Deep Learning-Based Signal Processing Framework for Real-Time Anomaly Detection in CCTV Video Surveillance

Author(s)	Chirag Aggarwal, Navin Kumar Tyagi
Country	India
Abstract	Video surveillance systems generate vast amounts of video signals, making manual monitoring impractical for real-time threat detection8. This paper presents a deep learning-based anomaly detection system for CCTV surveillance, formulated as a signal processing problem over spatiotemporal video data9. The proposed method combines Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs) to extract spatial and temporal features from video signals within a weakly supervised multiple instance learning (MIL) framework10. We incorporate sparsity and temporal consistency constraints in the loss function to enhance anomaly localization accuracy and signal smoothness11. The system is evaluated on a large-scale CCTV dataset comprising 1,900 untrimmed videos spanning 13 anomaly categories, demonstrating superior accuracy and robustness compared to conventional baselines and attention-based approaches such as COVAD12. Our method achieves real-time inference capability while maintaining high detection accuracy, making it practical for real-world video surveillance applications13
Field	Engineering
Published In	Volume 7, Issue 6, November-December 2025
Published On	2025-12-27
DOI	https://doi.org/10.36948/ijfmr.2025.v07i06.64744

About IJFMR Fees & Payment Current Issue Publication Archive	Submit Research Paper Track Submission Status Publication Guidelines Publication Ethics Peer Review & Plagiarism	Join as a Reviewer Editors & Reviewers Reviewer Referral Program Get Reviewer Membership Certi.	Website/Journal Policies Usage Policy Content Policies Privacy Policy

Contact Us		+91-9687-828-838	editor@ijfmr.com

International Journal For Multidisciplinary Research