Green Guard: AI Camera for Detection and Evidence Collection of Illegal Waste Dumping

Abstract

Illegal waste dumping is a significant environmental issue affecting urban cleanliness and public health. This paper presents Green Guard, an AI-based surveillance system designed to detect illegal dumping activities and automatically collect visual evidence. The system utilizes YOLOv8 for real-time object detection and integrates tracking and event-based validation to identify dumping behavior. The proposed framework captures time-stamped evidence and provides a dashboard for monitoring. Experimental results demonstrate strong detection performance and reliability, making the system suitable for smart city applications.

Introduction

The text presents an AI-based system called Green Guard designed to detect and document illegal waste dumping using computer vision and deep learning.

Illegal dumping causes environmental and health hazards, while traditional monitoring methods like manual inspection and CCTV are inefficient. To address this, the proposed system uses YOLOv8-based object detection, tracking, and event analysis to automatically identify dumping incidents in real time.

The system works by processing video frames to detect people and waste objects, assigning unique IDs for tracking, and calculating distances between them. An illegal dumping event is confirmed when waste is left behind after a person moves away. Once detected, the system automatically captures and stores timestamped evidence images.

Key features include real-time detection, person tracking, event-based logic, evidence collection, a dashboard interface, and scalability for multi-location deployment. The methodology involves dataset preparation with augmentation, frame-by-frame detection, tracking, distance computation, and rule-based event validation.

Performance results show good effectiveness, with strong precision (1.0), recall (0.86), F1 score (0.75), and mAP@0.5 of 0.762. The system achieves optimal performance at a confidence threshold of around 0.317.

Conclusion

This paper presented Green Guard, an AI-powered system for detecting illegal waste dumping and collecting evidence. The integration of deep learning and event-based logic enables accurate and reliable detection. The system is scalable and suitable for real-world deployment in smart city environments.

References

[1] S. Majchrowska et al., ”Deep learning-based waste detection in natural and urban environments,” Waste Management, vol. 1, 2022. [2] L. Chen and J. Zhu, ”Water surface garbage detection based on lightweight YOLOv5,” Scientific Reports, vol. 14, no. 6133, pp. 1–15, Mar. 2024, doi: 10.1038/s41598-024-55051-3. [3] P. Dubey et al., ”Enhanced autonomous water garbage collection system using deep learning-based object detection and path planning,” Bulletin of Electrical Engineering and Informatics, vol. 14, no. 4, pp. 2593–2602, Aug. 2025, doi: 10.11591/eei.v14i4.9399. [4] E. Alharbi, G. Alsulami, S. Aljohani, W. Alharbi, and S. Albaradei, ”Real-time detection and monitoring of public littering behavior using deep learning for a sustainable environment,” Scientific Reports, vol. 15, no. 3000, pp. 1–11, Jan. 2025, doi: 10.1038/s41598-024-77118-x. [5] I. Shukhratov et al., ”Optical detection of plastic waste through computer vision,” Intelligent Systems with Applications, vol. 22, Art. no. 200341, 2024, doi: 10.1016/j.iswa.2024.200341. [6] S. Shahab and M. Anjum, ”Solid Waste Management Scenario in India and Illegal Dump Detection Using Deep Learning: An AI Approach towards Sustainable Waste Management,” Sustainability, vol. 14, no. 15896, 2022, doi: 10.3390/su142315896. [7] J. S. Walia, K. Haridass, and L. K. Pavithra, ”Deep Learning Innovations for Underwater Waste Detection: An In-Depth Analysis,” IEEE Access, vol. 13, pp. 88917–88929, May 2025, doi: 10.1109/ACCESS.2025.3569344. [8] S. M. Usha and H. B. Mahesh, ”Accurate and High Speed Garbage Detection and Collection Technique using Neural Network and Machine Learning,” IOP Conf. Series: Materials Science and Engineering, vol. 1258, no. 1, p. 012055, 2022. [9] M. B. Ilyas et al., ”Deep learning-based waste classification and detection for a sustainable environment,” Waste Management, vol. 135, pp. 318–329, 2021, doi: 10.1016/j.wasman.2021.08.014. [10] D. Bhatnagar et al., ”BEHAVE: Dataset and method for tracking human object interactions,” in Proc. IEEE/CVF Conf. Computer Vision and Pattern Recognition (CVPR), New Orleans, LA, USA, Jun. 2022, pp. 15935–15946. [11] A. M. Shaikh, M. V. Kadam, and S. D. Shinde, ”Deep Learning Approach for Vehicle Number Plate Recognition System with Image Enhancement Technique,” International Journal of Research in Engineering, Science and Management, vol. 6, no. 6, pp. 17–21, Jun. 2023. [12] L. Du et al., ”Assessing and predicting the illegal dumping risks in relation to road characteristics,” Waste Management, vol. 169, pp. 332– 341, Jul. 2023, doi: 10.1016/j.wasman.2023.07.031. [13] M. H. Dipo et al., ”Real-Time Waste Detection and Classification Using YOLOv12-Based Deep Learning Model,” Digital, 2025. [14] S. Shinde, A. Kothari, and V. Gupta, ”YOLO based Human Action Recognition and Localization,” Procedia Computer Science, vol. 133, pp. 831–838, 2018, doi: 10.1016/j.procs.2018.07.112. [15] P. Zhou et al., ”Uncollected Solid Waste Detection and Reporting Using Machine-learning and Geotagging,” in European Conference on Innovation and Entrepreneurship, vol. 19, no. 1, p. 2581, 2024, doi: 10.34190/ecie.19.1.2581. [16] Y. Yun, C. Park, J. Lee, and J. Kim, ”Vision-based garbage dumping action detection for real-world surveillance platform,” ETRI Journal, vol. 41, no. 5, pp. 616–627, Oct. 2019, doi: 10.4218/etrij.2019-0186.

Copyright

Copyright © 2026 Amritalakshmi M, Hridya Raju George, Jayalakshmi M, Joseph Mathew, Dr. Deepa S Kumar, Jisha A K. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.