<div dir="ltr">Yes, this does result in the same problems as before.</div><div class="gmail_extra"><br><br><div class="gmail_quote">2014-08-12 15:06 GMT+02:00 Bill Lorensen <span dir="ltr"><<a href="mailto:bill.lorensen@gmail.com" target="_blank">bill.lorensen@gmail.com</a>></span>:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Have you tried:<br>
UseImageSpacingOn();<br>
<div class="HOEnZb"><div class="h5"><br>
<br>
On Tue, Aug 12, 2014 at 8:47 AM, Tobias Gass <<a href="mailto:gass@vision.ee.ethz.ch">gass@vision.ee.ethz.ch</a>> wrote:<br>
> Further update: when manually increasing the spacing of the input image<br>
> (without resampling, thus not changing the size), the filter works well.<br>
> This probably indicates numerical instability depending on image spacing<br>
> somewhere in the code, which would probably be great if it could be fixed.<br>
><br>
> Best,<br>
> Tobias<br>
><br>
><br>
> 2014-08-12 14:33 GMT+02:00 Tobias Gass <<a href="mailto:gass@vision.ee.ethz.ch">gass@vision.ee.ethz.ch</a>>:<br>
><br>
>> Update: meanwhile I also tried cropping the image to investigate the<br>
>> impact of the overall size (#voxels) of the image. It appears to not have an<br>
>> effect, the same artifacts occur when processing a significantly smaller<br>
>> region (1/2 size).<br>
>><br>
>><br>
>> 2014-08-12 10:44 GMT+02:00 Tobias Gass <<a href="mailto:gass@vision.ee.ethz.ch">gass@vision.ee.ethz.ch</a>>:<br>
>><br>
>>> Hi,<br>
>>><br>
>>> changing the PixelType did not yield any change. However, I did some<br>
>>> further tests which seem to point at image spacing or total size to be the<br>
>>> culprit. My original image is fairly high-res, with 0.25mm in-plane<br>
>>> resolution and 0.75mm slice thickness and 512x512x409 voxel size. I tried<br>
>>> resampling the image to different isotropic resolutions (0.25, 0.3, 0.34,<br>
>>> 0.35, 0.4, 0.5, 0.75, 1)mm isotropic spacing. For all spacings down to<br>
>>> 0.35mm, the AA filter converged nicely and the output looks as I'd expect<br>
>>> it. Lower than 0.35mm spacing, and it fails in the way I described (from<br>
>>> this spacing on the #vox is equal or greater than the #voxels in my original<br>
>>> image).<br>
>>><br>
>>> Any idea what might cause this, or what to further investigate?<br>
>>><br>
>>> Thanks,<br>
>>> Tobias<br>
>>><br>
>>><br>
>>> 2014-08-11 22:17 GMT+02:00 Bill Lorensen <<a href="mailto:bill.lorensen@gmail.com">bill.lorensen@gmail.com</a>>:<br>
>>><br>
>>>> Paraview should do the right thing.<br>
>>>><br>
>>>> On Mon, Aug 11, 2014 at 3:38 PM, Tobias Gass <<a href="mailto:gass@vision.ee.ethz.ch">gass@vision.ee.ethz.ch</a>><br>
>>>> wrote:<br>
>>>> > I will try that tomorrow morning (living in GMT+1). I used paraview to<br>
>>>> > extract the isosurface, but also looked at the output of the AA-filter<br>
>>>> > directly (which looked fuzzy). I can also post an image of that<br>
>>>> > tomorrow.<br>
>>>> ><br>
>>>> > Thanks for your help!<br>
>>>> > Tobias<br>
>>>> ><br>
>>>> ><br>
>>>> > 2014-08-11 21:22 GMT+02:00 Bill Lorensen <<a href="mailto:bill.lorensen@gmail.com">bill.lorensen@gmail.com</a>>:<br>
>>>> ><br>
>>>> >> How are you extracting the final isosurface?<br>
