IOT-enabled universal authentication platform

Number of Students : 2
Guides : T V Prabhakar, Rajesh Sundaresan
Sponsor : UIDAI

Aadhaar is the largest biometric identity system deployed in practice anywhere in the world. The idea here is to develop an advanced universal authentication system for Aadhaar. Universal authentication is challenging because of uncontrolled conditions under which data is captured. For instance, there are a variety of camera sensors, processing pipeline capabilities, capture poses, and lighting conditions. All these make face recognition a hard task. Similarly, Iris, voice, and fingerprints have their own challenges which we must overcome. Biometric authentication solutions will be designed by leveraging advances in artificial intelligence and machine learning.

In this project, we will build a hardware platform that interfaces sensors such as RGB cameras, Iris (IR), voice (Mic), and fingerprint sensors to support universal authentication. The platform should have sufficient data processing capability. For example, heart rate and its variability from a series of images might have to be required to affirm to the liveness of the subject.
Since the device will be used for “remote authentication”, it has to be secure and resilient to spoofing attacks.

Skills – Hardware platform building, Sensor interfacing, software drivers, applications such as student attendance.