HomeHelp Center › Working in the App

Working in the App

Checkpoints and accuracy verification

Prove your data's accuracy against surveyed points instead of just claiming a number.

A checkpoint is a surveyed point that was not used to build the model. Comparing the model's value at that point to the known value tells you how accurate the result actually is.

A one-and-a-half-metre uncertainty disc drawn to true scale on an orthomosaic beside a three-centimetre RTK dot
What a residual is measured against. Both circles are drawn to true scale: 1.50 m of consumer-GPS uncertainty against a 0.03 m RTK fix. A checkpoint tells you which of these your project actually achieved.

How it works

Upload your checkpoint coordinates and DroneFlow reports the residual at each one — the difference between measured and known — horizontally and vertically.

Reading the residuals

  • Small, random residuals mean the data is well-controlled.
  • A consistent bias (everything off in the same direction) points to a datum mismatch, not scatter — check your vertical datum first.
  • Compare vertical checks against the bare-earth model, not the surface, where the shot is on open ground.

Checkpoints verify accuracy; they don't grade or certify a survey. They give you the evidence to stand behind a number — or to catch a problem before a client does.

Ready to try it on your own site?

Start a free trial — no credit card required.

Start Free Trial

Still stuck? Contact us and a human will help.