Researchers from New York University (NYU) are employing the use of satellite imagery to identify urban green spaces.

The proposed data augmentation paradigm uses a technique known as ‘green augmentation’ to better improve the detection of varying vegetation through the use of Google Earth satellite imagery.

Learning models highlighted in the paper provide a more in-depth analysis of green spaces, taking into account vegetation, object shapes, textures, and colors to name a few.

The paper underscores the role in which green spaces play in tackling the climate crisis, noting that recently “shrinkage” of green spaces as a result of population growth, industrial expansion, developmental activities, and land encroachment has impacted “many” urban environments.

Testing for the new system took place across Karachi – the largest city in Pakistan – it was found that on average there is 4.17 sqm of green space per person.

This is less than half the World Health Organisation’s recommended minimum of nine sqm.

The full paper can be found here.