality control of thermally modified wood and identifying heat treatment intensity using nondestructive testing methods are critical tasks. This study used near-infrared (NIR) spectroscopy and machine learning modeling to classify thermally modified wood. NIR spectra were collected from the surfaces of untreated and thermally treated (at 170 °C, 212 °C, and 230 °C) western hemlock samples. An explainable machine learning approach was practiced using a TreeNet gradient boosting machine. No dimensionality reduction was performed to better explain the feature ranking results obtained from the model and provide insight into the critical wavelengths contributing to the performance of classification models. NIR spectra in the ranges of 1100–2500 nm, 1400–2500 nm, and 1700–2500 nm were fed into the TreeNet model, which resulted in classification accuracy values (test data) of 94.35%, 89.29%, and 84.52%, respectively. Feature ranking analysis revealed that when using the range of 1100–2500 nm, the changes in wood color resulted in the highest variation in NIR reflectance amongst treatments. As a result, associated features were given higher importance by TreeNet. Limiting the wavelength range increased the significance of features related to water or wood chemistry; however, these predictive models were not as accurate as the one benefiting from the impact of wood color change on the NIR spectra. The developed framework could be applied to different applications in which NIR spectra are used for wood characterization and quality control to provide improved insights into selected NIR wavelengths when developing a machine learning model. [ABSTRACT FROM AUTHOR]

Hyperspectral imaging is a technique that combines spectroscopy and imaging. Originally utilized in the 1980's by the remote sensing community it is now utilized in a wide variety of applications. Spectral imaging was first used for the detection of compression wood in the late 1990's and since that time research focused on wood and wood products has steadily increased with a variety of applications reported. While there are several reviews of wood related research utilizing near infrared spectrometers a comprehensive summary of wood—hyperspectral imaging research is lacking. Near infrared hyperspectral imaging systems (NIR-HSI) typically have a wavelength range of 900–1700 nm, whereas short-wave infrared hyperspectral imaging (SWIR-HSI) systems range from 1000 to 2500 nm. We provide a detailed account of the various studies that have been published utilizing both camera types. [ABSTRACT FROM AUTHOR]

Rubberwood (Hevea brasiliensis Muell. Arg) is underutilized in most developing countries but has potential to be a solution to the shortage of wood for pulp, paper and wood composite products at the end of its production cycle. Determining and understanding its fibre properties (fibre dimensions and morphology) is key in its utilization for pulp and paper as well as composites. This study assessed the effect of age on the fibre properties of rubberwood. Samples of wood at four ages (10, 15, 20 and 25 years) were obtained at both the axial (base and top of merchantable length) and radial direction (innerwood, middlewood and outerwood). Slivers were obtained from the samples and macerated to provide individual fibres for optical measurement. Fibre dimensions at the four ages were measured, and their morphology was determined. Fibre properties and morphology were significantly affected by age at p = 0.05. Average fibre length was 1.47 ± 0.12 mm, with non-constant variation observed across the different ages, and from pith to bark. Average lumen width was 16.87 µm, and fibre diameter ranged from 25.02 µm to 27.23 µm. The fibre properties of rubberwood potentially make it suitable for pulp and paper production as well as wood composite boards. [ABSTRACT FROM AUTHOR]

There is a growing interest in improving wood properties through the appropriate selection of seed provenances within species. In this study, wood density and mechanical properties were investigated for Acacia mangium trees from six different provenances, planted in Quang Tri, Vietnam. Radial and among provenance variation in stress wave velocity (SWV), wood density (WD), dynamic modulus of elasticity (MOEd), modulus of rupture (MOR), and modulus of elasticity (MOE) were determined on a total of 480 small clear specimens (20 × 20 × 300 mm) cut from 30 trees (five per provenance). SWV and selected wood properties near the pith were significantly lower than those near the bark. Differences in all selected mechanical properties among provenances were significant. The highest static properties (MOR and MOE) were found for the Long Thanh provenance indicating its potential suitability for breeding programs in Vietnam focused on improving A. mangium wood quality. A high positive correlation coefficient was found between MOEd and MOE (r = 0.93, P < 0.001) and our results indicate that the stiffness of A. mangium can be predicted using stress wave method provided the density of measured element is known. [ABSTRACT FROM AUTHOR]

