AI-Powered Real-Time Surveillance: An Intelligent Threat Detection System for Security Enhancement
Main Article Content
Abstract
Intelligent real-time surveillance systems are upending the status quo in high-risk and complicated contexts due to the growing need for advanced solutions to improve public safety. But their performance is frequently hindered by insufficient training data diversity, class imbalance, and limited generalization due to the varying environmental conditions. The YOLOv8 object detection model is the foundation of this study's AI-powered surveillance framework, which integrates hybrid dataset construction with a data preparation workflow that includes image preprocessing, data augmentation, dataset splitting, and SMOTE-based dataset balancing applied only to the training subset prior to YOLOv8 training. To increase environmental variety and model generalization, a hybrid dataset was created by merging the Microsoft COCO dataset and the CCTV Surveillance Image Dataset. Additionally, in order to enhance the representation of minority object categories prior to YOLOv8 training, SMOTE-based dataset balancing was only applied to the training subset during data preparation. The proposed approach was evaluated using precision, recall, F1-score, [email protected], [email protected]:0.95, AUC-ROC, confusion matrix analysis, and execution time. In the selected experimental configuration, precision increased from 0.87 to 0.94, recall from 0.84 to 0.96, and F1-score from 0.85 to 0.95. Additionally, the average execution time dropped from 0.55 to 0.35 s/image, and the suggested framework produced a [email protected] of 0.94, a [email protected]:0.95 of 0.71, and an AUC-ROC of 0.99. In conclusion, the recommended framework shows that it will open the door to intelligent real-time surveillance in general-purpose ad hoc real-world situations and applications while simultaneously improving detection performance and maintaining computing economy.
Article Details
Section
This work is licensed under a Creative Commons Attribution 4.0 International License.
How to Cite
References
[1] O. El Ghati, O. Alaoui-Fdili, O. Chahbouni, N. Alioua, and W. Bouarifi, “Artificial intelligence-powered visual Internet of Things in smart cities: A comprehensive review,” Sustainable Computing: Informatics and Systems, vol. 43, p. 101004, 2024. https://doi.org/10.1016/j.suscom.2024.101004
[2] J. Liang, “A review of the development of YOLO object detection algorithm,” Applied and Computational Engineering, vol. 71, pp. 39-46, 2024. https://doi.org/10.54254/2755-2721/71/20241642
[3] J. Terven and D. Cordova-Esparza, “A comprehensive review of YOLO: From YOLOv1 to YOLOv8 and beyond,” Machine Learning and Knowledge Extraction, vol. 5, no. 4, pp. 1680-1716, 2023. https://doi.org/10.3390/make5040083
[4] Z. Jingchun, E. S. Goh, and M. S. Sunar, “Low-light image enhancement: A comprehensive review on methods, datasets and evaluation metrics,” Journal of King Saud University – Computer and Information Sciences, vol. 36, no. 10, p. 102234, 2024. https://doi.org/10.1016/j.jksuci.2024.102234
[5] Y. Gong, Y. Hou, C. Zhang, and M. Jiang, “Beyond augmentation: Empowering model robustness under extreme capture environments,” in Proceedings of the International Joint Conference on Neural Networks (IJCNN 2024), 2024. https://doi.org/10.1109/IJCNN60899.2024.10650150
[6] W. Chen, K. Yang, Z. Yu, Y. Shi, and C. L. P. Chen, “A survey on imbalanced learning: Latest research, applications and future directions,” Artificial Intelligence Review, vol. 57, p. 137, 2024. https://doi.org/10.1007/s10462-024-10759-6
[7] P. Yadav, N. Gupta, and P. K. Sharma, “WeaponVision AI: A software for strengthening surveillance through deep learning in real-time automated weapon detection,” International Journal of Information Technology, 2025. https://doi.org/10.1007/s41870-024-02375-y
[8] I. Vijoli and C.-C. Vancea, “Towards improving real-time surveillance systems with enhancement techniques for deep learning models,” in 2024 IEEE 20th International Conference on Intelligent Computer Communication and Processing (ICCP), pp. 1-8, 2024. https://doi.org/10.1109/ICCP63557.2024.10793047
[9] Y. Sun, Z. Sun, and W. Chen, “The evolution of object detection methods,” Engineering Applications of Artificial Intelligence, vol. 133, p. 108458, 2024. https://doi.org/10.1016/j.engappai.2024.108458
[10] M. S. K. Namana and B. U. Kumar, “An Efficient and Robust Night-Time Surveillance Object Detection System Using YOLOv8 and High-Performance Computing,” International Journal of Safety and Security Engineering, vol. 14, no. 6, pp. 1763-1773, 2024. https://doi.org/10.18280/ijsse.140611
[11] J. M. A. Dass, A. Suresh, K. Swathi, P. Shalini, and N. Y. Sinha, “Enhancing public safety through real-time weapon detection: A deep learning approach,” in 2024 3rd International Conference for Advancement in Technology (ICONAT), pp. 1-4, 2024. https://doi.org/10.1109/ICONAT61936.2024.10775051
[12] M. Deshpande, A. Agarwal, and R. Kamathe, “Development of an energy-efficient CCTV camera system for real-time human detection using YOLOv8 model,” Eastern-European Journal of Enterprise Technologies, vol. 6, no. 9(132), pp. 47-59, 2024. https://doi.org/10.15587/1729-4061.2024.310401
[13] V. Dubey, B. Kaur, and P. Goel, “Enhancing accuracy through data augmentation using variational autoencoders in machine learning techniques,” in 2024 Control Instrumentation System Conference (CISCON), pp. 1-6, 2024.
[14] J. H. Joloudari, A. Marefat, M. A. Nematollahi, S. S. Oyelere, and S. Hussain, “Effective class-imbalance learning based on SMOTE and convolutional neural networks,” Applied Sciences, vol. 13, no. 6, p. 4006, 2023. https://doi.org/10.3390/app13064006
[15] N. V. Chereddy and B. K. Bolla, “Evaluating the utility of GAN-generated synthetic tabular data for class balancing and low-resource settings,” in Multi-disciplinary Trends in Artificial Intelligence, Lecture Notes in Artificial Intelligence, vol. 14078, pp. 48-59, Springer, 2023.
[16] F. Luo, J. Ma, and G. T. S. Ho, “An instance-level data balancing method for object detection via contextual information alignment,” Image and Vision Computing, vol. 149, p. 105155, 2024. https://doi.org/10.1016/j.imavis.2024.105155
[17] S. U. A. Shovo et al., “Advancing low-light object detection with YOLO models: An empirical study and performance evaluation,” Cognitive Computation and Systems, 2024. https://doi.org/10.1049/ccs2.12114
[18] R. W. Liu, C. Han, and Y. Huang, “Low-light visibility enhancement for improving visual surveillance in intelligent transportation systems,” IET Intelligent Transport Systems, 2024. https://doi.org/10.1049/itr2.12534
[19] P. Shrivastav and A. K. T. Andu, “Developments in Deep Learning for Low Light Object Detection: A Comprehensive Review of YOLO and YOLOv8,” Journal of Information Systems Engineering and Management, vol. 10, no. 46s, 2025. https://doi.org/10.52783/jisem.v10i46s.9003
[20] P. Mittal, “A comprehensive survey of deep learning-based lightweight object detection models for edge devices,” Artificial Intelligence Review, vol. 57, p. 242, 2024. https://doi.org/10.1007/s10462-024-10877-1
[21] A. Tharwat, “Classification assessment methods,” Applied Computing and Informatics, vol. 17, no. 1, pp. 168-192, 2021. https://doi.org/10.1016/j.aci.2018.08.003