Debris Deputy: A Real-Time AI Surveillance with Database State Tracking System

The Debris Deputy system demonstrates the application of artificial intelligence and computer vision for real-time monitoring of suspected disposal events. The proposed system combines YOLOv8n object detection, ByteTrack multi-object tracking, person-object association, temporal event analysis, evidence generation, duplicate verification, and MongoDB-based incident state tracking into a unified surveillance framework. Instead of relying only on the presence of a candidate object in an individual frame, the system considers the interaction between a person and an object across consecutive frames to identify a suspected disposal event. The implemented system successfully generates evidence snapshots and maintains incident information including the detected object, event confidence, tracking IDs, timestamp, and incident status. The integration of Telegram-based human verification provides an additional review stage, while the web-based interface enables centralized monitoring of live video, records, notifications, and system status. The results demonstrate the practical integration of real-time computer vision with database and notification mechanisms for waste-disposal monitoring. Although the system identifies suspected events rather than determining disposal intent with certainty, it provides a structured foundation for automated surveillance and incident management.

1.     B. G. Cabungcal, N. D. Hidalgo, A. V. U. Lorzano, P. Ona, K. P. Ferraro and A. Santos, "Detection of Violators of Garbage Disposal Schedule with Monitoring System," TENCON 2024 - 2024 IEEE Region 10 Conference (TENCON), Singapore, Singapore, 2024, pp. 1602-1605, doi: 10.1109/TENCON61640.2024.10903035. 2.    A. Kumar et al., "Vision Based Littering Detection and Face Tagging using multi-model Architecture," 2024 4th International Conference on Advance Computing and Innovative Technologies in Engineering (ICACITE), Greater Noida, India, 2024, pp. 1795-1800, doi: 10.1109/ICACITE60783.2024.10616763. 3.   P. Wibowo, A. Z. Wajdi, T. H. Muliawati, A. Fariza and Y. Kiyoki, "Illegal Waste Dump Detection Using a Deep Learning Approach: Performance Evaluation of YOLO Architectures," 2025 International Electronics Symposium (IES), Surabaya, Indonesia, 2025, pp. 802-807, doi: 10.1109/IES67184.2025.11161008. 4.   Rania and D. Kumar, "Smart Garbage Detection System for Sustainable Waste Management using Deep Learning Techniques," 2024 IEEE International Conference for Women in Innovation, Technology & Entrepreneurship (ICWITE), Bangalore, India, 2024, pp. 253-257,doi: 10.1109/ICWITE59797.2024.10503301. 5.   Kim, Y.; Cho, J. AIDM-Strat: Augmented Illegal Dumping Monitoring Strategy through Deep Neural Network-Based Spatial Separation Attention of Garbage. Sensors 2022, 22, 8819. https://doi.org/10.3390/s22228819. 6.    Y. Zhang et al., “ByteTrack: Multi-object tracking by associating every detection box,” in Computer Vision – ECCV 2022, S. Avidan, G. Brostow, M. Cissé, G. M. Farinella, and T. Hassner, Eds., Lecture Notes in Computer Science, vol. 13682. Cham, Switzerland: Springer, 2022, pp. 1–21. doi: 10.1007/978-3-031-20047-2_1. 7.   J. Redmon, S. Divvala, R. Girshick, and A. Farhadi, “You only look once: Unified, real-time object detection,” in Proc. IEEE Conf. Computer Vision and Pattern Recognition (CVPR), 2016, pp. 779–788. 8.   E. Cook, K. Ionkova, P. Bhada-Tata, S. Yadav, and F. Van Woerden, What a Waste 3.0: Global Snapshot of Solid Waste Management Toward Circularity until 2050. Washington, DC, USA: World Bank, 2026.