MUMBAI, India, Jan. 23 -- Intellectual Property India has published a patent application (202641001773 A) filed by Karpagam Academy Of Higher Education; Karpagam Institute Of Technology; C. Sasti Kumar; and Mukesh M, Coimbatore, Tamil Nadu, on Jan. 7, for 'real time secure clickbait and biometric atm user authentication and multiple bank transaction system.'
Inventor(s) include C. Sasti Kumar; and Mukesh M.
The application for the patent was published on Jan. 23, under issue no. 04/2026.
According to the abstract released by the Intellectual Property India: "Automated Teller Machines (ATMs) continue to play a vital role in modern banking systems by providing convenient access to financial services. However, the increasing prevalence of identity theft, card skimming, and unauthorized access highlights the limitations of traditional authentication mechanisms that rely solely on debit cards and Personal Identification Numbers (PINs). To address these vulnerabilities, this invention proposes an advanced ATM security model integrating Deep Convolutional Neural Network (DCNN)-based facial recognition and intelligent environmental monitoring to provide secure, contactless, and highly reliable user authentication. The system employs a high-resolution camera embedded within the ATM kiosk to capture real-time facial images during user interaction. These images undergo rigorous preprocessing operations including face detection, alignment, illumination correction, and noise filtering to ensure the generation of high-quality inputs for the DCNN model. The trained deep learning architecture extracts robust and discriminative facial feature embeddings, enabling rapid and accurate identity matching against an encrypted biometric user database. To mitigate presentation attacks, the system incorporates a dedicated spoof detection module capable of identifying printed photos, replayed videos, and three-dimensional mask attempts using texture analysis, depth estimation, and motion-based cues. In addition to primary facial authentication, the system integrates an environmental surveillance layer that continuously analyzes the ATM surroundings to detect suspicious activities such as multiple-person presence, camera obstruction, or abnormal movement patterns. Successful authentication grants the user access to ATM functionalities, whereas failed or malicious attempts activate secondary security procedures, including session lock, alert notifications, or referral to alternative verification mechanisms. The system communicates securely with the bank server using encrypted protocols to prevent data interception or tampering. By combining biometric authentication, deep learning inference, and intelligent surveillance, the proposed invention significantly enhances ATM security, reduces fraud risks, and improves overall user experience in financial transaction environments."
Disclaimer: Curated by HT Syndication.
