How DeepLearning, and semantic segmentation, can be an efficient way to detect and spot inconsistency in OSM dataset ?
Machine and DeepLearning can succeed to tackle some old issues, in a far more convenient and efficient way than ever before… For instance DeepLearning, with aerial imagery semantic segmentation can improve features detection ability and allow us to spot dataset inconsistencies.
In this presentation we will focus on how an OpenStreetMap subset dataset (for instance roads and buildings on an area), can be evaluated to produce quality metric.
We wanna focus on:
Deep Learning vision, and specific Satellite imagery considerations (high and low resolutions, multispectral dimensions, dataset aggregation…)
How to qualify a good enough labelled DataSet (to allow supervised learning)
FOSS4G (PostGIS and Grass) integration with Python ML/DL framework
Concrete solution for efficient treatments for wide coverages