For Prelims: Artificial Intelligence, Right to Information, Facial Recognition Technology

For Mains: Use of facial recognition technology, Benefits of facial recognition technology, risks of facial recognition. 

Why in the news?

The Delhi Police treats matches of more than 80% similarity produced by its facial recognition technology (FRT) system as positive results, according to responses to right to information (RTI) requests received by the Internet Freedom Foundation, a New Delhi-based digital rights organization.

What do we need to know about 2022 RTI responses?

  • The threshold for facial recognition:
    • According to the Delhi Police, comparisons with more than 80% similarity are considered positive results, while comparisons with less than 80% similarity are considered false positives and call for additional "corroborative evidence."
  • Collection of Data:

In accordance with Sections 3 and 4 of the Criminal Procedure (Identification) Act of 2022, which has since taken the place of the earlier Identification of Prisoners Act, the Delhi Police is comparing the photos/videos to photos gathered under those provisions.

Issues with an 80% Threshold:

  • The selection of 80% as the cutoff between true and false positives is not entirely clear.
  • The Delhi Police may still look into results that are below 80% because they are classified as false positives rather than negatives.
  • Individuals with similar facial features, such as those in extended families or communities, may become targets.
  • Communities that have historically experienced excessive policing and discrimination at the hands of law enforcement officials might be singled out as a result.

Criminal Procedure (Identification) Act, 2022:

An act authorizing the measurement of prisoners and other people for identification and investigation purposes in criminal matters, record preservation, and for matters related to and incidental to such investigations.

  • The Criminal Procedure (Identification) Act, 2022, is feared to result in the excessive collection of personal data, going against generally accepted best practices for data collection and processing.
  • It offers legal authorization for taking accurate body measurements of those who are required to provide them.
  • This includes, among other things, physical and biological samples, their analysis, photographs, iris and retina scans, finger, palm prints, as well as other prints.

What is Facial Recognition Technology?

In order to authenticate users through ID verification services, facial recognition systems are typically used. These systems work by identifying and measuring facial features from an image.

  • Algorithm-based facial recognition software creates a digital map of a person's face by identifying and cataloging their facial features, which it then compares to a database it has access to.
  • The Automated Facial Recognition System (AFRS) uses information from a sizable database that contains pictures and videos of people's faces to match and identify the person.
  • In order to find patterns and matches, artificial intelligence software compares a CCTV image of an unidentified person to the database.

Technology used for facial recognition.

  • Biometrics are used by a facial recognition system to map facial features from a picture or video.
  • It searches a database of recognized faces for a match by comparing the information. 
  • Although it can be used to confirm someone's identity, facial recognition also raises privacy concerns.

Benefits of facial recognition

The rise in public security:

One major benefit of facial recognition is the fair management of crimes, burglaries, and trespassers. 

  • It is easier to find burglars, thieves, and trespassers when face recognition is used. Technology exists to analyze CCTV camera networks, both open and closed.
  • The technology can be used for more than just finding criminals. For instance, it might also make it simpler to locate elderly people and missing children.
  • Airport security checkpoints might be less invasive for travelers if face recognition is used.

For the police, the outcry from the public over illegal stops and searches is a constant source of trouble and controversy—yet another situation where face recognition technology may be useful. Face recognition technology might assist in reducing the number of stops and searches of law-abiding citizens by identifying suspects among crowds.

Additionally, the mere knowledge of a face recognition system's presence deters crime, particularly minor offenses.

Fast and Non-Invasive Identity Verification

The quick processing speed and lack of user interaction offered by face recognition technology are additional advantages.  The speed and lack of direct user interaction of face recognition technology are additional benefits. With current identity verification methods, users must remember passwords, present I.D. cards, and other inconveniences.

Facial Recognition in Banking

The use of one-time passwords by banks to access accounts or approve transfers has advanced, but there is still room for improvement. Biometric identification through facial recognition technology is one potential remedy.

Better Worker Attendance Systems

Time theft, which continues to be one of the most prevalent breaches of work ethics, is a problem on industrial floors all over the world. Nevertheless, a small percentage of people use buddy favors from employees or security guards to skip work. 

  • To check-in for work, everyone must pass face-scanning devices. From now until checkout (again using facial recognition), there are paid hours.
  • Employees don't need to present identification or use plastic cards to clock in or out, making the process quick.

Why is the use of facial recognition technology harmful?

  • Misuse and Inaccuracy:
    • Issues related to “Misidentification” due to inaccuracy of the technology.
    • Issues related to “Mass Surveillance” due to misuse of the technology.
  • Gender & Race:
    • This may lead to a false positive, where a person is mistakenly identified as someone else, or a false negative, where a person's identity is not confirmed.
    • False positive outcomes can occasionally lead to prejudice against the person who was incorrectly identified.
    • Furthermore, based on race and gender, it has been reported that its accuracy rates sharply decline.