Honors College Thesis
 

Toward an Optimal Working Distance for Hyperspectral Imaging of Wood

Público Deposited

Conteúdo disponível para baixar

Baixar PDF
https://ir.library.oregonstate.edu/concern/honors_college_theses/v405sj01c

Descriptions

Attribute NameValues
Creator
Abstract
  • The forest products industry relies on the production of high-quality wood to satisfy consumer requirements. Improving wood quality requires an effective and inexpensive technique for measuring critical properties such as density and microfibril angle. An emerging technology known as near-infrared hyperspectral imaging (NIR HSI) essentially captures surface images of samples that can be used to generate NIR spectra, allowing for the development of models which predict various wood properties for the given sample. The objectives of this research were to determine an optimal working distance (from the imaging camera to the sample surface) for wood analysis applications of NIR HSI as well as investigate the benefits of an increased number of sample scans in terms of calibration model performance. Experimental methods primarily consisted of scanning samples with a SPECIM HSI system and using partial least squares (PLS) regression to develop calibration models. The results of this research showed that higher-performing models were obtained at a shorter working distance and that multiple scans added minimal value to the modeling process. While certain parameters may still require further optimization, NIR HSI technology shows considerable potential as an alternative wood property analysis technique.
  • Key Words: density, near-infrared, hyperspectral imaging, wood property analysis
License
Resource Type
Date Issued
Degree Level
Degree Name
Degree Field
Degree Grantor
Commencement Year
Advisor
Committee Member
Non-Academic Affiliation
Declaração de direitos
Publisher
Peer Reviewed
Language

Relações

Parents:

This work has no parents.

Em Collection:

Itens

Miniatura Título Data de carga Acesso Ações
file details JayawickramaMichelleT2020.pdf 2020-12-02 Público Baixar