Edge AI for Smart Surveillance: Real-time Human Activity Recognition on Low-power Devices
Keywords:
Computer vision, Edge AI, Human activity recognition, Low-power computing, Real-time processing, Smart surveillanceAbstract
The proliferation of internet of things (IoT) devices and the increasing demand for intelligent surveillance systems have created a compelling need for edge-based artificial intelligence (Edge AI) solutions. This study presents a comprehensive Edge AI framework for real-time human activity recognition (HAR) on low-power devices, achieving 87.3% classification accuracy with an average power consumption of 2.1W on ARM-based edge platforms. The system integrates optimized deep-learning models with quantization and pruning techniques to balance accuracy, latency, and power efficiency. Comparative analysis against existing approaches highlights our system’s superior balance of real-time performance and energy efficiency, outperforming traditional cloud-based and motion detection systems in terms of latency and false-positive rates. The results validate the feasibility of deploying sophisticated HAR at the edge, paving the way for scalable, privacy-preserving smart surveillance systems.
References
S. Singh, K. V. Arya, C. R. Rodriguez, and A. O. Mulani, Emerging trends in artificial intelligence, Data Science and Signal Processing. Switzerland: Cham: Springer Nature, 2025.
V. Godase, and A. Jagadale “Three element control using PLC, PID & SCADA interface,” International Journal for Scientific Research & Development, vol. 7, no. 2, pp. 1105-1109, 2019, Available: https://www.ijsrd.com/articles/IJSRDV7I21002.pdf
V. Godase, “Optimized algorithm for face recognition using Deepface and multi-task cascaded convolutional network (MTCNN),” Optimum Science Journal, vol. 3, pp. 66-74, doi: https://doi.org/10.5281/zenodo.15341560
V. Godase, “A comprehensive study of revolutionizing EV charging with solar-powered wireless solutions,” Advance Research in Power Electronics and Devices, vol. 2, no. 1, pp. 23-37, Apr. 2025, Available: https://matjournals.net/engineering/index.php/ARPED/article/view/1752
V. Godase, A. Mulani, A. Pawar and K. Sahani, “A comprehensive review on PIR sensor-based light automation systems,” International Journal of Image Processing a Smart Sensors, vol. 1, no. 1, pp. 22-29, May 2025, Available: https://www.researchgate.net/publication/391462475
V. Godase, A. Mulani, S. Takale, and R. Ghodake, “Comprehensive review on automated field irrigation using soil image analysis and IoT,” Journal of Advance Electrical Engineering and Devices, vol. 3, no. 1, pp. 46-55, Apr. 2025, Available: https://matjournals.net/engineering/index.php/JAEED/article/view/1713
V. Godase, A. Mulani, S. Takale and R. Ghodake, “A holistic review of automatic drip irrigation systems: Foundations and emerging trends,” Journal of Instrumentation and Innovation Sciences, vol. 10, no. 1, pp. 38-47, Apr. 2025, Available: https://matjournals.net/engineering/index.php/JIIS/article/view/1703
S. K. A, A. Kathiravan, P. Mannam, D. Akila, B. S. Kumar and V. Vilas Godase, “Advanced neural network models for optimal energy management in microgrids with integrated electric vehicles,” 2025 5th International Conference on Trends in Material Science and Inventive Materials (ICTMIM), Kanyakumari, India, 2025, pp. 1869-1874, doi: https://doi.org/10.1109/ICTMIM65579.2025.10988248
V. K. Jamadade, M. G. Ghodke, S. S. Katakdhond and V. Godase, “A comprehensive review on scalable Arduino radar platform for real-time object detection and mapping,” Journal of Microprocessor and Microcontroller Research, vol. 2, no. 2, pp. 1-12, May 2025, Available: https://matjournals.net/engineering/index.php/JoMMR/article/view/1888
V. Godase, “Navigating the digital battlefield: An in-depth analysis of cyber-attacks and cybercrime”, International Journal of Data Science, Bioinformatics and Cyber Security, vol. 1, no. 1, pp. 16-27, May 2025., Available: https://matjournals.net/engineering/index.php/IJDSBCS/article/view/1896
