Oceanic Plastic Waste Detection with AI and Remote Sensing

• Understand satellite-based detection of ocean plastic.
• Apply AI models for debris identification and segmentation.
• Develop predictive models for plastic drift tracking.
• Explore optimization techniques for cleanup planning.
• Gain hands-on implementation experience.

🛰️ Day 1 — Detection from Orbit

• Floating Plastic Index (FPI) using Sentinel-2 SWIR bands
• Multi-scale detection with satellite and UAV imagery (Super-Resolution GANs)
• Debris segmentation using U-Net and DeepLabV3+
• Hands-on: Build a spectral classifier for plastic identification

🛰️ Day 2 — Movement & Drift Prediction

• Ocean transport modeling: Eulerian vs. Lagrangian approaches
• RNN-based drift forecasting for marine debris
• ConvLSTM heatmap prediction for debris accumulation zones
• Hands-on: Develop a 72-hour plastic trajectory predictor

🛰️ Day 3 — Cleanup Strategy Optimization

• Reinforcement learning for autonomous vessel routing
• YOLO-based waste classification
• Drone swarm coordination for debris scouting
• Hands-on: Implement an AI-based cleanup path optimizer

• Analyze satellite data for marine plastic detection.
• Develop AI models for debris identification and drift prediction.
• Apply optimization methods for cleanup planning.
• Implement practical AI workflows for ocean monitoring.

• Postgraduate students, Ph.D. scholars, researchers, and academicians in relevant fields
• Industry professionals working in sustainability, climate tech, or geospatial analytics
• Basic knowledge of Python and machine learning is recommended
• Familiarity with remote sensing or satellite data is beneficial but not mandatory