[ITK-users] TileImageFilter SideBySide

Wed Aug 20 07:59:40 EDT 2014

Thanks Jim for the great explanation, I understand now.
So in simple CLI programs where an ITK pipeline is only updated once by the 
final writer: In order to limit maximum memory used during execution, would it 
be best to always store the output of a filter in the pipeline in an image 
object and then call DisconnectPipeline (and setting ReleaseDataFlagOn for the 
filter)?

Many thanks again for the answer.
Roman

On 20/08/14 13:16, Jim Miller wrote:
> Once DisconnectPipeline is called the Image object is completely
> disassociated with the source filter (in this case a reader). The source
> filter then constructs a new a Image object to serve as its output. For
> instance, if you call GetOutput() after calling DisconnectPipeline, you will
> get a different Image object than one retrieved before DisconnectPipeline is
> called.
>
> The pixel buffer is managed by the Image object, not the
> source/filter/reader. So after a call to DisconnectPipeline(), the reader has
> a new Image object as it its output and that new Image object has a distinct
> pixel buffer. Initially this pixel buffer is not even allocated. The act of
> reading the next image causes the reader to request the Image object to
> allocate the pixel buffer to an appropriate size.
>
> If DisconnectPipeline() was not called and the reader was asked to read
> another image of the same size or smaller, its output Image object would
> already have an allocated pixel buffer. The reader would still ask the Image
> object to allocate a pixel buffer of the appropriate size. But in this case,
> the Image object would recognize that it already has a block of memory of
> sufficient size and not do another allocation.
>
> Jim
>
>> On Aug 20, 2014, at 3:37 AM, "Dr. Roman Grothausmann"
>> <grothausmann.roman at mh-hannover.de> wrote:
>>
>> Hi
>>
>>
>> As this is a nice case of a question I have always had: In the case of the
>> c++ example
>> (http://www.itk.org/Wiki/ITK/Examples/ImageProcessing/TileImageFilter_CreateVolume)
>>
>>
>>
InputImageType::Pointer inputImageTile;
>>
>> for (int i = 1; i < argc - 1; i++) { reader->SetFileName( argv[i] );
>> reader->UpdateLargestPossibleRegion(); inputImageTile =
>> reader->GetOutput(); inputImageTile->DisconnectPipeline(); tiler->SetInput(
>> inputImageNumber++, inputImageTile ); }
>>
>> When the reader output is assigned to inputImageTile, I thought this is
>> only a pointer assignment and not a full copy of the image. However, I
>> wonder, what makes sure the address space in the memory is not used
>> (partially) again for the next image read? Is there a full copy created
>> "behind the scene" at any point in the code? Or does the reader remember
>> the address space of former images read even though DisconnectPipeline was
>> called?
>>
>> Thanks for any explanations Roman
>>
>>> On 18/08/14 17:01, Michka Popoff wrote: Hi
>>>
>>> Going the way described in the CreateVolume example is the way to go.
>>>
>>> Why did you uncomment the DisconnectPipeline() call and Update() call in
>>> the for loop ? They need to stay there, else you will always use the last
>>> image. That’s why you see the same image, the single update call will be
>>> triggered at the end of the script and read only one image.
>>>
>>> You don’t need to call tileFilter->Update(), you can remove this line.
>>> The writer->Update() will take care of the pipeline update. As a bonus,
>>> you may wrap your writer->Update() call, to fetch errors at the end of
>>> the pipeline:
>>>
>>> try { writer->Update(); } catch( itk::ExceptionObject & error ) {
>>> std::cerr << "Error: " << error << std::endl; return EXIT_FAILURE; }
>>>
>>> I made a little Python prototype, the syntax is a little bit different
>>> than in C++ but it gives you the main idea.
>>>
>>> Michka
>>>
>>>
>>>
>>>
>>> On 18 août 2014, at 16:05, Dan Mercik <mercikd at thomas.edu
>>> <mailto:mercikd at thomas.edu>> wrote:
>>>
>>>> Good Morning,
>>>>
>>>>
>>>>
>>>> The demonstration you showed me works great, however if I wanted to do
>>>> 20 images that would take  some time. So, I tried doing a for loop
>>>> which didn't work exactly how I wanted it to work. I would like the
>>>> program to take any amount of images and output them side by side , any
>>>> ideas? The attachment shows the for loop I used similar to
>>>> TileImageFilter CreateVolume example however when I use it for some odd
>>>> reason only one image is being display side by side multiple times.
>>>>
>>>>
>>>>
>>>> Daniel Mercik
>>>>
>>>> Computer Science Major
>>>>
>>>> Service Desk Technician
>>>>
>>>> Thomas College
>>>>
>>>> ________________________________ From: Michka Popoff
>>>> <michkapopoff at gmail.com <mailto:michkapopoff at gmail.com>> Sent:
>>>> Wednesday, August 13, 2014 2:32 AM To: Dan Mercik Cc: ITK Mailing
>>>> List;community at itk.org <mailto:community at itk.org> Subject: Re:
>>>> [ITK-users] TileImageFilter SideBySide
>>>>
>>>> Hi
>>>>
>>>>
>>>> Moving this mail to the new community mailing list
>>>>
>>>>
>>>> For your problem, you should use the setLayout method [1]
>>>>
>>>>
>>>> For 4 images, that would be:
>>>>
>>>>
>>>>
>>>> itk::FixedArray< unsigned int, 2 > layout;
>>>>
>>>> layout[0] = 4; layout[1] = 0;
>>>>
>>>>
>>>> Michka
>>>>
>>>>
>>>> [1]
>>>> http://www.itk.org/Wiki/ITK/Examples/ImageProcessing/TileImageFilter_SideBySide
>>>>
>>>>
>>>>
On 13 ao?t 2014, at 01:06, Dan Mercik <mercikd at thomas.edu
>>>> <mailto:mercikd at thomas.edu><mailto:mercikd at thomas.edu>> wrote:
>>>>
>>>> Good Afternoon,
>>>>
>>>>
>>>>
>>>> I want to know if it's possible to have more than two image be side by
>>>> side vertically? If so what would I have to change in the algorithm,
>>>> would I have to make a loop for the amount of input images? Any help
>>>> would be greatly appreciative.
>>>>
>>>>
>>>>
>>>> Daniel Mercik
>>>>
>>>> Computer Science Major
>>>>
>>>> Service Desk Technician
>>>>
>>>> Thomas College _____________________________________ Powered
>>>> bywww.kitware.com <http://www.kitware.com/><http://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
>>>>
>>>> <TileImageFilter_SideBySide.cxx><Tilesidebyside.JPG>
>>>
>>>
>>>
>>> _____________________________________ 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
>>
>> -- Dr. Roman Grothausmann
>>
>> Tomographie und Digitale Bildverarbeitung Tomography and Digital Image
>> Analysis
>>
>> Institut für Funktionelle und Angewandte Anatomie, OE 4120 Medizinische
>> Hochschule Hannover Carl-Neuberg-Str. 1 D-30625 Hannover
>>
>> Tel. +49 511 532-9574 _____________________________________ 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

-- 
Dr. Roman Grothausmann

Tomographie und Digitale Bildverarbeitung
Tomography and Digital Image Analysis

Institut für Funktionelle und Angewandte Anatomie, OE 4120
Medizinische Hochschule Hannover
Carl-Neuberg-Str. 1
D-30625 Hannover

Tel. +49 511 532-9574