City Watch
The project aims to leverage the capabilities of Unmanned Aerial Vehicles (UAVs) to create an automated surveillance system. Drones, equipped with high-resolution cameras and AI software, will be used for accurate object detection and tracking, including vehicles and people, even under challenging conditions such as partial occlusion and varied viewpoints. The project specifically targets two critical objectives: (i) real-time tracking of vehicles and individuals on roads and (ii) precise estimation of vehicle queue lengths in dense traffic. Emphasis will be placed on developing efficient algorithms for embedded systems, ensuring optimal performance without overloading the UAV’s hardware. By addressing these challenges, the project seeks to contribute to the advancement of effective UAV-based surveillance systems for diverse applications, including traffic monitoring and security.
Parikshit Singh Rathore, Romit Bhaumik, Vishwajeet Pattanaik, Aravinda Reddy
Computer Vision, Edge-Computing, Deep Learning
Aerial View Videos
Bharat Electronics Limited (BEL)