Object Detection and Tracking for Smart Surveillance with Automated Email Alerts

This paper is about a surveillance system that can detect humans and weapons in real time. It can also track objects. Send automated email alerts. The system uses the YOLOv8 object detection model and the DeepSORT tracking algorithm to make a security monitoring solution. The system can get input from three sources: live webcam feeds, prerecorded video files and CCTV camera streams. This makes it good for both testing and real-world use. The YOLOv8 model is used to detect humans and weapons in each video frame. The DeepSORT algorithm keeps track of the detected objects from one frame to the next. When a weapon is found the system takes a snapshot of the frame. Sends an email with the picture to the person who is supposed to get it. This way potential threats can be reported away and people do not have to watch the videos all the time. The system was. It worked well. It detected objects accurately tracked them reliably and sent alerts on time. The system is good for surveillance because it can detect, track and notify all in one. It worked well with kinds of input and kept up with real-time performance, which is what security environments need. This system can be used in places like airports, railway stations, schools, shopping malls, banks and government buildings where it's really important to detect threats early. There are still some challenges, like detecting weapons or weapons that are partly hidden and reducing false alarms.. The results show that this system is a good start for making advanced surveillance systems. In the end using YOLOv8, DeepSORT and automated email alerts is a way to detect threats in real time and respond to them. The system makes surveillance better by making it possible to monitor, track and notify people accurately and immediately which makes people safer and more secure, in todays world.

[1] S. P. Singh, S. Suryanarayana, P. Kumari, and V. Lakshmi, “Object Detection and Recognition Using Machine Learning Techniques from Images,” Department of Computer Science and Engineering, Birla Institute of Technology Mesra, Ranchi, India, 2022. [2] Joseph Redmon, S. Divvala, R. Girshick, and A. Farhadi, “You Only Look Once: Unified, Real-Time Object Detection,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016, pp. 779–788. [3] J. Redmon and A. Farhadi, “YOLO9000: Better, Faster, Stronger,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017, pp. 7263–7271. [4] J. Redmon and A. Farhadi, “YOLOv3: An Incremental Improvement,” arXiv:1804.02767, 2018. [5] A. Bochkovskiy, C.-Y. Wang, and H.-Y. M. Liao, “YOLOv4: Optimal Speed and Accuracy of Object Detection,” arXiv:2004.10934, 2020. [6] Ultralytics, “Ultralytics YOLOv8 Documentation,” 2023. Available: Ultralytics Documentation [7] N. Wojke, A. Bewley, and D. Paulus, “Simple Online and Realtime Tracking with a Deep Association Metric,” in Proceedings of the IEEE International Conference on Image Processing (ICIP), 2017, pp. 3645–3649. [8] A. Bewley, Z. Ge, L. Ott, F. Ramos, and B. Upcroft, “Simple Online and Realtime Tracking,” in Proceedings of the IEEE International Conference on Image Processing (ICIP), 2016, pp. 3464–3468. [9] R. Girshick, “Fast R-CNN,” in Proceedings of the IEEE International Conference on Computer Vision (ICCV), 2015, pp. 1440–1448. [10] R. Girshick, J. Donahue, T. Darrell, and J. Malik, “Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2014, pp. 580–587.