
AI Powered Satellite System Reveals Stark Green Space Inequality in Karachi
A research team led by NYU Associate Professor Rumi Chunara has developed an advanced artificial intelligence (AI) system that accurately tracks urban green spaces using satellite imagery offering a crucial tool for healthier city planning. The system significantly outperforms traditional methods, which have historically failed to capture up to 37% of urban vegetation.
Breaking New Ground in Urban Mapping
To validate their approach, researchers tested the system in Karachi, Pakistan’s largest city, known for its mix of dense urban sprawl and highly varied vegetation. The AI, built on enhanced DeepLabV3+ segmentation models, was trained using a technique called ‘green augmentation,’ which adjusts training data to account for different lighting and seasonal conditions. This innovation ...