Multiple blocks, Ortho discoloration issue

PIX4Dmapper
1

Hello,

I just processed some data in pix4dmapper that was collected with a Micasense RedEdge Dual camera rig. I’m facing a couple issues:

  1. I notice that in the quality report it says “2 blocks” next to Quality Check >> Dataset. Is this due to not having enough manual tie points in the areas of sparse overlap? I’m going to try to put more tie points in those areas and run Rematch + Optimize to see if that helps. I also notice my RMS error is 0.3, which isnt great… so I suppose redoing those to get the error down is my next option. Open to any suggestions you may have as well.
  2. This may not even be a issue with the processing in pix4d but once I process all of the reflectance mosaics, I bring those into arcpro to create a 10 band composite image. I notice alot of discoloration and I think its due to the bands not aligning correctly. Is there anything I can do to help this in pix4d? Maybe placing more tie points? Let me know… Thanks in advance.
    MicrosoftTeams-image (1)
    MicrosoftTeams-image (1)924×777 78.4 KB
2

Hi lgerber,

Can you attach your quality report to this posting? It will help us better understand your project.

3

SC_WL19_SouthTibwin_report.pdf (1.4 MB)

4

FYI this collection was missing reflectance panels, so I had to use panels from a collection that was done shortly after…

5

Hello @lgerber, The multiple block is due to few uncalibrated images where the image acquisition (overlap) was not uniform. It is in the edge of your project so it shouldn’t be any issue to your project. The second issue about the visualization, when you merge the 10 bands, you will get the output like the one in the screenshot, it will not be like RGB orthomosaic.
From the screenshot, I can’t see if there is any offset in the bands or not, however, I guess when you process the two camera, you created your own rig ang then processed it. If you have processed it without creating a custom rig then there might also be some offset in the results between two cameras. For more information about process 10 bands camera as a rig, visit the below support article