A Helmert fit onto control points, returning rotation, scale and shift together with RMS and per-point residuals. Plus a grid ↔ ground factor, so you know how far a projected distance sits from the real one on site.
Localization handles an arbitrary local system; grid ↔ ground handles the gap between the projection plane and the actual ground.
For each point, give the source coordinate (what you measured) and the control coordinate (the true, known value for the same point).
With at least two pairs you get the rotation, scale and shift that best fit the source points onto the control points.
RMS shows the overall fit; the residual shows how far each point deviates. Red means over 5 cm — that point is suspect.
Once the fit looks right, type any source coordinate below and read it back in the control system.
For a chosen system, position and height you get the CSF, and the distance converter then works in both directions.
When you are working in an arbitrary local system — a site grid, an old report, a network someone laid out with no tie to the national system. A Helmert fit ties that system to known points.
A residual is how far one point deviates once the fit is computed. If every point has a small residual except one, that one is usually mistyped or its marker has moved — the fix is to check that point, not to add more.
A projection distorts distances, and height changes them again. At 500 m elevation and well away from the central meridian, the difference over a kilometre runs to several decimetres — enough to throw out a long stakeout.
When you are working in an official system (Gauss-Krüger) and need the real distance you will measure on site with a tape or EDM, rather than the computed one from coordinates.
At least two, which is enough to solve rotation, scale and shift. With three or more you also get residuals and an RMS, so you can judge how good the fit is.
That the point deviates by more than 5 cm from the fit. The usual cause is a mistyped coordinate or a disturbed marker, rather than a bad transformation.
The product of the grid factor (projection effect) and the elevation factor (height effect). It converts a grid distance from coordinates into the true ground distance and back.
No. Coordinate transformation moves between defined systems with known parameters. Localization derives the parameters from your own measurements, for a system that has no official definition.
This tool ships inside the GeoKantar app. Available on the App Store and Google Play.