Leong, Pou Yee (2023) Development and evaluation of a wood classification system usinga low-cost miniature nir sensor. Final Year Project, UTAR.
Abstract
Wood classification is a pivotal stage in wood processing, as a significant portion of commercially utilized wood species, known as "lesser-known", lackadequate visual characteristics based on anatomical features. NIRS has emerged as an effective and non-destructive technique for classifying woodspecies, surpassing the limitations of conventional methods. NIRS measures the absorption or emission of spectroscopic signals within the electromagnetic wave range of 800 nm to 2500 nm. In this study, the feasibility of employingNIRS for discriminating diverse wood species was assessed. A meticulouslycalibrated sensory system incorporating a sophisticated NIR sensor modeledAS7263 was constructed to capture and analyze spectral data, facilitating the identification of three wood species: Hevea, Jelutong, and Chengal. The data were acquired using the mode of diffuse reflectance, laying the foundation for subsequent comprehensive analysis and classification. The findings have proven that NIRS is a highly effective method for accurately categorizingvarious wood species based on the analysis of their distinct spectral data, irrespective of ambient conditions. Nonetheless, it is imperative toacknowledge that potential alterations in the wood surface may impact the reliability of the results.
Actions (login required)
![[img]](http://eprints.utar.edu.my/style/images/fileicons/application_pdf.png)
