Real-Time River Health Monitoring and Crowdsourced Solutions

Abstract

Urban rivers in India, such as the Mula-Mutha in Pune, are increasingly burdened by surface-level garbage pollution due to rapid urbanization and inadequate waste management. Our project presents an integrated, technology-driven solution to address this issue through real-time monitoring, detection, and community-driven action. We employ the YOLOv8 deep learning model for object detection to identify and localize surface garbage in real-time images captured from the Mula-Mutha River. The model is trained on a custom dataset to ensure high accuracy in detecting floating waste. These insights are then used to trigger cleanup actions facilitated through a decentralized, blockchain-based crowdfunding platform. The platform allows local citizens to contribute funds transparently and securely, while NGOs, schools, and volunteers are invited to participate in scheduled cleanliness drives. Additionally, a dedicated website has been developed to share updates, event schedules, impact statistics, and encourage greater community participation. This project demonstrates the power of combining artificial intelligence, blockchain technology, and civic engagement to tackle urban environmental challenges in a scalable and transparent manner.

Introduction

Urban rivers like Pune’s Mula-Mutha are crucial for ecology and human needs but face severe pollution, particularly from surface garbage such as plastics. Traditional cleanup efforts are manual and unsustainable, prompting the need for a smarter, scalable solution. The project proposes an end-to-end system that uses mobile phone cameras and advanced AI (YOLOv8) for real-time surface garbage detection. This data is linked to a blockchain-based crowdfunding platform that enables transparent, traceable donations for funding organized cleanup drives involving NGOs, schools, and volunteers. A supporting website displays detection data, event schedules, and fund usage, fostering community engagement and trust. The solution combines AI, mobile tech, and blockchain to bridge detection with action, aiming to be modular, scalable, and replicable across other polluted water bodies, exemplifying youth-led innovation in environmental tech.

Literature Review Highlights:

• Crowdfunding reduces spatial constraints but social connections still matter for funding success.

• YOLO-based models have been used effectively for water surface garbage detection and real-time marine animal monitoring, showing promising accuracy and efficiency improvements.

• Lightweight YOLO models enable deployment on devices like unmanned boats and mobile phones.

• YOLO’s evolution (from v3 to v12) shows trade-offs between accuracy, efficiency, and complexity.

• Real-time object detection frameworks using YOLO variants (v3, v7) are widely applied in diverse real-world scenarios.

• Success factors in crowdfunding include founder communication, website presence, and community engagement rather than just founder’s past experience.

• Integration of decentralized finance (DeFi) with systemic risk frameworks is crucial for financial stability, with implications for blockchain crowdfunding transparency and governance.

Conclusion

In conclusion, our project brings together advanced technologies and civic responsibility to address the growing problem of river pollution in urban areas, with a specific focus on the Mula-Mutha River. By utilizing a mobile phone camera to capture real-time images and employing the YOLOv8 deep learning model for accurate surface garbage detection, we have created a system that offers quick, scalable, and automated environmental monitoring. However, detection alone is not enough, which is why we integrated a blockchain-based crowdfunding platform that allows the local public to contribute transparently and securely to cleanup efforts. Through smart contracts and decentralized applications, this platform ensures trust and accountability while encouraging public participation. The funds collected are used to organize cleanliness drives involving schools, NGOs, and community volunteers, making the process inclusive and action-oriented. Our website acts as a bridge, displaying real-time data, campaign updates, and volunteer opportunities, thus closing the loop between detection, funding, and execution. This project is more than a technical solution—it is a replicable model for community-driven environmental restoration, demonstrating how emerging technologies can be harnessed for the greater good of society and nature.

References

[1] Ultralytics, “YOLOv8 Documentation,” 2023. Online. Available: https://docs.ultralytics.com [2] MongoDB Inc., “MongoDB Manual,” Online. Available: https://www.mongodb.com/docs/manual/ [3] J. Brownlee, Deep Learning for Computer Vision, Machine Learning Mastery, 2019. [4] Express.js, “Express – Node.js Web Application Framework,” Online. Available: https://expressjs.com/ [5] Node.js Foundation, “Node.js Documentation,” Online. Available: https://nodejs.org/en/docs/ [6] Bcrypt.js, “A library to help you hash passwords,” Online. Available: https://www.npmjs.com/package/bcrypt [7] OpenCV.org, “Open Source Computer Vision Library,” Online. Available: https://opencv.org/ [8] E. Bertolini and B. Schmitt, “Understanding Blockchain Technology,” IBM Developer, 2020. Online. Available: https://developer.ibm.com/articles/cl-blockchain-basics-intro-bluemix-trs/ [9] Mongoose, “Elegant MongoDB object modeling for Node.js,” Online. Available: https://mongoosejs.com/ [10] 101010 T. Ahmed, S. Begum, and R. Dey, “Garbage Detection and Classification Using Deep Learning,” International Journal of Computer Applications, vol. 176, no. 24, pp. 1–6, 2020. [11] Solidity Docs, “Solidity – Ethereum Smart Contract Language,” [Online]. Available: https://docs.soliditylang.org/ [12] TensorFlow, “TensorFlow: An Open Source Machine Learning Framework,” [Online]. Available: https://www.tensorflow.org/ [13] M. Z. A. Bhuiyan et al., “A Blockchain-Based Crowdsourcing Platform: Decentralized and Secure Payments for Smart Community Services,” IEEE Access, vol. 7, pp. 9674–9684, 2019. [Online]. Available: https://ieeexplore.ieee.org/document/8633911 [14] G. Wood, “Ethereum: A Secure Decentralised Generalised Transaction Ledger,” Ethereum Project Yellow Paper, 2014. [Online]. Available: https://ethereum.org/en/whitepaper/ [15] M. Yaseen, “What is YOLOv8: An In-Depth Exploration of the Internal Features of the Next-Generation Object Detector,” arXiv preprint, arXiv:2408.15857, 2024. [Online]. Available: https://arxiv.org/abs/2408.15857

Copyright

Copyright © 2025 Prof. Ujwala Wanaskar, Riddhi Khot, Yashi Solanki, Aakanksha Wadekar, Pradnya Nalawade. 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.