The result itself is quite impressive, watch mine from 2.6km^2.
Price: this is a service. The answer is 'it depends' on so many ways.
You click the resolution and it displays the price, but what it displays it depends on your contract terms.
Their system looks bulletproof for square areas, I had a few problems only with very elongated map but we are working on it.
PLS keep in mind that well georeferenced ortophoto is around thousand of euros per sq km if it has accuracy required for road construction works - but this is something like data product with atests norms etc. IMO pix4d has output that allows mnaking rigorously those data, on a remote server, without human intervention, some 12h uploading and 24h processing (I got very low priority there yet). This is not something you can make easily on your home computer.
Nice thing it shows the fit accuracy so you have a valid positional report, and you get kml file plus georeffed bitmap as output, + lots more (DEM etc).
You can deduce precision form there.
This is a link to the report form our sulphur2 case:
This was 70min flight and we did it with double overpass to make it compatible with other, less capable services. We have also reduced crab angle to a few degrees using our realtime wind estimator (self-adjusting flight pattern), just for testing (but this is not helpful for 99% of processing services).
The turbulence and thermals were wild, the altitude was 300m+/- 40m because of 3-4m/s updrafts, very sunny. The area is recultivated sulphur mine, south of Tarnobrzeg, south-east Poland. Looked so uniform till the horison, but from above, no two lakes had similar colour not to mention about UFO marks all around.
MarcS my experience of similar systems is that the 'hand work' is minimal as long as the system works with the photos you have. The SfM type systems only need a few GCPs to transform, scale and rotate to the user grid. If there are more than three GCPs the best you can hope for is that the system will minimise residuals at the GCPs - in general they wont adjust the model shape to suit the GCPs as the underlying algorithms are expected to be pixel accurate and in some cases - sub-pixel.
Hi Rory, not so sure if they broke restrictions. Swiss aviation regulations seem to be quite good regarding UAV. At least you can talk to the agencies and are not blocked at first...
I´m quite impressed by the results, too. Question is how much "hand work" goes into the final 3D result? Thats more of an interest then the actual time it takes calculating... The projects I know and have worked on needed a significant amount of ground control points when flying and manual pointing to them when calculating the data This can take several hours or more.... This could be a good point of a DIY competition: Usability without photogrammetric knowledge!
Comments
Friend of mine ordered the following from Sensefly
Pix4D Premium 5000 euros
Sensefly eBee Drone
Near InfraRed Camera 1k euros
Grand total was 19k euros
I'd like to know more about pricing as well
Anyone found a price on the desktop version?
The autopilot saves position stamps in the logs. Then depending on autopilot you have one way or another to match jpeg filenames to their positions.
how do you get the GPS tags ?? do you need a special camera ??
The result itself is quite impressive, watch mine from 2.6km^2.
Price: this is a service. The answer is 'it depends' on so many ways.
You click the resolution and it displays the price, but what it displays it depends on your contract terms.
Their system looks bulletproof for square areas, I had a few problems only with very elongated map but we are working on it.
PLS keep in mind that well georeferenced ortophoto is around thousand of euros per sq km if it has accuracy required for road construction works - but this is something like data product with atests norms etc. IMO pix4d has output that allows mnaking rigorously those data, on a remote server, without human intervention, some 12h uploading and 24h processing (I got very low priority there yet). This is not something you can make easily on your home computer.
Nice thing it shows the fit accuracy so you have a valid positional report, and you get kml file plus georeffed bitmap as output, + lots more (DEM etc).
You can deduce precision form there.This is a link to the report form our sulphur2 case:
http://ftp.aerialrobotics.eu/pub/sulphur2-2011-04-18/sulphur2cloud_...
This was 70min flight and we did it with double overpass to make it compatible with other, less capable services. We have also reduced crab angle to a few degrees using our realtime wind estimator (self-adjusting flight pattern), just for testing (but this is not helpful for 99% of processing services).
The turbulence and thermals were wild, the altitude was 300m+/- 40m because of 3-4m/s updrafts, very sunny. The area is recultivated sulphur mine, south of Tarnobrzeg, south-east Poland. Looked so uniform till the horison, but from above, no two lakes had similar colour not to mention about UFO marks all around.
Hi Rory, not so sure if they broke restrictions. Swiss aviation regulations seem to be quite good regarding UAV. At least you can talk to the agencies and are not blocked at first...
I´m quite impressed by the results, too. Question is how much "hand work" goes into the final 3D result? Thats more of an interest then the actual time it takes calculating... The projects I know and have worked on needed a significant amount of ground control points when flying and manual pointing to them when calculating the data This can take several hours or more.... This could be a good point of a DIY competition: Usability without photogrammetric knowledge!