[vtkusers] Creating a contour from volume data

[Mathieu Malaterre]

Hi Rene,
  Could you try a :
skinExtractor->Update()

before displaying any results.

HTH
mathieu


Rene Tschirley wrote:

> Dear VTK gurus,
>
> I'm quite near to despairation. With upgrading to VTK 4.0 (I used to work
> with 2.3 and 3.2 by now) I rearranged some applications. But now, I seem
> to be unable to create a simple isosurface from a volume dataset as it
> worked quite well with 2.3 and 3.2. It's a pity that there are so few
> examples for 4.0, especially the most interesting examples have been
> removed, e.g. medical[1-3].cxx, so I have nothing to look at how things
> work well.
>
> Here's the relevant C++ code:
>
>   vtkImageReader *reader = vtkImageReader::New();
>   reader->SetFilePrefix(filename);
>   reader->SetFileNameSliceOffset(sliceoffset);
>   reader->SetFilePattern(pattern);
>   reader->SetDataExtent(0, xsize-1, 0, ysize-1, 0, zsize-1);
>   reader->SetDataScalarTypeToUnsignedChar();
>   reader->SetDataSpacing (xspacing, yspacing, zspacing);
>   reader->SetDataMask(mask);
>   reader->Update();
>   cout << "Reader has "
>        << reader->GetOutput()->GetNumberOfPoints()
>        << "points.\n"; // at runtime this tells me that there are
>                        // 13107200 data cells available which is
>                        // correct for 512*512*50 @ 8bpp
>
>   vtkImageShrink3D *shrinker = vtkImageShrink3D::New();
>   shrinker->SetShrinkFactors(xshrink,yshrink,zshrink);
>   shrinker->SetInput(reader->GetOutput());
>   shrinker->Update();  // a PrintSelf() makes me sure that I didn't loose
>                        // any points yet.
>
>   vtkMarchingCubes *skinExtractor = vtkMarchingCubes::New();
>   skinExtractor->SetInput(shrinker->GetOutput());
>   skinExtractor->SetValue(0, isolevel);
>   cout << "Creating " << skinExtractor->GetOutput()->GetNumberOfPolys()
>        << " polygons.\n";
>                        // at runtime this tells me that there are
>                        // 0 polygons created which is obviously not what
>                        // I wanted.
>
> After that I use vtkDecimatePro and vtkPolyDataMapper to continue the
> pipeline, but without any polygons, the pipeline is not of any use.
>
> Below follows the PrintSelf() output of reader->GetOutput() and Marching
> Cubes object if this is of any help.
>
> This part of the pipeline worked well with VTK 2.3 and 3.2 so I wonder if
> there is any little nifty mistake in it. I hope somebody of you will
> point it out.
>
> Yours,
>         René
>
> PrintSelf of reader->GetOutput:
> Debug: Off
> Modified Time: 128
> Reference Count: 1
> Registered Events: (none)
> Source: 0x8055328
> Release Data: Off
> Data Released: False
> Global Release Data: Off
> MaximumNumberOfPieces: -1
> PipelineMTime: 122
> UpdateTime: 129
> UpdateExtent: Not Initialized
> Update Number Of Pieces: 1
> Update Piece: 0
> Update Ghost Level: 0
> RequestExactExtent: Off
>  UpdateExtent: 0, 511, 0, 511, 0, 49
> WholeExtent: 0, 511, 0, 511, 0, 49
> Field Data:
>   Debug: Off
>   Modified Time: 71
>   Reference Count: 1
>   Registered Events: (none)
>   Number Of Arrays: 0
>   Number Of Components: 0
>   Number Of Tuples: 0
> Locality: 0
> NumberOfConsumers: 0
> ExtentTranslator: (0x8055680)
> MaximumNumberOfPieces: -1
> Number Of Points: 13107200
> Number Of Cells: 12794929
> Cell Data:
>   Debug: Off
>   Modified Time: 75
>   Reference Count: 1
>   Registered Events: (none)
>   Number Of Arrays: 0
>   Number Of Components: 0
>   Number Of Tuples: 0
>   Copy Flags: ( 1 1 1 1 1 )
>   Scalars: (none)
>   Vectors: (none)
>   Normals: (none)
>   TCoords: (none)
>   Tensors: (none)
> Point Data:
>   Debug: Off
>   Modified Time: 128
>   Reference Count: 1
>   Registered Events: (none)
>   Number Of Arrays: 1
>   Array 0 name = ImageValues
>   Number Of Components: 1
>   Number Of Tuples: 13107200
>   Copy Flags: ( 1 1 1 1 1 )
>   Scalars:
>     Debug: Off
>     Modified Time: 125
>     Reference Count: 1
>     Registered Events: (none)
>     Name: ImageValues
>     Number Of Components: 1
>     Number Of Tuples: 13107200
>     Size: 13107200
>     MaxId: 13107199
>     LookupTable: (none)
>     Array: 0x4254a008
>   Vectors: (none)
>   Normals: (none)
>   TCoords: (none)
>   Tensors: (none)
> Bounds:
>   Xmin,Xmax: (0, 511)
>   Ymin,Ymax: (0, 511)
>   Zmin,Zmax: (0, 49)
> Compute Time: 0
> Release Data: Off
> ScalarType: 3
> NumberOfScalarComponents: 1
> Spacing: (1, 1, 1)
> Origin: (0, 0, 0)
> Dimensions: (512, 512, 50)
> Increments: (1, 512, 262144)
> Extent: (0, 511, 0, 511, 0, 49)
> WholeExtent: (0, 511, 0, 511, 0, 49)
>
> PrintSelf of skinExtractor (vtkMarchingCubes):
> number of contours = 1
> Debug: Off
> Modified Time: 153
> Reference Count: 2
> Registered Events: (none)
> Number Of Required Inputs: 1
> Input 0: (0x80553c0)
> AbortExecute: Off
> Progress: 1
> Progress Text: (None)
> Output 0: (0x8058a98)
> Contour Values:
>   Value 0: 30
> Compute Normals: On
> Compute Gradients: Off
> Compute Scalars: On
> Locator:0x8059060
>   Debug: Off
>   Modified Time: 153
>   Reference Count: 1
>   Registered Events: (none)
>   DataSet: (none)
>   Automatic: On
>   Tolerance: 0.001
>   Level: 8
>   MaxLevel: 8
>   Retain Cell Lists: On
>   Build Time: 0
>   Number of Points Per Bucket: 3
>   Divisions: (42, 42, 42)
>
> --
> Dipl.-Inform. René Tschirley             http://www.cs.tu-berlin.de/~pooh
> TU Berlin, Computer Graphics and Computer Assisted Medicine research group
>
>   ------------------------------------------------------------------------
>    Part 1.2Type: application/pgp-signature

--
If all else fail read the instructions
Malaterre, Mathieu
The John P. Robarts Research Institute
Imaging Research Laboratories
http://www.imaging.robarts.ca/~mmalat