5th Annual Graduate Student Research Symposium - March 21, 2013
Congratulations to Darryl D'Souza, 2nd Place Poster Presentation winner and Roger Ouch, 3rd Place Poster Presentation winner! The Annual Graduate Student Research Symposium is a 3-minute oral presentation where graduate students discuss their research project and why it is important. More information and complete list of winners.
The poster introduces Avatar CAPTCHA, an image based approach to distinguish human users from computer programs (bots). The proposed CAPTCHA asks users to identify avatar faces from a set of 12 grayscale images comprised of a mix of human and avatar faces. Experimental results indicate that it can be solved 91% of the time by human users with an average success time of 23.86 seconds and a positive user rating of 90%. It is designed to provide security against computer programs (bots).
Governments, organizations and companies lose billions of dollars due to cyber-attacks every year. Biometrics systems have been used to enhance security via user recognition and authentication. However, the use of a single biometric is limited by low accuracies, noisy data, vulnerability to attacks and lack of scalability. So in my research I have developed a robust multimodal biometric system using face and voice recognition. The fusion of multiple biometrics increases the overall recognition rate and helps building more secure systems.