I want to provide some example data from a Riegl miniVUX system on a DJI M600.?ÿ I prefer the Riegl systems over the velodyne because of the accuracy from the sensor itself and the excellent penetration of vegetation and also the 5 returns it is capable of capturing.?ÿ I have been flying systems for all the major manufacturers, Phoenix, Swiss, GreenValley and have had the opportunity to capture and process data with the Riegl Vux, Riegl miniVux, Velodyne HDL 32, Velodyne ultra Puck, Velodyne VLP-16.?ÿ I am happy to answer all things unmanned LiDAR!
?ÿ
I want to start using drone based LIDAR on smaller area projects where my main focus id getting bare earth data in forested areas. What would you recommend in terms of both a sensor or what specifications I should focus on. Things like number of returns, resolution and so forth. Remember this is for fairly small sites of anywhere from 1 to 20 acres in size.
Can a single return system even work? It's seems it would not. I know the mini has 5 returns, but that is not a cost effective sensor for doing small projects.
Thanks!
I am curious about the maximum project size of a drone-mounted LiDAR. At what?ÿ acreage does it not make sense for drones and instead use a manned aerial system? An addition to this question could apply to photos instead of LiDAR?
Size does matter, but mostly it's a matter of the type of site for economic purposes. For images, you can only map what you can see. And conversely, there are some things you cannot see with lidar that can readily seen in images. Each has it's own strengths.?ÿ
Sorry I answered a different question than you asked. Part of an answer is that that it is not only size, but point density needed..
I've been flying a MiniVUX DL and I can tell you from experience that in the dense canopy we run into on undeveloped land you are getting very few returns from the actual ground. There are several things you can try to do to mitigate that, including tightening your overlap and flying opposing lines, but the single best thing you can do is wait until the leaves fall off deciduous trees. This is not to say that you can't get good results, but if you're looking for small variations in elevation and terrain it's a challenge. If anyone knows of a software that does an excellent job of stripping out tall grass and weeds and getting to the bare earth I'd love to know what it is.
What kind of points / sq meter are you getting?
Typically around 850, varies with height, speed and overlap of course.
What we discovered very quickly is you need a lot of data storage when flying this package. Just the processed LAS files from most flights are between 1.2GB - 1.5GB. Then you have the raw files, photos, processed photogrammetry, etc.
Thanks Lee. How many end up being bare earth?
I'm possibly going to rent one of these sensors if I by chance get a contract I am applying for, so it's good to know the details. I have someone that has flown it before, but I will be the lead on the proposal.
My main question was about a less expensive sensor that might be usable for small forested areas such as steep coastal hillsides.
When you fly in areas with more dense vegetation, what do you typically use for Ground Control Points?
There's no single answer to the question how many end up being ground, it's completely dependent on the environment. Sometimes none of them, but usually enough to get a reasonably accurate surface. As far as GCP, with the package we fly we have an Applanix APX-20 and we set an R10 base logging at one second for control, so it's really ground truth points rather than GCP. And we'll use anything that we know we'll be able to easily pick out in the point cloud.
