MUMBAI, India, July 24 -- Intellectual Property India has published a patent application (202641086055 A) filed by Vallurupalli Nageswara Rao Vignana Jyothi Institute Of Engineering And on July 14, 2026, for Intelligent Cctv Surveillance Framework For Anomaly Detection And Instant Alert Generation.
Inventors include Dr. D. N. Vasundhara; Dr. B. V. Kiranmayee; and S. Sudeshna.
The application for the patent was published on July 17, 2026, under issue no. 29/2026.
Abstract: Improving real-time monitoring is crucial for contemporary security systems, which require advanced anomaly detection systems and quick response mechanisms. This research investigates an end-to-end methodology for video summarization and anomaly detection using deep learning paradigms and automated alerting mechanisms. The framework combines a DenseNet-based deep neural model to conduct real-time anomaly classification of CCTV video frames. Each frame is processed independently, with inferential predictions summed to arrive at a global evaluation of video anomalies. A web interface powered by Flask provides smooth video ingestion, where OpenCV-based processing extracts frames and normalizes them to a 64×64 resolution for model inference. Upon identifying anomalous activities like unauthorized entry, violent confrontations, or theft the system automatically initiates alert notifications through the SMTP protocol, including evidentiary video segments to allow immediate situational evaluation. By aligning machine learning-based anomaly detection with automatic notification schemes, this design complements real-time monitoring functionality and encourages a more proactive security framework. Our suggested system has the potential to evolve real-time security systems through an efficient and scalable architecture for smart surveillance.
Disclaimer: Curated by HT Syndication.
