Im putting together a UAV for aerial surveying using stereo photogrammetry.
Id like to get some input for the experts before going ahead.
My plan is to buy a high wing trainer coupled with an APM2.
Electric powered with a gyro camera mount.
The software Im using is photomodeler scanner. It can do DSM.
Im quite a n00b when it comes to fixed wing, but I have got a fair amount of expericence on multi-rotors.
Im doing this on a fairly tight budget and need to know a good camera to use.
Any advice would be greatly appreciated and will gladly buy any1 a beer if you're ever in South Africa :)
Yes, it is ~1 frame per second, we currently use IR and obtain more than enough "vertical overlap", the side-lap is strictly kept at 70-75%, although we will be testing the setup at 18m/s because of the benefits on endurance.
A zoom lens set at 35mm would defy the purpose, prime/standard lenses provide less distortion, better light transmission due to the fact they are constructed by fewer elements.
I couldn't recommend enough a good quality prime lens no matter what camera you are using.
Although ground resolution with the current A7s and 35mm at 120m AGL is lower than using Nex5N with 24mm flown at 60m AGL, even if we keep sidelap at 70-75% for both setups, model reconstruction is far better on the A7s setup. I think it should be no different with the higher resolution A7r although it would increase memory consumption for the reconstruction.
Hope this help
I did a few tests with a rented A7R on my Skywalker last November.
Zeiss 2.8/35mm, 90m AGL, 15m/s.
Great pictures but I had to go for 1/4000s or higher. Not scientific
work but for this resolution a cam with only 1/4000s may not be enough.
Did someone make similar experiences?
Why did you have to shoot at 1/4000?
I wonder if the problem was with vibrations and set of the aircraft rather than required shutter speed at that altitude/speed and lens
I couldn't succeed with my first test flights. Then I stepped back to the ground.
Here I could do larger series of setup combinations. From within my car I tested with similar speed and distance. Lower shutter speed did not make perfect pictures on the pixel level. In the air this setup gave me series of more than 100 pictures without a single fail.
I started with 1/2000s. But this didn't hold for me.
As the car and plane tests behave similar I vote for the shutter speed.
But may be someone can convince me that it can't be true.
For the next time I don't have an A7R to test in more detail.
just found my original calculation I started with
AGL: 90m => ground: 61.7m x 92.6m
pixel: 61.7m/4912pix => 1.26cm
1/2 pix size: 1.26cm/2 => 0.63cm
15m/s: 0.63cm/15m/s => 1/2381 sec
For mapping you need a small quad and something like the Samsung NX1000 with a 16mm lens. No need for a gimbal! Since you are in South Africa, consider steadidrone.com equipped with APM2 or Pixhawk. This platform has a wide user base (for support) and is easy to repair and maintain. This advice is from practical experience.
Another very good camera is the Sony Alpha 6000. See http://diydrones.com/profiles/blog/list?user=2amo6urztvqt6 for a robust quality test. We use Agisoft's Photoscan Professional.
Make sure you have lots of well surveyed ground control - consider purchasing dual frequency GPS receivers such as the V-Map systems for this purpose - cheap and extremely easy to use. Can also be carried on the UAV (weight is 130g) to fix camera stations.
If I ever do swing by, make mine a Windhoek Lager - I don't drink Bud, Miller or any of the SAB varieties.
Walter Volkmann (ex Namibian Land Surveyor - now in the US)