I am trying to interpret the .asc data files in Rhino and use them to generate a 3D building model (Town scale), for which I am using the DTM plugin to visualize the data but it seems it is not a good approach for doing this. I know it must have a better approach for doing this and I appreciate it if there is any idea.
From that screenshot, there simply isn’t enough resolution to your mesh to be able to extract anything useful from it. You’ll have to find a better data source somehow.
-wim
I download the data from the UK gov. I just tried out four of them (there are around 12 data files there on the drive). I am wondering if there are other methods to visualize it rather than using DTM.
On the other hand, my goal is to classify the buildings in the same group. I am not sure how to do it and still finding reference about this part.
I’m afraid that, in the scope of just using Rhino, I have no idea what that means.
As for that last file you posted, it certainly has more resolution but that comes with a size as well. At 1.3 GB, it empties my PC’s RAM after a short period of time. Perhaps you have a better spec’ed machine.
If your goal is to create NURBS representations of all the buildings in London, you have quite a bit of work ahead of you. I haven’t come across anything that will do that automatically.
It is conceivable to write a Grasshopper definition that creates a rectangular box and, perhaps, a sloped roof, but (1) that probably requires several mouse-clicks per building, and (2) as it is, it’s impossible to navigate in that file (on my machine).
It looks like most, if not all, buildings have footprints in Openstreetmap:
… but it doesn’t look like many of these have any height information attached.
Perhaps it’s possible to use, e.g., Elk to import the outlines and combine these 2D footprints with the 3D mesh but I’m sure you’ll find that the projection of these two datasets are different and you’ll have to somehow reproject at least one of them to get them lined-up.
Apart from all that, I’m pretty sure it’s possible to buy a dataset with 3D buildings completely finished.
-wim
Thanks for your insights. I just tried the MeshPatch but it seems that it is not good enough for metadata. Elk is great but it lacks height data in many areas.
Also, I have tried different point cloud processing software these days (such as ContextCapture, Pix4DMapper), both of them focus on transferring images to point cloud data but cannot classify data/ import point cloud data directly.
My goal actually is to classify the buildings, road, ground from the point cloud data and generate geometry from these objects to perform different analyses. After trial and error these days, I think the critical step is to classify the building, road, and terrain data from the point cloud or regenerate these data from the DSM (like this post: https://www.grasshopper3d.com/group/gismo/forum/topics/building-heights-trick)
The GIS plugin is the ideal tool as the shapefile had classified the objects into different categories so that I think the alternative workflow could be extracted the height data from the point cloud data to the shapefile then generate the geometry in Rhino.
I had reduced the size of the point cloud data and hence we can use it to test different workflow.
