Raphael was one of the most famous Italian Renaissance masters and lived from 1483 to 1520.
Hassan Ugail, director of the Center for Visual Computing and Intelligent Systems at the University of Bradford, revealed that he had developed an algorithm to recognize real Raphael paintings with 98% accuracy.
It analyzed 4,000 parameters, such as brushstrokes, color gamut and hues, to determine whether a painting was indeed by Raphael, CNN reported.
The Madonna of the Rose painting, which is in Madrid's Prado Museum, has long been the subject of debate, said study co-author Howell Edwards, professor of molecular spectroscopy at the University of Bradford.
"Artificial intelligence analysis of our work has shown conclusively that while the three figures are unequivocally painted by Raphael, that of Saint Joseph is painted by someone else," Edwards said.
The algorithm is based on previous research that found the painting, known as the Tondo de Bresi, is likely the work of the real Raphael, contradicting some art historians who believe the painting is a copy made in the 19th century.
The findings have drawn criticism from some art historians, who Ugail says are "a bit against" the study.
"I was a little surprised," he said, explaining that the algorithm analyzes details that are beyond the capabilities of the human eye.
"It's going to be very, very difficult, no matter how good someone is, to go into that level of detail and create something like this," Ugail said.
The de Bresi Tondo debate fits into broader discussions about the role of artificial intelligence in the authentication of works of art, which Ugail says complements other forms of analysis, such as tracing a work's provenance.
"It's just another tool," said the director of the Center for Visual Computing, who noted that the algorithm could be used to determine whether a mysterious piece of art merits further investigation.
"It's a very quick way to see if you need to look deeper into a painting," he said.
Next, Ugail plans to develop an algorithm capable of recognizing other artists' works, which he says will bring science to art authentication.
"The potential of this type of instrument is great," he emphasized./BGNES
