NIR PLSR results obtained by calibration with noisy, low-precision reference values: Are the results acceptable?

Abstract

Both spectral noise and reference method noise affect the accuracy and the precision NIR predicted values. The reference noise is often neglected, and the few reports dealing with it only consider random noise artificially added to the original sound reference data. A calibration for lignin content of maritime pine (Pinus pinaster Ait.) wood meal was developed, but due to low precision and accuracy in the reference data set, NIR partial least-squares regression (PLSR) yielded a slope of 0.51 and an intercept at 14% Klason lignin. We demonstrate with an independent data set for external validation, obtained with higher precision and accuracy, that the NIR PLSR model based on the noisy reference data led to better results. The slope of the correlation between predicted and reference values was 0.89 and the intercept was 3.9. Thus, the model performed much better than expected from the cross-validation results. The predictability can be explained by the facts that the loadings of the first principal component (PC) of the calibration and test samples are very similar and dominated by lignin-related bands, and that most of the variation in the test set can be explained by the first PC. This only explains why the Klason lignin content could be predicted with the model without giving many spectral outliers, but not the good result of the external validation. We show that the latter can be explained by the inverse calibration used for PLSR and that predicted values can be more accurate and precise than the reference values used for calibration.

Estimation of physical and mechanical properties of agro-based particleboards by near infrared spectroscopy

“…Although it is well known that both spectral noise and reference method affect the accuracy and the precision of NIR predicted values (Rodrigues et al 2006), in this study, different samples (cut from the same board) were used for NIR spectra acquisition and for determination of the reference data. In principle this procedure would cause a representativeness problem disabling the use of NIRS.…”

Near infrared (NIR) spectroscopy for estimating the chemical composition of (Acacia mangium Willd.) wood

“…The PLS method is typically used to identify and quantify chemical components in wood based on NIR spectra and provides better predictive diagnostics than other methods (e.g., principal components regression) in wood chemistry (Rodrigues et al 2006;Via et al 2014). The present study was aimed at developing a fast and accurate method to measure the chemical components of mangium wood based on NIRS.…”

Challenges in the use of Near Infrared Spectroscopy for improving wood quality: A review

“…The best way to test a calibration model, whether it is a quantitative or a qualitative model, is to have some samples in reserve, that are not included among the ones on which the calibration calculations are based, and use those samples as validation samples, sometimes called test or prediction samples (Mark & Workman, 2007). In regard to the validation of the NIR-based models for wood properties, several studies have used independent test sets to validate their NIRbased calibration for wood traits (Rodrigues et al, 2006;Sousa-Correia et al, 2007;Hein et al, 2009). Rodrigues et al (2006) presented an interesting paper where they demonstrated that predicted values can be better than expected from cross-validation results using Klason lignin content estimations in wood meal samples of 15-year-old maritime pine (Pinus pinaster).…”

“…In regard to the validation of the NIR-based models for wood properties, several studies have used independent test sets to validate their NIRbased calibration for wood traits (Rodrigues et al, 2006;Sousa-Correia et al, 2007;Hein et al, 2009). Rodrigues et al (2006) presented an interesting paper where they demonstrated that predicted values can be better than expected from cross-validation results using Klason lignin content estimations in wood meal samples of 15-year-old maritime pine (Pinus pinaster). They developed calibration for lignin content, but due to low precision and accuracy in the reference data set, NIR-based regression yielded a slope of 0.51 and an intercept at 14% lignin.…”

“…Despite many papers have demonstrated that NIR models present good performance in cross-validations or independent validations for wood density (Via et al, 2003(Via et al, , 2005Cogdill et al, 2004;Mora et al, 2008;Hein et al, 2009), mechanical properties (Schimleck et al, 2001;Kelley et al, 2004;Fujimoto et al, 2008) and chemical composition (Gierlinger et al, 2002;Kelley et al, 2004;Rodrigues et al, 2006;Sousa-Correia et al, 2007), there is still a lack for applications in real situations. While a number of studies have demonstrated that the NIR spectroscopy and multivariate analysis works well in laboratories where the conditions are controlled, limited studies have been done simulating in real industrial conditions, like pulp mills or wood processing factories.…”