Face value of facial recognition system

Police organisations have started to deploy facial recognition systems for investigation.

In Minority Report, a 2002 sci-fi film set in 2054, Tom Cruise plays a Washington DC police officer going by the name John Anderton. In a scene from the movie, billboards recognise and solicit people strolling in a mall by scanning the eyeballs. As John Anderton walks through the mall, interactive ads scream and shout at him trying to lure him into enjoying a Guinness.

Back in 2002, these digital ads seemed like a far cry, but we have witnessed this sci-fi prophecy become a reality in less than a decade. In 2010, Japanese company NEC developed electronic billboards embedded with facial recognition technology capable of displaying tailor-made ads on the billboard by identifying the age and gender of passers-by, by comparing it with its database.

IBM researchers went even further to develop billboards with sensors that interacted with RFID chips built into credit cards and mobile phones even when tucked away inside wallets and purses of people. Sensors embedded on the billboard would scoop up the RFID signals of the credit card holder as he passed by.

After capturing the information available on the credit card, such as name, age, gender shopping preferences, etc., a customised advertisement is displayed.
Shop owners today are resorting to facial recognition software to identify potential thieves by comparing faces of customers against a database of known shoplifters. FaceFirst software scans faces, as customers pass through a store entrance, and captures multiple pictures of each shopper to compare it to a database of shoplifters that the retailer has compiled; if there is a match, the software sends an alert to store employees within seconds of that person walking through the door.

A similar system deployed by Hilton hotels uses facial recognition to scan the faces of all guests, allowing employees to greet visitors by name, especially VIP Gold Card members.

A software company based in Austin, US has developed an app using face recognition software that gives a snapshot of the scene inside bar/pub that one can see on a downloadable free iPhone or Android app called SceneTap. The app provides information such as male-to-female ratio, average age, and crowd size, all in real time so that a bar-goer can determine, if there are enough women to hit on in a bar, before heading out to it. SceneTap is reported to be up and running in more than a dozen US cities.

Another app which runs on face recognition software called NameTag scans the faces, compares them with millions of publicly available online records to generate the person’s name and social media profiles, including those on Facebook, Twitter, and Instagram. Google has acquired a facial recognition software company PittPatt (Pittsburgh Pattern Recognition), that can match people across photos, videos, and more.

PittPatt can search the web and identify an image in less than 60 seconds. NEC designed NeoFace facial recognition solutions captures and extracts faces of individuals from video feeds, performs quality matches in real-time with unprecedented accuracy and speed – handling up to 3.02 million searches per second. Its high performance also allows stadiums and public venues to keep thousands of sports fans or crowds safe by accurately identifying the bad guys and alerting security personnel.

SnapChat’s animated lenses also uses facial recognition technology and so does Apple introduced Face ID on the flagship iPhone X. A face recognition software called 'FindFace' in Russia can identify faces with about 70% accuracy using the social media app called VK. This app was allegedly used to harass women allegedly involved in online pornography.

Facebook acquired for $100 million in 2012, since then it has been performing facial recognition on every picture uploaded on its system.
The acquisition has helped Facebook to improve its tag suggestions feature, which identifies all the images posted on Facebook using algorithms. Facebook has uploaded more than a quarter of a trillion photos on its site, making it the single largest repository of biometric data on this planet outstripping even Aadhaar. This represents a database that governments could abuse for face recognition purposes.

DeepFace is a deep learning facial recognition system created by a research group at Facebook. It identifies human faces in digital images. DeepFace is reported to be 97% accurate, when compared to 85% accuracy of the FBI's Next Generation Identification system. Facebook's DeepFace has been attracting several lawsuits under the Biometric Information Privacy Act.
Edward Snowden, in his revelations on NSA leaks, has alleged that his agency had directly infiltrated into the servers of the nine biggest internet companies including Facebook, and helped the intelligence agencies to gain full access to Facebook's biometric data.

Snowden has also divulged that NSA was having the capability of processing 55,000 images a day and was aspirating hundreds of photographs posted online.

The UK Police has been experimenting with facial recognition technology at public events since 2015, but they found the systems up to 98 percent inaccurate.

