Over the last decade we have investigated the possible benefits of graph representations of ridge pattern biometrics (such as fingerprints and ball prints) and vascular biometrics (such as retina and hand vein). For automatic recognition of imposter matches, and the use of dissimilarity vectors to avoid storing the actual biometric (its distances from a set of prototypes are stored instead), it has become clear that vascular biometrics benefit from graph representation far more than ridge pattern biometrics do. However, our most recent application has been in gamifying forensic fingerprint analysis. We have developed Delta Core which is a web application where players engage in the first stage (Analysis) of the ACE-V process used by law enforcement to establish the source of fingerprints left at the scene of a crime, disaster or act of terrorism. The game mechanics rely on noisy graph matching techniques to compare the performance of players to that of human experts and create an environment wherein players develop forensic skills.

About the speaker: Stephen is a mathematician who can be spotted teaching Calculus to hapless first-year students at RMIT University when he is not researching or designing Serious Games like Delta Core. His passion for applying mathematics first led him to work on models of infectious disease and he has first-author papers in Nature and Science contributing to the theory and practice of managing disease outbreaks. Following postdoctoral appointments abroad at the University of Antwerp, the University of Utrecht and Yale University, he returned to Melbourne in 2009 to join RMIT University. He’s currently fascinated by pattern matching problems in the forensic sciences and has collaborated with Victoria Police for the last 5 years.

How to participate in this seminar:

1. Book your nearest ACE facility;

2. Notify Vera Roshchina at RMIT (rmitopt@rmit.edu.au) to notify you will be participating.

No access to an ACE facility? Contact Maaike Wienk to arrange a temporary Visimeet licence for remote access (limited number of licences available – first come first serve)