Improving Ptz Camera Control for Surveillance Using Proximal Policy Optimization
| DOWNLOAD | DOI: 10.62897/COS2024.2-1.167 |
|
Author: Adonisz Dimitriu, Balázs Kósa, Viktor Remeli, Viktor Tihanyi Széchenyi István University, Hungary dimitriu.adonisz@techtra.hu |
Abstract: This paper presents advancements in camera-only surveillance systems, focusing on the inte-gration of Proximal Policy Optimization (PPO) and its recurrent variant to improve the detec-tion and tracking capabilities of PTZ cameras. Building on previous research, we refine our PPO model by adjusting observation representations and hyperparameters, significantly improving target coverage and tracking accuracy. The study evaluates two simulated scenarios, an abstract grid-like environment and a more realistic scenario with predefined intruder paths. The results demonstrate that the standard PPO outperforms both the heuristic and recurrent models. The findings confirm the potential of Reinforcement Learning (RL) in improving autonomous sur-veillance systems but also point to the need for more complex simulation environments to fully exploit the capabilities of recurrent neural networks.
REFERENCES
Bisagno N., Xamin A., De Natale F., Conci N., Rinner B., 2020, Dynamic camera reconfiguration with rein-forcement learning and stochastic methods for crowd surveillance. Sensors, 20(17), 4691.
Chen , Zhao J., Yang M., Zhou W., Li H., 2024, Optimizing Camera Motion with MCTS and Target Motion Modeling in Multi-Target Active Object Tracking. ACM Transactions on Multimedia Computing, Communi-cations and Applications., 20(7), 1-19
Darazs , Bukovinszki M., Kósa B., Remeli V. Tihanyi V.R., 2024, Comparison of Barrier Surveillence Algorithms for Directional Sensors and UAVs. Sensors, 24(14), 4490
Fahim , Papalexakis E., Krishnamurthy S.V., Roy Chowdhury A.K., Kaplan L., Abdelzaher T., 2023, AcTrak: Controlling a Steerable Surveillance Camera using Reinforcement Learning. ACM Transactions on Cy-ber-Physical Systems, 7(2), 1-27.
Fang Z., Zhao J., Yang M., Zhou W., Lu Z., Li H., 2022, Coordinate-aligned multi-camera collaboration for active multi-object tracking. arXiv.org, <https://arxiv.org/abs/2202.10881>.
Gupta J.K., Egorov M., Kochenderfer M., 2017, Cooperative multi-agent control using deep reinforcement learning. In Autonomous Agents and Multiagent Systems: AAMAS 2017 Workshops, Best Papers, São Pau-lo, Brazil, May 8-12, 2017, Springer International Publishing, Revised Selected Papers, 16, 66-83.
Guvensan M.A., Yavuz A.G., 2011, On coverage issues in directional sensor networks: A survey. Ad Hoc Networks, 9(7), 1238-1255.
Pan , Liu M., Zhong F., Yang Y., Zhu S. C., Wang Y., 2022. Mate: Benchmarking multi-agent reinforcement learning in distributed target coverage control. Advances in Neural Information Processing Systems, 35, 27862-27879.
Ramshaw L., Tarjan R.E., 2012, On minimum-cost assignments in unbalanced bipartite graphs. HP Labs, Palo Alto, CA, USA, Rep. HPL-2012-40R1, 20.
Schulman J., Wolski F., Dhariwal P., Radford A., Klimov O., 2017, Proximal policy optimization algorithms. arXiv preprint arXiv:1707.06347.
Wang , Cao X., Lin X., Zhang W., Qin L., 2018, Efficient computing of radius-bounded k-cores. IEEE 34th International Conference on Data Engineering (ICDE), Paris, France, 2018, 233-244, DOI: 10.1109-ICDE.2018.00030.
Wang , Zhong F., Xu J., Wang Y., 2021, Tom2c: Target-oriented multi-agent communication and cooper-ation with theory of mind. arXiv preprint arXiv:2111.09189.