Detecting swimming pools from satellite imagery employing multiple modalities
Problem Statement:
The proliferation of residential swimming pools presents significant challenges for water resource management, urban planning, and public health monitoring. This project addresses the critical need for automated, large-scale pool detection by leveraging multiple satellite modalities. Students will design a system that fuses high-resolution optical and multispectral satellite imagery to identify pools across diverse environments. By accurately mapping this dispersed infrastructure, the developed solution will empower data-driven decision-making for sustainable water policy, equitable taxation, and climate resilience planning.
Task:
This assignment aims for:
- Development of an AI model that detects swimming pools based on satellite imagery with high precision and recall;
- The AI-based system should be robust enough to work reliably across diverse environments and landscapes.
Work:
10% Theory, 60% Modelling, Coding and Testing, 10% Evaluation and Validation, 20% Writing
Contact:
Andreas Kamilaris: a.kamilaris@utwente.nl