Re: [vmtk-users] vmtkcenterlines never return
Brought to you by:
davidsteinman,
lucantiga
From: Luca A. <luc...@or...> - 2011-12-23 10:14:00
|
Hi Lingfeng, for those cases you might want to add a subdivision step right before centerlines (note that this might lead to a large surface and longer computation times): vmtksurfaceconnectivity -ifile airway.vtp --pipe vmtksurfacesmoothing -passband 0.01 -iterations 30 --pipe vmtksurfacesubdivision -method butterfly --pipe vmtkcenterlines -ofile airway_centerlines.vtp There are a few degenerate cases that still lead to infinite loops in certain occasions. When I get to it, this issue will likely be solved at its root. I'll keep you posted on this one. Thanks a lot for your input. Luca On Dec 23, 2011, at 11:04 AM, Lingfeng Zhuang wrote: > Thank you Luca, > > It is much better than before. In most cases, it can produce > centerline. However, if I select all branch distal points, it would > not return. > > Does vmtk support timeout? If the time took for calculating centerline > between two points is longer than expected, vmtk will ignore the two > points and continue to calculate others. > > Best regards, > > Jerry > > On Fri, Dec 23, 2011 at 5:11 PM, Luca Antiga <luc...@or...> wrote: >> Hi Linfeng, >> the surface probably leads to degenerate Voronoi polygons somewhere in the >> smaller branches. >> Try smoothing the surface first, using >> vmtksurfacesmoothing -ifile airway.vtp -passband 0.01 -iterations 30 -ofile airway_smooth.vtp >> >> maybe try with -passband 0.1 if the smoothing is too strong with 0.01. >> >> I also suggest to get rid of the disconnected segments using vmtksurfaceconnectivity. >> >> You can do everything in a single pipe >> >> vmtksurfaceconnectivity -ifile airway.vtp --pipe vmtksurfacesmoothing -passband 0.01 -iterations 30 --pipe vmtkcenterlines -ofile airway_centerlines.vtp >> >> Let me know if this helps. >> >> >> Luca >> >> >> On Dec 22, 2011, at 4:29 PM, Lingfeng Zhuang wrote: >> >>> Hi >>> >>> I am trying to extract centerline for the polydata attached. But after >>> picking start point and end points, the program did not return in two >>> hours. Python took 100% CPU usage in this period. >>> >>> The command used is vmtkcenterlines -ifile airway.vtp -ofile foo_centerlines.vtp >>> >>> Please help. >>> >>> Best regards, >>> >>> Jerry >>> <airway.vtp><endpoints.JPG><startpoints.JPG>------------------------------------------------------------------------------ >>> Write once. Port to many. >>> Get the SDK and tools to simplify cross-platform app development. Create >>> new or port existing apps to sell to consumers worldwide. Explore the >>> Intel AppUpSM program developer opportunity. appdeveloper.intel.com/join >>> http://p.sf.net/sfu/intel-appdev_______________________________________________ >>> vmtk-users mailing list >>> vmt...@li... >>> https://lists.sourceforge.net/lists/listinfo/vmtk-users >> > > ------------------------------------------------------------------------------ > Write once. Port to many. > Get the SDK and tools to simplify cross-platform app development. Create > new or port existing apps to sell to consumers worldwide. Explore the > Intel AppUpSM program developer opportunity. appdeveloper.intel.com/join > http://p.sf.net/sfu/intel-appdev > _______________________________________________ > vmtk-users mailing list > vmt...@li... > https://lists.sourceforge.net/lists/listinfo/vmtk-users |