Characterizing the Pigments in Wall Paintings: Comparing Portable and Stationary Multiband Remote Sensing Imaging Systems

Wall paintings are art works with valuable information on ancient technologies, materials and cultural expressions of past civilizations, and assessing wall paintings’ state of preservation allows to differentiate between the original materials and degradation products. Imaging techniques such as portable spectral cameras allow studying the materials on a large-scale, and characterizing the pigments and assess their preservation state. A relatively cost effective and easy-to-use method is based on modified digital camera equipped with external filters in the UV-VIS-NIR. The multiband images need to be aligned and normalized to estimate the reflected light’s intensity, which makes characterizing pigments difficult, due to noise from uneven illumination, surface degradation, and the physical handling of the external filters. Here, we present a comparison between multiband images from a stationary camera and portable modified digital camera of painted wall fragments, showing the influence of environmental conditions on the quality of the images. The stacked photos of the different filters, representing the reflected intensity at different band widths, were analyzed as multispectral data to study the pigments in the pictorial layer. We combine the data with chemical information from X ray fluorescence (XRF), and high resolution structural information from fiber optic reflectance spectroscopy (FORS), showing portable cameras with external filters could be used as spectral documenting tool for characterizing iron-oxide pigments such as yellow and red ochres on-site.