interpretation of check point errors

[konopiski@angst.at]

Dear community,

I used markers on the ground to georeference my model. After picking the locations I run the camara optimization.

Now I uncheck 1 marker to define it as a check point.

My question: How can I interpret the result of the error, and when would it be too high?


Cheers,
Markus

[jrp]

Hi,

I usually work on small objects, or occasional buildings, not landscpes, so I'm not an expert. If I got errors that small on a house, I'd feel happy about it.

For a site of ~500m x ~1000m with a ~25m elevation change, having 3 centemeters of error is around 0.003%, or 1 part in ~30000 -- I can't think of any method I have to measure anything that accurately.

Ask yourself how large an error would you accept the offset of any point to be (by that I mean the simple distance between where Metashape thinks any location is, versus the real world) -- measured in meters.

If any of the errors in the error column are above that, you may have an issue.

That's the method I use, and it seems to work for me, others who work with landscapes may have a more refined method.