Presenter(s)
2014 ISIT Plenary Lecture
Signal Analysis Helping Art Historians and Conservators
Ingrid Daubechies
Duke University
Topic
Abstract
Signal analysis and information theory can help art historians and art conservators in studying and help understand art works, their manufacture process and their state of conservation. The presentation will review several instances of such collaborations in the last decade or so, and then focus on one particular example: virtual cradle removal. Between the 12th to the 17th century, European artists typically painted on wooden boards. To remediate or prevent structural or insect damage, conservators in the 19th and first half of the 20th century first thinned the panels to a few millimeters, and then strengthened the much thinner wood structures by (permanently) attaching to their backs, hardwood lattices called cradles. These cradles are highly visible in X-ray images of the paintings. X-rays of paintings are a useful tool for art conservators and art historians to study the condition of a painting, as well as the techniques used by the artist and subsequent restorers. The cradling artifacts obstruct a clear "reading" of the X-rays by these experts. We introduce approaches to remove these artifacts, using a variety of mathematical tools, including Bayesian algorithms.