findinEsegui ricerca

Art gets its digital passport

Preventing art theft has become a high-tech activity that involves diving into the sub-micrometric scale to record a fingerprint based on each art piece’s unique roughness and colour

Art gets its digital passport

Imagine that a hundred years ago it was possible to take the ‘digital fingerprint’ of Leonardo Da Vinci’s Mona Lisa and store it in an international database. If that was the case, the painting would probably have never been stolen as it would become virtually impossible to resell.

Each painting has characteristics of roughness and colour, or chromatism, that observed at the sub-micron scale make it unique and easily identifiable, just like someone’s fingerprint. Based on that concept, an international scientific team focuses on creating a ‘digital signature’ for art masterpieces, under the EU funded FINGaRtPRINT project.

To establish a painting’s fingerpring, the team of researchers use a type of sophisticated microscope, called white light confocal profilometer, mounted on a robotic arm. This equipment, once positioned at distance of 10mm from a painting, measures its roughness at the sub-micrometric scale in a target area of one square centimetre.

Then, the scientists create what they call a spectral fingerprint by pointing at the same area a special camera to record how the light pointed at the painting bounces off. This gives valuable information about the pigments and dyes used by the artist. Finally, the scientists also take pictures of the art piece with a webcam to document the exact position of the fingerprints for future reference.

"[We] carried out experiments to see if micro-scale 3D data could be combined with accurate colour measurements of the same area in order to make a fingerprint, which was unique to that object.” Kirk Martinez, one of the project partner and a Senior Lecturer in Electronics and Computer Science at the University of Southampton, UK, says, adding: “these two technologies had not been tried together previously, so you should be aware that it was highly experimental work."

Applications of this technology will be either to prevent forgery or theft. It could also be used to authenticate a piece after a long period of loan or restoration. This can be done by comparing the digital image of the signature before and after the loan.

Although it seems to be a brilliant idea some expressed doubts. Michele Nappi, Associate Professor and Director of the Biometrics and Image Processing Laboratory at the University of Salerno, Italy says: “Features like roughness and chromatism, both [used] by scientists to make a fingerprint of an object, are not permanent, but may vary due to factors such as environmental degradation or because the object is not original but... a fake.” He adds: “If an item [is returned] to its owner with some features changed, does it mean that he's looking at a fake or a simply degraded object? It would make it very difficult to establish the truth."

The nature of the material used matters too. "We should consider that paintings are made of linseed oil, acrylic colours, organic and inorganic pigments", explains Maria Laura Santarelli, researcher at the Department of Chemical Engineering, Materials and Environment of the University of Rome La Sapienza, Italy. "All of these materials are normally affected by solar or artificial lights, dust and pollutants, which are able to change the colour of the painting—for example, linseed oil turns very dark—as well as to form micro-fissures that will necessarily change the roughness of the painted surface.” Thus, fingerprinting could be a good identification method to work on, but does not address all the issues. She adds: “I think more and more experimentation has still to be done on it".

(15 October 2012)

Marco Merola

The information on the INNOVATIONSEEDS platform can be used provided that the source is acknowledged.
This project has received funding from the European Union’s Horizon 2020 Programme for research, technological development and demonstration under grant agreement No 6419747th FWP