I’m looking for software that provides the capability to orthorectify imagery from planes. Does ODM support that or does it only support imagery from drones? I’ve stumbled across OSSIM, but the ODM community seems much more active.
We should be able to support it, but not all flight plans provide sufficient overlap or sidelap, and we’ve got room to grow with supporting plane-borne sensors upstream in OpenSFM.
So, if you can, provide us some test data at dronedb.app and give it a go so we can all work on it together!
We have definitely stitched imagery from planes before (provided it has sufficient overlap).
How about just generating orthos, not necessarily stitching or mosaicing the imagery? For every input image, I’d like to generate a corresponding ortho.
There is early experimental support for doing something similar in the DroneDB app:
I’m not sure if that is actually generating an Orthophoto for each image or just visualizing it on a map. Would be great if it could generate an ortho. Do you all have any recommendations on open source software that could do this? GDAL?
Check this out:
What’s OSSIM? I only find security software.
Open Source Image Processing Toolkit a few of my colleagues have mentioned. I’m having a rough time getting it installed and working properly…
I found this Dockerfile which is working but the elevation is not configured properly. I also found an old installer for Windows but still can’t get the elevation set up properly.
It seems like a pretty old and inactive project so I’m not sure if it’s worth investing more time.
Was hoping to use this tool
Yeah, this approach would do most of that.
In addition to Piero’s example, I have tested with a variety of other airborne data.
That approach is using ODM correct? What is different about that approach?
This thread was pretty helpful to me! I had a dataset that had reduced overlap due to changes in ground height not accounted for in the mission that would not give me a good output in other settings.
After reading the linked post about aerial imagery I decided to run the dataset through using the “Fast-Ortho” parameters, and it worked! A few small holes in the final output, but much better than I was getting before!
This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.