A. C. Cob-Parro, C. Losada-Gutiérrez, M. Marrón-Romera, A. Gardel-Vicente, and I. Bravo-Muñoz, “Smart video surveillance system based on edge computing,” Sensors, vol. 21, no. 9, p. 2958, Apr. 2021, doi: https://doi.org/10.3390/s21092958
V. Godase, P. Pawar, S. Nagane, and S. Kumbhar “Automatic railway horn system using node MCU.” Journal of Control & Instrumentation, vol. 15 no. 1, pp. 11-19, May 2024, Available: https://www.researchgate.net/publication/381965272
Aimé Cedric Muhoza, E. Bergeret, C. Brdys, and F. Gary, “Power consumption reduction for IoT devices thanks to Edge-AI: Application to human activity recognition,” Internet of Things, vol. 24, pp. 100930–100930, Dec. 2023, doi: https://doi.org/10.1016/j.iot.2023.100930
V. Godase, A. Mulani, R. Ghodak, G. Birajadar, S. Takale, M. Kolte, “A MapReduce and Kalman Filter based secure IIoT environment in Hadoop,” Sanshodhak, vol. 25, Jun. 2024, Available: https://www.researchgate.net/publication/383941977
A. D. and M. R.I., “Edge computing based surveillance framework for real-time activity recognition,” ICT Express, vol. 7, no. 2, pp. 182–186, Jun. 2021, doi: https://doi.org/10.1016/j.icte.2021.04.010
V. Dhope, A. Chavan, N. Hadmode, and V. Godase, Smart plant monitoring system,” International Journal of Creative Research Thoughts, vol. 12, no. 5, pp. b844–b849, May 2024, Available: https://ijcrt.org/papers/IJCRT2405203.pdf
Vittorio Fra, E. Forno, Riccardo Pignari, T. C. Stewart, Enrico Macii, and Gianvito Urgese, “Human activity recognition: Suitability of a neuromorphic approach for on-edge AIoT applications,” Neuromorphic Computing and Engineering, vol. 2, no. 1, pp. 014006–014006, Feb. 2022, doi: https://doi.org/10.1088/2634-4386/ac4c38
V. Godase, A. Lawande, K. Mane, K. Davad, and S. Gangonda, “Pipeline survey robot,” International Journal for Scientific Research and Development, vol. 12, no.3, 2024, Available: https://www.ijsrd.com/articles/IJSRDV12I30108.pdf
A. C. Cob-Parro, C. Losada-Gutiérrez, M. Marrón-Romera, A. Gardel-Vicente, and I. Bravo-Muñoz, “A new framework for deep learning video based Human action recognition on the edge,” Expert Systems with Applications, vol. 238, no. Part E, pp. 122220–122220, Mar. 2024, doi: https://doi.org/10.1016/j.eswa.2023.122220
V. Godase, Y. Jadhav, K. Vishal, V. Metkari, and S. Gangonda, “IoT based greenhouse monitoring and controlling system,” International Journal for Scientific Research and Development, vol. 12, no. 3, 2024, Available: https://www.ijsrd.com/articles/IJSRDV12I30107.pdf
A. Ullah, K. Muhammad, W. Ding, V. Palade, I. Ul Haq, and S. W. Baik, “Efficient activity recognition using lightweight CNN and DS-GRU network for surveillance applications,” Applied Soft Computing, vol. 103, pp. 107102–107102, May 2021, doi: https://doi.org/10.1016/j.asoc.2021.107102
H. H. Nguyen, T. N. Ta, N. C. Nguyen, V. T. Bui, H. M. Pham and D. M. Nguyen, “YOLO based real-time human detection for smart video surveillance at the edge,” 2020 IEEE Eighth International Conference on Communications and Electronics (ICCE), Phu Quoc Island, Vietnam, 2021, pp. 439-444, doi: https://doi.org/10.1109/ICCE48956.2021.9352144
S. K. Nagane, S. Pawar, and V. V. Godase, “Cinematica sentiment analysis,” Journal of Image Processing and Intelligent Remote Sensing, vol. 2, no. 3, pp. 27–32, May 2022, doi: https://doi.org/10.55529/jipirs.23.27.32
R. Dange, E. Attar, P. Ghodake, and V. Godase, “Smart agriculture automation using ESP8266 Node MCU,” Journal of Electronics Computer Networking and Applied Mathematics, vol. 3, no. 5, pp. 1–9, Sep. 2023, doi: https://doi.org/10.55529/jecnam.35.1.9
V. Godase and J. Godase, “Diet prediction and feature importance of gut Microbiome using machine learning”, Evolution in Electrical and Electronics Engineering, vol. 5, no. 2, pp. 214–219, Nov. 2024, Available: https://publisher.uthm.edu.my/periodicals/index.php/eeee/article/view/16120
