MUMBAI, India, Aug. 31 -- Intellectual Property India has published a patent application (202621074590 A) filed by Indian Institute Of Technology Indore on June 16, 2026, for A Video Incident Detection System And Method For Real Time Highway Surveillance And Traffic Monitoring.
Inventors include Aruna Tiwari; Dr. Sanjay Singh; Dr. Sumeet Saurav; Dr. Rituraj; Aditya Chandle; and Anikeit Sethi.
The application for the patent was published on August 28, 2026, under issue no. 35/2026.
Abstract: ABSTRACT A VIDEO INCIDENT DETECTION SYSTEM AND METHOD FOR REAL TIME HIGHWAY SURVEILLANCE AND TRAFFIC MONITORING The present invention relates to an intelligent video incident detection system (VIDS) for real time traffic monitoring and highway surveillance. The system comprises a video acquisition module configured to capture live video streams from roadside surveillance cameras, an Automatic Traffic Counter and Classifier (ATCC), and an anomaly detection module operating in parallel. The ATCC detects, tracks, and classifies vehicles, estimates vehicle speed, performs axle based classification, and generates structured traffic metadata in real time. The anomaly detection module employs a constrained video anomaly detection generative adversarial network (CVAD GAN) to learn normal spatio temporal traffic patterns and to identify abnormal traffic events such as sudden stoppage, wrong way movement, and stalled vehicles. A key aspect of the invention is the integration of anomaly detection results with vehicle level metadata to improve detection accuracy and reduce false alarms. Upon detection of anomalies, real time alerts and visual analytics are generated through a centralized dashboard. The system enables scalable, real time, and automated traffic incident detection under practical highway conditions.
Disclaimer: Curated by HT Syndication.
