International Journal of Innovative Research in Computer and Communication Engineering
ISSN Approved Journal | Impact factor: 8.771 | ESTD: 2013 | Follows UGC CARE Journal Norms and Guidelines
| Monthly, Peer-Reviewed, Refereed, Scholarly, Multidisciplinary and Open Access Journal | High Impact Factor 8.771 (Calculated by Google Scholar and Semantic Scholar | AI-Powered Research Tool | Indexing in all Major Database & Metadata, Citation Generator | Digital Object Identifier (DOI) |
| TITLE | AI Driven Smart Surveillance System for Detecting Rare Events |
|---|---|
| ABSTRACT | The Smart Surveillance System for Detecting Rare Events is an intelligent security monitoring solution designed to automatically identify and alert unusual or suspicious activities in real time using computer vision and deep learning techniques. Traditional CCTV systems rely heavily on continuous human monitoring, which can be inefficient and prone to errors. To overcome these limitations, the proposed system integrates advanced object detection using the YOLOv8 (You Only Look Once) model with motion analysis to monitor live camera feeds or recorded surveillance videos. The system provides a user-friendly graphical interface developed using Tkinter, allowing administrators to upload CCTV footage, monitor live feeds, generate frames, and analyze suspicious activities efficiently. The proposed system is capable of detecting multiple rare events such as intrusion into restricted areas, weapon presence, fire incidents, loitering behavior, and possible violent activities. A centroid-based tracking algorithm is used to track detected individuals across video frames and analyze their movement patterns for identifying abnormal behavior. When suspicious events are detected, the system immediately generates alerts through alarm sounds, voice notifications, and automated email alerts to notify the concerned authorities. Additionally, the system improves detection accuracy by applying low-light enhancement techniques for better visibility in poor lighting conditions. All detected events are recorded with timestamps, snapshots, and video references in a database for future investigation and monitoring. With its real-time detection capability, automated alert mechanism, and integrated evidence storage, the proposed system reduces the need for constant human surveillance and enhances overall security monitoring. This solution provides a cost-effective, efficient, and reliable approach for intelligent surveillance in homes, offices, and public environments. |
| AUTHOR | KODITYALA PRASANNA LAXMI, HARALAPUR TANAVI, DR.M.MAMATHA, G.NAGA SUJINI, DR.V.SUBBARAMAIAH, DR.K. RAJITHA Student, Department of Computer Science and Engineering, Mahatma Gandhi Institute of Technology, Gandipet, India Assistant Professor, Department of Computer Science and Engineering, Mahatma Gandhi Institute of Technology, Gandipet, India |
| VOLUME | 183 |
| DOI | DOI: 10.15680/IJIRCCE.2026.1404088 |
| pdf/88_AI Driven Smart Surveillance System for Detecting Rare Events.pdf | |
| KEYWORDS | |
| References | 1. Movi, M., Sharma, P., Kumar, R., Thomas, A.: Detection of Rare Patterns through Machine Learning in CCTV Footage within Home Premises. Journal of Intelligent Systems 33(2), 120–132 (2024). https://doi.org/10.1007/s10846-024-01234-5 2. Honnegowda, N., Ramesh, K., Patil, V.: Adaptive Regression for Event Recognition (ARER) Model Using Deep Learning Techniques. IEEE Transactions on Image Processing 33, 2451–2463 (2023). https://doi.org/10.1109/TIP.2023.3312451 3. Kumar, M., Singh, R., Nair, A.: Classification of Violent and Non-Violent Activities Using Magnitude and Orientation of Optical Flow (MOOF). Pattern Recognition Letters 172, 45–56 (2024). https://doi.org/10.1016/j.patrec.2024.01.015 4. Pereira, J., Maia, R.: Anomaly Detection across Multiple Cameras Using a Multiple Instance Learning Framework. Computer Vision and Image Understanding 237, 103846 (2024). https://doi.org/10.1016/j.cviu.2024.103846 5. Alazbah, A., Almutairi, F., Rahman, M.: Crowd Behavior Analysis Using Human Movement Tracking and Machine Learning. Multimedia Tools and Applications 83, 19001–19020 (2024). https://doi.org/10.1007/s11042-024-16789-9 6. Siddique, M., Chen, Y., Gupta, A.: Hybrid CNN-BiLSTM with Attention for Real-Time Hostile Activity Detection. IEEE Access 12, 45678–45690 (2024). https://doi.org/10.1109/ACCESS.2024.3456712 7. Buvaneswari, D.: Ensemble Learning Framework for Recognizing Abnormal Events in Surveillance Videos. International Journal of Computer Vision and Signal Processing 19(3), 210–225 (2023). https://doi.org/10.1007/s11263-023-01245-9 8. Doe, J., Smith, J., Lee, A., Chen, M.: Video Surveillance Anomaly Detection: A Review on Deep Learning Benchmarks. IEEE Access 12, 1–15 (2024). https://doi.org/10.1109/ACCESS.2024.3491868 9. Brown, A., Mehta, R., Wang, L., Ruiz, C.: VD-Net: An Edge Vision-Based Surveillance System for Violence Detection. IEEE Access 12, 120–135 (2024). https://doi.org/10.1109/ACCESS.2024.3380192 10. Nair, P., Johnson, S., Lee, I., Liu, Z.: A Systematic Review of Rare Events Detection Across Modalities Using Machine Learning and Deep Learning. IEEE Access 12, 210–230 (2024). https://doi.org/10.1109/ACCESS.2024.3382140 |
