CANELA: An AI-Driven Edge Camera System for Active Object Tracking in Intelligent Surveillance Systems

• José Manoel Silva UFRN https://orcid.org/0009-0001-9534-1267
• Leonardo Lima UFRN https://orcid.org/0009-0008-4840-4342
• Stefano Loss UFRN https://orcid.org/0000-0003-0093-4930
• Thais Batista UFRN https://orcid.org/0000-0003-3558-1450
• Everton Cavalcante UFRN https://orcid.org/0000-0002-2475-5075
• Aluizio Rocha Neto UFRN https://orcid.org/0000-0003-1531-1488
• Daniel Araújo UFRN https://orcid.org/0000-0001-5572-0505
• Eduardo Nogueira Cunha UFRN https://orcid.org/0000-0001-6991-7296
• Nélio Cacho UFRN https://orcid.org/0000-0002-4706-5616
• Frederico Lopes UFRN https://orcid.org/0000-0002-1108-8875

DOI: https://doi.org/10.5753/semish.2026.22201

Resumo

Intelligent surveillance systems have been increasingly relying on edge-based Artificial Intelligence for real-time object detection and tracking. However, fixed cameras suffer from a limited field of view and degraded image quality at long distances, while resource-constrained edge devices struggle to run complex vision models without incurring high latency and thermal issues. This paper presents CANELA (Camera with Autonomous Navigation and Envisioning Led by AI), a low-cost pan-tilt edge camera system for active object tracking. CANELA combines a custom gimbal, parallel control algorithms, and a Raspberry Pi 5 with a Hailo-8L M.2 Entry-Level accelerator to enable efficient on-device inference and dynamic camera repositioning to keep targets centered. Experimental results in a face-detection pipeline leveraging hardware acceleration demonstrate that CANELA maintains high frame rates while preserving CPU availability by offloading neural network inference to the accelerator. Compared to fixed cameras, it increases the duration of high-quality target capture, improving detection performance and tracking stability.

Referências

Adrian, A. I., Ismet, P., and Petru, P. (2018). An overview of intelligent surveillance systems development. In 2018 International Symposium on Electronics and Telecommunications (ISETC), USA. IEEE.

Bashir, F. and Arief, S. (2025). Utilization of deep learning in PTZ (pan-tilt-zoom) camera control systems for geospatial-based intelligence surveillance. Remote Sensing Technology in Defense and Environment, 2(2):117–131.

Benoit-Cattin, T., Velasco-Montero, D., and Fernández-Berni, J. (2020). Impact of thermal throttling on long-term visual inference in a CPU-based edge device. Electronics, 9(12).

Garcia, D. R., Serna, A. A., Crespo, D. S., Yu, X., and Saniie, J. (2024). AI smart security camera for person detection, face recognition, tracking and logging. In 2024 IEEE International Conference on Electro Information Technology (eIT), USA. IEEE.

Guo, J., Deng, J., Lattas, A., and Zafeiriou, S. (2022). Sample and computation redistribution for efficient face detection. In 2022 International Conference on Learning Representations. [link].

Hsia, S.-C., Wang, S.-H., Wei, C.-M., and Chang, C.-Y. (2022). Intelligent object tracking with an automatic image zoom algorithm for a camera sensing surveillance system. Sensors, 22(22).

Kim, I. S., Choi, H. S., Yi, K. M., Choi, J. Y., and Kong, S. G. (2010). Intelligent visual surveillance – A survey. International Journal of Control, Automation and Systems, 8(5):926–939.

Niu, H., Liu, J., Yu, Z., Zheng, D., He, P., and Wang, F. (2022). Real-time object tracking system using PTZ camera. In 2022 IEEE 4th International Conference on Civil Aviation Safety and Information Technology (ICCASIT), pages 471–478, USA. IEEE.

Schroff, F., Kalenichenko, D., and Philbin, J. (2015). FaceNet: A unified embedding for face recognition and clustering. In 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 815–823, USA. IEEE.

Sharma, H. and Kanwal, N. (2024). Video surveillance in smart cities: Current status, challenges & future directions. Multimedia Tools and Applications, 84(16):15787–15832.

Singh, S., Saini, R., Saurav, S., and Saini, A. K. (2017). Real-time object tracking with active PTZ camera using hardware acceleration approach. International Journal of Image, Graphics and Signal Processing, 9(2):55.

Wang, X., Tang, Z., Guo, J., Meng, T., Wang, C., Wang, T., and Jia, W. (2025). Empowering edge intelligence: A comprehensive survey on on-device AI models. ACM Computing Surveys, 57(9).

Yu, C., Wang, M., Chen, S., Wang, W., Fang, W., Chen, Y., and N.Xiong, N. (2025). Efficient NPU–GPU scheduling for real-time deep learning inference on mobile devices. Journal of Real-Time Image Processing, 22.