Historical story

Fake Van Goghs recognized

Counterfeiters should try even harder to imitate paintings. PhD student Laurens van der Maaten of Tilburg University has developed a computer program that analyzes paintings, pots and coins at the smallest level. If a painting has not been copied exactly in the style of the painter, Van der Maaten's technique will immediately pick out the forgery. Laurens van der Maaten will receive his PhD on 23 June.

The Cultural Heritage Agency of the Netherlands has a new tool:a new analysis technique that sorts archaeological objects according to baking method (pottery), color and size (seeds) and painting style. The computer program of PhD student Laurens van der Maaten can sort out all the measurements that archaeologists can come up with. This creates a clear 'map' of a series of objects and it quickly becomes clear whether a particular object belongs in the series or not.

For example, if archaeologists want to know the function of a series of pots, they look at the profile of the pottery. A pot with a wide neck profile was probably used as a cooking pot, while a narrow neck was more convenient for storing fruits. Van der Maaten's program puts all individual measurements next to each other in a diagram. All wide-necked pots are in one place in the diagram, thin-necked pots are in another place. If an archaeologist finds one new pot, he only has to compare the profile with the diagram of the pots found earlier. This way, the researcher immediately knows whether it is a cooking pot or a storage pot.

mercilessly critical

Van der Maaten's technique has already proven itself among art connoisseurs. Indeed, it was possible to distinguish false paintings by Van Gogh from real ones. “We use high-resolution images of Van Gogh paintings for this,” says Van der Maaten. “The strokes of the brush create a specific texture. You then use the same techniques as those for the fracture surfaces of the pottery. With this technique, for example, I saw the difference between the Wacker forgeries and real Van Goghs. In 1930, everyone thought these forgeries were real.”

According to Van der Maaten, art connoisseurs need not be afraid of their job. The technology just gives them additional evidence for their claims. The program itself does not determine whether a painting is real or fake. In any case, the message is clear to forgers:learn to paint exactly as Van Gogh did, because the computer is mercilessly critical.

Read more at Kennislink:

  • Fake betrays itself
  • Mathematics distinguishes art from kitsch
  • A real Pollock:fractal or not?
  • Hidden under the paint
  • Painting shot reveals pigments