Hawk eye
Loading...
Date
2023-08-13
Journal Title
Journal ISSN
Volume Title
Publisher
UMT, Lhr
Abstract
The goal of this project is to develop a system that can detect, identify, and track a person using non-overlapping cameras. The system utilizes computer vision techniques to detect and identify the target object in each camera's field of view. Once the target is detected and identified, the system tracks the target as it moves between cameras by analyzing the motion patterns and identifying unique features of the target. The system then uses this information to seamlessly link the tracks across multiple non-overlapping cameras, providing a comprehensive view of the target's movement. The system is tested on various scenarios with different camera angles and lighting conditions, and the results show that it can accurately detect, identify, and track a person with high precision and recall. The system can be used in various applications, such as surveillance, traffic monitoring, and crowd management