Hyperspectral imaging (HSI) in the near-infrared (NIR) range is a non-destructive technique that can be used for material characterization and several other applications. Different wood species exhibit characteristic spectral signatures in this range, enabling their classification with chemometric methods [1].
Wood species identification is crucial for enforcing regulations aimed at preventing illegal trade of protected species [2] and in cultural heritage conservation, where this information can guide restoration and support authenticity assessments [3]. Prior research has focused mainly on tropical species, with little attention paid to European ones.
This study addresses these gaps by applying NIR HSI to classify selected Slovenian wood species. Alongside optimising classification accuracy using various preprocessing and chemometric approaches, the work contributes to building a reference dataset that, with further research, could support the non-destructive identification of aged or archaeological wood in cultural heritage contexts.
Keywords: HSI, NIR, wood species, classification, chemometrics
Authors
Luka Kopač
University of Ljubljana, Biotechnical faculty, Department of wood science and technology, Slovenia
Žiga Špiclin
University of Ljubljana, Faculty of electrical engineering, Laboratory of imaging technologies, Slovenia
Miha Humar
University of Ljubljana, Biotechnical faculty, Department of wood science and technology, Slovenia
Login to download the PDF