Abstract
Providing content-based access to archives of historical watermarks involves a number of problems. This paper describes the development and evaluation of SHREW, a shape-matching system for watermark images based on techniques developed for the ARTISAN trademark retrieval system. Encouraging retrieval results have been obtained using tracings of watermark images, whether whole-image features or component-based measures are used for matching. Further development of the system is continuing.
Cite
CITATION STYLE
Riley, K. J., & Eakins, J. P. (2002). Content-based retrieval of historical watermark images: I-tracings. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 2383, pp. 253–261). Springer Verlag. https://doi.org/10.1007/3-540-45479-9_27
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.
