Остання редакція: 2026-06-07
Анотація
This paper examines methods for recognizing anomalous behavior in intelligent video analytics systems. The main approaches to detecting non-standard situations in video recordings are analyzed, including classical algorithmic methods and modern deep learning-based approaches. Special attention is given to convolutional neural networks, LSTM models, and transformer architectures, which enable automated detection of suspicious events in real time.
Ключові слова
Посилання
1. Kiran B. R. et al. Deep Learning Based Anomaly Detection in Video Surveillance: A Survey // IEEE Access. 2022. Vol. 10. P. 28509–28524.
2. Stauffer C., Grimson W. E. L. Adaptive Background Mixture Models for Real-Time Tracking // Proc. IEEE CVPR. Fort Collins, 1999. P. 246–252.
3. Dalal N., Triggs B. Histograms of Oriented Gradients for Human Detection // Proc. IEEE CVPR. San Diego, 2005. Vol. 1. P. 886–893.
4. Simonyan K., Zisserman A. Two-Stream Convolutional Networks for Action Recognition in Videos // Advances in Neural Information Processing Systems. 2014. Vol. 27. P. 568–576.
5. Luo W. et al. Video Anomaly Detection with Sparse Coding Inspired Deep Neural Networks // IEEE Transactions on Pattern Analysis and Machine Intelligence. 2021. Vol. 43. No. 3. P. 1070–1084.
6. Liu Z. et al. Video Swin Transformer // Proc. IEEE CVPR. 2022. P. 3202–3211.
7. Liu W. et al. Future Frame Prediction for Anomaly Detection – A New Baseline // Proc. IEEE CVPR. Salt Lake City, 2018. P. 6536–6545.