Georeferencing
The process of aligning spatial data (images, maps, or datasets) to a known coordinate reference system so it can be accurately positioned on the Earth's surface. Uses ground control points to establish the relationship between image pixels and real-world coordinates.
Overview
Georeferencing is the process of aligning spatial data — typically a scanned map, aerial photograph, or unreferenced image — to a known coordinate reference system so that each pixel or point corresponds to a real-world geographic location. This is accomplished by identifying Ground Control Points (GCPs): identifiable features (road intersections, building corners, survey markers) whose coordinates are known, then computing a mathematical transformation that maps image coordinates to geographic coordinates.
How It Works
The analyst identifies matching points between the unreferenced image and a referenced source (orthoimage, GPS survey, or existing map). A minimum of three GCPs is needed for an affine transformation (which handles translation, rotation, scaling, and skewing). More GCPs enable higher-order polynomial transformations that can correct for lens distortion and non-linear warping. A rubber-sheet (spline) transformation forces exact alignment at each GCP but can introduce distortion between points.
The Root Mean Square Error (RMSE) of the transformation quantifies accuracy — it measures how far GCPs deviate from their predicted positions after the transformation is applied. Lower RMSE indicates a better fit. After transformation, the image is resampled onto the target coordinate grid using nearest-neighbor, bilinear, or cubic interpolation.
Key Facts
- Minimum 3 GCPs for affine transformation; more GCPs enable polynomial and spline corrections.
- RMSE measures transformation accuracy — the residual positional error at GCPs.
- Georeferencing does NOT correct for terrain relief — that requires orthorectification.
- Common in QGIS (Sketcher plugin), ArcGIS (Georeferencing toolbar), and GDAL (gdal_translate).
Applications
Historical Map Digitization
Aligning scanned historical maps to modern coordinate systems enables comparison of past and present landscapes — urban growth, coastline change, deforestation.
Drone Imagery Integration
Georeferencing raw drone photographs allows them to be overlaid with satellite imagery and GIS layers for site analysis.
Field Data Alignment
Aligning hand-drawn field sketches, floor plans, or site photographs to coordinate systems for GIS integration.
Limitations & Considerations
Georeferencing with affine or polynomial transformations does not correct for terrain-induced distortion — features at different elevations will still be displaced. This is the key difference from orthorectification, which uses a DEM to correct relief displacement. Accuracy depends on GCP quality and distribution — poor GCP placement (clustered in one area) leads to unreliable transformation elsewhere. Historical maps may have inherent cartographic errors that cannot be fully corrected. Higher-order transformations require more GCPs and can introduce unrealistic warping if points are not well distributed.
History & Background
Georeferencing has been practiced since the earliest days of GIS in the 1970s, when aerial photographs needed to be registered to map coordinate systems. The process was originally performed using analog instruments (zoom transfer scopes) before transitioning to digital methods in the 1980s-90s. Today, georeferencing is a standard function in all GIS software and is most commonly applied to scanned historical maps, unreferenced imagery, and CAD drawings.
Analyze Georeferencing data with LYRASENSE
Use our agentic notebook environment to work with satellite data and compute indices like Georeferencing — no setup required.