[Insight-developers] Failing Tests

[Blezek, Daniel J (Research)]

I think the itkToboganImageFilterTest output should be fine.
-dan

-----Original Message-----
From: Luis Ibanez [mailto:luis.ibanez at kitware.com]
Sent: Wednesday, January 19, 2005 2:14 PM
To: Lorensen, William E (Research)
Cc: Insight Developers List
Subject: Re: [Insight-developers] Failing Tests



Bill,

 From the 24 tests that are failing in zion, here is the list
of tests for which the new output seems to be reasonable. Meaning
that the output differs from the previous baseline just by a couple
of pixels, or by features that are in the uncertainty area where we
cannot say that one image is more correct than the baseline.


- CurvesLevelSetImageFilterTest
- DeformableRegistration4Test
- DeformableRegistration5Test
- DeformableRegistration6Test
- FastMarchingLeftVentricleTest
- FastMarchingRightVentricleTest
- GeodesicActiveContourGrayMatterTest
- GeodesicActiveContourLeftVentricleTest
- GeodesicActiveContourRightVentricleTest
- GeodesicActiveContourShapePriorCallosumTest
- GeodesicActiveContourWhiteMatterTest
- ImageRegistration12Test
- ImageRegistration5Test1
- ImageRegistration5Test2
- ImageRegistration9Test
- itkLaplacianRecursiveGaussianImageFilterTest
- ShapeDetectionLevelSetGrayMatterTest
- ShapeDetectionLevelSetLeftVentricleTest
- ShapeDetectionLevelSetRightVentricleTest (here the new image
                           is better than the previous baseline)
- ShapeDetectionLevelSetWhiteMatterTest





The tests that have some issues are:



- FastMarchingWhiteMatter deserves a bit of fine tunning
    since the segmentation is extracting pieces that are
    visibly of graymatter or csf.

- ImageReigstration3Test is stopping too early...

- itkRecursiveGaussianImageFiltersTest needs to change
   a numerical test in the code. The new number is ok,
   it just differs by some decimals.

- itkToboganImageFilterTest looks quite different, but
   I'm not sure if the new output is better or worst than
   the original baseline.





   Luis




--------------------------------------
Lorensen, William E (Research) wrote:

> Luis,
> I would like to clear up the failing tests introduced by the gaussian
> changes. Do you think the new results are correct. If so, I'll populate
the
> baseline regression test results with new images.
> 
> Bill
> _______________________________________________
> Insight-developers mailing list
> Insight-developers at itk.org
> http://www.itk.org/mailman/listinfo/insight-developers
> 
> 




_______________________________________________
Insight-developers mailing list
Insight-developers at itk.org
http://www.itk.org/mailman/listinfo/insight-developers