Interview with Anil Jain at Michigan State University on his latest biometric research
12 September 2017
Isabelle Moeller, Chief Executive of the Biometrics Institute in conversation with Anil Jain of Michigan State University:
I was lucky to speak to Anil recently about our Academic Research and Innovation Group and asked him what his areas of research currently are as we are aiming to build better links between our user, supplier and academic community. Here is what he told me. Anil will be in Australia in February 2018 and is planning to speak at one of our Member Meetings.
1. What is your current area of research?
Three main research topics that I have worked over the past 25 years are: (i) fingerprint recognition, (ii) face recognition, and (iii) biometric fusion. The specific research problems that we work on these three topics have changed over time from law enforcement to social media and now to mobile phone applications. I have had a sequence of Ph.D. students and postdocs who have made significant contributions in these areas. Their scientific papers are widely cited and the resulting technology has been licensed by major biometric companies. For a list of our publications, see our Google scholar page; https://scholar.google.com/citations?user=g-_ZXGsAAAAJ
2. What research are you currently doing on biometrics and children?
We conducted the first systematic and rigorous longitudinal study that addresses the following questions: (i) Do fingerprints of young children possess the salient features required to uniquely recognize a child? (ii) If so, at what age can a child’s fingerprints be captured with sufficient fidelity for recognition? (iii) Can a child’s fingerprints be used to reliably recognize the child as he ages? For our study, we collected fingerprints of 309 children (0-5 years old) four different times over a 12-month period. We showed, for the first time, that fingerprints acquired from a child as young as 6 hours old exhibit distinguishing features necessary for recognition, and that state-of-the-art fingerprint technology can deliver high recognition accuracy (98.9% true accept rate @ 0.1% false accept rate) for children older than 6 months.
3. What are the main challenges you face with your research?
One of the main challenges we face is access to biometric data. I understand the data privacy issues which make it very difficult for any organization to share data in their archives. There are also strict guidelines in the human subject data collection protocol on sharing data. Despite this limitation, we have been fortunate to have access to a variety of face and fingerprint data, under a very restrictive NDA. This resulted in technology transfer for tattoo image matching, altered fingerprint detection, composite to mugshot matching, and large-scale unconstrained face search. Our fundamental research on fingerprint and face longitudinal studies would not have been possible without access to large scale operational data.
4. Are there gaps in research that you would like to see covered?
I am afraid that a bulk of academic research is becoming irrelevant to biometric deployments in consumer products (e.g. mobile phones), government deployments (e.g., international border crossing), and law enforcement (e.g., latent fingerprint recognition). This is evident form the lack of growth in the number of participants at major conferences devoted to biometrics research. The research community needs to be in tune with the application requirements and find their niche in making the impact.