[ITK-users] itkImageFileWriter + ExtractSliceImageFilter: largest region exception
Marcos
fotosentido at gmail.com
Sun Sep 13 15:58:37 EDT 2015
Hi,
I can't use unsigned char reading files, cause I'm seeing horrible views on
my viewer for some files.
I'm gonna have to avoid using the itk exporter, and start using the vtk
exporter, which doesn't have these problems.
2015-09-09 18:21 GMT+02:00 Marcos <fotosentido at gmail.com>:
> Ok, finally, I got it.
>
> I left the duplicated thing, but it wasn't the pipeline.
>
> I didn't see a warning from compiler, and the exception message didn't
> give me that clue:
> the pixeltype I was using.
>
>
> A lot of examples are using *unsigned short *or *signed short*, right? So
> I was.
> Reviewing the source code on the imageio classes:
>
> "BMPImageIO supports unsigned char only"
> "JPEG supports unsigned char/int only"
> "PNG supports unsigned char and unsigned short"
>
> I just changed to *unsigned char*, and everything works fine.
> Thanks for all the ideas, I really appreciate it. It's not easy trying to
> help without all the info.
> I didn't realise the problem was outside my method. Sorry.
>
> It takes some time to get things working, but it's great when it happens.
>
>
> 2015-09-09 17:36 GMT+02:00 Marcos <fotosentido at gmail.com>:
>
>> Hi,
>>
>> GetBufferedRegion gives me the same numbers as GetLargestPossibleRegion:
>> index: 0, 0, 0
>> size: 512, 512, 12
>>
>> I also tried, just after reading the image, using ImageDuplicator:
>>
>> reader = ReaderType::New();
>>
>> reader->SetFileName(fileName);
>>
>>
>> dicomIO = itk::GDCMImageIO::New();
>>
>>
>> typedef itk::ImageDuplicator< ImageType > DuplicatorType;
>>
>> DuplicatorType::Pointer duplicator = DuplicatorType::New();
>>
>>
>>
>> reader->SetImageIO(dicomIO);
>>
>> try
>>
>> {
>>
>> reader->Update();
>>
>> image = reader->GetOutput();
>>
>>
>> // duplicar imagen
>>
>> duplicator->SetInputImage(image);
>>
>> duplicator->Update();
>>
>> imageToSave = duplicator->GetOutput();
>>
>> }
>>
>> catch
>>
>> ...
>>
>>
>>
>> None worked.
>>
>> I saw the ExtractImageFilter example, so I changed to extracter->SetDirectionCollapseToSubmatrix() instead of SetDirectionCollapseToIdentity().
>>
>> The filename is "C:/Users/M/r.png", in Windows.
>> Nothing. Same error.
>>
>> bmp: itkbmpimageio.cxx:727
>> jpg: itkjpegimageio.cxx:454
>> png: itkpngimageio.cxx:521
>>
>> By the way, it would be better if the exception messages were captured in my writer's try catch. I have to dive into these sources.
>>
>>
>> 2015-09-08 22:05 GMT+02:00 Bradley Lowekamp <blowekamp at mail.nih.gov>:
>>
>>> Hello,
>>>
>>> You can use the Image::GetBufferedRegion method and print the resulting
>>> ImageRegion.
>>>
>>> This looks like there is a pipelining issue to me. I'd recommend trying
>>> to simplify by not reusing the Image data object. Yes you will need to
>>> recompute, but it may help to narrow down the problem.
>>>
>>> Brad
>>>
>>>
>>>
>>> On Sep 8, 2015, at 3:06 PM, Marcos <fotosentido at gmail.com> wrote:
>>>
>>> Hi,
>>>
>>> I know the input and desired regions, not sure how to look for the
>>> buffered region.
>>>
>>> 1. Reviewing my code I saw I didn't set the input image for the
>>> extracter, so I added:
>>> extracter->SetInput(image);
>>>
>>> Good, right? But no. Still the same error.
>>>
>>> 2. Then I tried switching to itkExtractImageFilter, and I got:
>>> Exception at 0x7633c42d, code: 0xe06d7363: C++ exception, flags=0x1
>>> (execution cannot be continued) (first chance) at
>>> c:\itk-4.8.0-installed\include\itk-4.8\itkextractimagefilter.hxx:242
>>>
>>> So I had to look inside this code (why don't throw me more info in the
>>> try catch?) to see what's happening, and I read this:
>>> itkExceptionMacro( << "It is required that the strategy for collapsing
>>> the direction matrix be explicitly specified. " << "Set with either
>>> myfilter->SetDirectionCollapseToIdentity() or
>>> myfilter->SetDirectionCollapseToSubmatrix() " << typeid( ImageBase<
>>> InputImageDimension > * ).name() );
>>> 3. I also tried to insert a:
>>> image->Update();
>>> or
>>> image->UpdateOutputData();
>>> with same results.
