Assessment of multispectral and hyperspectral imaging systems for digitisation of a Russian icon

Abstract

In a study of multispectral and hyperspectral reflectance imaging, a Round Robin Test assessed the performance of different systems for the spectral digitisation of artworks. A Russian icon, mass-produced in Moscow in 1899, was digitised by ten institutions around Europe. The image quality was assessed by observers, and the reflectance spectra at selected points were reconstructed to characterise the icon's colourants and to obtain a quantitative estimate of accuracy. The differing spatial resolutions of the systems affected their ability to resolve fine details in the printed pattern. There was a surprisingly wide variation in the quality of imagery, caused by unwanted reflections from both glossy painted and metallic gold areas of the icon's surface. Specular reflection also degraded the accuracy of the reconstructed reflectance spectrum in some places, indicating the importance of control over the illumination geometry. Some devices that gave excellent results for matte colour charts proved to have poor performance for this demanding test object. There is a need for adoption of standards for digitising cultural heritage objects to achieve greater consistency of system performance and image quality.

Hyperspectral imaging of art: Acquisition and calibration workflows

“…Hyperspectral systems produce geometric distortions due to the scan motion, to the arrangement of dispersion optics, from the lens itself and also from alignment errors with respect to the work of art. The RRT results showed that not only were the spectral results highly variable, but the resulting image geometries were often equally so (MacDonald et al 2017).…”

Review of Methods for Documentation, Management, and Sustainability of Cultural Heritage. Case Study: Museum of King Jan III’s Palace at Wilanów

“…The distinction between multispectral and hyperspectral imaging is not clearly defined but it is agreed that hyperspectral imaging provides higher spectral resolution collecting several thousand (200 or more) contiguous, narrow bands, thus allowing for the reconstruction of almost continuous reflectance spectrum [116] (Figure 19). In contrast, an imaging sensor is considered as multispectral if it registers a small number (usually not more than 20) of discrete spectral bands of bandwidths ranging from 20 to 50 nm [117], which may not be equal in bandwidth and are not necessarily adjacent to each other [115]. Figure 19.…”

“…On the other hand, this knowledge allows the art historians for verification of the object's authenticity. Pigment identification commonly involves the use of VIS-NIR range of spectrum [117,120,124,126,147,148] as the signatures of most inorganic pigments vary significantly in this range, so the spectral information is sufficient to differentiate them. However, to tell apart organic materials like binders, varnishes, or organic pigments, the spectrum of longer wavelengths like SWIR and MWIR is usually needed [149].…”

Configuration and Registration of Multi-Camera Spectral Image Database of Icon Paintings

“…Although various general guidelines exist for performing these steps [32], the lack of a standard protocol is recognized by the community. Capturing the spectral images of icons with several cameras was studied by the authors of [33][34][35] in an imaging campaign with a round robin format among ten European institutes. MacDonald et al [33] compared multi-and hyperspectral measurements from manually-selected points.…”

“…Capturing the spectral images of icons with several cameras was studied by the authors of [33][34][35] in an imaging campaign with a round robin format among ten European institutes. MacDonald et al [33] compared multi-and hyperspectral measurements from manually-selected points. They reported high variations in the image quality and poor spectral estimation performance due to specular reflections from the glossy and metallic areas.…”