Abnormal Behavior Detection in Surveillance Systems Using a Hybrid EfficientNet-Transformer Model

  • Hesham A. Alberry Department of Computer Science, Faculty of Computers & Artificial Intelligence, Benha University
  • M. E. Khalifa Basic Science Department, Faculty of Computer & Information Sciences, Ain Shams University, Cairo, Egypt
  • Ahmed Taha Department of Computer Science, Faculty of Computers \& Artificial Intelligence, Benha University, Cairo, Egypt
DOI: https://doi.org/10.19139/soic-2310-5070-2259
Keywords: Anomaly detection Deep learning Unsupervised learning Transformers

Abstract

Anomaly detection in video surveillance is vital for public safety, but challenges arise from the unpredictability of abnormal behaviors and large-scale systems. We propose a hybrid architecture combining EfficientNetV2S for efficient feature extraction with a transformer encoder to capture long-range dependencies through self-attention. This model robustly detects abnormal events by modeling local and global patterns in video frames. Evaluated on UCSD Ped1, UCSD Ped2, and Avenue datasets, our approach achieved accuracies of 99.51, 99.80, and 94.82, outperforming existing methods and proving their suitability for real-time smart surveillance applications.
Published
2025-01-09
How to Cite
Alberry, H. A., Khalifa, M. E., & Taha, A. (2025). Abnormal Behavior Detection in Surveillance Systems Using a Hybrid EfficientNet-Transformer Model. Statistics, Optimization & Information Computing, 13(4), 1610-1622. https://doi.org/10.19139/soic-2310-5070-2259
Issue
Vol 13 No 4 (2025)
Section
Research Articles
Authors who publish with this journal agree to the following terms:

    1. Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
    2. Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
    3. Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).