Researchers from the School of Computer Sciences at Mahatma Gandhi University have been granted an Indian Patent (No. 568699) for their innovative smart surveillance system that uses artificial intelligence to detect rare and suspicious activities in real-time CCTV footage.
The patented invention titled “Method and System for Rare Pattern Detection in Surveillance Video, and Super-Resolved Real-Time Alerts” significantly enhances the functionality of traditional surveillance systems by automating the detection, classification, and reporting of unusual visual patterns from both live and archived video feeds.
The research team includes PhD scholars Mintu Movi and Nisha Shamsudin, guided by Dr. Abdul Jabbar P and Prof. (Dr.) Bindu V R, who also heads the department. Mintu is pursuing research in rare pattern mining, while Nisha focuses on video super-resolution technologies.
How It Works
The system leverages AI to analyze human-centric visual features such as posture, movement, proximity between individuals, facial details, and environmental context. By applying advanced clustering algorithms and data reduction techniques, the system classifies behaviors as common or rare with high accuracy.
According to the inventors, the technology reduces memory usage by 75%, hard disk access by 90%, and processing time by 83%, making it a highly efficient alternative to traditional CCTV systems.
The system also features a super-resolution module that enhances low-light and long-distance footage, allowing for clearer visuals. Real-time alerts with enhanced images are sent via WhatsApp, enabling faster response during emergencies and improving post-event investigation.
One of the standout features is its hardware independence—the technology can be integrated with any existing CCTV infrastructure, making it widely scalable. A smart user interface also enables content-based video search, streamlining review and analysis.
Recognition and Support
The patent was granted after comparison with several internationally approved patents, validating its novelty and technical advancement. This marks the first patent awarded to the School of Computer Sciences at MGU.
The invention also earned the Best Research Innovative Practice Award from the university’s Research Incubation Programme (RINP) and received seed funding from the Research Innovation Network Kerala (RINK) under the Kerala Startup Mission.
MG University Vice-Chancellor Dr. C.T. Aravindakumar lauded the achievement, stating that the invention strengthens India’s capabilities in AI-powered surveillance and supports public safety through intelligent, scalable solutions.