cedthered
lvl.2
Flight distance : 1225827 ft
Sweden
Offline
|
I have tested mapping with an Inspire.
I currently use an Aibotix to do surveying and I have a Phantom 4 as well. (For fun )
Due to the heavy price of the Aibotix, I was interested to see if a cheaper alternative was availiable for high risk jobs.
I got a friend to fly the same mission as I did with the Aibotix, equipped with a Nikon Coolpix A with APS CMOS sensor at 18 megapixel.
I measured in my GCP (Ground Controll Points) with a Leica MS 60 totalstation, and CS10 RTK gps to establish the local grid.
I also scanned in the area with the MS60 (It has a Laser scanning function)
I used Agisoft to generate the point clouds acuired from both the Inspire as well as the Aibotix.
The MS 60 created it's own point cloud internally.
As it is difficult (impossible) to compare point clouds with each other, I created a mesh in SBG Geo from each respective point cloud. (Inspire/Aibotix/MS 60)
My flight heigt on both the copters was 40 meters above ground, the speed was 3 meters per second and the shutter interval was at 2 seconds.
I had the same ISO, shutter speed and F-stop settings on both copters.
My findings were:
On the X-and Y-axis the Aibotix data was within 2mm of the MS 60 data and on the Z-axis (Height) wich is THE MOST CRITICAL for my type of work, the data was within 3mm on flat surface, and 4mm on undulating surface.
The Inspire with the Zenmuse X5, the data was on X-and Y-axis within 5mm, but on the Z-axis it had differences of up to 150mm!
I think I can get the Inspire to get better data, but I will have to play around with camera settings, and flight heigt/speed etc.
I am also keen to do this test with my P4.
What results are you guys getting?
What can I do to get better results with the Inspire/P4? They are both really sweet copters to fly, and a hell of a lot cheaper!
Or is it a case of you get what you pay for? |
|