Researcher at UMass Amherst Studying How to Conduct Computer Searches on Handwritten Manuscript Collections

AMHERST, Mass. - University of Massachusetts computer scientist Raghavan Manmatha has received a $450,000 grant from the National Science Foundation (NSF) and the National Endowment for the Humanities (NEH) aimed at finding ways to conduct computer searches on historical, handwritten document collections.

The project will develop techniques for indexing digitized copies of handwritten historical manuscripts composed by a single author. Using the Library of Congress''s digitized collection of George Washington''s papers, comprising 6,400 scanned manuscript pages, Manmatha and his research team plan to develop a fast-indexing algorithm that will locate similarities in the images of handwritten words. The result will be a prototype, showing how to search for individual words in digitized handwritten manuscript collections.

"Many libraries and other organizations are putting scanned versions of historical manuscript collections on the Web or on CD-ROMS," Manmatha said. "Today these documents require expensive and time-consuming manual annotations if they are to be searched. The technology proposed here will enable us to look up words in scanned manuscripts in the same way we can look up indexed terms in a printed book."

The grant is one of five awarded by the NEH aimed at providing high-tech solutions to research problems in the humanities. The five projects are being developed under the Digital Libraries Initiative, a federal program conducted by the NSF in partnership with NEH and other federal agencies. The initiative provides national leadership and support for the next generation of digital libraries. Other recipients are Indiana University; University of California, Los Angeles; University of Hawaii; and Stanford University.

Manmatha obtained his Ph.D. in computer science at UMass in 1997. He has been a research assistant professor since 1998. Besides document imaging, Manmatha''s research interests are in the areas of image and video retrieval, and in the application of statistical techniques to these areas.

NOTE: Raghavan Manmatha can be reached at 413/545-3623 or