MUET logo

Efficient and Deployable Violence Detection using Transfer Learning Framework for Public Safety

Authors

  • Sanam Narejo

    Department of Computer Systems Engineering, Mehran University of Engineering and Technology, Pakistan
    Author

  • Shabana Hajano

    Department of Computer Systems Engineering, Mehran University of Engineering and Technology, Pakistan
    Author

  • Anoud Shaikh

    Department of Software Engineering, Mehran University of Engineering and Technology, Pakistan
    Author

  • Jibran Virk

    Department of Computer Systems Engineering, Mehran University of Engineering and Technology, Pakistan
    Author

  • Maheen Shaikh

    Department of Computer Systems Engineering, Mehran University of Engineering and Technology, Pakistan
    Author

DOI:

https://doi.org/10.22581/muet1982.0125

Keywords:

Smart city , Violence detection , Transfer learning , Closed-Circuit Television , Convolution Neural Networks , Artificial Intelligence

Abstract

One growing concern in urban areas is public safety in gathering spaces. Traditional Closed-Circuit Television (CCTV) monitoring systems depend almost entirely on human operators, making them prone to error due to fatigue and distraction. Due to the growing number of violent events, it is essential to develop automated intelligent systems to facilitate quick and accurate detection of potential violence, specifically in public spaces.  This research aims to strengthen traditional surveillance systems by reducing human reliance in favour of AI-driven automation and implementing transfer learning (TL), a subset of Artificial Intelligence (AI), for developing an automated violence detection system. TL allows models trained on much larger, diverse datasets to be further fine-tuned for specific problems, e.g., violent actions in CCTV footage. Thus, the proposed system intends to efficiently classify violent and non-violent actions at a relatively lower computation cost. The final system is expected to be deployable in diverse public environments, thus contributing to faster incident response and improved public safety. 

Downloads

References

[1] Y. Bi, D. Li, and Y. Luo, “Combining Keyframes and Image Classification for Violent Behavior Recognition,” Applied Sciences, vol. 12, no. 16, p. 8014, Aug. 2022, doi: 10.3390/app12168014.

[2] “US Homicides Increased Nearly 30% in 2020, FBI Reports.” Accessed: Feb. 02, 2025. [Online]. Available: https://www.voanews.com/a/us-homicides-increased-nearly-30-in-2020-fbi-reports-/6247529.html

[3] R. A. Taha, A. A. H. Youssif, and M. M. Fouad, “Transfer learning model for anomalous event recognition in big video data,” Scientific Reports 2024 14:1, vol. 14, no. 1, pp. 1–14, Nov. 2024, doi: 10.1038/s41598-024-78414-2.

[4] A. S. Keçeli and A. Kaya, “Violent activity detection with transfer learning method,” Electron Lett, vol. 53, no. 15, pp. 1047–1048, Jul. 2017, doi: 10.1049/el.2017.0970.

[5] A. J. Naik and M. T. Gopalakrishna, “Violence Detection in SurveillanceVideo-A survey,” International Journal of Latest Research in Engineering and Technology (IJLRET) www.ijlret.com ǁ PP, pp. 11–17, [Online]. Available: www.ijlret.com

[6] K. Lloyd, P. L. Rosin, D. Marshall, and S. C. Moore, “Detecting violent and abnormal crowd activity using temporal analysis of grey level co-occurrence matrix (GLCM)-based texture measures,” Mach Vis Appl, vol. 28, no. 3–4, pp. 361–371, May 2017, doi: 10.1007/s00138-017-0830-x.

[7] Y. Watanabe, M. Okabe, Y. Harada, and N. Kashima, “Real-World Video Anomaly Detection by Extracting Salient Features in Videos,” IEEE Access, vol. 10, pp. 125052–125060, 2022, doi: 10.1109/ACCESS.2022.3224952.

