Computers expose fake Bruegels
An American mathematician has developed a computer programme that can spot a fake Bruegel from a real one by analysing the artist’s use of the brush. Daniel Rockmore and his team at Dartmouth College developed the technique of sparse coding, which breaks a work down into the most basic elements used by an artist to represent reality – in the case of Bruegel, the brushstroke.
The team, which included experts on visual perception, scanned eight works by the Flemish artist Pieter Bruegel the Elder (1525- 1569), as well as five known imitations, and mapped the brushstrokes according to direction, orientation, length and thickness, thus building up a sort of “vocabulary” of brushstrokes particular to Bruegel. The technique is similar to handwriting analysis except that brushstrokes are infinitely more variable than the 26 letters of the alphabet.
Armed with Bruegel’s brushstroke-fingerprint, it should now be possible to spot a fake simply by showing that it doesn’t carry Bruegel’s stamp.
The technique has limitations, Rockmore admitted. It requires sufficient source material to draw up a profile of the artist’s brushhand, which of course need to be of unimpeachable provenance. However, over and above their application to the tracing of fakes, the techniques of sparse coding are likely to be of interest to art historians to measure, for example, if parts of a painting are by different hands – by members of an artist’s studio or by later restorers.
"Our model is purely quantitative and data driven and does not take into account the vast experience of connoisseurs," said Daniel Graham, a post-doctoral fellow at Dartmouth. "Our work is complementary to the judgment of connoisseurs and to physical methods of authentication, like chemical signatures and analysis of canvas weaving. It is also complementary to what is considered the gold standard of attribution, chain-of-custody records. We see our method as another form of evidence, which can be weighed against traditional forms, and it has the distinct advantage of being rigorously objective."
