UIDAI’s Biometrics highly accurate, Scales to 10 lakh Aadhaar cards a day!
If implemented and executed properly, the landmark Unique Identification (UID) project could serve as a single biggest reform initiative in the country with a population of 1.2 billion residents of diverse demographics and identity.
Ever since the Aadhaar project’s conception, questions have often been raised about the reliability and efficacy of the UID data collected and the effectiveness of biometric technology in the enrolment scheme in Indian context.
In fact, often the index finger has been raised from within the government’s ministerial departments alleging the biometric census by the UID project as flawed and deviating from the security assurances required under the National Population Register. Of late, even the Home Secretary had stated that the UID data is not reliable as anyone can get themselves registered under any name with any address.
In order to dispel fears on the reliability of the UID data, the UIDAI has come out with a report titled “The Role of Biometric technology in Aadhaar Enrollment” to highlight the high degree of accuracy of biometrics used in the UID project in the Indian context.
The major goal of the UID project is to achieve de-duplication – which is to ensure that each person gets a unique Aadhaar number captured and matched against every other resident who has previously enrolled under the scheme, and build a clean base for the purposes of a National Identity system.
The report says that it is not possible to de-duplicate 1.2 billion residents by relying merely on demographic identities which can be surrogated through falsification; but it also requires differentiating through biometric system such as fingerprint and Iris scanning.
In fact, the accuracy levels achieved with a combination of fingerprint and Iris are said to be 10x better than an order of magnitude using either of the two individually.
As of December 31, the UIDAI project has enrolled 15 crore resident at a peak enrollment processing rate of 10 lakhs/day, with 10.25 crore Aadhaars already generated by the authority.
The report claims a negligible biometric failure-to-enroll rate of 0.14%, which means that 99.86% of the population can be uniquely identified by the biometric system.
As far as false positive identification rate is concerned, from a run rate of 10 lakh enrolments a day, only about 570 cases need to be manually reviewed daily to ensure that no resident is erroneously denied an Aadhaar number.
Besides, a fractional false negative identification rate of 0.035% has also been reported, wherein a wholesome 99.965% of all duplicates submitted to the biometric de-duplication system are correctly caught by the system as duplicates.
Certainly, with such precise analysis and past track record based on the proof-of-concept studies conducted in India, there is no need to rely on small size tests from other projects, to ascertain the reliability, accuracy and scalability of the UID data generated.
Over to you… Mr Nilekani…