>>>> >><br>
>>>> >> On Mon, Aug 11, 2014 at 3:22 PM, Bill Lorensen<br>
>>>> >> <<a href="mailto:bill.lorensen@gmail.com">bill.lorensen@gmail.com</a>><br>
>>>> >> wrote:<br>
>>>> >> > If you change<br>
>>>> >> > typedef short PixelType;<br>
>>>> >> > to<br>
>>>> >> > typedef double PixelType;<br>
>>>> >> ><br>
>>>> >> > do the results change.<br>
>>>> >> ><br>
>>>> >> ><br>
>>>> >> > On Mon, Aug 11, 2014 at 2:47 PM, Tobias Gass<br>
>>>> >> > <<a href="mailto:gass@vision.ee.ethz.ch">gass@vision.ee.ethz.ch</a>><br>
>>>> >> > wrote:<br>
>>>> >> >><br>
>>>> >> >>> Can you post your code. The results do not look correct.<br>
>>>> >> >><br>
>>>> >> >><br>
>>>> >> >> Sure. The {write,read}Image functions are just convenient wrappers<br>
>>>> >> >> for<br>
>>>> >> >> the<br>
>>>> >> >> respective ITK tools.<br>
>>>> >> >><br>
>>>> >> >><br>
>>>> >> >> #include <stdio.h><br>
>>>> >> >><br>
>>>> >> >> #include <iostream><br>
>>>> >> >><br>
>>>> >> >> #include "argstream.h"<br>
>>>> >> >><br>
>>>> >> >> #include "ImageUtils.h"<br>
>>>> >> >><br>
>>>> >> >> #include "itkAntiAliasBinaryImageFilter.h"<br>
>>>> >> >><br>
>>>> >> >> using namespace std;<br>
>>>> >> >><br>
>>>> >> >> using namespace itk;<br>
>>>> >> >><br>
>>>> >> >><br>
>>>> >> >><br>
>>>> >> >><br>
>>>> >> >> int main(int argc, char ** argv)<br>
>>>> >> >><br>
>>>> >> >> {<br>
>>>> >> >><br>
>>>> >> >><br>
>>>> >> >> feenableexcept(FE_INVALID|FE_DIVBYZERO|FE_OVERFLOW);<br>
>>>> >> >><br>
>>>> >> >> typedef short PixelType;<br>
>>>> >> >><br>
>>>> >> >> typedef double OutputPixelType;<br>
>>>> >> >><br>
>>>> >> >> const unsigned int D=3;<br>
>>>> >> >><br>
>>>> >> >> typedef Image<PixelType,D> ImageType;<br>
>>>> >> >><br>
>>>> >> >><br>
>>>> >> >><br>
>>>> >> >> typedef ImageType::Pointer ImagePointerType;<br>
>>>> >> >><br>
>>>> >> >><br>
>>>> >> >><br>
>>>> >> >> typedef Image<OutputPixelType,D> OutputImageType;<br>
>>>> >> >><br>
>>>> >> >> typedef OutputImageType::Pointer OutputImagePointerType;<br>
>>>> >> >><br>
>>>> >> >><br>
>>>> >> >><br>
>>>> >> >> argstream * as=new argstream(argc,argv);<br>
>>>> >> >><br>
>>>> >> >> string inFile, outFile;<br>
>>>> >> >><br>
>>>> >> >> double thresh=0.0;<br>
>>>> >> >><br>
>>>> >> >> (*as) >> parameter ("in", inFile, " filename...", true);<br>
>>>> >> >><br>
>>>> >> >> (*as) >> parameter ("out", outFile, " filename...", true);<br>
>>>> >> >><br>
>>>> >> >><br>
>>>> >> >> (*as) >> help();<br>
>>>> >> >><br>
>>>> >> >> as->defaultErrorHandling();<br>
>>>> >> >><br>
>>>> >> >><br>
>>>> >> >> ImagePointerType img =<br>
>>>> >> >> ImageUtils<ImageType>::readImage(inFile);<br>
>>>> >> >><br>
>>>> >> >><br>
>>>> >> >> typedef itk::AntiAliasBinaryImageFilter <ImageType,<br>
>>>> >> >> OutputImageType><br>
>>>> >> >><br>
>>>> >> >> AntiAliasBinaryImageFilterType;<br>
>>>> >> >><br>
>>>> >> >><br>
>>>> >> >><br>
>>>> >> >> AntiAliasBinaryImageFilterType::Pointer antiAliasFilter<br>
>>>> >> >><br>
>>>> >> >> = AntiAliasBinaryImageFilterType::New ();<br>
>>>> >> >><br>
>>>> >> >> antiAliasFilter->SetInput(img);<br>
>>>> >> >><br>
>>>> >> >> antiAliasFilter->SetNumberOfIterations(5);<br>
>>>> >> >><br>
