DOI: 10.5593/SGEM2014/B23/S10.025


E. Matouskova, K. Pavelka, M. Faltynova
Wednesday 1 October 2014 by Libadmin2014

References: 14th International Multidisciplinary Scientific GeoConference SGEM 2014, www.sgem.org, SGEM2014 Conference Proceedings, ISBN 978-619-7105-12-4 / ISSN 1314-2704, June 19-25, 2014, Book 2, Vol. 3, 199-206 pp

Hyperspectral imaging (also called imaging spectroscopy) is a relatively new method that has been used multidisciplinarily in last years. This process has been speeded up by technological developments and opens new topics of analyses and research. Spectroscopy has been used as a powerful quantitative tool in laboratories for many years now, and a lot of investigative methods have been developed. These methods can now be transferred into the cultural heritage field and may lead to new findings in this branch of study. This scientific discipline studies light as a function of wavelength, where light can be emitted, deflected and scattered by a specific material. Unlike multispectral imaging, which is commonly known from satellite imagery when investigating the Earth’s surface and provides a small amount of wide, non-continuous bands, hyperspectral imaging shifts the imagery a level up. Tens or hundreds continuous narrow bands are detected which allow one to see all the spectral information in the explored spectral range. This enables spectral reflectance extraction on a pixel level and when compared to laboratory measurements, material type and characteristics can be determined. A hyperspectral sensor (Hyperspec VNIR A-series by Headwall Photonics) is available at the Czech Technical University in Prague, Faculty of Civil Engineering, Department of Geomatics will be used for this analysis. Illumination plays a key role in spectral characteristics extraction because reflectance (the amount of reflected light) is measured. It is required that the quantity of reflected light should be just enough to reach (but not exceed) the saturation level of the hyperspectral instrument. If the amount of light detected by the instrument is too low, significant noise will appear in the data. This noise cannot be mathematically corrected because it affects dark parts of the investigated object more than the light ones. One can dispose of the noise via careful and fine illumination. Several sets of illuminations will be tested. Another important issue is the white reference material. White and black reference imagery is used for calibration in order to extract the actual response of analysed object. This paper deals with white reference materials used for the calibration of the hyperspectral sensor. It is an image of white material with a maximum reflectance. In this study, several white reference materials (classic office paper, technical white reference material and SpectralonTM) will be analysed and the noise level will be shown. A cultural heritage object (painting) will be used for this test. The illumination influence will be closely observed due to the possibility of harming the painting. Old paintings are very susceptible to temperature, thus higher temperatures (>35°C) can make the bee wax (present in the painting) melt. This paper will provide useful information regarding the hyperspectral set-up for the future investigation of cultural heritage objects.

Keywords: hyperspectral imaging, cultural heritage, illumination, white reference material, woodcut graphics