Copyright © 2025 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{179782,
        author = {N. SRUTHILAYA and M. THAARANI},
        title = {VIDEO ANOMALY DETECTION USING MACHINE LEARNING},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {11},
        number = {12},
        pages = {8467-8471},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=179782},
        abstract = {Video anomaly detection (VAD) is a critical 
task in intelligent surveillance, industrial automation, 
and autonomous systems, where timely identification 
of unusual events in video streams can significantly 
enhance operational safety and efficiency. Traditional 
anomaly detection methods often fail to capture com
plex spatiotemporal patterns due to their limited ca
pacity for contextual understanding. This study pro
poses a machine learning-based approach that employs 
Convolutional Neural Networks (CNNs) for spatial 
feature extraction and Recurrent Neural Networks 
(RNNs), specifically Long Short-Term Memory 
(LSTM) networks, for modeling temporal dependen
cies in video sequences. By leveraging unsupervised 
learning on extensive datasets representing normal 
behaviors, the system can detect deviations indicative 
of anomalous events, such as intrusions, abnormal 
motion, and irregular object interactions. The integra
tion of attention mechanisms further refines the mod
el’s focus on critical video regions, enhancing accuracy 
while minimizing false detections. Experimental results 
demonstrate the system's robust performance in real
time scenarios, making it suitable for deployment in 
dynamic environments. This research underlines the 
potential of deep learning in developing scalable and 
accurate video anomaly detection systems without 
reliance on extensive manual labeling.},
        keywords = {Video Anomaly Detection, Machine Learn ing, Deep Learning, CNN, RNN, LSTM, Unsupervised  Learning, Attention Mechanism, Surveillance, Real Time Detection},
        month = {May},
        }