Agisoft Metashape

Agisoft Metashape => General => Topic started by: Fkybarless on August 05, 2020, 08:04:33 AM

Title: GCPs not affecting accuracy after optimization - only on Phantom 4 RTK data set
Post by: Fkybarless on August 05, 2020, 08:04:33 AM

Hello,

I am getting constant 5CM accuracy (xyz) with only running Align and using GCPs as check points only.
But, I was told that on Phantom 4 RTK I can get great results using significantly less GCPs than in "regular" drone.

When inserting the GCPs in to the solution, then hit Optimize (all the parameters checked F,K1,K2 etc...)

Nothing changes, or changes very little.

Maybe it's because the photos has high accuracy weight?

Whats the right way of doing this?

Thanks

Title: Re: GCPs not affecting accuracy after optimization - only on Phantom 4 RTK data set
Post by: Fkybarless on August 17, 2020, 05:21:07 PM

Nothing?

Title: Re: GCPs not affecting accuracy after optimization - only on Phantom 4 RTK data set
Post by: stihl on August 28, 2020, 10:40:17 AM

The approach you should follow in order to reduce the deviation on the GCPs is by filtering the tie-points ór heavily reducing the 'weight' (standard deviation) of the camera positions. Since you're using a P4 RTK it's best to use the gradual selection tool to reduce the amount of (bad) tie-points.

Filtering the tie-points in essence also reduces the weight of the camera positions as less measurements (in the form of tie-points) are available, which increases the importance of the GCPs.

Use the gradual selection tool to filter tie-points visible on only 2 images with the 'Image count' tool, then use the other tools; 'reprojection error', 'reconstruction uncertainty' and 'projection accuracy' to reduce the bad tie-points further. Right click on your chunk in the workspace pane and select 'Show Info' to check the RMSE of the tie-points. Preferably get this RMSE value below 0.8 pixels.

Title: Re: GCPs not affecting accuracy after optimization - only on Phantom 4 RTK data set
Post by: xabierr on September 02, 2020, 05:27:36 AM

Hi Stihl.

Are not the weight of cameras determined based on (RTK derived) XYZ positional errors? Still not good enough?

Also, what values would you suggest for the other criterion 'reprojection error', 'reconstruction uncertainty' and 'projection accuracy'?

cheers,

Javier