This is a more of a noob GIS question. How are all the features on the maps we come use know and love (Google Maps, Bing Maps) generated from imagery? Does somebody actually look at the imagery and draw feature layers on top using something like ArcGIS or is there algorithms for automatic extraction?

  • multiple sources, data suppliers (tomtom, navteq) streetview is a large source for google and some data extraction from aerial imagery and automatically with lidar) – Mapperz Dec 9 '15 at 18:34
  • I think it would be better to focus your question on a particular dataset because there are multiple ways that spatial data is captured. – PolyGeo Dec 9 '15 at 19:39

Generally speaking, for vector geometry (e.g. roads, buildings, polygons, areas of interest) these features are created by outside contractors that specialize in geospatial data production. However, the tech companies like Google/Apple/Microsoft must also QA/QC the massive datasets for inaccuracies. A lot of this can be done programmatically by their engineers, but other features must be edited manually (e.g. to correct inaccurate features). As we saw with the rollout of Apple Maps, sometimes this process can be cumbersome.

The best mapping platforms use "ground-truthing" to improve their datasets beyond just what exists in vendor datasets, which may have incomplete or inaccurate features. Ground-truthing consists of sending on the ground teachnicians to map roadways and high-profile landmarks. A similar process exists for raster imagery as well.

Not the answer you're looking for? Browse other questions tagged or ask your own question.