Alternating Empty Areas in Orthomosaic

Hello Pix4Dmatic community! I am processing imagery from intertidal oyster reefs and am having a strange recurring issue with my orthomosaics. Namely, I receive a DSM but an orthomosaic with regularly alternating images, with black squares in between. I receive no errors during processing, and am marking 3 GCPs prior to creating my DSM and ortho. My quality report indicates poor camera optimization, which I know can be caused by low overlap. This flight was done with 80% front and 75% side overlap, so I am not convinced that is the issue. I have uploaded my quality report, which includes images of the DSM and ortho. Has anyone else encountered this issue? Thank you!

[ECK_Site5_080124_V1-quality_report.pdf|attachment]

Hello Emory,

I don’t seem to be able to download the quality report, could you upload it once again, please?

Thank you and kind regards,
Alexey

ECK_Site5_080124_V1-quality_report.pdf (1.2 MB)

Hi Alexey - shoot I am sorry about that! I just tried again, hopefully it worked this time!

Hello! Thank you for sharing the quality report. As you noticed, there is a large difference in the camera internal parameters before and after the calibration.

What you can try is to rerun the calibration with the Map template + Standard pipeline, and, if this doesn’t help then try setting Internal confidence to High.

Please let me know if it helps.

Kind regards,
Alexey

1 Like

Hi Alexey - thank you for these recommendations! Calibrating using the map template + standard pipeline did not help, but using high internal confidence did! Do you know why my original camera optimization was so poor? I would love to tweak my future flight plans going forward to avoid the problem again if I can! Thank you for your help!

Hello! I am glad that you could process your dataset successfully. From the information that you shared I have an impression that the flight was correctly planned.

Maybe you can check the horizontal/vertical accuracies for the cameras in the camera table. If they are set to small values and if the camera coordinates do not agree with the GCPs then it might be that the calibration results go too far from the default internal parameters; setting the internal confidence to high prevents this. We can make a more detailed analysis of your dataset, if you agree to share the images, for this you can reach out to our support team.

Hope this helps.
Kind regards,
Alexey