The FBI has launched a Next Generation Identification program which includes facial recognition besides biometrics such as iris scans and fingerprints with access to both criminal and civil database. The FBI uses the images as an investigative tool.

The U.S. Department of State operates one of the most extensive face recognition systems with a database of 117 million American adults with pictures drawn from the drivers' licence database.

As of 2016, the police of San Diego and Los Angeles were using facial recognition to identify people. At Maryland USA, there was controversy when police resorted to facial recognition to identify and arrest unruly protestors making up a mob by comparing mob images with a driver licence photo database. According to a report, China has installed facial recognition cameras in several cities and facial recognition checkpoints at several public areas such as gas stations, markets, etc.

Police organisations have started to deploy facial recognition systems for investigation. For instance, police by installing facial recognition systems at the Qingdao International Beer Festival in China could identify 25 wanted suspects. Police in China are boasting of 98.1 percent accuracy in identification while in the UK the use of this technology yielded up to 98 percent inaccuracy.

The Telangana Police have developed a facial recognition system that helps them identify offenders by comparing the suspect's face with digital photographs available in a central database called Crime and Criminal Tracking Networks and Systems, (CCTNS).

In Chennai, a face recognition software called FaceTagr developed by a Chennai based company was used in a few areas with mixed results. Similarly, Amritsar Police using a face recognition technology developed by a Gurugram AI company Staqu Technologies called Punjab Artificial Intelligence System (PAIS) could detect a murder case in 24 hours. The Staqu-developed PAIS claims that it can match images with an accuracy of 98% if the database has five images of the person. Elsewhere, the Surat Police is using the state-of-the-art NeoFace technology of NEC for solving crimes. In July 2018, Andhra Pradesh launched e-Pragati, a searchable database of millions of people containing e-KYC Aadhaar numbers. Uttar Pradesh Police in December 2018 launched ‘Trinetra’ an AI-based application which has face recognition capabilities and a database containing 5 lakh criminals. In April 2018, the Delhi Police could identify almost 3,000 missing children in just four days during a trial of a facial recognition system.

The Australian Border Force and New Zealand Customs Service have an automated border checking system called SmartGate that uses facial recognition. All Canadian airports in their Primary Inspection Kiosk program use facial recognition to compare the traveller with the photo available in their ePassport. GMR-operated Hyderabad International Airport Ltd. (GHIAL) has successfully tested facial recognition for staff entry at the passenger terminal building. The government of India appears to have plans to introduce paperless boarding in Indian airports. The IT firm SITA is planning to use a technology called Smart Path to provide a paperless boarding system for Indian airports by leveraging the Aadhaar biometric identity system. The Tocumen International Airport in Panama has an airport surveillance system which incorporates hundreds of live facial recognition cameras to identify wanted individuals passing through the airport.

Facial recognition technology is less accurate on people of colour. Further, the error rate of facial recognition technology is higher for men than women; for instance, the accuracy rate for men was 91.9 per cent while for women it was only 79.4per cent. CyberExtruder, a reputed company supplying facial recognition software to some law enforcement agencies, has accepted the fact that some skin colours give high error rates. Face recognition is also less effective when facial expressions vary. A big smile can render the system less effective. Further, profile pictures give a high error when compared to frontal images.

Facial recognition is a powerful technology; It should be used only for law enforcement and national security that too with adequate safeguards. Aadhaar at present has iris and biometric information; there appears to be a move to strengthen it with facial recognition. Once that is done, Aadhaar will have total surveillance infrastructure. Usage of facial recognition technology in the absence of any data protection or data privacy law could lead to misuse of the technology.

There is no legal provision to stop the misuse of facial recognition technology in India. The Information Technology Act, 2000 does not have provisions to deal with the misuse of technology. Besides, there is no framework in India even to regulate the storage of such data. Cyber criminals appear to be taking advantage of this situation by making such data available on the darknet.

There is a saying that face is the index of the mind, Our face in that sense is the outward embodiment of the mind. The fears of a surveillance state and loss of privacy that are spawned by facial recognition systems in our mind are scrambling our inner connections and making our faces and lives look miserable and paranoid. We may be able to overcome all this with Inner Spiritual Technology. When we all connect to the divine blueprint of our soul - There can only be utopia and no dystopia!

Next Story