Art curators can say goodbye to an age-old problem, thanks to Duke University’s Mathematics Department. The department has collaborated with the North Carolina Museum of Art to develop a software solution for art curators analyzing old paintings using X-ray images.
The problem associated with cradling – the wooden framework on the back of many old paintings – is not new. Curators have been frustrated by the distracting lattice-like pattern it causes on X-ray images for years. That is, until Duke research assistant Rujie Yin decided to solve the problem with mathematical algorithms while researching a 14th-century altarpiece by Francescuccio Ghissi for an upcoming exhibition at the NCMA.
Rujie, along with Duke professors Ingrid Daubechies and David Dunson, developed the Platypus program, which uses an algorithm to remove the cradle pattern from the X-ray image. Compared to manually altering images like this in Photoshop, Platypus saves both time and effort. The three-stage procedure starts with locating the cradled regions. The second step consists of separating the X-ray image into a textural and image component. In the last step, the algorithm learns to distinguish between the texture caused by the wooden cradle and the texture belonging to the original painted wooden panel.
“We have been fortunate to work with world leaders in the fields of time-frequency analysis, informatics and image processing to come up with a practical solution to a difficult problem,” says William Brown, chief conservator at the NCMA.
The Platypus program was introduced in August 2015 and is now available as a free Photoshop plug-in online.
The project is a product of the Information Initiative at Duke (iiD), and is also part of an ongoing collaboration with the NCMA’s Art + Science Initiative, which partners with local universities to research topics that have to do with both art and science.
