Skip to content

PlasticWatch, ML based web platform for detecting plastic pollution in water bodies using YOLOv8 deep learning model with real-time monitoring, GPS tracking, and interactive dashboards.

License

Notifications You must be signed in to change notification settings

sathish-k7/PlasticWatch-ML-based-Water-Monitoring-System

Repository files navigation

PlasticWatch - AI-Powered Water Pollution Monitoring

An AI-powered web platform for detecting plastic pollution in water bodies using YOLOv8 deep learning model with real-time monitoring, GPS tracking, and interactive dashboards.

Features

  • AI Detection - Uses YOLOv8m deep learning model to automatically identify plastic pollution in water with 92% accuracy
  • Multiple Sources - Upload individual images, batch process multiple files, or use live webcam for real-time monitoring
  • Interactive Map - GPS-enabled map visualization showing pollution hotspots across Chennai water bodies with clickable markers
  • Trend Analysis - Dynamic charts displaying detection patterns, pollution levels, and trends over time periods
  • Export Reports - Generate detailed CSV data files and formatted PDF reports for sharing and documentation
  • Professional UI - Clean, responsive web interface with gradient themes, animations, and intuitive navigation

Quick Start

Prerequisites

  • Python 3.13.7
  • Virtual environment support

Quick Start

  1. Install dependencies
pip install -r requirements.txt
  1. Run the application
python app.py
  1. Open browser
http://localhost:5001

🤖 Model Details

  • Model: YOLOv8m custom trained
  • Dataset: 70 augmented training images
  • Performance: 92% confidence detection
  • Classes: Plastic pollution detection

License

This project is licensed under the MIT License.

About

PlasticWatch, ML based web platform for detecting plastic pollution in water bodies using YOLOv8 deep learning model with real-time monitoring, GPS tracking, and interactive dashboards.

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published