>>>> >> >> antiAliasFilter->SetMaximumRMSError(0.02);<br>
>>>> >> >><br>
>>>> >> >><br>
>>>> >> >> antiAliasFilter->Update();<br>
>>>> >> >><br>
>>>> >> >><br>
>>>> >> >> // For increased code coverage. Does nothing.<br>
>>>> >> >><br>
>>>> >> >> antiAliasFilter->GetMaximumRMSError();<br>
>>>> >> >><br>
>>>> >> >><br>
>>>> >> >><br>
>>>> >> >><br>
>>>> >> >> antiAliasFilter->Update();<br>
>>>> >> >><br>
>>>> >> >><br>
>>>> >> >> std::cout << "Maximum RMS change value threshold was: 0.02 " <<<br>
>>>> >> >> std::endl;<br>
>>>> >> >><br>
>>>> >> >> std::cout << "Last RMS change value was: " <<<br>
>>>> >> >> antiAliasFilter->GetRMSChange() << std::endl;<br>
>>>> >> >><br>
>>>> >> >><br>
>>>> >> >> std::cout<< antiAliasFilter->GetElapsedIterations() <<endl;<br>
>>>> >> >><br>
>>>> >> >><br>
>>>> >> >> std::cout<<antiAliasFilter->GetLowerBinaryValue()<<" "<<<br>
>>>> >> >> antiAliasFilter->GetUpperBinaryValue()<<std::endl;<br>
>>>> >> >><br>
>>>> >> >><br>
>>>> >> >> OutputImagePointerType out=antiAliasFilter->GetOutput();<br>
>>>> >> >><br>
>>>> >> >><br>
>>>> >> >> ImageUtils<OutputImageType>::writeImage(outFile,out);<br>
>>>> >> >><br>
>>>> >> >><br>
>>>> >> >> return 1;<br>
>>>> >> >><br>
>>>> >> >> }<br>
>>>> >> >><br>
>>>> >> >><br>
>>>> >> >> I tried different values for maxIter (1,2,3,4,5,10,100,1000) and<br>
>>>> >> >> maxRMSError<br>
>>>> >> >> (0.02 0.07 0.08 0.8), and the best visual results are achieved<br>
>>>> >> >> after<br>
>>>> >> >> one<br>
>>>> >> >> iteration where RMSE changes by about 0.08. This increases for all<br>
>>>> >> >> following<br>
>>>> >> >> iterations, which have an RMSE-change of about 0.25 for all number<br>
>>>> >> >> of<br>
>>>> >> >> iterations tested. All this is using ITK 4.5 pulled from the GIT<br>
>>>> >> >> repository<br>
>>>> >> >> a while ago, running on an intel i7 on Debian.<br>
>>>> >> >><br>
>>>> >> >> Thanks a lot!<br>
>>>> >> >> Tobias<br>
>>>> >> >><br>
>>>> >> >>><br>
>>>> >> >>><br>
>>>> >> >>> On Mon, Aug 11, 2014 at 12:18 PM, Tobias Gass<br>
>>>> >> >>> <<a href="mailto:gass@vision.ee.ethz.ch">gass@vision.ee.ethz.ch</a>><br>
>>>> >> >>> wrote:<br>
>>>> >> >>> > Dear all,<br>
>>>> >> >>> ><br>
>>>> >> >>> > I'm having problems using the anti-alias filter, which I want<br>
>>>> >> >>> > to use<br>
>>>> >> >>> > in<br>
>>>> >> >>> > order to smooth a binary segmentation volume. I'm using<br>
>>>> >> >>> > essentially<br>
>>>> >> >>> > the<br>
>>>> >> >>> > example code from the repository, which for my input file seems<br>
>>>> >> >>> > to<br>
>>>> >> >>> > start<br>
>>>> >> >>> > oscillating after 2 iterations, leading to weird 'pointy'<br>
>>>> >> >>> > artifacts<br>
>>>> >> >>> > on<br>
>>>> >> >>> > the<br>
>>>> >> >>> > surface as can be seen from the attached screenshot (AA filter<br>
>>>> >> >>> > output is<br>
>>>> >> >>> > on<br>
>>>> >> >>> > the left, input on the right).<br>
>>>> >> >>> ><br>
>>>> >> >>> > Is there any parameter to set/implement which could improve the<br>
>>>> >> >>> > behavior<br>
>>>> >> >>> > of<br>
>>>> >> >>> > this filter? I'm assuming that my input is reasonable, or is<br>
>>>> >> >>> > this<br>
>>>> >> >>> > something<br>
