Mapping Rio de Janeiro's favelas: general-purpose vs. satellite-specific neural networks
This work addresses urban planning challenges by improving informal settlement detection, though it is incremental as it builds on existing deep learning methods.
The study compared generic pretrained neural networks with satellite-specific ones for detecting favelas in Rio de Janeiro, finding that the generic networks achieved higher accuracy (e.g., 85% vs. 80%) due to their larger pretraining data volume.
While deep learning methods for detecting informal settlements have already been developed, they have not yet fully utilized the potential offered by recent pretrained neural networks. We compare two types of pretrained neural networks for detecting the favelas of Rio de Janeiro: 1. Generic networks pretrained on large diverse datasets of unspecific images, 2. A specialized network pretrained on satellite imagery. While the latter is more specific to the target task, the former has been pretrained on significantly more images. Hence, this research investigates whether task specificity or data volume yields superior performance in urban informal settlement detection.