Prediction of pulp yield of Eucalyptus globulus wood samples based on partial least squares (PLS) regression can be optimized by utilizing specific near infrared (NIR) wavelengths. A critical feature of this approach is the weighting of constraint conditions. Equal weighting balances optimization in terms of calibration and prediction; however, there is a lack of knowledge regarding prediction performance of wood property models when different weight factors are used. In this study, pulp yield models were developed using two E. globulus data sets characterized by narrow (5%) and extreme (22.6%) yield ranges and represented by untreated and second derivative NIR spectra. The global optimization solver pySOT was used to optimize the performance of a PLS regression model in terms of wavelengths selected and number of latent variables. A linear function of R-squares for calibration ( R c 2 ) and prediction ( R p 2 ) sets was utilized as the objective function with the aim of maximizing α R c 2 + 1 - α R p 2 for all values of α between 0 (maximizing R p 2 without concern for R c 2 ) and 1 (only maximizing R c 2 ). Values of α ≤ 0.8 provided good predictive performance, whereas α ≥ 0.9 tended to overfit the calibration data indicating that models are robust for values of α from 0 to 0.8. Representative wavelengths for each data set were identified and assigned to corresponding wood components through a band assignment process. Strong agreement was observed for α ≤ 0.8 ; however, for α ≥ 0.9 , identified wavelengths generally occurred in regions unrelated to vibrations arising from specific wood components. [ABSTRACT FROM AUTHOR]

Near-infrared wavelengths selected by genetic algorithm were used to optimize partial least squares (PLS) regression models for loblolly pine (Pinus taeda L.) from the southeastern United States. Wood properties examined included density (D), microfibril angle, modulus of elasticity and tracheid coarseness (C), radial diameter (R), tangential diameter (T), and wall thickness (w)—measured by SilviScan. The optimization process was run for each property with Agenda 2020 samples utilized for PLS model development and the other sets used for prediction. The number of variables (i.e. wavelengths) varied from 10 to 100 with an optimum number identified by genetic algorithm. When compared to a full data set model (based on 700 wavelengths), calibration and prediction performance of optimized PLS regression models were superior for all properties. Importantly, representative wavelengths for each property were consistently related to recognized bond vibrations observed in specific wood components demonstrating that optimization targets wavelengths directly related to changes in wood chemistry within the examined loblolly pine samples. [ABSTRACT FROM AUTHOR]

Acoustic tools are increasingly used to estimate standing-tree (dynamic) stiffness; however, such techniques overestimate static stiffness, the standard measurement for determining modulus of elasticity (MOE) of wood. This study aimed to identify correction methods for standing-tree estimates making dynamic and static stiffness comparable. Sixty Pinus taeda L. trees, ranging from 14 to 19 years old, obtained from genetic tests established in the southeastern United States, were analyzed. Standing-tree acoustic velocities were measured using the TreeSonic tool. Acoustic velocities were also recorded in butt logs cut from the same trees using the Director HM200. A strong but biased relationship between tree and log velocities was observed, with tree velocities 32% higher (on average) than the corresponding log velocities. Two correction methods, one for calibrating tree velocities and one for accounting for differences in wood moisture content, were used to determine an adjusted MOE. After correction, adjusted MOE estimates were in good agreement with static longitudinal MOE values measured on clearwood specimens obtained from the trees, and no systematic bias was observed. The results of this study show that acoustic estimates of MOE on standing trees largely depend on how the data are processed and the reference method used. [ABSTRACT FROM AUTHOR]

Copyright of Canadian Journal of Forest Research is the property of Canadian Science Publishing and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Loblolly pine (Pinus taeda L.) is one of the most important commercial timber species in the world. While the species is native to the southeastern United States of America (USA), it has been widely planted in southern Brazil, where it is the most commonly planted exotic species. Interest exists in utilizing nondestructive testing methods for wood property assessment to aid in improving the wood quality of Brazilian grown loblolly pine. We used near-infrared hyperspectral imaging (NIR-HSI) on increment cores to provide data representative of the radial variation of families sampled from a 10-year-old progeny test located in Rio Negrinho municipality, Santa Catarina, Brazil. Hyperspectral images were averaged to provide an individual NIR spectrum per tree for cluster analysis (hierarchical complete linkage with square Euclidean distance) to identify trees with similar wood properties. Four clusters (0, 1, 2, 3) were identified, and based on SilviScan data for air-dry density, microfibril angle (MFA), and stiffness, clusters differed in average wood properties. Average ring data demonstrated that trees in Cluster 0 had the highest average ring densities, and those in Cluster 3 the lowest. Cluster 3 trees also had the lowest ring MFAs. NIR-HSI provides a rapid approach for collecting wood property data and, when coupled with cluster analysis, potentially, allows screening for desirable wood properties amongst families in tree improvement programs. [ABSTRACT FROM AUTHOR]