I've been successfully testing CGAL Efficient RANSAC using this provided example to detect planes in a point cloud. The only piece missing from the CGAL Efficient RANSAC examples is a way to write the detected planes into a mesh (something like what polyfit offers at the end of this example)
I've noted that I can use the planes() method from Efficient_RANSAC to get an iterator of Plane but I'm a bit lost after that. Do I need to read all points from each plane and try and manually create a mesh from it ?
Thanks in advance for any help
Related
I recently got started with CGAL for implementing some models which required 2D and 3D periodic regular triangulations (weighted Delaunay or power diagram). I have made some progress with the 3D periodic regular triangulations with this class.
But I couldn't any similar class from CGAL for the 2D case and since I am new to the CGAL package, I am quite unsure on how to adapt the existing code database to help me with my case. Any help with adapting the code or pointers to the packages which can help me with implementing 2D Periodic regular triangulations will be appreciated!
There is currently no Periodic_2_regular_triangulation_2 class, and it is not straightforward to adapt Periodic_2_triangulation_2 to add weights. It is incidentally being done in https://github.com/CGAL/cgal/pull/4914 because that pull request completely rewrites Periodic_2_triangulation_2 in a way that will make it easy to add a weighted version, but this is WIP and without any ETA (I am missing time to finish it).
A possible way to go around that limitation would be to add a "fake" third dimension (z = 0) to your 2D data and use a weighted, periodic 3D triangulation?
I'm trying to use Polygonal Surface Reconstruction with building point cloud to create simplified building models.
I did first tests with this CGAL code example and got first promising results.
As an example, I used this point cloud with vertex normals correctly oriented and got the following result from PSR. Some faces are clearly inverted (dark faces with normals pointing inside the watertight mesh and therefore not visible).
I was wondering if there a way to fix this face orientation error. I've noticed orientation methods on Polygon mesh but I don't really know to apply them to the resulting PSR surface mesh. As far as logic is concerned making normal point outwards should not be too complicated I guess.
Thanks in advance for any help
You can use the function reverse_face_orientations in the Polygon mesh processing package.
Note that this package has several functions that can help you to correct/modify your mesh.
I am attempting to mesh a complicated design (~80,000 faces) for a microchannel heat sink, as pictured, and I would appreciate some advice. I have tried a range of different mesh controls (especially face sizing and body sizing), mesh settings and element sizes, and all have failed to produce a working mesh. The most common errors are shown in the linked picture, in particular the one regarding "The following surfaces cannot be meshed with acceptable quality. Try using a different element size or virtual topology." However, I have already reduced the element size to 2x10^-6 m, which takes two days to resolve before failure.
Unfortunately I cannot alter the geometry significantly, as it is imported from generation in SolidWORKS as either a STEP or an x.t file. As such, any advice for how I can successfully mesh the geometry for CFD analysis in FLUENT would be greatly appreciated.
I can provide more details or the geometry file itself if required.
Thanks in advance.
Meshing Attempt
Probably your cad design is not clean at all. But it is impossible to notice from this image. If you don't have control over the geometry source it is trouble. Because you might ask somebody else about check and fix something. First check you can do with your model it's trying to reduce the number of elements until the minimum possible value. Then if the mesh runs properly you can relay in the surfaces of your cad model. After that, you can refine the mesh, but the refining process is something that you have to do following some error criteria. If you are also the designer why not try to simplify a bit the geometry if you consider it is really hard to mesh? Meshing properly is a hard task, you should go step-by-step until you reach some solution. Also, you must not allow the preprocessor mesh automatically, without giving some criteria. Probably the first thing you have to answer even before apply any mesh is, what is your Reynolds number? And what is the most valuable result in which you can base the goodness of your discretization?
Thank you for your suggestions. In the end I solved the issue by importing the original mesh generated by COMSOL into SpaceClaim, then employing both the "Smooth" and "Reduce Faces" tools in tandem to simplify the geometry, before finally using SolidWORKS to turn the smoothed mesh into a solid body. This body retained many of the same features as the original, but was much less complex, having two orders of magnitude fewer faces. In turn, this permitted both meshing and heat transfer analysis in FLUENT.
I would like to generate visually appealing surface reconstruction from the the point clouds.
I am using point cloud library. I tried creating a mesh using poisson reconstruction method but later found that it gives a water tight reconstruction.
For example: In my case I have a point cloud of a room
Using the code at http://justpaste.it/code1 , I was able to get a reconstruction like this
(source: pcl-users.org)
The above picture has the surface which is covering the top view. This was visualized using MeshLab.
Then later on the MeshLab GUI when I press points, it looks like this.
(source: pcl-users.org)
But in the second picture there are points on its surface too(Not clearly visible in the attached picture).
Can you help in creating a model that has no points on the top and just has the inside structure ?
Any other suggestions to improve the reconstruction quality ?
The point cloud of the room and generated ply file can be downloaded from https://dl.dropboxusercontent.com/u/95042389/temp_pcd_ply_files.tar.bz2
One solution that works for me is obtaining a convex/concave hull of your point cloud. Then you can use this hull to filter/crop your mesh after Poisson reconstruction. If you use the PCL you can try ConvexHull or ConcaveHull together with CropHull and test the results. Hope this solves your issue, it did for me.
As far as my experience is concerned (meshing caves), meshing with Poisson will result in watertight model/mesh, which is why your model was covered entirely. I only deal with meshes using MeshLab but I am guessing it is the same thing. What I did try is using Ball-Pivoting meshing algorithm in MeshLab which result in non-watertight model. Maybe that is what you are looking for.
I have a set of points in 3D that lie on a surface and I also have the normals at every point.
I would like to generate a surface triangulation with this information. In addition I could tell the algorithm to use what points lie on the boundary if that is needed.
So, I have quite a bit of information:
* points
* normals
* boundary
How do I triangulate a surface with this information using vtk?
A surface reconstruction algorithm is like using a bomb for this problem since I have all this information that I would like to use. This information comes from a simulation so I know the surface exists and that is quite smooth.
I would like the answer to be cast in terms of either what vtk function to use and if available (and that would be great) examples using this function.
Thank you so much in advance.
You can use the vtkSurfaceReconstruction filter to create a surface from a set of 3D points.
You could try the point cloud library
Point Cloud Library
Just the 3D points would be good enough. Since you know that your surface is smooth, you can perform a Delaunay triangulation of the points (vtkDelaunay3D) and apply a subdivision filter for smoothening (vtkButterflySubdivisionFilter).
Delaunay3D triangulation