Push-broom NIR-HSI scanning of painting reconstruction, inspired by Sandro Botticelli's "Venus"

The dataset contains the push-broom scanning NIR-HSI of Sandro Botticelli’s Venus painting detail, produced in the Microchemistry and Microscopy Art Diagnostic Laboratory (M2ADL) of the University of Bologna (2018). The painting was executed with egg tempera on a wood panel prepared with a layer of gypsum and glue. Ancient and modern pigments were selected and employed according to their vibrational signals, detectable by NIR spectroscopy. In a more detail, earth-based pigments were used for the hair and for the flesh tone, while zinc white (ZnO) was applied in the white areas of the eyes and in the hair ribbon. Finally, dammar varnish was applied to half of the painted surface, according to ancient practices which involved the use of a terpenic varnish to improve color rendering and protect the underlying painted layer (Figure 8A). The painting was scanned with a SWIR3 hyperspectral push-broom camera working in the 1000–2500 nm spectral range, at 5.6 nm spectral resolution (Specim Ltd, Finland). The scanning system is also characterized by three halogen lamps (35 W, 430 lm, 2900 K, each) used as illumination sources and a horizontal moving stage (40 × 20 cm) on which the painting was scanned. The scanning parameters were set as follows: scan rate equal to 0.7 mm/s, push-broom camera frame rate equal to 50.00 Hz and exposure time equal to 9 ms. The hyperspectral system was controlled with the Lumo Scanner software (Specim Ltd, Finland). Before the scan, dark (with the camera shutter closed) and white (using a Spectralon reference tile) reference images were acquired, to compute reflectance values.

ENVI hyperspectral files (.raw and .hdr) related to the scanned sample, white reference, and dark reference are included. Please, note that ALL the files must be downloaded and included in the same folder.

Please, cite as:

R. Rocha de Oliveira, C. Malegori, G. Sciutto, P. Oliveri
PoliBrush – A user-friendly software to aid multivariate image analysis dissemination
Chemometrics and Intelligent Laboratory Systems, 240 (2023) 104918
https://doi.org/10.1016/j.chemolab.2023.104918

You may also be interested in:

   XRF-HSI Vermeer dataset
   10.5281/zenodo.8143464

   PoliBrush
   A freely distributed software for exploratory multivariate analysis in RGB and spectral imaging
   10.5281/zenodo.8143341