Identification of twelve frequently processed wood species from Ivorian forests by carpenters using near-infrared spectroscopy
Abstract
This study demonstrates the efficacy of near-infrared (NIR) spectroscopy combined with chemometric techniques for identifying twelve tropical wood species commonly processed in Ivorian carpentry workshops. Principal Component Analysis (PCA) was employed to explore spectral variability, revealing distinct groupings corresponding to differences in chemical composition, wood density, and anatomical characteristics. Although some overlap was observed between certain species, three major trends could be distinguished, indicating meaningful structural or chemical similarities within each group. Subsequently, a Partial Least Squares Discriminant Analysis (PLS-DA) mode! was developed using 864 spectral samples. T he mode! achieved a classification accuracy of 97.8% with raw spectra, which increased to 98.96% after applying the first derivative as a preprocessing step. This improvement confirmed the utility of spectral preprocessing for enhancing signal separation and reducing scattering effects. Severa! species were classified with perfect accuracy, demonstrating highly distinctive spectral signatures. Others showed improved discrimination after preprocessing, suggesting doser chemical similarities that were better resolved through data treatment. The strong overall performance confirms the robustness of this approach. The results affirm that NIR spectroscopy, when combined with appropriate chemometric tools, offers a reliable, rapid, and non-destructive method for tropical wood identification. This approach is particularly valuable in contexts requiring species-level discrimination, such as timber traceability, quality control, or combating illegal logging in biodiverse tropical regions.
Related
- Solution-processed Ti3C2Tx MXene coatings for three-dimensional microelectrodes in label-free single-cell impedance spectroscopy
- Application of partial least squares discriminant analysis to mid-infrared spectral data for discriminating green and roasted coffee processed by different post-harvest methods: MATLAB- and R-based computational tools and modeling pipelines
- Near-infrared spectroscopy for rapid and non-destructive assessment of PFAS effects in blue mussels (Mytilus spp.): a proof-of-concept study
- Monitoring a Polyurethane Synthesis by Fiber-Coupled Attenuated Total Reflection Fourier Transform Infrared Spectroscopy and Multivariate Curve Resolution-Alternating Least Squares
- Solvent-Dependent Self-Association of Chiral 2-Acetoxy-2-Phenylacetic Acid Revealed by Infrared and Vibrational Circular Dichroism Spectroscopy
BibTeX
@article{nguessan2026identification,
title = {{Identification of twelve frequently processed wood species from Ivorian forests by carpenters using near-infrared spectroscopy}},
author = {Jean Louis Lepetit N’guessan and Bobelé Forence Niamké and Hakim Abdel Aziz Ouattara and Appolinaire Bley-Atse Bley and N'Guessan Jean Claude Yao and Nadine Amusant and Amissa Augustin Adima},
journal = {Sciences des Structures et de la Matiere},
year = {2026},
month = oct,
doi = {10.70974/mat10126001},
url = {https://hal.science/hal-05820450v1},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Identification of twelve frequently processed wood species from Ivorian forests by carpenters using near-infrared spectroscopy
AU - Jean Louis Lepetit N’guessan
AU - Bobelé Forence Niamké
AU - Hakim Abdel Aziz Ouattara
AU - Appolinaire Bley-Atse Bley
AU - N'Guessan Jean Claude Yao
AU - Nadine Amusant
AU - Amissa Augustin Adima
PY - 2026
DA - 2026/10/26
JO - Sciences des Structures et de la Matiere
DO - 10.70974/mat10126001
UR - https://hal.science/hal-05820450v1
AB - This study demonstrates the efficacy of near-infrared (NIR) spectroscopy combined with chemometric techniques for identifying twelve tropical wood species commonly processed in Ivorian carpentry workshops. Principal Component Analysis (PCA) was employed to explore spectral variability, revealing distinct groupings corresponding to differences in chemical composition, wood density, and anatomical characteristics. Although some overlap was observed between certain species, three major trends could be distinguished, indicating meaningful structural or chemical similarities within each group. Subsequently, a Partial Least Squares Discriminant Analysis (PLS-DA) mode! was developed using 864 spectral samples. T he mode! achieved a classification accuracy of 97.8% with raw spectra, which increased to 98.96% after applying the first derivative as a preprocessing step. This improvement confirmed the utility of spectral preprocessing for enhancing signal separation and reducing scattering effects. Severa! species were classified with perfect accuracy, demonstrating highly distinctive spectral signatures. Others showed improved discrimination after preprocessing, suggesting doser chemical similarities that were better resolved through data treatment. The strong overall performance confirms the robustness of this approach. The results affirm that NIR spectroscopy, when combined with appropriate chemometric tools, offers a reliable, rapid, and non-destructive method for tropical wood identification. This approach is particularly valuable in contexts requiring species-level discrimination, such as timber traceability, quality control, or combating illegal logging in biodiverse tropical regions.
SN - 2630-1180
ER -