As part of the Smart Cities Mission and efforts to amp up surveillance in Bihar, the Bihar government is looking to deploy facial recognition system (FRS) connected with the Crime and Criminal Tracking Network & Systems (CCTNS) and other databases available with the police in Bhagalpur and Muzaffarpur.
On April 13, the Telecommunications Consultants of India Limited (TCIL) issued a tender in this regard inviting bids from a single master system integrator company or a consortium of companies to set up an Integrated Command and Control Centre (ICCC), which apart from deploying FRS, also includes controlling public WiFi, smart street lights, installing CCTVs and so on. According to the tender, reviewed by MediaNama, around 1,500 and 400 CCTVs will be installed in Bhagalpur and Muzaffarpur, respectively.
The tender notes that the facial recognition system should be able to detect attributes in real time from a CCTV camera or from a recorded video, photographs, or a sketch wherein skin colour, hair styles, texture of clothes can be altered. It is to be installed in public areas such as railway stations, bus stands, religious places, high security area and so on.
An important part of this facial recognition drive of the Bihar government is to increase the police’s surveillance capabilities. “The system shall be able to broadly match a suspect/criminal photograph with database created using photograph images available with passport, CCTNS and Prisons, State or National Automated Fingerprint Identification System or any other image database available with police/other identity,” the tender read.
Other than the conventional databases mentioned above, the images will also be collated from a “private or public organisation” video feeds, photographs from newspapers, raids and so on for helping the police in criminal investigations. “The system should have ability to handle initial real-time watch list of 1,000,000 Faces (should be scalable to at least 100 Million faces) and 50 Camera Feeds simultaneously per server and generate face matching alerts (sic),” it added (emphasis added).
Bihar amping up surveillance measures
Not just Bhagalpur and Muzaffarpur, the Bihar administration recently installed over 16k CCTV cameras in Patna. Around 15,122 cameras were installed in residential apartments, 1,345 at malls and 341 at police stations. The Hindustan Times quoted Patna divisional commissioner Sanjay Kumar Agarwal as saying, “We often receive complaints from apartment residents and societies regarding poor security in their buildings. To check on unscrupulous activities, it is mandatory to install CCTV cameras at the main entry, stairs, lift, parking, every floor and at an outer location from the gate. For future constructions, it is mandatory to indicate CCTV camera points while passing the construction map for apartments and commercial complexes.”
Apart from this in 2019, the Bihar Police issued a tender for an application called Prahaar which would have an image search module that can assist user to search criminals through “advanced facial search module”. The tender said, “The image can be captured either through mobile camera or image gallery of the phone.”
Privacy concerns regarding usage of facial recogntion
These developments come at a time when there has been considerable outrage regarding the introduction of facial recognition by the Central government and various State governments. Internet researchers and privacy activists have time and again pointed out that the usage of databases such as CCTNS in the absence of a personal data protection law, is problematic.
Speaking to MediaNama, Shweta Mohandas, a policy officer with the Bengaluru-based Centre for Internet and Society, said, “The introduction of these systems before the Personal Data Protection Bill becomes an Act is a cause for worry and could cause confusion once the Act is out. The introduction of this system without letting people know about how this data will be used, and without the provision of consent will lead to a power asymmetry between the people whose data is being processed and the authorities who are processing this data.”
Mohandas also pointed out that there have been multiple instances where facial recognition systems from around the world have failed. “To have such a system been put up without a means of recourse could severely harm people who might be falsely identified, sometimes leaving them with no means to questions or rectify the error,” Mohandas added.
Similar concerns were raised in 2019 by Apar Gupta, executive director of the Internet Freedom Foundation in a letter to the National Crime Records Bureau over a tender issued by NCRB to automate facial recognition system. Quite akin to the contents of the tenders for Bhagalpur and Muzaffarpur, ‘Automated Facial Recognition System’ tender too talked about matching “a suspect/criminal photograph with database created using photograph images available with passport, CCTNS, ICJS.. or any other image database available with police/other identity,” the tender read.
Gupta pointed out that linkage of several databases to identify individuals violated the principles of consent. “The specific features of the RFP aims to identify individuals across various databases… Additionally it is not restricted to databases, it also requires the system to be compatible with biometric solutions such as iris and AFIS.” He opined that the inclusion of biometric could mean a reliance on the Aadhar database.