Conference Paper2025

Accelerating Violence Detection: A Comparative Study of Frame Difference and Traditional Methods on UCF Crime Dataset

Saniul Islam Sani, Rashedur Rahman, Zahidul Islam, Siam Tahsin Bhuiyan, Mahmudul Haque, Sefatul Wasi, Sajed Imtenanul Haque, Saadia Binte Alam

2025 International Conference on Wavelet Analysis and Pattern Recognition (ICWAPR)

IEEE, pp. 78-83

  • Peace, Justice and Strong Institutions — Relevance: 55%
DOI: 10.1109/icwapr66259.2025.11306761URL

Abstract

Effective violence detection is crucial for enhancing public safety and preventing harmful events. This paper evaluates two approaches using a ResNet50-LSTM model: normal frame input and frame difference input. Data preprocessing involved using normal frames for one method and computing differences between consecutive frames for the other. Performance was assessed using ROC curves and AUROC metrics. The findings indicate that the frame difference method improves AUROC scores and reduces computational time, making it a more efficient and accurate solution for real-time violence detection.

Keywords

  • violence detection
  • frame difference
  • computational efficiency
  • real-time video analysis
  • UCF-Crime dataset
  • video surveillance

CCDS Authors

Md. Rashedur Rahman, D. Eng.
Md. Rashedur Rahman, D. Eng.

Md. Rashedur Rahman, D. Eng.

  • Co-Director, MIRA & NEST Wings

Assistant Professor

Department of Computer Science and Engineering

Independent University, Bangladesh

Published at

CHI, HCII, Scientific Reports, Cognitive Computation, Neural Computing and Applications

Siam Tahsin Bhuiyan
Siam Tahsin Bhuiyan

Siam Tahsin Bhuiyan

  • Intern, CCDS

Bachelors

Independent University, Bangladesh

Published at

Scientific Reports

Saadia Binte Alam, PhD
Saadia Binte Alam, PhD

Saadia Binte Alam, PhD

  • Director, MIRA & WiSE Wings

Associate Professor

Department of Computer Science and Engineering

Independent University, Bangladesh

Published at

HCII, Scientific Reports, Cognitive Computation, Neural Computing and Applications

References

  1. 1.Kaiming He, Xiangyu Zhang, Shaoqing Ren, Jian Sun. (2016). Deep Residual Learning for Image Recognition. , 770–778[10.1109/cvpr.2016.90]
  2. 2.Sepp Hochreiter, Jürgen Schmidhuber. (1997). Long Short-Term Memory. Neural Computation, 9(8), 1735–1780[10.1162/neco.1997.9.8.1735]
  3. 3.Wei Liu, Dragomir Anguelov, Dumitru Erhan, Christian Szegedy, Scott Reed, Cheng-Yang Fu, Alexander C. Berg. (2016). SSD: Single Shot MultiBox Detector. Lecture notes in computer science, 21–37[10.1007/978-3-319-46448-0_2]
  4. 4.Limin Wang, Yuanjun Xiong, Zhe Wang, Yu Qiao, Dahua Lin, Xiaoou Tang, Luc Van Gool. (2016). Temporal Segment Networks: Towards Good Practices for Deep Action Recognition. Lecture notes in computer science, 20–36[10.1007/978-3-319-46484-8_2]
  5. 5.Francis Sahngun Nahm. (2022). Receiver operating characteristic curve: overview and practical use for clinicians. Korean Journal of Anesthesiology, 75(1), 25–36[10.4097/kja.21209]
  6. 6.Licheng Jiao, Fan Zhang, Fang Liu, Shuyuan Yang, Lingling Li, Zhixi Feng, Rong Qu. (2019). A Survey of Deep Learning-Based Object Detection. IEEE Access, 7, 128837–128868[10.1109/access.2019.2939201]
  7. 7.Kensho Hara, Hirokatsu Kataoka, Yutaka Satoh. (2017). Learning Spatio-Temporal Features with 3D Residual Networks for Action Recognition. , 3154–3160[10.1109/iccvw.2017.373]
  8. 8.Soheil Vosta, Kin-Choong Yow. (2022). A CNN-RNN Combined Structure for Real-World Violence Detection in Surveillance Cameras. Applied Sciences, 12(3), 1021[10.3390/app12031021]
  9. 9.Yanjinlkham Myagmar-Ochir, Wooseong Kim. (2023). A Survey of Video Surveillance Systems in Smart City. Electronics, 12(17), 3567[10.3390/electronics12173567]
  10. 10.Javaria Amin, Muhammad Almas Anjum, Kainat Ibrar, Muhammad Sharif, Seifedine Kadry, Ruben González Crespo. (2023). Detection of anomaly in surveillance videos using quantum convolutional neural networks. Image and Vision Computing, 135, 104710[10.1016/j.imavis.2023.104710]
  11. 11.Muhammad Javed Iqbal, Muhammad Munwar Iqbal, Iftikhar Ahmad, Madini O. Alassafi, Ahmed S. Alfakeeh, Ahmed Alhomoud. (2021). Real-Time Surveillance Using Deep Learning. Security and Communication Networks, 2021, 1–17[10.1155/2021/6184756]
  12. 12.Sarthak Sharma, B. Sudharsan, Saamaja Naraharisetti, Vimarsh Trehan, Kayalvizhi Jayavel. (2021). A fully integrated violence detection system using CNN and LSTM. International Journal of Power Electronics and Drive Systems/International Journal of Electrical and Computer Engineering, 11(4), 3374[10.11591/ijece.v11i4.pp3374-3380]
  13. 13.Soheil Vosta, Kin-Choong Yow. (2023). KianNet: A Violence Detection Model Using an Attention-Based CNN-LSTM Structure. IEEE Access, 12, 2198–2209[10.1109/access.2023.3339379]
  14. 14.Naz Dündar, Ali Seydi Keçeli, Aydın Kaya, Hayri Sever. (2024). A shallow 3D convolutional neural network for violence detection in videos. Egyptian Informatics Journal, 26, 100455[10.1016/j.eij.2024.100455]
  15. 15.Israel Mugunga, Junyu Dong, Eric Rigall, Shaoxiang Guo, Amanuel Hirpa Madessa, Hafiza Sadia Nawaz. (2021). A Frame-Based Feature Model for Violence Detection from Surveillance Cameras Using ConvLSTM Network. , 55–60[10.1109/icivc52351.2021.9526948]
  16. 16.Maria Gadelkarim, Mazen Khodier, Walid Gomaa. (2022). Violence Detection and Recognition from Diverse Video Sources. 2022 International Joint Conference on Neural Networks (IJCNN), 1–8[10.1109/ijcnn55064.2022.9892660]
  17. 17.Ioffe, Sergey, Szegedy, Christian. (2015). Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift. arXiv (Cornell University)[10.48550/arxiv.1502.03167]
  18. 18.Francisco Melo. (2013). Area under the ROC Curve. , 38–39[10.1007/978-1-4419-9863-7_209]
← All publications