MUMBAI, India, July 7 -- Intellectual Property India has published a patent application (202641080413 A) filed by Cmr Institute Of Technology on June 30, 2026, for An Intelligent Real-Time Examination Monitoring And Proctoring System Using Artificial Intelligence, Facial Recognition, And Behavioural Analytics.
Inventors include Kumbala Pradeep Reddy; Andavolu Prakash; A. Srinish Reddy; N. Srivani; Gopireddy Purna Chandu Reddy; and Gundareddy Bala Yogeswar Reddy.
The application for the patent was published on July 03, 2026, under issue no. 27/2026.
Abstract: The present invention discloses an intelligent real-time examination monitoring and proctoring system employing artificial intelligence, facial recognition, eye gaze analysis, and behavioral analytics to ensure the integrity of online examinations conducted in remote environments. The system comprises a secure React.js-based examination interface that restricts candidate browser activity to the examination context, an AI proctoring module utilizing OpenCV, TensorFlow, and MediaPipe libraries for continuous analysis of webcam video feeds at real-time frame rates, an audio monitoring subsystem for speech detection and ambient sound classification, and a screen activity tracker that intercepts unauthorized navigation attempts. The proctoring module detects and logs violations including face absence, identity mismatch, unauthorized person presence, sustained gaze deviation, lip movement, verbal communication, and browser tab switching, each timestamped and preserved with evidentiary captures. A live streaming dashboard provides invigilators with real-time visibility into all active examination sessions, while an automated alert system triggers priority notifications upon detection of high-severity violations. The automated evaluation engine assesses objective and code-based examination submissions and generates candidate scorecards. Upon examination completion, the system produces a comprehensive violation report and an integrity score computed as a weighted composite of detected anomalies, enabling data-driven administrative decisions. Built on a cloud-native Node.js, Express.js, and MongoDB Atlas architecture, the system is scalable, interoperable with Learning Management Systems, and designed for broad institutional adoption across academic, professional certification, and government examination contexts.
Disclaimer: Curated by HT Syndication.
