If you flew say a 20-acre area at 200 feet and flew two 1-acre areas inside the 20-acre area at 100 ft, what is the GSD of the highest resolution orthomosaic export option? Let’s say the GSD for the 200-ft photos are 1in/pixel and the GSD for the 100-ft photos are 0.5in/pixel. Is the Orthomosaic exported a uniform 1in or 0.5 in/pixel image? Or is the effective GSD variable and the higher shots are 1in/pixel in the exported ortho and the small areas with lower shots have 0.5 GSD?
1 Like
I do exactly this most of the time. The parts I want to have very accurate are flown low as well. I guess, but am not 100% sure!! that it is a different accuracy though.
Means the spots I fly quite low are very highly detailed, the parts flown with a higher altitude are less detailed.
@MichaelL what do you know about this?
1 Like
I do the same. Buildings and other vertical structures get much lower and closer attention. All these stats we get are just estimates on static values so the map can definitely be different resolutions in different parts so the GSD doesn’t really have a true value. It will report the value of the flight that you upload it under. Visual resolution increases, but so does point cloud density. The higher images capture mass details which help stitch the areas that you flew much lower together. I haven’t flown a ton of them, but flying really large (500ac +) tracts I would fly at max altitude with minimal overlaps in addition to my normal flight just to get a better mass stitch and minimize warping.
1 Like
We’re really only talking about resolution of the export files coming out of DD. Stitching, 3D model accuracy, etc are really not the topic.
Here’s what DD says:
Thanks for reaching out to our team with your question - we enjoy a hard question that stumps us every now and then! We consulted our processing engineers and they confirmed that you will get a single resolution for the entire map. In other words, mixing and matching is not allowed.
For a bit more context, we do a median calculation of your overall pixel resolution and our processing is based on the majority of your map. We compute the weighted average of the GSD. I hope that makes sense. Let us know if there’s anything else we can help answer!
So basically I translate this to, the output GSD of the ortho is going to be closer to the GSD of the majority of the source photos. In a way the answer seems a bit contradictory as first they same mixing is not allowed, but then they refer to the “weighted average” If mixing isn’t allowed a weighted average wouldn’t be necessary - theoretically, but only if the ground was perfectly flat or you were using terrain following to keep the altitude above ground consistent. So I think they are just confirming you won’t get better resolution in the subareas if the majority of the photos, including ones for the exact same area, were at a higher altitude. Better stitching and elevation modeling, yes, but better GSD in the ortho export for those areas, no.
1 Like
As I tried to explain before you were going to get the resolution of the map that you upload, being the majority of the photos, but that is not a true value. The resolution will be better in the areas that were captured at a lower altitude. I would suggest you try it for yourself. In my experience you cannot export the data to a file and see it as in high of a resolution as what is seen on the web browser. That is the native resolution.
Also, the point cloud and 3D model accuracy are a factor as this is an ortho-rectified image.
Online viewing is irrelevant for this thread as I specifically said I am inquiring about the export file.
I understand that, but I am trying to help the understanding of how the system works. I am glad you got the question resolved.
1 Like