Ph.D. student takes second in National Security Innovation Competition

Author: Nina Welding

(Left to right) Estefan Ortiz; Maj. Gen. (USAF, Ret.) Mark Volcheff, National Security Innovation Competition executive director; Jay Brasseur, chief technology officer of Neumann Systems Group

Estefan Ortiz, a doctoral student in electrical engineering at the University of Notre Dame, took second place in the 2011 National Security Innovation Competition (NSIC).

Sponsored by the National Homeland Defense Foundation, the competition is designed to link college students conducting cutting-edge research with government and industries who can help commercialize their innovations.

The competition helps advance new concepts and technologies in national security which have commercial value while identifying future innovators.

Student participants are judged on their unique and appropriate solutions to national security issues, how well they described a national security problem and if the problem was a current or future need.

In addition, the participants are required to demonstrate an understanding of market need, including end users and competing products, as well as their comprehension of trends in industry and the forces that affect those trends.

Their presentation also must outline a reasonable and attainable action plan through which to pursue their innovation.

Ortiz was awarded second place and $5,000 for his efforts in “Dilation Aware Multi-image Enrollment for Iris Biometrics.”

Iris recognition technologies

In the proposal, Ortiz proposed a solution to the weaknesses that many iris recognition technologies are affected by. Most systems encounter issues as a result of a subject’s pupil changes, usually as a result of dilation, between the original time the iris image was introduced to the recognition program and the time a recognition test is administered.

For example, the current “enrollment” method stores a single best image at the time of enrollment, which does not allow for environmental or physical changes in the subject.

Ortiz’ method suggests a revision in the enrollment process by dividing the initial information introduced into a system into equal intervals called “quantiles.” Those quantiles can then be compared individually against a number of enrollment images per person.

Testing data showed that Ortiz’ method consistently outperformed the current method used in most commercial iris recognition systems.

Kevin W. Bowyer, Schubmehl-Prein Chair of the Department of Computer Science and Engineering and Patrick J. Flynn, professor of computer science and engineering, served as Ortiz’ advisers.

Contact: Nina Welding 574-631-7169,