Road Traffic Accident Detection Using Autoencoder-Based Models: A Performance Analysis of CAE and VAE
DOI:
https://doi.org/10.32628/IJSRST261330Keywords:
Traffic accident detection, Convolutional Autoencoder, Variational Autoencoder, Anomaly detection, Video surveillance, Intelligent transportation systemsAbstract
Road traffic accidents are a major public health and safety issue worldwide, and thus there is a need to design systems that can automatically detect accidents in time to provide emergency response and manage traffic. In this paper, a detailed performance analysis of Convolutional Autoencoder (CAE) and Variational Autoencoder (VAE)-based architectures for the detection of road traffic accidents in surveillance videos is presented. Unlike supervised approaches that require extensive labelled datasets, autoencoders learn normal traffic patterns and identify accidents as anomalies through reconstruction error analysis. We evaluate both models on two benchmark datasets: the IITH road accident dataset from Hyderabad, India, and the UCF-Crime dataset's road accident subset. Our experimental results demonstrate that CAE consistently outperforms VAE across multiple metrics, achieving 86.78% accuracy on IITH dataset compared to VAE's 83.42% and achieving 81.63% accuracy on UCF-Crime Dataset compared to VAE's 78.56%. The deterministic latent representations of CAE prove more effective for accident detection than VAE's probabilistic modelling, though VAE's uncertainty quantification offers potential advantages for confidence estimation. Furthermore, we provide an analytical discussion on the implications of reconstruction-based anomaly scoring for downstream applications such as traffic video summarization, where high reconstruction errors can serve as saliency indicators for keyframe selection. This study contributes empirical evidence for selecting appropriate autoencoder architectures in intelligent transportation systems and highlights the trade-offs between deterministic and probabilistic latent space modelling for real-world traffic surveillance applications.
Downloads
References
World Health Organization. (2023). Road traffic injuries. Retrieved December 13, 2023, from https://www.who.int/news-room/fact-sheets/detail/road-traffic-injuries
Ministry of Road Transport and Highways. (2022). Road accidents in India 2022. https://morth.nic.in/sites/default/files/RA_2022_30_Oct.pdf
Ministry of Road Transport and Highways. (2024, July 24). Deaths due to road accidents in India. Press Information Bureau. https://www.pib.gov.in/PressReleasePage.aspx?PRID=2036268
Chauhan, A., & Vegad, S. (2022). Smart surveillance based on video summarization: A comprehensive review, issues, and challenges. In V. Suma, X. Fernando, K. L. Du, & H. Wang (Eds.), Evolutionary computing and mobile sustainable networks (Vol. 116). Springer, Singapore. https://doi.org/10.1007/978-981-16-9605-3_29
Saini, P., Kumar, K., Kashid, S., et al. (2023). Video summarization using deep learning techniques: A detailed analysis and investigation. Artificial Intelligence Review, 56, 12347–12385. https://doi.org/10.1007/s10462-023-10444-0
Pawar, K., & Attar, V. (2019). Deep learning approaches for video-based anomalous activity detection. World Wide Web, 22, 571–601. https://doi.org/10.1007/s11280-018-0582-1
Chauhan, A., & Vegad, S. (2026). CVAE-ADS: A deep learning framework for traffic accident detection and video summarization. Journal of Electronic, Electromedical Engineering and Medical Informatics, 8(1), 185–205.
Singh, D., & Mohan, C. K. (2019). Deep spatio-temporal representation for detection of road accidents using stacked autoencoder. IEEE Transactions on Intelligent Transportation Systems, 20(3), 879–887. https://doi.org/10.1109/TITS.2018.2835308
Srinivasan, A., Srikanth, H., Indrajit, H., & Narasimhan, V. (2020). A novel approach for road accident detection using DETR algorithm. In Proceedings of the International Conference on Intelligent Data Science Technologies and Applications (IDSTA) (pp. 75–80). https://doi.org/10.1109/IDSTA50958.2020.9263703
Wang, C., Dai, Y., Zhou, W., & Geng, Y. (2020). A vision-based video crash detection framework for mixed traffic flow environment considering low-visibility condition. Journal of Advanced Transportation. https://doi.org/10.1155/2020/9194028
Robles-Serrano, S., Sanchez-Torres, G., & Branch-Bedoya, J. (2021). Automatic detection of traffic accidents from video using deep learning techniques. Computers, 10(11), 148. https://doi.org/10.3390/computers10110148
Khan, S. W., Hafeez, Q., Khalid, M. I., Alroobaea, R., Hussain, S., Iqbal, J., Almotiri, J., & Ullah, S. S. (2022). Anomaly detection in traffic surveillance videos using deep learning. Sensors, 22(17), 6563. https://doi.org/10.3390/s22176563
Pawar, K., & Attar, V. (2021). Deep learning based detection and localization of road accidents from traffic surveillance videos. ICT Express. https://doi.org/10.1016/j.icte.2021.11.004
Pathak, R., & Elster, A. C. (2022). Applying transfer learning to traffic surveillance videos for accident detection. In Proceedings of the International Conference on Applied Artificial Intelligence (ICAPAI) (pp. 1–7). https://doi.org/10.1109/ICAPAI55158.2022.9801568
Thakare, K. V., Dogra, D. P., Choi, H., Kim, H., & Kim, I.-J. (2022). Object interaction-based localization and description of road accident events using deep learning. IEEE Transactions on Intelligent Transportation Systems, 23(11), 20601–20613. https://doi.org/10.1109/TITS.2022.3170648
Adewopo, V. A., & Elsayed, N. (2024). Smart city transportation: Deep learning ensemble approach for traffic accident detection. IEEE Access, 12, 59134–59147. https://doi.org/10.1109/ACCESS.2024.3387972
Thomas, S. S., Gupta, S., & Subramanian, V. K. (2017). Event detection on roads using perceptual video summarization. IEEE Transactions on Intelligent Transportation Systems, 19(9), 2944–2954.
