MUMBAI, India, Aug. 24 -- Intellectual Property India has published a patent application (202641097939 A) filed by Sri Eshwar College Of Engineering on August 12, 2026, for Facial Ai - Physiological Facial Recognition System.
Inventors include Ms. Kalaivani T; Devaranjan J; K S Abishek; and Vainavi N.
The application for the patent was published on August 21, 2026, under issue no. 34/2026.
Abstract: This invention introduces a multi-modal biometric monitoring system and methods for the non-invasive collection of emotional and physiological data using a specialised surveillance setup. The system combines high-resolution optical imaging sensors with long-wave infrared (LWIR) thermographic sensors in a single CCTV unit. It captures synchronised data streams showing both external facial muscle movements and internal thermal radiation patterns. The main feature of this invention is a sensor fusion engine that employs a deep learning framework for facial expression recognition (FER). It also analyses certain thermal areas in real time, such as the periorbital, nasal, and forehead regions. Unlike standard visual-only systems, this dual-layer approach connects micro-expressions to involuntary bodily responses, like changes in blood flow or fever. This offers a dependable way to assess a person's emotional state and physical condition. Additionally, the system includes a processing module that normalises thermal data based on environmental conditions and distance, ensuring accuracy in various lighting and weather situations. By mapping thermal patterns onto specific facial features, the system can detect signs of stress, deception, or medical issues that might be concealed by intentional facial control. This innovation provides a strong solution for automated health checks, high-security behaviour analysis, and cognitive load monitoring in public and private areas. The design aims to send automatic alerts when the fusion algorithm detects a significant difference between a person's displayed emotion and their core body temperature. This effectively addresses the limitations of traditional single-mode surveillance. Consequently, the invention offers a scalable, non-contact diagnostic tool that enhances standard surveillance systems by linking computer vision with biological monitoring. This results in a detailed behavioural profile for use in healthcare, law enforcement, and emergency response.
Disclaimer: Curated by HT Syndication.
