U.S. Visit ID system may be missing the mark
- By Wilson P. Dizard III
- May 17, 2005
The U.S. Visitor and Immigrant Status Indicator Technology program's method for matching travelers' fingerprints with images in massive federal databases has fallen into question as a result of research published by the National Academy of Sciences.
Two Stanford University specialists in biometrics and game theory published an article, "Using Fingerprint Image Quality to Improve the Identification Performance of the U.S. Visitor and Immigrant Status Indicator Technology Program," in the Proceedings of the National Academy of Science
's current issue. The publication accepts articles after they have been vetted, or "refereed," by other scientists in the field.
The Stanford scientists found that the U.S. Visit program's image matching system would reach a detection probability of only about 53 percent for poor-quality fingerprints. The U.S. Visit system uses a two-fingerprint matching method.
"About five percent of the general public and 10 percent of those on the watch list have bad quality fingerprints due to either genetics or hard labor," according to a statement by Lawrence Wein, lead scientist on the project.
"We assume that terrorist organizations will eventually defeat the U.S. Visit program by employing a majority of people whose fingerprint quality is either naturally bad or deliberately made so," he said.
The scientists proposed software modifications that would increase the system's success in matching poor-quality fingerprint images from 53 percent to 73 percent.
Homeland Security Department officials were not immediately available to respond to the scientists' statements.