Image processing could help to identify artists by their brushstrokes

對於非油畫品呢

Art forgeries

The computer says no

Aug 5th 2008
From Economist.com

Image processing could help to identify artists by their brushstrokes

Who framed thy fearful asymmetry?

THE ability of computers to analyse complex digital images is growing rapidly. Robots are being fitted with powerful vision systems that allow them to recognise and hold things. Satellite images of the Earth can be scanned for tiny features, or pictures from deep space searched for strange objects. Medical images can be analysed to find out what might be going on inside a human body. Now digital imaging is learning how to spot art forgeries too.

Scientific methods have long been used to help authenticate works of art: for instance, paint can be dated from its chemical composition or canvasses X-rayed to reveal what lies below the surface. In recent years, however, the art itself has come under more scientific scrutiny, especially through the analysis of brushstrokes. The idea is to establish the equivalent of an artist’s “handwriting” to help experts attribute paintings.

One of the most comprehensive studies using such methods was published recently in Signal Processing Magazine from the Institute of Electrical and Electronics Engineers. It involved analysing the paintings of Vincent van Gogh and was carried out by James Wang of Penn State University and his colleagues from other universities. It was done with the help of the Van Gogh and Kröller-Müller museums in the Netherlands.

The researchers used high-resolution scans, made in shades of grey, of 101 paintings which were either by van Gogh or in his style. Of these, 82 of the pictures have consistently been attributed to the Dutchman, while six have long been known to have been painted by others. Experts cannot agree on who painted the remaining 13.

From the larger group of paintings, 23 were selected because they were known to be unquestionably by van Gogh and because they represented different periods of his life, during which his style changed. These paintings were used to “train” computers running image-analysis systems in the way the artist painted.

Scans of small areas of the paintings were taken for individual analysis, so that recurrent use of certain brushstroke patterns, texture and other features could be identified. The data were used to build a mathematical model of van Gogh’s style against which the other 78 paintings could be tested.

The computers had some success. For instance, when two known van Goghs, “The Plough and the Harrow” and “Wheatfield with Crows”, were used for training, the system indicated that among the paintings that resembled them according to a standard measure of brushstrokes was one of the sea at Saintes-Maries, a fake commissioned or sold by Otto Wacker, a German art dealer. But when the image analysis was intensified and carried out at a greater level of detail, the Wacker forgery was shown to be different.

The researchers describe their results as “encouraging, but not perfect”. Computerised image-processing systems should get better at detecting forgeries the more they can learn about artists’ techniques and styles, says Dr Wang. His colleagues are now using other wavelengths, including ultraviolet, to analyse more of van Gogh’s brushwork.

Interest has been shown in using image analysis to help find forgeries of other artists. The bigger the body of an artist’s known work, the more accurate the system would be, because it would have more to learn from. Yet however good computers get at classifying paintings, they are likely to remain only one of the tools used to detect forgeries, albeit an increasingly powerful one. Although the researchers are certain that technology can provide some answers to riddles about whose hand was on the canvas, they also concede that art experts will have the final say.