What is it about?

Top-tier video systems employ (Pan/Tilt/Zoom) PTZ cameras, but these cameras are currently controlled by humans or by predetermined scanning tours. This paper addresses the following two main research problems: how to design an overall system for autonomously controlling the PTZ cameras and how to control these cameras in a manner that seeks to optimize the overall threat detection or recognition accuracy. Specifically, considering a security system with multiple PTZ cameras, we seek to determine a plan of camera settings (pan, tilt, and zoom) over time for each camera to increase the accuracy. Our proposed system employs wide-angle cameras, PTZ cameras, and a centralized parallel processor. The camera control works with realistic 3D environments and considers many factors, including the direction of the subject’s movement and its location, distances from the cameras, occlusion, as well as the movements of cameras and their capabilities and limitations. In addition, the paper utilizes clustering to group subjects, thereby enabling the system to focus on the areas that are more densely populated. Moreover, it proposes a dynamic mechanism for controlling the time spent on running the solution. Furthermore, it develops a parallel algorithm, allowing the most time-consuming phases to be parallelized and thus run efficiently by the centralized processor. We analyze through simulation the effectiveness of the overall solution, including the clustering approach, dynamic mechanism, and parallel implementation in terms of overall recognition performance and the running time of the solution, considering the impacts of numerous parameters.

Featured Image

Why is it important?

This project is important because it relates to the design of a fully autonomous security system. It also increases public safety by providing the ability to autonomously recognize malicious acts and enabling a fast response.

Read the Original

This page is a summary of: An Autonomous System for Efficient Control of PTZ Cameras, ACM Transactions on Autonomous and Adaptive Systems, June 2021, ACM (Association for Computing Machinery),
DOI: 10.1145/3507658.
You can read the full text:

Read

Contributors

The following have contributed to this page