Triangulation Between Hydro and Topo Survey Data

Delauney triangulation algorithm is used to create the terrain, and this does not allow 'overhang' features. There is settings to be able to adjust this, so you will have to manually amend the data.

https://en.wikipedia.org/wiki/Delaunay_triangulation 

I am not understanding your second sentence; there are, or are not settings to adjust the method or otherwise account for overhang features in a terrain?  I have clipped the pointcloud to remove the overhanging points, and will have to add contingency to my quantities to account for this; is there a better way to resolve this issue in the future?

Sorry - missed a word. 'There are NO settings . .. '

Hi Joseph,

To complement Mark's answer, DTM - Terrains are 2,5 surfaces and are not appropriate to model more complex surfaces with caves or overhangs.
To effectively model such surfaces, I had tested successfully 2 worflows depending on available data/Software:

1. Contextcapture (from Bentley) will enable you to create a "real" and accurate 3D surface, but for this you would need to have original LIDAR data that is including the position of scanner along with scanned point - and this is rarely the case.
   you may "cheat" and "invent" scanner location but it might not be that easy depending on shape of surface to reconstruct.
2. If you just have the XYZ points, then MESHLAB (free) will enable you to reconstruct a "real" 3D surface.
   I had tested it and got sensible results in a case similar to yours, see this post (sorry for French..).
   It was 5 years ago so I don't remember the exacts steps in MESHLAB but it's quite easy to learn and there is a lot of material available on the Internet.

In both cases you will get a mesh you will be able to assign a Feature definition to, then target or compute quantities from.

HTH

Hi Joseph,

To complement Mark's answer, DTM - Terrains are 2,5 surfaces and are not appropriate to model more complex surfaces with caves or overhangs.
To effectively model such surfaces, I had tested successfully 2 worflows depending on available data/Software:

1. Contextcapture (from Bentley) will enable you to create a "real" and accurate 3D surface, but for this you would need to have original LIDAR data that is including the position of scanner along with scanned point - and this is rarely the case.
   you may "cheat" and "invent" scanner location but it might not be that easy depending on shape of surface to reconstruct.
2. If you just have the XYZ points, then MESHLAB (free) will enable you to reconstruct a "real" 3D surface.
   I had tested it and got sensible results in a case similar to yours, see this post (sorry for French..).
   It was 5 years ago so I don't remember the exacts steps in MESHLAB but it's quite easy to learn and there is a lot of material available on the Internet.

In both cases you will get a mesh you will be able to assign a Feature definition to, then target or compute quantities from.

HTH

Hi Vincent, I'm running into an issue where I believe I need to assign a feature definition to a 3D mesh as you describe above; I'm trying to take the subsurface data integrated from gINT to update my quantities by the material type being excavated.  We've constructed 3D meshes, but I'm not finding any options to apply a Feature Definition to them...  I'm also not finding much info on this workflow, beyond sales pitches saying it can be done.  Let me know if you have any advice on this.

Thanks,

Joseph

Hi Joseph,

To assign a feature you need to use the "Set Feature Definition" tool

Then choose "Mesh".
I'm sure there is a SIG that covers similar workflow, will try to find something.