Eckerd College receives an award for further development of the DARWIN (Digital Analysis and Recognition of Whale Images on a Network) project. Marine mammologists studying the behavior and ecology of wild dolphins often employ photo identification as a means of associating observational data with individual dolphins. After the researchers return from the field, the best photograph of each individual is compared to a catalog of previously observed and identified photographs of dolphin dorsal fins. DARWIN (http://darwin.eckerd.edu) is a computer program that simplifies photo-identification by applying computer vision and signal processing techniques to automate much of the manual process. The DARWIN system allows a researcher to query a database of previously identified dolphin dorsal fin images with an unidentified image of a dolphin's fin. The system responds with a rank ordered list of database fn images most closely resembling the query image. In this award the goals are 1) to investigate and incorporate methods for more intuitive comparisons of dolphin dorsal fns, 2) to more closely emulate the manual photo-identification process and provide additional tools to augment standard methods currently in use, 3) to minimize the level of effort required to create a database of dorsal fn images and query that database to identify unknown dolphins, 4) to provide additional import and export formats which better match the needs of the user community, and 5) to evaluate and further enhance the user interface.

The broader impact of this proposal is to provide members of the marine mammal research community with a powerful tool for the identification of individual dolphins. The time savings potentially realized with such a program would enable researchers to spend more time on the thoughtful analysis of their data rather than on its acquisition, thus helping to advance understanding of the subject. The software will be developed as well documented as open source, and to made available at SourceForge to facilitate its broad dissemination. This project will include significant research opportunities for undergraduate students and an educational component that focuses on middle and high school students from under-represented groups through our participation with the National Youth Sports Program.

Agency
National Science Foundation (NSF)
Institute
Division of Biological Infrastructure (DBI)
Type
Standard Grant (Standard)
Application #
0445126
Program Officer
Reed Beaman
Project Start
Project End
Budget Start
2005-04-01
Budget End
2010-09-30
Support Year
Fiscal Year
2004
Total Cost
$237,286
Indirect Cost
Name
Eckerd College
Department
Type
DUNS #
City
Saint Petersburg
State
FL
Country
United States
Zip Code
33711