Highly distorted calibration results - Thermal images of PV plants

Hi,
I am processing thermal images of solar plants with 80% overlaps (I know that the recommended values are 90% but I can’t re-collect the data anew). I am obatining highly distorted calibration results whatever the calibration method I use, is there someone who can help with this problem ? Thank you very much.

Can you send us the quality reports?