Design and Implementation of a Smart Scenic Area Tourist Behavior Safety Monitoring System based on HiSilic

Authors

  • Dan Yin Wuhan University of Technology and Business, Wuhan, Hubei, China
  • Siwen Sun South-Central Minzu University, Wuhan, Hubei, China
  • Ting Zhang Wuhan University of Technology and Business, Wuhan, Hubei, China
  • Jianli Gong Wuhan University of Technology and Business, Wuhan, Hubei, China

DOI:

https://doi.org/10.54097/hy7qwb90

Keywords:

HiSilicon, Tourist Behavior Monitoring, Crowd Flow Statistics, Edge-cloud Collaboration, Edge Computing

Abstract

Inspired by the technical capability of the HiSilicon Hi3516DV300 camera to efficiently run the YOLOv3-tiny model on embedded devices, this solution offers significant advantages of high accuracy, low latency, and low power consumption, providing a valuable reference for the application of lightweight edge intelligence devices in scenic area monitoring. This study analyzes and designs a smart scenic area tourist behavior safety monitoring system based on an “edge-cloud” collaborative architecture. The hardware terminal adopts the HiSilicon Hi3516DV300 smart camera, equipped with an optimized YOLOv3-tiny lightweight object detection model accelerated by TensorRT, to achieve local real-time detection of human bodies and key targets within the scenic area. Simple tracking and trajectory analysis are combined to complete crowd flow statistics and density warning. The model is also tailored for head/body detection and deployed at the edge for real‑time counting. On the cloud side, the Alibaba Cloud IoT platform provides device management, data storage, visual analytics, and remote interaction functionalities.

Downloads

Download data is not yet available.

References

[1] Ren, K., et al. (2020). A systematic study of tiny YOLO3 inference: Toward compact brainware processor with less memory and logic gate. IEEE Access, 8, 142931–142955. https://doi.org/10.1109/ACCESS.2020.2972284.

[2] Adewumi, A., & Saraswat, D. (2025). Assessing TensorRT optimization for real-time object detection in agriculture. In 2025 ASABE Annual International Meeting (Paper No. 2500469). ASABE. https://doi.org/10. 13031/aim.2025. 2500 469.

[3] Qian, B., Xuan, Y., Wu, D., et al. (2025). Edge-cloud collaborative streaming video analytics with multi-agent deep reinforcement learning. IEEE Network, 39(1). https://doi. org/ 10. 1109/MNET.2025.3492761.

[4] Dong, Q. (2024). Safety monitoring system for tourist scenic spots based on crowd scene type recognition. International Journal of Security and Networks, 19(3), 128–137. https://doi.org/10.1504/IJSN.2024.139652.

[5] Huang, S., Ji, J., Wang, Y., Li, W., & Zheng, Y. (2023). A machine vision-based method for crowd density estimation and evacuation simulation. Safety Science, 167, 106285. https:// doi. org/10.1016/j.ssci.2023.106285.

[6] Tariq, M. A., Ajmal, M., Jo, E., et al. (2024). Revolutionizing surveillance: A brief survey of edge AI terminals. In 2024 6th International Conference on Artificial Intelligence in Information and Communication (ICAIIC) (pp. 281–285). IEEE. https://doi.org/10.1109/ICAIIC60668.2024.10446789.

[7] Calì, R., Falaschetti, L., & Biagetti, G. (2025). Optimized implementation of YOLOv3-Tiny for real-time image and video recognition on FPGA. Electronics, 14(20), 3993. https://doi.org/10.3390/electronics14203993.

[8] Tariq, M. A., Ajmal, M., Jo, E., et al. (2024). Revolutionizing surveillance: A brief survey of edge AI terminals. In 2024 6th International Conference on Artificial Intelligence in Information and Communication (ICAIIC) (pp. 281–285). IEEE. https://doi.org/10.1109/ICAIIC60668.2024.10446789.

[9] Lema, D. G., Usamentiaga, R., & García, D. F. (2024). Quantitative comparison and performance evaluation of deep learning-based object detection models on edge computing devices. Integration, 95, 102127. https://doi.org/ 10. 1016/j. integration.2024.102127.

[10] Martin-Salinas, I., Badia, J. M., Valls, O., et al. (2024). Evaluating and accelerating vision transformers on GPU-based embedded edge AI systems. The Journal of Supercomputing, 80, 28461–28494. https://doi.org/10.1007/s11227-024-05987-9.

[11] Wijayanti, K., Mutiara, G. A., Suryawardani, B., Ervina, E., & Kusuma, G. P. (2025). Non-intrusive real-time tourist crowd monitoring for overtourism mitigation using YOLOv8-based head detection and tracking. Journal of Robotics and Control, 6(4). https://doi.org/10.18196/jrc.v6i4.26396.

[12] Dionis-Ros, A., Vila-Francés, J., Magdalena-Benedito, R., Mateo, F., & Serrano-López, A. J. (2024). Multimodal video analysis for crowd anomaly detection using open access tourism cameras. Applied Sciences, 14(23), 11075. https://doi. org/ 10.3390/app142311075.

[13] Veesam, S. B., Rao, B. T., Begum, Z., et al. (2025). Multi-camera spatiotemporal deep learning framework for real-time abnormal behavior detection in dense urban environments. Scientific Reports, 15, 26813. https://doi.org/10.1038/s41598-025-12388-7.

[14] Li, J., & Fu, A. (2025). Anomaly behavior detection based on deep learning in an IoT environment. Sensors, 25(24), 7605. https://doi.org/10.3390/s25247605.

Downloads

Published

29-06-2026

Issue

Vol. 17 No. 1 (2026)

Section

Articles

License

Copyright (c) 2026 Frontiers in Computing and Intelligent Systems

Creative Commons License

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.

How to Cite

Yin, D., Sun, S., Zhang, T., & Gong, J. (2026). Design and Implementation of a Smart Scenic Area Tourist Behavior Safety Monitoring System based on HiSilic. Frontiers in Computing and Intelligent Systems, 17(1), 22-27. https://doi.org/10.54097/hy7qwb90