Share Email Print
cover

Proceedings Paper

Analytical robustness of quantitative NIR chemical imaging for Islamic paper characterization
Author(s): Hend Mahgoub; John R. Gilchrist; Thomas Fearn; Matija Strlič
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

Recently, spectral imaging techniques such as Multispectral (MSI) and Hyperspectral Imaging (HSI) have gained importance in the field of heritage conservation. This paper explores the analytical robustness of quantitative chemical imaging for Islamic paper characterization by focusing on the effect of different measurement and processing parameters, i.e. acquisition conditions and calibration on the accuracy of the collected spectral data. This will provide a better understanding of the technique that can provide a measure of change in collections through imaging. For the quantitative model, special calibration target was devised using 105 samples from a well-characterized reference Islamic paper collection. Two material properties were of interest: starch sizing and cellulose degree of polymerization (DP). Multivariate data analysis methods were used to develop discrimination and regression models which were used as an evaluation methodology for the metrology of quantitative NIR chemical imaging. Spectral data were collected using a pushbroom HSI scanner (Gilden Photonics Ltd) in the 1000-2500 nm range with a spectral resolution of 6.3 nm using a mirror scanning setup and halogen illumination. Data were acquired at different measurement conditions and acquisition parameters. Preliminary results showed the potential of the evaluation methodology to show that measurement parameters such as the use of different lenses and different scanning backgrounds may not have a great influence on the quantitative results. Moreover, the evaluation methodology allowed for the selection of the best pre-treatment method to be applied to the data.

Paper Details

Date Published: 11 July 2017
PDF: 8 pages
Proc. SPIE 10331, Optics for Arts, Architecture, and Archaeology VI, 103310P (11 July 2017); doi: 10.1117/12.2271971
Show Author Affiliations
Hend Mahgoub, Univ. College London (United Kingdom)
John R. Gilchrist, Gilden Photonics Ltd. (United Kingdom)
Thomas Fearn, Univ. College London (United Kingdom)
Matija Strlič, Univ. College London (United Kingdom)


Published in SPIE Proceedings Vol. 10331:
Optics for Arts, Architecture, and Archaeology VI
Luca Pezzati; Piotr Targowski, Editor(s)

© SPIE. Terms of Use
Back to Top