Similar to the unique make-up of a human fingerprint, each artist has a technique inimitable to only them. Scientists have recently applied this idea to finding a new way of determining the authenticity of an art piece.
Using Van Gogh’s paintings and a digital analysis system, Penn State researchers James Z. Wang and Jia Li, are doing their part to help solve counterfeit painting schemes using computer science and engineering.
The system used by Wang and Li involves high resolution scans of Van Gogh paintings from the Van Gogh and Kröller-Müller Museums in the Netherlands. By breaking each image down into sections the professionals studied the artist’s distinctive brush strokes based on patterns and geometric contours.
With reference to “handwriting” as the unique style attributed to Van Gough and other distinguished painters, the two scientists studied the “writing” involved in 101 scans of Van Gogh images.
The in-depth results, published in the July issue of IEEE Signal Processing, revealed that art historians asserted 23 of the scans to be real Van Gogh. With the information provided Wang and Li were able to statistically compile the information in a computerized training database they could use to read Van Gogh’s handwriting and make solid predictions about the authenticity the art.
The other images scanned were labeled as the 78 whose legitimacy experts weren’t 100 percent sure of. These were compared with the remaining bona fide 23 and recorded in an online data system used to store and compare such information.
Wang says he is optimistic about the future of digital painting analysis through its importance in the study of arts and cultural heritages and the prevention of forgery. “Through tackling these tough problems, we can advance the core technologies at the same time," he said also voicing his prediction about the augmented collaboration of computer scientists, art historians and mathematicians.