Height

We are flying a large excavation with a DJI Phantom 2 Vision+ to calculate volume production and are in process of evaluating accuracy. What happens to the Point Cloud when we use the Capture App with a fixed height of 50m and we transvers across 200ft to 300ft near vertical elevation changes at the boundaries of the excavation. How does this impact the accuracy of the points, resolution, focal length, etc.?

When will the distance restrictions on the App be lifted so that we can cover more area as battery life permits?

Very good question Mearon.
I would have liked the Subject title to be bit more explanatory so we get more comments… Panning Pit walls (?)
I have just returned from a very steep quarry 250m. Using an eBee, For VLOS reasons I had to be near the top. Therefore when I launch the ebee for NIDAR with a ATO of say 100m, the bottom of the pit is 350m away…poor GSD
Unfortunately I could not change the flight line heights (eg using satellite elev data) else I would collide with the sides on turn arounds.
SO…
I want to know if the software can handle panoramas? Combining NIDAR and the obliques of a multirotor.
Most of the discussion focuses on a poi that the multirotor circles (at various heights)
BUT can it handle situations where the UAV is the centre and pivots n degrees i.e. no real gps change.
I cannot see how it can. I cannot see how we can fly (accurately) in and arc (to get changing geotags) pointing out and ensuring overlap without very short interval shots. Keeping in mind over concentration of images in a particular spot can reduce the output quality such as with excessive overlap on large data sets.
The only alternative I can see is…
fly the copter up the walls trying to maintain an offset from it, move across, descend and guess the overlap, which mind you would increase towards the bottom given a benched /coned wall.
Very keen to hear others ideas on how to improve GSD/resolution in pits.
Cheers
Troy

Edit NADIR not NIDAR oops

Hi,

If the distance with the floor change, then the GSD will be different in the different parts, more accuracy where less distance.
you can take one flight over the terrain and one inside to not lose accuracy.

Then you can process all images in one project, and if you want very good model even in the walls of the terrain, then you take and oblique flight and process all having manual tie points.

check this: https://support.pix4d.com/hc/en-us/articles/204656209-Images-acquisition-plan-for-terrain-with-height-variations