>>>> >> >>> > the filter is expected to have problems with? It is notable<br>
>>>> >> >>> > that<br>
>>>> >> >>> > such<br>
>>>> >> >>> > errors<br>
>>>> >> >>> > seem to mainly appear in 'flat' regions of the volume, and are<br>
>>>> >> >>> > not<br>
>>>> >> >>> > apparent<br>
>>>> >> >>> > in the curved shaft of the bone.<br>
>>>> >> >>> ><br>
>>>> >> >>> > Any help, or pointers to further resources are appreciated,<br>
>>>> >> >>> > thanks,<br>
>>>> >> >>> > Tobias<br>
>>>> >> >>> ><br>
>>>> >> >>> > _____________________________________<br>
>>>> >> >>> > Powered by <a href="http://www.kitware.com" target="_blank">www.kitware.com</a><br>
>>>> >> >>> ><br>
>>>> >> >>> > Visit other Kitware open-source projects at<br>
>>>> >> >>> > <a href="http://www.kitware.com/opensource/opensource.html" target="_blank">http://www.kitware.com/opensource/opensource.html</a><br>
>>>> >> >>> ><br>
>>>> >> >>> > Kitware offers ITK Training Courses, for more information<br>
>>>> >> >>> > visit:<br>
>>>> >> >>> > <a href="http://www.kitware.com/products/protraining.php" target="_blank">http://www.kitware.com/products/protraining.php</a><br>
>>>> >> >>> ><br>
>>>> >> >>> > Please keep messages on-topic and check the ITK FAQ at:<br>
>>>> >> >>> > <a href="http://www.itk.org/Wiki/ITK_FAQ" target="_blank">http://www.itk.org/Wiki/ITK_FAQ</a><br>
>>>> >> >>> ><br>
>>>> >> >>> > Follow this link to subscribe/unsubscribe:<br>
>>>> >> >>> > <a href="http://public.kitware.com/mailman/listinfo/insight-users" target="_blank">http://public.kitware.com/mailman/listinfo/insight-users</a><br>
>>>> >> >>> ><br>
>>>> >> >>><br>
>>>> >> >>><br>
>>>> >> >>><br>
>>>> >> >>> --<br>
>>>> >> >>> Unpaid intern in BillsBasement at noware dot com<br>
>>>> >> >><br>
>>>> >> >><br>
>>>> >> ><br>
>>>> >> ><br>
>>>> >> ><br>
>>>> >> > --<br>
>>>> >> > Unpaid intern in BillsBasement at noware dot com<br>
>>>> >><br>
>>>> >><br>
>>>> >><br>
>>>> >> --<br>
>>>> >> Unpaid intern in BillsBasement at noware dot com<br>
>>>> ><br>
>>>> ><br>
>>>><br>
>>>><br>
>>>><br>
>>>> --<br>
>>>> Unpaid intern in BillsBasement at noware dot com<br>
>>>> _____________________________________<br>
>>>> Powered by <a href="http://www.kitware.com" target="_blank">www.kitware.com</a><br>
>>>><br>
>>>> Visit other Kitware open-source projects at<br>
>>>> <a href="http://www.kitware.com/opensource/opensource.html" target="_blank">http://www.kitware.com/opensource/opensource.html</a><br>
>>>><br>
>>>> Kitware offers ITK Training Courses, for more information visit:<br>
>>>> <a href="http://www.kitware.com/products/protraining.php" target="_blank">http://www.kitware.com/products/protraining.php</a><br>
>>>><br>
>>>> Please keep messages on-topic and check the ITK FAQ at:<br>
>>>> <a href="http://www.itk.org/Wiki/ITK_FAQ" target="_blank">http://www.itk.org/Wiki/ITK_FAQ</a><br>
>>>><br>
>>>> Follow this link to subscribe/unsubscribe:<br>
>>>> <a href="http://public.kitware.com/mailman/listinfo/insight-users" target="_blank">http://public.kitware.com/mailman/listinfo/insight-users</a><br>
>>><br>
>>><br>
>><br>
><br>
<br>
<br>
<br>
--<br>
Unpaid intern in BillsBasement at noware dot com<br>
</div></div></blockquote></div><br></div>