Georeferencing
Definition
Georeferencing assigns spatial coordinates to unreferenced images or drawings—historical maps, scanned plans, drone mosaics—so they align with real‑world locations. Control points tie pixels to known coordinates; transformations (affine, polynomial, projective, or rubber‑sheeting) correct distortions. Proper georeferencing preserves scale and orientation while documenting residual error and chosen CRS.
Application
Planners align legacy land‑use maps to modern parcels, archaeologists place historic charts over current terrain, and utilities register as‑builts to the grid. In disaster response, rapid georeferencing of field sketches enables immediate overlay with authoritative layers.
FAQ
How many control points are enough?
Minimums depend on the transform—three for affine—but more well‑distributed points improve accuracy. Evaluate RMS error and inspect local residuals, not just the global mean.
What if the source map has cartographic exaggerations?
Use higher‑order or piecewise transforms and record limitations in metadata. Sometimes you must digitize features approximately and flag areas of poor fit for cautious use.
How do we pick the right CRS?
Choose a projection appropriate to the map’s extent—local state plane/UTM for cities, equal‑area for regional land‑cover change—and declare it explicitly in the GeoTIFF or world file.
Can we georeference in 3D?
Yes, by registering imagery to a surface (DEM) or using photogrammetry to derive camera poses. This supports accurate draping over terrain and volumetric measurements.