I have drone images from a mavic that I’ve processed using brown 3, brown 3 with tangential 2 and k+brown 4 with tangential 2 and no matter what it comes out curved/banana effect at the end. I thought merging it to the laser scan would help correct it and I ran just an alignment and image wise they merged perfectly but I still had the curvature at the end. I then added GCPs and ran the alignment again, I have also tried changing the distortion models when merging the laser scan with the drone imagery. I’ve also said yes to merge georeferenced components and every time the images merge perfectly but i still have the curvature in my model. Any ideas? is there a way to fix the drone images as they seem to be the problem?
Hello,
The curvature of the model is usually connected to the distortion which can be found on the edges of your images. It piles up and creates a curvature.
I see you already tried some of the recommendations we offer to fix this issue.
When using GCPs how were they distributed? Did you distribute them evenly over your project? How many of them were there? Did you disable the images’ location information when you were trying to flatten out your model with GCPs? Did you align few more times after that (cache files may be your friend here)?