Johnson details computational art history techniques

Johnson, the Geoffrey S.M. Hedrick Senior Professor of Electrical and Computer Engineering at the College of Engineering and the Jacobs Fellow in Computational Arts and Humanities at the Jacobs Technion-Cornell Institute at Cornell Tech, developed a process using digital imaging to address questions in art history scholarship. He is credited with creating the field of computational art history.

“You become a detective,” he said. “The art historical questions are often linked to artist’s intent, and materials can assist in providing information that can corroborate a hypothesis about an artists’ intent.”

In his presentation, Johnson detailed how canvases can be analyzed through thread counting and weave mapping, aided by use of technology. Computers create detailed color scans of the art based on thread counting, which are later used for analysis. These scans are used to authentic and date the paintings, by showing which paintings came from the same rolls of canvas.

Johnson teaches undergraduates to use the applied mathematics techniques he has developed, providing a rare opportunity for them to participate in projects that are revolutionizing a field of study.

His projects include the analysis of Rembrandt’s drawings on paper by creating digital scans that reveal the watermark on the paper under the drawings.

In the WIRE (Watermark Identification in Rembrandt’s Etchings) project at Cornell in collaboration with the Herbert F. Johnson Museum of Art, Johnson worked with a team of undergraduate student researchers to create a decision tree of data that lists variants of possible watermarks previously encountered in Rembrandt’s work. Students majoring in industrial and labor relations and electrical and computer engineering are part of the team.

This article also apeared in the Cornell Chronicle.