On 8 December 2010 13:53, Wolfgang Bangerth <[email protected]> wrote:
> > > I could then go through every possible voxel. If the data indicates that > it > > is a cell, I could then use the list of vertices in the equivalent cell > in > > the generated grid above and store this list in the appropriate place in: > > > > int cell_vertices[][GeometryInfo<dim>::vertices_per_cell > > <http://www.dealii.org/6.3.1/doxygen/deal.II/structGeometryInfo.html>] > > > > > > Is there a way to obtain the list/numbering of vertices of a cell? > > Yes: > for (unsigned int v=0; v<GeometryInfo<dim>::vertices_per_cell; ++v) > std::cout << cell->vertex_index(v) << std::endl; > If you want the coordinates, use cell->vertex(v) instead. > > > > Is this a reasonable way to do it or is there a cleaner and better way? > > Yes, though you would still first create an extremely large mesh of 86M > cells > if I understand you correctly. That's going to take forever. How many of > these 86M cells are actually present in your MRI data? > > In the MRI data, there're approximately 15M valid cells. You're right. It took forever when I tried to create the mesh with 86M cells. So, this avenue is definitely not the way to go. It did work on a very small test mesh, though. Ted
_______________________________________________ dealii mailing list http://poisson.dealii.org/mailman/listinfo/dealii
