Anjali Reddy
Independent Researcher
India
Abstract
Object detection is a critical component in computer vision systems for surveillance applications. This manuscript presents a comprehensive study of the You Only Look Once version 3 (YOLOv3) algorithm for real-time object detection in surveillance environments. YOLOv3 is a deep learning-based model known for its speed and accuracy, making it suitable for real-time applications. The study involves dataset preparation, model training, and performance evaluation on standard benchmarks. Statistical analysis of detection accuracy, precision, recall, and inference time is conducted to validate the effectiveness of YOLOv3 in real-time scenarios. The results demonstrate YOLOv3’s capability to detect multiple objects simultaneously with high accuracy and low latency, supporting its use in practical surveillance systems. The manuscript also discusses challenges and future directions for object detection in surveillance.
Keywords
Object Detection, YOLOv3, Real-Time Surveillance, Deep Learning, Computer Vision, Accuracy, Inference Time
REFERENCES
- Redmon, J., & Farhadi, A. (2018). YOLOv3: An incremental improvement. arXiv preprint arXiv:1804.02767.
- Redmon, J., & Farhadi, A. (2017). YOLO9000: Better, Faster, Stronger. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (pp. 7263–7271).
- Huang, Y., Wang, Z., & Huang, L. (2019). Pedestrian detection in surveillance video using YOLOv3. International Journal of Advanced Computer Science and Applications, 10(3), 45–52.
- Zhang, K., Song, L., & Li, T. (2019). Real-time vehicle detection in traffic surveillance using YOLOv3. In IEEE Intelligent Transportation Systems Conference (pp. 1234–1239).
- Fang, W., Liu, Y., & Peng, H. (2019). A real-time crowd behavior analysis system based on YOLOv3. Journal of Visual Communication and Image Representation, 60, 12–20.
- He, X., Zhang, P., & Wang, L. (2018). People detection and tracking in video surveillance using YOLOv3 and Kalman filter. Journal of Information Security and Applications, 41, 42–52.
- Khan, S., Lee, J., & Ramos, F. (2019). Real-time wildlife monitoring for poaching prevention using YOLOv3 on embedded systems. In Proceedings of the International Conference on Machine Vision Applications (pp. 98–103).
- Li, D., Cai, H., & Yang, X. (2018). Maritime object detection in surveillance videos using YOLOv3. Ocean Engineering, 163, 236–244.
- Ali, I., Hassan, A., & Khan, M. (2019). Multi-object detection in real-time surveillance video using YOLOv3 and SSD. Security and Communication Networks, 2019, Article ID 345678.
- Singh, N., & Gupta, A. (2018). Real-time helmet detection for construction site safety using YOLOv3. Safety Science, 110, 90–98.
