MUMBAI, India, July 7 -- Intellectual Property India has published a patent application (202641077241 A) filed by Koneru Vijaya Lakshmi; Koneru Lakshmaiah Education Foundation; Srilatha M; and Dr Srinivasu N on June 23, 2026, for An Edge Computing Based Fr-Cnn-Cso System For Real-Time Human Detection And Tracking From Live Cctv And Mobile Camera Video Streams.
Inventors include Srilatha M; and Dr Srinivasu N.
The application for the patent was published on July 03, 2026, under issue no. 27/2026.
Abstract: The present invention discloses an edge computing-based FR-CNN-CSO system for real-time human detection and tracking from live CCTV and mobile camera video streams. The system utilizes a hybrid deep learning framework integrating Faster Region-Based Convolutional Neural Network (FR-CNN) with Crow Search Optimization (CSO) for accurate human detection, localization, and continuous tracking. The system receives live video streams from CCTV cameras, mobile phone cameras, IP cameras, and IoT- enabled imaging devices and performs real-time processing through an edge computing architecture. The invention includes preprocessing, HOG-based feature extraction, FR-CNN based detection, CSO-based parameter optimization, and tracking modules for generating human identification, bounding box coordinates, confidence scores, and movement trajectories. The proposed system reduces latency, improves computational efficiency, and enhances tracking stability under challenging surveillance conditions including occlusion, illumination variation, and crowded environments, thereby enabling intelligent surveillance and video analytics applications.
Disclaimer: Curated by HT Syndication.
