Well, I've passed my faa exam, registered my phantom 4 pro, and getting ready for a trial flight.
I can't find anything telling me how high to fly and the accuracy I could expect with a 20 mp camera.
I realize it will be more accurate at 100' than 300', but isn't there a formula for calculating the approximate accuracy one could expect?
I will probably set ground control points on each job, I assume.
thanks,?ÿ JTS
Yes, there is a formula. But it is much easier to use an App like Map Pilot. Just draw your flight area. Change the flight altitude above ground level and you should see the calculated?ÿGround?ÿSample Distance (GSD).
The accuracy of yourpoint cloud and other deliverables will be more dependant?ÿupon the
- image overlap
- clarity?ÿof images
- type of targets?ÿand number ground control points and proper?ÿplacement (GCP)
- the accuracy of you marking the GCP's
- terrain?ÿtype, elevations changes, and features
- obstructions like trees and buildings
- Post-processing software works best with a lot a change in each photo to?ÿcreate a mosaic. If you have snow on the ground, it likely will NOT process. Too many similar photos.
Lower altitude flights will have smaller GSD, but at the expense of longer flight time and with more photos.?ÿ?ÿ
Sounds you may need?ÿassistance and practice. Feel free to contact me over the weekend.
I was in a seminar about drone mapping on Wednesday. The presenter had a table that said that at 100' AGL I should expect 0.1' accuracy, at 200' AGL 0.2' accuracy, and at 400', 0.4' accuracy.?ÿ?ÿ
I just finished a project with a 4 pro.
We flew at 140 with 75 percent overlap. There was about 60 percent snow cover on one half of the site. GCPs were at nominal 1000' intervals. We are seeing about a tenth in our data.
We are extracting data with Virtual Surveyor. So far its pretty awesome...
Virtual Surveyor software looks very interesting.
The overall Accuracy of your deliverable is not derived from GSD or GCP.?ÿ
NSPS has a document describing methods and procedure needed to apply an accuracy statement to your deliverable.
It is "The National Standard for Spatial Data Accuracy".
In summary.
You need to measure sample data throughout your project (atleast 20 points, 4 in each quadrant) using equipment and procedures that are at least 3x more accurate the final accuracy statement. So if you are using RTK methods that are accurate to ?ñ0.04' , then you are already limited to an accuracy statement no better than ?ñ0.12'.
Thanks a bunch, everybody.
Lee, you are right, I do need assistance and practice. I'll email you.
