A team of Indian researchers has created a first-of-its-kind human authentication system for mobile and other handheld devices using teeth.
The app, according to the team, captures biometric samples using a mobile smartphone camera. Similar to existing apps that record the full face, the app features specialized markers that detect human teeth for authentication.
The research titled “Deepteeth: A Teeth-Photo Based Human Authentication System for Mobile and Hand-Held Devices”, authored by Ms. Geetika Arora, Mr. Rohit K Bharadwaj, and Mr. Kamlesh Tiwari from the Birla Institute of Technology and Science (BITS), Pilani explains the functioning of the app, the workforce and that the region of interest (RoI) is extracted from markers and the obtained sample is then enhanced using contrast limited adaptive histogram equalization (CLAHE) for higher visibility.
According to the team, this is the first work on teeth-photo-based verification for any mobile device, and the findings have shown “perfect accuracy.” The research paper also includes a diagram that explains how teeth-photo authentication works.
The app first takes an impression of your teeth using the front camera on your smartphone and this is followed by ROI extraction and enhancement. The app’s next feature is “deep feature extraction,” which is followed by “enroll or verify and identify.”
The authentication process begins in the next step after comparing the teeth impression with the database, the software makes a “decision” on whether or not it matches with the right person.
Researchers also stated that, while it takes a bit longer to train at first, it is highly efficient for identification and verification once deployed. The proposed model, according to the study, works properly with a small sample size and is thus power-efficient and ideal for mobile devices.