With museums increasingly digitizing their art collections, it becomes pretty
easy to forge paintings. Now, two researchers are working on a digital system to
help detect original works from counterfeit ones. The findings of the study,
which was led by James Z. Wang, associate professor of information sciences and
technology, Jia Li, associate professor of statistics, were based on 101
high-resolution grayscale scans of van Gogh paintings provided by the Van Gogh
and Kroller-Muller Museums in the Netherlands.
and Li broke each scan down into sections measuring 512 by 512 pixels, or about
2.5 by 2.5 inches in canvas size, and analyzed them based on patterns and
geometric characteristics of the brush strokes.
the 101 scans received from the museums, art historians identified 23 as
unquestionably authentic van Gogh works.
were used by the computer system as a training database for van Gogh's
models were created to capture the unique style, or "handwriting," that became
the artist's signature in 23 of the scans.
other 78 - either works of van Gogh, works of van Gogh's peers or paintings that
had at one time been attributed to him but later found to be unauthentic - were
compared against the generated models to test the algorithms.
and Li, along with computer science and engineering doctoral student Weina Ge,
compiled those findings into an online system that allows any painting to be
compared against existing data to help determine its authenticity.
painting analysis project results were first presented at a workshop at the Van
Gogh Museum in May 2007.
study "Image Processing for Artist Identification: Computerized Analysis for
Vincent van Gogh's Painting Brushstrokes" has been published in the July issue
of IEEE Signal Processing.