I'm desperately searching for a page I've recently seen on the forum (or was it the Wiki?), describing the maths and process on how to determine a good distance between GCP;s on a large land survey project.
- Is there anyone out there who know what page I might be thinking of and is able to point it out?
We're running a full city project with a couple of thousands of drone photos and realized we might be short on control points. Our goal is to accive approximately 1" ground resolution after flying at 80m with 80/75% overlap.GCP;s are now spread out with 250 - 500 meters distance and according to the quality report that wasn't good enough.
/Anders Theodorsson - Ludvika Municipal GIS-office
You could use less GCPs if you have RTK GNSS and also reusing calibration from 3D flights for Nadir only.
Contextcapture doesn't adjust camera calibration based on GCPs so if The model appears to be curved instead of being horizontal after AT without GCPs then trying to warp it flat using GCPs will result in lower quality and requiring more points than running with good calibration as less re-warping is needed. So try to reuse calibration only which after AT look good and flight paths are flat.
Thank you for guiding me to that document! In fact I did find it late last night myself and it turns out I've seen something similar explained somewhere els, with some drawings aswell. Don't know where, but that's not so important since the manual you pointed out here gives the answer too.
Thanks a lot!
What you are looking for can be found in the "Drone Capture Guide" (pdf file) Page 4. This document can be found in C:\Program Files\Bentley\ContextCapture\doc
Answer Verified By: Anders Theodorsson