>>>
>>> I was thinking about passing the reader, but I'd like to apply changes
>>> to my image, so this method would receive a modified copy (no DeepCopy
>>> method like in vtk?)
>>>
>>>
>>> Yes, the input image is connected to a pipeline:
>>> itk::Image -> itk::ImageToVTKImageFilter -> vtk::vtkImageFlip ->
>>> vtkImageData -> vtkImageViewer2
>>>
>>>
>>> Thanks again for your interest.
>>>
>>>
>>> 2015-09-08 16:16 GMT+02:00 Bradley Lowekamp <blowekamp at mail.nih.gov>:
>>>
>>>> Hello,
>>>>
>>>> What is the buffered region for the input image? Is the input image
>>>> connected to a pipeline?
>>>>
>>>> Perhaps you just forgot to update the input image before running this
>>>> function?
>>>>
>>>> Brad
>>>>
>>>> On Sep 7, 2015, at 5:21 PM, Marcos <fotosentido at gmail.com> wrote:
>>>>
>>>> Hi,
>>>>
>>>> I have a method that was exporting fine one frame from my itk::Image.
>>>>
>>>> My method:
>>>>
>>>> void Exporter::exportOriginal(QString fileName, ImageType::Pointer image, int frame)
>>>>
>>>> {
>>>>
>>>> typedef itk::ImageFileWriter<FrameImageType> WriterType;
>>>>
>>>> WriterType::Pointer writer = WriterType::New();
>>>>
>>>>
>>>> writer->SetFileName(fileName.toStdString().c_str());
>>>>
>>>>
>>>> ImageType::RegionType inputRegion = image->GetLargestPossibleRegion();
>>>>
>>>>
>>>> ImageType::IndexType index = inputRegion.GetIndex();
>>>>
>>>> ImageType::SizeType size = inputRegion.GetSize();
>>>>
>>>>
>>>> index[0] = 0;
>>>>
>>>> index[1] = 0;
>>>>
>>>> index[2] = frame;
>>>>
>>>>
>>>> // 3D to 2D
>>>>
>>>> size[2] = 0;
>>>>
>>>>
>>>> ImageType::RegionType outputRegion;
>>>>
>>>> outputRegion.SetIndex(index);
>>>>
>>>> outputRegion.SetSize(size);
>>>>
>>>>
>>>> typedef itk::Testing::ExtractSliceImageFilter<ImageType, FrameImageType> ExtractType;
>>>>
>>>> ExtractType::Pointer extracter = ExtractType::New();
>>>>
>>>> extracter->SetExtractionRegion(outputRegion);
>>>>
>>>>
>>>> writer->SetInput(extracter->GetOutput());
>>>>
>>>> try
>>>>
>>>> {
>>>>
>>>> writer->Update();
>>>>
>>>> }
>>>>
>>>> catch (itk::ExceptionObject &ex)
>>>>
>>>> {
>>>>
>>>> std:cerr << ex << std::endl;
>>>>
>>>> }
>>>>
>>>> }
>>>>
>>>>
>>>> as inputRegion, I have:
>>>> index: [0, 0, 0]
>>>> size: [512, 512, 12]
>>>>
>>>> as outputRegion, I have:
>>>> index: [0, 0, 1] # first frame
>>>> size: [512, 512, 0]
>>>>
>>>>
>>>>
>>>> But I'm not sure what happened, now with every image I got an exception:
>>>> itkimagefilewriter.hxx:287: "Largest possible region does not fully
>>>> contain requested paste IO region"
>>>>
>>>> By the way, my try catch is not getting the exception message, I had to
>>>> look inside the writer class.
>>>> Any ideas? Thank you.
>>>> _____________________________________
>>>> Powered by www.kitware.com
>>>>
>>>> Visit other Kitware open-source projects at
>>>> http://www.kitware.com/opensource/opensource.html
>>>>
>>>> Kitware offers ITK Training Courses, for more information visit:
>>>> http://www.kitware.com/products/protraining.php
>>>>
>>>> Please keep messages on-topic and check the ITK FAQ at:
>>>> http://www.itk.org/Wiki/ITK_FAQ
>>>>
>>>> Follow this link to subscribe/unsubscribe:
>>>> http://public.kitware.com/mailman/listinfo/insight-users
>>>>
>>>>
>>>>
>>> _____________________________________
>>> Powered by www.kitware.com
>>>
>>> Visit other Kitware open-source projects at
>>> http://www.kitware.com/opensource/opensource.html
>>>
>>> Kitware offers ITK Training Courses, for more information visit:
>>> http://www.kitware.com/products/protraining.php
>>>
>>> Please keep messages on-topic and check the ITK FAQ at:
>>> http://www.itk.org/Wiki/ITK_FAQ
>>>
>>> Follow this link to subscribe/unsubscribe:
>>> http://public.kitware.com/mailman/listinfo/insight-users
>>>
>>>
>>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/insight-users/attachments/20150913/68a58c58/attachment.html>
More information about the Insight-users
mailing list