Automated Facial Recognition

The National Crime Records Bureau (NCRB) has released a Request for Proposal for an Automated Facial Recognition System (AFRS) to be used by police officers across the country.

How Automated Facial Recognition System Works?

• The AFRS will maintain a large database with photos and videos of peoples faces.
• Whenever there is a need to identify an unidentified person, an image or a CCTV footage is compared to the existing database to find a match and identify the person using artificial intelligence.
• The artificial intelligence technology used for pattern-finding and matching is called “neural networks”.

A similar facial recognition technology has been adopted by the Civil Aviation Ministries DigiYatra initiative for airport entry is under the trail at the Hyderabad airport.

NCRB Request of Proposal

• NCRB wants to use automated facial recognition to identify criminals, missing people, and unidentified dead bodies, as well as for “crime prevention”.
• NCRB also aims the AFRS to be compatible with other biometrics such as iris and fingerprints
• It proposes gathering CCTV footage, as well as photos from newspapers, raids, and sketches.
• The proposal involves a mobile and web application hosted in NCRB’s Data Centre in Delhi to be used by all police stations in the country.
• Further, the NCRB calls for integrating the facial recognition with NCRB-managed Crime and Criminal Tracking Network & Systems (CCTNS).
• NCRB also wants to integrate facial recognition system with Integrated Criminal Justice System (ICJS), as well as state-specific systems, the Immigration, Visa and Foreigners Registration & Tracking (IVFRT), and the Koya Paya portal on missing children.

Concerns against Facial Recognition

• The AFRS may result in indiscriminate searching by tapping into publicly and privately installed CCTVs pan-India. Even though facial recognition technology (FRT) has proliferated globally, only a few countries have systems that use footage from CCTVs installed in public areas (as proposed by NCRB). This is the most excessive use of FRT.
• The CCTV cameras are already been haunted with cybersecurity issues. The integration with the AFRS will expand the “attack surface” for exploiting vulnerabilities in the AFRS.
• The AFRS will also allow for real-time querying, enabling “continuous” mass surveillance. A glaring example of misuse of continuous surveillance has been witnessed in China, with the Uighurs being persecuted as an ethnic minority.
• The AFRS  is a non-consensual and covert surveillance technology.  It is designed to operate at a distance, without any knowledge of the targeted individual(s).
• AFRS would also result in potential infringements on the right to privacy, a fundamental right as it allows for continuous and ongoing identification.
• The AFRS will also violate the legal test of proportionality with constant surveillance being used as a strategy for crime detection was articulated in the landmark judgment of Justice Puttaswamy vs Union of India.
• Further variations in pose, illumination, and expression, among other factors could adversely impact the accuracy of automated facial analysis.
• Researches show that default camera settings better expose light skin than dark. As a result, there could be differential results across racial groups.

Moreover, the absence of a Data Protection Law in India may give investigative authorities unhindered access and power. This could turn India into a police state. Hence there have been demands for a complete prohibition on FRT due to the history of political surveillance.