Support Website Contact Support Blog

Enhancement to Simply GCP linking

In my maps with GCPs, it always seems the drone altitudes and computed point cloud is way different that the measures GCP data.  As a result the GCPs are not “near” any of the images so it is not possible to mark the GCP’s on images.  Your support pages provide a solution - basically override all of the image EXIF altitudes to fudge the values to about the GCP altitude plus flight height.  I have tried this and it has worked, but it is big waste of time as it requires completely rerunning step 1.

There would seem to be a very simply alternative solution - allow matching GCPs using only X, Y data, then then reoptimize to fix the point cloud altitudes.

Another fix would be to allow me to apply a fixed offset to all computed points to allow GCP marking.

 

 

 

Hello Mark,

Thank you for your comments and suggestions.

There are indeed some solutions which could be implemented in the software.

For now, the best workaround is to change the geolocation altitude as you describe. However, that can be done before Step 1. As soon as the images are imported, the altitude can be checked and changed in case it is not correct. That will allow to not process Step 1 again.

Regards.

 

If I do have good GCP data that I input and link to images, does this completely override all of the altitude data in the EXIF fields from the drone during map construction?  In other words if I edit the EXIF data on the image input dialog, does it matter how accurate I am in this data or is this simply a starting point to help place the map closer to the GCPS?

Hello Mark,

There are three different calibration methods:

 

Depending on what method you select, the geotags will influence more (or less) in the adjustment. If you choose standard, it means that the initial photo positions (x,y,z) are not very precise so they will not have much influence. If you use Alternative, it means that your x,y,z are precise and if you choose Accurate Geolocation and Orientation means that both position and orientation are precise. These two last option will make the geotags have more influence in the final reconstruction.

More information can be found here.

The accuracy that you type in the Image Properties Editor will make the geotags to have more (or less influence). The more accuracy, the more weight they will have. The value selected must be realistic.

As for your question, as long as the geotags are not precise (coming from an RTK/PPK drone),  if you edit the EXIF data on the image input dialog, it does not matter how accurate it is, it is simply a starting point to help place the map closer to the GCPS.

Please let me know if it is clear now.

Regards.