A Short Definition of Geospatial Registration Accuracy from GeoVAR, a Geospatial Data and Information Technology Solutions Provider

DEFINITION: Geospatial Registration Accuracy

Geospatially referenced ortho-rectified imagery offer photographic information combined with accurate geodetic models to support compilation of scaled mapping themes.

The actual geospatial registration accuracy (i.e.: production processing expense) required for any given project, really depends on our customers' specific end-use application. Some applications do not require ortho-rectification. This is why our all of our geospatially referenced products are always initially quoted and ultimately delivered with one of the following industry standard measurement values ...

CE90, (Circular Error of 90%). CE90 is commonly used for quoting and validating geodetic image registration accuracy. A CE90 value is the minimum diameter of the horizontal circle that can be centered on all photo-identifiable Ground Control Points (GCPs) and also contain 90% of their respective twin counterparts acquired in an independent geodetic survey.
LE90, (Linear Error of 90%). LE90 is commonly used for quoting and validating DEMs. A LE90 value represents the linear vertical distance that 90% of control points and their respective twin matching counterparts acquired in an independent geodetic survey should be found from each other.

RMSE, (Root Mean Squared Error). RMSE is commonly used for quoting and validating geodetic image registration accuracy. A RMSE value is a single summary statistic that describes the square-root of the mean horizontal distance between all photo-identifiable GCPs and their respective twin counterparts acquired in an independent geodetic survey.
1-Sigma, (Standard Deviation Error). 1-Sigma is used for quoting and validating geodetic image registration accuracy. 1-Sigma is the minimum diameter of the horizontal circle that, when centered on all of the photo-identifiable GCPs, would contain one Standard Deviation (i.e.: ~68%) of the population of all available twin counterparts acquired in an independent geodetic survey. This is provided that the GCP population is sufficiently large for their relationship to be "normally" distributed.
Approximate mapscale equivalencies based on the US NMAS
(Unitied States National Map Accuracy Standard) ...

Map Scale	CE90	RMSE	1-Sigma
1:2,400	2 m	1 m	1 m
1:4,800	4 m	2 m	2 m
1:12,000	10 m	5 m	2 m
1:24,000	12 m	6 m	3 m
1:50,000	25 m	15 m	12 m

During deployment of an earth-imaging system, imagery are recorded simultaneously with data from Global Positioning Systems (GPS) and Internal Measurement Units (IMU). In order to produce highly accurate ortho-rectified planimetric-mapping imagery sources, imagery must be adjusted by Digital Elevation Models (DEMs), and surveyed Ground Control Points (GCPs), or Replacement Sensor Models (RSMs).

Our product quality derives from experience with the following methods...

Ortho-rectification - This method yields our most accurate results by correcting distortion, (or bias), caused by...

Terrain relief by using digital elevation models.
The object-to-image transformation by using a rational polynomial coefficients (RPC) mathematical description derived from star tracking observations for sensor position, satellite ephemeris for sensor orientation, and photo-identifiable ground control points (GCPs) for optional RPC bundle adjustment calibration.

Reference Ellipsoid or Geoid Registration - Many requirements are met with image products resulting from a less rigorous and less expensive registration to a reference ellipsoid instead of a terrain model. This geo-referenced product is commonly referred to as "Geo". Its accuracy can equal that of ortho-metric products for small near-nadir subsets covering fairly flat terrain. We offer Geo products made from two basic methods: RPC or GCPs.

Linear Co-ordinate Translation - This product is commonly accepted with a separate "world file" for each image describing: geospatial position, scale, and rotation. Pixel re-sampling or data transformation do not necessarily occur. It is well suited for geo-referencing large volumes of scanned paper maps, or aerial or street-level photography.

Database Linking - Panoramic aerial or street-level imagery can also be geo-referenced via database linking. OAP GIS efficiently perform this task with large volumes of geo-referenced imagery.

GeoVAR