Facial Recognition with Biometric Encryption in Match-on-Card Architecture for Gaming and Other Computer Applications (Feasibility Study)
Last year, as part of improvements in the Responsible Gambling program, the Ontario Lottery and Gaming Corporation (OLG) completed the implementation of facial recognition technology utilizing Biometric Encryption techniques in most of its casino facilities [1, 2]. Biometric Encryption or BE (a.k.a. biometric template protection, biometric crypto system, fuzzy extractor, etc.) is a process that binds a digital key to, or generates a key from, a biometric so that no biometric image or template is stored. This work was a result of a partnership with the University of Toronto, Ontario’s Information and Privacy Commissioner (IPC) and OLG. The process employed Privacy by Design (PbD) techniques to ensure the privacy of the public and self-excluded players, while at the same time improving the overall detection rate of self-excluded patrons.
While the facial recognition with BE has delivered the best practice solutions for authentication, responsible gambling, player protection and privacy, there is an opportunity to determine if it is feasible to apply biometric authentication for other contexts. Some examples include authentication for online gaming, strengthening the lottery retail sales security (i.e. making sure that the person logged into the retail terminal is actually being verified for conducting each sales transaction), etc. Outside of the gaming industry, there are a countless number of applications where strong remote or local authentication is required, such as accessing electronic health records or other government-held personal records, online banking, etc.
Ann Cavoukian, Ph.D.
Michelle Chibba, M.A.
Alex Stoianov, Ph.D.
Office of the Information and Privacy Commissioner of Ontario, Canada
Tom Marinelli, P.Eng.
Klaus Peltsch, M.Sc., M.B.A.
Ontario Lottery and Gaming Corporation
Hervé Chabanne, Ph.D.
Olivier Beiler, M.Sc.
Julien Bringer, Ph.D.
Vincent Despiegel, Ph.D.