Ok, so this may be a more advanced topic than normally found on this forum. But I am struggling to classify the point clouds generated with the UAV. After Pix4D, after Agisoft or others you are left with an unclassified point cloud If I want to classify the buildings, vegetation, and ground points so I can then generate a DTM. How is everyone doing this? What software have you used, what settings?
The main problem I have found is that most point cloud manipulation software is designed for LiDAR, and LiDAR has one thing we do not have, multiple returns. So for them clasifying the ground is a matter of getting the last return and then just cleaning up the buildings. But most of these don't work so well with the point clouds from Pix4D, Agisoft, etc... I have tried LASTools, ESRI, and recently DTMaster. Any other suggestions? or suggestions on the settings to use?
Would you be willing to share a dataset? I'm a photogrammetrist/remote sensing pro and currently researching the UAS market for reliable 3D acquisition. Let me know if interested. We could benefit.
I am just getting into point cloud classification. I am using Agisoft. They have a built in GUI for doing this. It is done by taking the lowest point elevation in a user defined cell and applying a masking angle to that along with a max height. I have successfully generated A DTM on an open area. Big whoop right. I am currently trying to build a DTM in a wooded area with leaf-off conditions. I will let you know how it turns out.
A little background about me and my rig. I am a professional land surveyor. Armature UAV pilot. I think I have had more crashes than successful flights. I am flying a Turbo Ace controlled by a Pixhawk. I am using a Sony A7r to capture images, and a Leica GS14 deferentially corrected GPS to acquire ground control points.
I am curious: what are you using to classify your points. I think this is where the future is.
I use Terrasolid's TerraScan product which sits on top of Microstation as an MDL app to process LiDAR points clouds.. TerraScan uses statistical based algorithms, so it is not dependent on Return/echo number. I'm in the process of generating point clouds from UAV photos using photoscan and I'l classify them with TerraScan. I'll let you know how they turn out. I'm having problems georefing the images using Mission Planner. I'm getting very high altitude numbers for the Z. I'm flying at 50 meters and it is saying they are at 2200 meters.
50 mts is probably relative to ground 2200 is from MSL.
I have tried a bunch of programs LASTools, DTMaster, And ESRI tools, but I have not gotten good results yet. Also there are a lot of parameters you can adjust on some of these tools and I may not have the right settings yet. So I keep trying.
I think I figured it out. When I setup my home position I didn't have the MP connected to the X-8 so it assigned an arbitrary value for the elevation. Newbee stupidity :)
Try Globalmapper, there is a way to classify pointclouds from photogrammetry using the LIDAR module...
Huh. I will look into that. Thanks
Happy to see this thread... I'm doing a photogrammetry and had successfully generate a very nice dtm, dem and orthophoto with acurate ground control points using agisoft. Maybe the same problem with Jaime but different perspective, i want to clasify the point cloud that has green colour (which represent trees) seperately from the red and grey clour point cloud (which represent actual ground) automatically. Is there any such function on any softwarr to do that automatically? I have managed to do that manually, but if the survey area is too big, it will be too much pain.
Hi Jaime, since you are a developer, you should not be scared to use the open source best photogrammetry software called "Micmac". It is not a GUI based soft, only command lines. This soft is used by the French National Mapping organization (IGN).
I thought I would share my experience with Agisoft. Here is what Agisoft looks like when you are classifying the dense point cloud. The software takes all the points generated from the dense point cloud and breaks them into cell size which the user defined as Cell size. Then specify from the lowest point, z, in that cell how high to come up from that point (max distance) and what the angle (max angle) should be from the lowest point to any point included in the DTM surface. Everything that is brown is now in the newly classified ground surface, and everything that is white is separated out.