a subject in transition; opportunities and problems
CHArt Seventeenth Annual Conference
|
|
DIGITAL
ART HISTORY a subject in transition; opportunities and problems |
John Eakins, A. Jean E. Brown, Margaret Graham, Richard Mulholland, Jonathan Riley, Jonathan Edwards, University of Northumbria at Newcastle, UK
A Shape Retrieval System for Watermark Images: The creation of a web-based archive of information on watermarks, paper history and paper conservation using Content-Based Image Retrieval
This paper will discuss the results of an eighteen-month research project representing a collaboration between The Conservation Unit, School of Humanities, and the Institute for Image Data Research (IIDR) at the University of Northumbria, Newcastle.
The project concerns the design and implementation of shape retrieval system for digital images of historic watermarks from works of art on paper. The intended database, which will become The Northumbria Watermarks Archive, will be published on the web and is intended to provide a freely accessible highly valuable research tool for art and paper historians, forensic scientists and paper conservators. It is intended to draw together and cross-reference existing records in addition to being the starting point for a progressive and interactive collation of new and related information. In addition to the provision of identification information for watermarks and paper, the archive will include extensive data on paper technology, paper history, working methods of artists, artists' materials and preservation/conservation issues.
Current watermark research is hindered by the fact that, in order to match the watermark to its reference image, researchers are required to manually and sequentially search through thousands of similar images. These references are often unreliable, inaccurate, and in many instances have been combined with an arbitrary approach to description and classification.
The Northumbria Watermark Archive will utilize content-based image retrieval (CBIR) software written at the university in order to recognize and search for similar watermark images using their shape characteristics. The software will draw heavily from our existing CBIR system for retrieval of trademark images (ARTISAN), unique in its exploitation of principles from Gestalt psychology.
We believe that this system will improve the accessibility of watermark images and related information and in doing so provide a fresh, interactive approach to art historical and technical research.