Lifting Wavelet Transform De-noising for Model Optimization of Vis-NIR Spectroscopy to Predict Wood Tracheid Length in Trees

Comparison of various chemometric methods on visible and near-infrared spectral analysis for wood density prediction among different tree species and geographical origins

Visible and near-infrared (Vis-NIR) spectroscopy has been widely applied in many fields for the qualitative and quantitative analysis. Chemometric techniques including pre-processing, variable selection, and multivariate calibration models play an important role to better extract useful information from spectral data. In this study, a new de-noising method (lifting wavelet transform, LWT), four variable selection methods, as well as two non-linear machine learning models were simultaneously analyzed to compare the impact of chemometric approaches on wood density determination among various tree species and geographical locations. In addition, fruit fly optimization algorithm (FOA) and response surface methodology (RSM) were employed to optimize the parameters of generalized regression neural network (GRNN) and particle swarm optimization-support vector machine (PSO-SVM), respectively. As for various chemometric methods, the optimal chemometric method was different for the same tree species collected from different locations. FOA-GRNN model combined with LWT and CARS deliver the best performance for Chinese white poplar of Heilongjiang province. In contrast, PLS model showed a good performance for Chinese white poplar collected from Jilin province based on raw spectra. However, for other tree species, RSM-PSO-SVM models can improve the performance of wood density prediction compared to traditional linear and FOA-GRNN models. Especially for Acer mono Maxim, when compared to linear models, the coefficient of determination of prediction set ( R p 2 ) and relative prediction deviation (RPD) were increased by 47.70% and 44.48%, respectively. And the dimensionality of Vis-NIR spectral data was decreased from 2048 to 20. Therefore, the appropriate chemometric technique should be selected before building calibration models.

Lifting wavelet transform for Vis-NIR spectral data optimization to predict wood density

Visible and near infrared (Vis-NIR) spectroscopy is a mature analytical tool for qualitative and quantitative analysis in various sectors. However, in the face of "curse of dimensionality" due to thousands of wavelengths for a Vis-NIR spectrum of a sample, the complexity of computation and memory will be increased. Additionally, variable optimization technique can be used to improve prediction accuracy through removing some irrelevant information or noise. Wood density is a critical parameter of wood quality because it determines other important traits. Accurate estimation of wood density is becoming increasingly important for forest management and end uses of wood. In this study, the performance of two-dimensional (2D) correlation spectroscopy between wavelengths of various spectral transformations, i.e., reflectance spectra (R), reciprocal (1/R), and logarithm spectra (log (1/R)), were analyzed before optimizing spectral variable. The spectra of optimal transformation were decomposed using biorthogonal wavelet family from 3rd to 8th decomposition level based on lifting wavelet transform (LWT). The optimal wavelet coefficients of LWT were selected based on the performance of calibration set using partial least squares (PLS). Two frequent variable selection methods including uninformative variable elimination (UVE) and competitive adaptive reweighted sampling (CARS) were also compared. The results showed that the dimensionality of spectral matrix was reduced from 2048 to 16 and the best density prediction results of Siberian elm (Ulmus pumila L.) were obtained (R_p²R = 0.899, RMSEP = 0.016) based on LWT.

Development of Low-Cost Portable Spectrometers for Detection of Wood Defects

Portable spectroscopic instruments are an interesting alternative for in-field and on-line measurements. However, the practical implementation of visible-near infrared (VIS-NIR) portable sensors in the forest sector is challenging due to operation in harsh environmental conditions and natural variability of wood itself. The objective of this work was to use spectroscopic methods as an alternative to visual grading of wood quality. Three portable spectrometers covering visible and near infrared range were used for the detection of selected naturally occurring wood defects, such as knots, decay, resin pockets and reaction wood. Measurements were performed on wooden discs collected during the harvesting process, without any conditioning or sample preparation. Two prototype instruments were developed by integrating commercially available micro-electro-mechanical systems with for-purpose selected lenses and light source. The prototype modules of spectrometers were driven by an Arduino controller. Data were transferred to the PC by USB serial port. Performance of all tested instruments was confronted by two discriminant methods. The best performing was the microNIR instrument, even though the performance of custom prototypes was also satisfactory. This work was an essential part of practical implementation of VIS-NIR spectroscopy for automatic grading of logs directly in the forest. Prototype low-cost spectrometers described here formed the basis for development of a prototype hyperspectral imaging solution tested during harvesting of trees within the frame of a practical demonstration in mountain forests.

Visible-Near Infrared Spectroscopy and Chemometric Methods for Wood Density Prediction and Origin/Species Identification

This study aimed to rapidly and accurately identify geographical origin, tree species, and model wood density using visible and near infrared (Vis-NIR) spectroscopy coupled with chemometric methods. A total of 280 samples with two origins (Jilin and Heilongjiang province, China), and three species, Dahurian larch (Larix gmelinii (Rupr.) Rupr.), Japanese elm (Ulmus davidiana Planch. var. japonica Nakai), and Chinese white poplar (Populus tomentosa carriere), were collected for classification and prediction analysis. The spectral data were de-noised using lifting wavelet transform (LWT) and linear and nonlinear models were built from the de-noised spectra using partial least squares (PLS) and particle swarm optimization (PSO)-support vector machine (SVM) methods, respectively. The response surface methodology (RSM) was applied to analyze the best combined parameters of PSO-SVM. The PSO-SVM model was employed for discrimination of origin and species. The identification accuracy for tree species using wavelet coefficients were better than models developed using raw spectra, and the accuracy of geographical origin and species was greater than 98% for the prediction dataset. The prediction accuracy of density using wavelet coefficients was better than that of constructed spectra. The PSO-SVM models optimized by RSM obtained the best results with coefficients of determination of the calibration set of 0.953, 0.974, 0.959, and 0.837 for Dahurian larch, Japanese elm, Chinese white poplar (Jilin), and Chinese white poplar (Heilongjiang), respectively. The results showed the feasibility of Vis-NIR spectroscopy coupled with chemometric methods for determining wood property and geographical origin with simple, rapid, and non-destructive advantages.