I am trying to merge two projects, I was following the steps from the online guide. I created manual tie points and made sure that the So^2 value was under 1 pixel but it still shows that there is very high error in the z - axis. I attached the quality report below.
M3_report.pdf (1.1 MB)
The same thing happened with two other projects, the quality report for that one is below as well.
M4_report.pdf (2.2 MB)
Hi hadenapenrod,
I see a couple areas that could be improved. First, you have a high camera optimization error. Anything above 5% I would suggest enabling All Prior under step 1.
The other item has to do with how you mark your MTPs. You made reference to this but it could be improved. I would suggest only marking 10-15 of the best images. Make sure the points you are marking are at the center of the image and zoom in as far as you can. As you said, you will want to keep the So^2 value under 1 pixel. If you use blurry images or the points are at the edge of an image then this value will increase. That should help merge your projects together.
Hi Mike_K,
Thank you for your response! I tried your suggestions and they worked great in reducing the camera optimization error. I updated the MTPs, some were very off so I reduced the number of them.
For some reason the y error decreased but is still quite high.
Any ideas or suggestions what I can do to reduce that?
M3_report.pdf (1.2 MB)
Hi ellis,
The geolocation has to do with the location of the images. Mapper will compare what the geolocation is on the exif to where it computes its location. Due to low grade GPS units on the drone it is difficult to rely on this data. It doesn’t give you a very good idea of the accuracy of your project. If you could add some GCPs then we can understand if the model isn’t fitting very well.
