Monika Kuffer: AI helping track development of informal settlements

Informal settlements have distinctive development features that can be accurately identified using satellite data and artificial intelligence, facilitating more effective urban planning, Monika Kuffer, an expert in mapping disadvantaged urban areas and professor at the IDEAMAPS network, said at WUF13 in Baku, Report informs.

According to her, informal settlements in different cities around the world, despite their differences, often share similar characteristics-high building density, small buildings located close to each other, and an irregular urban structure-which makes them easily visible in satellite imagery.

"As part of an international project supported by the European Space Agency (ESA), an approach was developed for creating balanced datasets for training artificial intelligence models. These models make it possible to determine not only the location but also the development dynamics of such settlements using medium-resolution satellite imagery," she said.

Kuffer emphasized that data quality is crucial, as it determines the effectiveness of AI models, as well as the need for continuous verification of results at the local level.

She noted that the developed approach involves not only creating models but also jointly verifying them with local communities, which improves data accuracy and takes into account real-world changes.