[8] F. D. M. De Souza, G. C. Chávez, E. A. Do Valle, and A. D. A. Araújo, “Violence detection in video using spatio-temporal features,” Proceedings - 23rd SIBGRAPI Conference on Graphics, Patterns and Images, SIBGRAPI 2010, pp. 224–230, 2010, doi: 10.1109/SIBGRAPI.2010.38.

[9] A. A. Sodemann, M. P. Ross, and B. J. Borghetti, “A review of anomaly detection in automated surveillance,” IEEE Transactions on Systems, Man and Cybernetics Part C: Applications and Reviews, vol. 42, no. 6, pp. 1257–1272, 2012, doi: 10.1109/TSMCC.2012.2215319.

[10] X. Fan, L. Xu, X. Zhang, and L. Chen, “The research and application of human detection based on support vector machine using in intelligent video surveillance system,” Proceedings - 4th International Conference on Natural Computation, ICNC 2008, vol. 2, pp. 139–143, 2008, doi: 10.1109/ICNC.2008.315.

[11] J. Albusac, J. J. Castro-Schez, L. M. Lopez-Lopez, D. Vallejo, and L. Jimenez-Linares, “A supervised learning approach to automate the acquisition of knowledge in surveillance systems,” Signal Processing, vol. 89, no. 12, pp. 2400–2414, Dec. 2009, doi: 10.1016/J.SIGPRO.2009.04.008.

[12] F. Kulsoom, S. Narejo, Z. Mehmood, H. N. Chaudhry, A. Butt, and A. K. Bashir, “A review of machine learning-based human activity recognition for diverse applications,” Neural Comput Appl, vol. 34, no. 21, pp. 18289–18324, Nov. 2022, doi: 10.1007/S00521-022-07665-9.

[13] E. Bermejo, O. Deniz, G. Bueno, and R. Sukthankar, “Violence Detection in Video Using Computer Vision Techniques.” [Online]. Available: http://visilab.etsii.uclm.es/

[14] K. Simonyan and A. Zisserman, “Very Deep Convolutional Networks for Large-Scale Image Recognition,” 3rd International Conference on Learning Representations, ICLR 2015 - Conference Track Proceedings, Sep. 2014, Accessed: Feb. 03, 2025. [Online]. Available: https://arxiv.org/abs/1409.1556v6

[15] P. Zhou, Q. Ding, H. Luo, and X. Hou, “Violence detection in surveillance video using low-level features,” PLoS One, vol. 13, no. 10, p. e0203668, Oct. 2018, doi: 10.1371/JOURNAL.PONE.0203668.

[16] J. Vidhya and R. Uthra, “Violence detection in videos using Conv2D VGG-19 architecture and LSTM network,” 2021.

[17] S. Narejo, B. Pandey, D. Esenarro Vargas, C. Rodriguez, and M. R. Anjum, “Weapon Detection Using YOLO V3 for Smart Surveillance System,” Math Probl Eng, vol. 2021, no. 1, p. 9975700, Jan. 2021, doi: 10.1155/2021/9975700.

[18] T. Aremu, L. Zhiyuan, R. Alameeri, M. Khan, and A. El Saddik, “SSIVD-Net: A Novel Salient Super Image Classification & Detection Technique for Weaponized Violence,” Lecture Notes in Networks and Systems, vol. 1018 LNNS, pp. 16–35, Jul. 2022, doi: 10.1007/978-3-031-62269-4_2.

[19] M. M. Mukto et al., “Design of a real-time crime monitoring system using deep learning techniques,” Intelligent Systems with Applications, vol. 21, Mar. 2024, doi: 10.1016/j.iswa.2023.200311.

Published

2025-10-28

Issue

Vol. 44 No. 4 (2025): October Issue

Section

Articles

License

Copyright (c) 2025 Mehran University Research Journal of Engineering and Technology

Creative Commons License

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

How to Cite

Efficient and Deployable Violence Detection using Transfer Learning Framework for Public Safety. (2025). Mehran University Research Journal of Engineering and Technology, 44(4), 01-16. https://doi.org/10.22581/muet1982.0125

Similar Articles

51-60 of 66

You may also start an advanced similarity search for this article.