On 21 Oct 2010, at 15:20, JuanPi wrote: > Hi c. > > It may be that there is a difference. I cannot fully understand the > way you use the ismember function. To me, it looks you are comparing a > matrix with integer numbers (the triangle vertices) against a logical > matrix with 1's where the vertex of a triangle is inside the region or > in the border. I just read the help of ismember and this will just > give you the triangles that have the vertex "1". Am I right? Here a > simplified example > > % an abstract triangulation in your format > A = [1 2 3; 2 4 3; 4 5 3]'; > % the result of the distance test: vertex 3 and 1 are inside the > region > res = [1 0 1 0 0]; > B = reshape(ismember(A,res),3,3) > B = > 1 0 0 > 0 0 0 > 0 0 0 > > then the operation the functions "all" and "find" give > find(all(B)) > ans = [](1x0) > > However, now that I now the function "ismember" I may think of a use > to it! Thanks a lot.
sorry, as I said that had not been tested, I had misplaced the 'find' command: try the following: >> pkg load msh fpl >> msh = msh2m_structured_mesh(linspace (0,1,100), linspace (0,1,100), 1, 1:4); >> xc = yc = .5; d = .1; >> triangles = all (reshape (ismember (msh.t (1:3, :), find ((msh.p(1,:) - xc).^2 + (msh.p(2,:) - yc).^2 <= d^2)), size (msh.t(1:3, :)))); >> msh.t (4, triangles) = 2; >> submsh = msh2m_submesh (msh, [], 2); >> subplot (1, 2, 1) >> pdemesh (msh.p, msh.e, msh.t) >> subplot (1, 2, 2) >> pdemesh (submsh.p, submsh.e, submsh.t) be careful, gnuplot would take a long time to plot this mesh, so I recommend doing >> close all >> backend ('fltk') before the plot part > I been checking the package mesh, it is pretty nice, but for my > objective, I would be force to create the .geo description of my > region (right?)... I'm not sure, I don't think so. > which is the result of a nonlinear mapping on the > unit hemisphere... no really easy, jeje. I am trying to merge alpha > shapes from CGAL or some functions from iso2mesh to get good meshes on > these deformed spherical sections; relying only on sampled points on > the region and its border. I'm not sure I understand what you want to do, so I am afraid I can't be of much help here. What I wanted to say is simply that, if you develop yourself some algorithms for handling triangular or tetrahedral meshes that are currently missing in msh, it woul be very nice if you contribute them to the community. > Thanks again > JPi c. ------------------------------------------------------------------------------ Nokia and AT&T present the 2010 Calling All Innovators-North America contest Create new apps & games for the Nokia N8 for consumers in U.S. and Canada $10 million total in prizes - $4M cash, 500 devices, nearly $6M in marketing Develop with Nokia Qt SDK, Web Runtime, or Java and Publish to Ovi Store http://p.sf.net/sfu/nokia-dev2dev _______________________________________________ Octave-dev mailing list Octave-dev@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/octave-dev