From UPSC perspective, the following things are important :
Prelims level: ASTR, AI
Mains level: Not Much
Central Idea: The Department of Telecommunications (DoT) has developed an artificial-intelligence-based facial recognition tool called Artificial Intelligence and Facial Recognition powered Solution for Telecom SIM Subscriber Verification (ASTR).
What is ASTR?
- ASTR is designed to check subscriber databases of telecom operators to identify multiple connections associated with the same person.
- The goal of ASTR is to detect and block fraudulent mobile connections, thereby reducing cyber frauds.
Development of ASTR
- In 2012, DoT issued an order requiring telecom operators to share their subscriber database, including users’ pictures, with the department.
- These images serve as the core database for facial recognition using ASTR.
- The ASTR project was conceptualized and designed by the DoT’s unit in Haryana between April 2021 and July 2021.
- A pilot project was conducted in Haryana’s Mewat region to test the feasibility of ASTR, where a significant number of fraudulent SIMs were detected.
How ASTR works?
- ASTR uses convolutional neural network (CNN) models to encode human faces in subscribers’ images, accounting for various factors like face tilt, angle, image opaqueness, and dark color.
- A face comparison is performed for each face against all faces in the database, grouping similar faces under one directory.
- ASTR considers two faces to be identical if they match to a minimum extent of 97.5%.
- It can detect all SIMs associated with a suspected face within 10 seconds from a database of 1 crore (10 million) images.
- After matching faces, ASTR’s algorithm utilizes “fuzzy logic” to find approximate matches for subscriber names, considering variations, typographical errors, and related results.
Impact and Results
- In the first phase, ASTR analyzed over 87 crore (870 million) mobile connections and detected more than 40 lakh (4 million) cases of people using a single photograph to obtain multiple connections.
- After verification, over 36 lakh (3.6 million) connections were discontinued by telecom operators.
- The list of fraudulent connections is also shared with banks, payment wallets, and social media platforms to disengage these numbers from their respective platforms.
- WhatsApp collaborated with the government to disable accounts created using such numbers, and similar efforts are being made with other social media platforms.
Facts for PrelimsConvolutional Neural Network (CNN): A type of deep learning algorithm commonly used for image recognition tasks, where it extracts features and patterns from images by applying convolution operations. Fuzzy Logic: A form of logic that deals with approximate or qualitative reasoning rather than strict binary true/false values. In the context of ASTR, it is used to find similarity or approximate matches for subscriber names, accounting for variations and typographical errors. |
Get an IAS/IPS ranker as your 1: 1 personal mentor for UPSC 2024
