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

Hein, Paulo Ricardo Gherardi; Pakkanen, Hannu; Santos, António A. Dos

doi:10.5424/fs/2017263-11892

Cited by 20 publications

(5 citation statements)

References 54 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In the past few years, near-infrared (NIR) spectroscopy has developed rapidly as a non-destructive testing technique that has the characteristics of convenient analytical instruments, 8 no sample preparation, 9,10 and real-time monitoring. 11,12 NIR is widely used in food, [13][14][15] medicine, 16,17 and other elds.…”

Section: Introductionmentioning

confidence: 99%

Detection and analysis of hyaluronic acid raw materials from different sources by NIR and aquaphotomics

Ma,

Guo,

Lin

et al. 2024

Anal. Methods

View full text Add to dashboard Cite

show abstract

Section: Introductionmentioning

confidence: 99%

Detection and analysis of hyaluronic acid raw materials from different sources by NIR and aquaphotomics

Ma,

Guo,

Lin

et al. 2024

Anal. Methods

View full text Add to dashboard Cite

show abstract

“…Despite the growing opportunities that NIR spectroscopy offers for wood characterization and QC, there are still limitations associated with this technique. Apart from the need for a robust calibration model, spectral data acquisition in industrial settings imposes practical challenges because of noise related to temperature and humidity variations (Hein et al 2017). Specifically, there is a significant gap between the performance of NIR spectroscopy in laboratory vs industrial conditions, where the high performance achieved under laboratory-controlled conditions using multivariate data analysis may not necessarily account for the high variation in data acquired in real time in a manufacturing environment (Hein et al 2017).…”

Section: Introductionmentioning

confidence: 99%

Fiber Quality Prediction Using Nir Spectral Data: Tree-Based Ensemble Learning VS Deep Neural Networks

Nasir

Mohammadpanah²,

Raut³

et al. 2023

WFS

View full text Add to dashboard Cite

The growing applications of near infrared (NIR) spectroscopy in wood quality control and monitoring necessitates focusing on data-driven methods to develop predictive models. Despite the advancements in analyzing NIR spectral data, literature on wood science and engineering has mainly utilized the classic model development methods, such as principal component analysis (PCA) regression or partial least squares (PLS) regression, with relatively limited studies conducted on evaluating machine learning (ML) models, and specifically, artificial neural networks (ANNs). This could potentially limit the performance of predictive models, specifically for some wood properties, such as tracheid width that are both time-consuming to measure and challenging to predict using spectral data. This study aims to enhance the prediction accuracy for tracheid width using deep neural networks and tree-based ensemble learning algorithms on a dataset consisting of 2018 samples and 692 features (NIR spectra wavelengths). Accordingly, NIR spectra were fed into multilayer perceptron (MLP), 1 dimensional-convolutional neural networks (1D-CNNs), random forest, TreeNet gradient-boosting, extreme gradient-boosting (XGBoost), and light gradient-boosting machine (LGBM). It was of interest to study the performance of the models with and without applying PCA to assess how effective they would perform when analyzing NIR spectra without employing dimensionality reduction on data. It was shown that gradient-boosting machines outperformed the ANNs regardless of the number of features (data dimension). All the models performed better without PCA. It is concluded that tree-based gradient-boosting machines could be effectively used for wood characterization utilizing a medium-sized NIR spectral dataset.

show abstract

“…Sendo, o NIRS um rápido diagnóstico quando comparado com análises convenconais, com um equipamento de fácil operação e boa precisão. Podendo ser utilizado para estimar e analisar parâmetros ou moléculas que contenham ligações CH, NH, SH ou OH, colaborando na padronização e diferenciação das amostras (Hein et al, 2017;.…”

Section: Introductionunclassified

Caracterização anatômica e espectroscópica para identificação de madeiras da família Fabaceae

Sobral,

Gouvêa,

Carvalho

2022

Br. J. of B. Sci.

View full text Add to dashboard Cite

O comércio ilegal madereiro gera grandes prejuízos econômicos e ambientais em nosso país, sendo necessário realizar uma proteção regulatória para impedir estas atividades ilegais. É comum a apreensão das madeiras já desdobradas, sem os componentes comumente usados para identificação das espécies, sendo de suma importância conhecer as propriedades anatômicas e químicas para a correta classificação. A adoção de metodologias de avaliação não destrutivas de madeira vem se tornando uma alternativa viável e eficaz, sendo a espectroscópia no infravermelho próximo (NIRS) uma ferramenta que colabora na caracterização das madeiras, reduzindo os custos e o tempo nas análises convencionais. O objetivo deste estudo foi avaliar a caracterização química e identificar as madeiras da família Fabaceae, a partir do NIRS. Avaliou-se seis espécies nativas: amburana, fava de anta, pau óleo, faveiro, angelim pedra e rouxinho. Realizou-se caracterizações químicas convencional (extrativos totais, lignina solúvel, lignina insolúvel e lignina total, holocelulose, cinzas e relação S/G) e anatômicas. Para a caracterização anatômica foi confeccionado corpos-de-prova do cerne periférico 2 x 2 x 2 cm, passando por etapas de cozimento, coloroção, descoloração, desidratação e realização de cortes histológicos para montagem das lâminas permanentes. Utilizou-se para mensuração dos elementos o microscópio óptico da Leica DM750 com uma câmera ICC50 (50x e 100x). Foram realizadas leituras das serragens no equipamento NIRS e obtidos os valores preditos a partir de um modelo pré-existente para caracterização química, e posteriormente correlacionados com os valores de referência. Os resultados obtidos demonstraram que o NIRS apresentou eficiência na identificação das amostras de maneira rápida, clara e objetiva, tendo eficiência na determinação do teor de holocelulose e de extrativos totais, porém para as demais características químicas os modelos precisam de melhores ajustes.

show abstract

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

Cited by 20 publications

References 54 publications

Detection and analysis of hyaluronic acid raw materials from different sources by NIR and aquaphotomics

Detection and analysis of hyaluronic acid raw materials from different sources by NIR and aquaphotomics

Fiber Quality Prediction Using Nir Spectral Data: Tree-Based Ensemble Learning VS Deep Neural Networks

Caracterização anatômica e espectroscópica para identificação de madeiras da família Fabaceae

Contact Info

Product

Resources

About