Kosambia, T., & Gheewala, J. (2021). Video synopsis for accident detection using deep learning technique. In Proceedings of the International Conference on Smart Data Intelligence (ICSMDI 2021). https://ssrn.com/abstract=3851250
Pramanik, A., Pal, S. K., Maiti, J., & Mitra, P. (2022). Traffic anomaly detection and video summarization using spatio-temporal rough fuzzy granulation with Z-numbers. IEEE Transactions on Intelligent Transportation Systems, 23(12), 24116–24125. https://doi.org/10.1109/TITS.2022.3198595
Tahir, M., Qiao, Y., Kanwal, N., Lee, B., & Asghar, M. N. (2023). Privacy preserved video summarization of road traffic events for IoT smart cities. Cryptography, 7(1), 7. https://doi.org/10.3390/cryptography7010007
Saxena, N., & Asghar, M. N. (2023). YOLOv5 for road events based video summarization. In K. Arai (Ed.), Intelligent computing (Vol. 739). Springer, Cham. https://doi.org/10.1007/978-3-031-37963-5_69
Zhang, K., Chao, W. L., Sha, F., & Grauman, K. (2016). Video summarization with long short-term memory. In Computer Vision – ECCV 2016 (Vol. 9911). Springer, Cham. https://doi.org/10.1007/978-3-319-46478-7_47
Huilin Liu, Xiaolong Hu, Guanghan Sun, Wenkang Zhang, Jialei Zhan, Haobo Fang, Yan Li, Wanqi Ma, Intelligent traffic accident detection system in complex dynamic scenarios based on the dual-stream spatiotemporal-fusion model, Engineering Applications of Artificial Intelligence, Volume 162, Part E, 2025, 112675, ISSN 0952-1976, https://doi.org/10.1016/j.engappai.2025.112675.
Sultani, W., Chen, C., & Shah, M. (2018). Real-world anomaly detection in surveillance videos. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) (pp. 6479–6488).
An, J., & Cho, S. (2015). Variational autoencoder based anomaly detection using reconstruction probability.
Nguyen, H. H., Nguyen, C. N., Dao, X. T., Duong, Q. T., Kim, D. P., & Pham, M. (2024). Variational autoencoder for anomaly detection: A comparative study. arXiv.
Iqbal, T., & Qureshi, S. (2022). Reconstruction probability-based anomaly detection using variational auto-encoders. International Journal of Computers and Applications, 45(3), 231–237. https://doi.org/10.1080/1206212X.2022.2143026
Aslam, N., & Kolekar, M. H. (2023). A-VAE: Attention based variational autoencoder for traffic video anomaly detection. In IEEE I2CT (pp. 1–7). https://doi.org/10.1109/I2CT57861.2023.10126296
Yu, W., & Huang, Q. (2022). A deep encoder-decoder network for anomaly detection in driving trajectory behavior under spatio-temporal context. International Journal of Applied Earth Observation and Geoinformation, 115, 103115. https://doi.org/10.1016/j.jag.2022.103115
Zhang, C., Wang, X., Zhang, J., et al. (2023). VESC: A new variational autoencoder based model for anomaly detection. International Journal of Machine Learning and Cybernetics, 14, 683–696. https://doi.org/10.1007/s13042-022-01657-w
Masci, J., Meier, U., Cireşan, D., & Schmidhuber, J. (2011). Stacked convolutional auto-encoders for hierarchical feature extraction. In Artificial Neural Networks and Machine Learning – ICANN 2011 (Vol. 6791). Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-21735-7_7
Yao, Y., Xu, M., Wang, Y., Crandall, D. J., & Atkins, E. M. (2019). Unsupervised traffic accident detection in first-person videos. In IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) (pp. 273–280). https://doi.org/10.1109/IROS40897.2019.8967556
First Author and Second Author. 2002. International Journal of Scientific Research in Science, Engineering and Technology. (Nov 2002), ISSN NO:XXXX-XXXX DOI:10.251XXXXX
Downloads
Published
Issue
Section
License
Copyright (c) 2026 International Journal of Scientific Research in Science and Technology
This work is licensed under a Creative Commons Attribution 4.0 International License.
https://creativecommons.org/licenses/by/4.0
