We captured GCP’s at ground level next to permanent objects on the landscape and also measured the height of the objects, so we were wondering if there is a way for Pix4D to utilize these known heights to (a) calibrate the DSM to produce that height when going from the object’s pixel to the pixel of the ground next to it and/or (b) contribute to the accuracy report?
It is not possible to directly add 1D GCP. When adding GCPs, then the software will expect either 2D dimensions or 3D, otherwise, the software will not know where to locate this point.
A possible workaround would for the 1D GCP is:
- Create a Manual Tie Point in the rayCloud, e.g. at the floor of the building or bridge. For more information about how to create an MTP you can check here: https://support.pix4d.com/hc/en-us/articles/202560349.
- Open the GCP/MTP Manager.
- Double click in the type field and change the type to 3D GCP.
- Double click in the altitude field and set the altitude for the vertex.
In order for the whole model to adjust to these values, you should Reoptimize.
I would also recommend you to have a look at the following articles that explain how to create surfaces in the software that can be later on used for DSM generation:
- How to draw a Surface in the rayCloud.
- How to correct Building Artifacts in the DSM and Orthomosaic with the rayCloud.
Thanks Blaz! Good answer for option (a). For option (b), I guess I just compare the known heights to the estimated heights in the DSM manually?
If the “building points” will be marked as GCPs or Control Points then the error estimation will be shown in the Quality report, section Ground Control Points or compare the results manually in some third party software (QGIS,…).
Thank you, Blaz!