Re: [Gdcm2] Buffer problem in Qt environment

[Mathieu M. <mat...@gm...>]

Hi Sylvain,

  congrats ! I realize this was my mistake, I quickly browse through
the documentation of the image format:

http://doc.trolltech.com/4.5/qimage.html#Format-enum

  and I thought QImage could handle 16bits monochrome image, while it does not.

  Please share your conversion code and I'll put it in Examples/Cxx
(did you use C++ ?).

  As a side note, you can use the command line gdcmimg to convert a
DICOM image to pgm, and then you could load the QImage using the load
API:

http://doc.trolltech.com/4.5/qimage.html#reading-and-writing-image-files

  so that you can check your conversion from 8bits is identical to
what 'load()' would do with an input 16bits file (I am surprise there
is no such filter already in Qt).

Eg:

  $ gdcmconv --raw  gdcmData/012345.002.050.dcm /tmp/dummy.dcm
  $ gdcmimg /tmp/dummy.dcm /tmp/dummy.pgm
  $ identify /tmp/dummy.pgm


        ~
  /tmp/dummy.pgm PNM 256x256 256x256+0+0 PseudoClass 65536c 16-bit 128.016kb

thanks for your contribution !

On Wed, Jun 17, 2009 at 10:39 AM, Sylvain ADAM<syl...@ho...> wrote:
> Hi there!
>
> Yataaaa. I did it !!
>
> Well, finally, I guess I'm done with it!
>
> Sorry guys, but I actually didn't know I had to convert the buffer myself. I
> thought there were a solution in the API. As I haven't much experience with
> all of it.
>
> I guess I'll write a few tutorials on how to use GDCM with Qt. Seems that
> none of the functions of images librairies allowed me to convert a buffer to
> what I needed.
>
> I finally managed to extract the 16-bits words with bitwise shifts, and
> convert them to 8 bits and add them to a RGB 24 bits (8-8-8) buffer, format
> which is known by the Qt image library.
>
> ===> However, is this normal: the strongest part of the 16-bits words in the
> char was AFTER the lightest one. What I'm saying is, I understand that, to
> put the 16-bits image in a char buffer, that is in 8-bits words, you
> obviously have to SPLIT the words. But how comes that the left part of the
> 16-bits word is set AFTER the right one? Is that normal?
>
> Anyway, thanks for all and I'm very grateful to Mathieu for his answers and
> his library :).
>
> Cheers all.
>
> Sylvain
>
>
>> Date: Tue, 16 Jun 2009 12:00:06 +0200
>> Subject: Re: [Gdcm2] Buffer problem in Qt environment
>> From: mat...@gm...
>> To: syl...@ho...
>> CC: gdc...@li...
>>
>> On Tue, Jun 16, 2009 at 10:45 AM, Sylvain ADAM<syl...@ho...>
>> wrote:
>> > Salut Mathieu.
>> >
>> > Thanks for the answers !!
>> >
>> > Yeah I finally got almost what I wanted by displaying my image in a
>> > QImage.
>> > But as I said earlier, well, I'm still using the RGB16 format of the
>> > QImage
>> > class.
>> >
>> > The RGB888 format doesn't work. I actually looked over the API (maybe
>> > not
>> > deep enough), both from GDCM and Qt, and I still don't have any clue on
>> > how
>> > to display my image correctly in Qt and how to access 16bits pixel
>> > information at coordinate (x, y) on the original image!
>> >
>> > For now, I can just get the grayscale value on my QImage (which is
>> > actually
>> > blue-green because of the format used) by calculating it using the RGB
>> > value. So I get a gray value from 0 to 140, which is not the usual gray
>> > value for a DICOM image in here. We usually have something from 0 to
>> > 2000
>> > ~ich.
>> >
>> > I tried the getPhotometricInterpretation() function, to see what does
>> > this
>> > return: MONOCHROME2. And that's it. The getSamplesPerPixel returns 1.
>> > I'm
>> > afraid my image knowledge is a bit limited!
>> >
>> > As the QImage is my display image, getting information from it is not
>> > what I
>> > intend to do ('cause for display reasons, we don't really need 16-bits,
>> > do
>> > we?).
>> >
>> >
>> > Another thing for now, does anyone have had this error before? This
>> > happens
>> > when I'm using the buffer. This is just displayed by my application
>> > output
>> > in debugging mode:
>> >
>> >
>> > HEAP[IntensityProfileGUI.exe]:
>> >
>> > Invalid Address specified to RtlFreeHeap( 05080000, 04DAA558 )
>> >
>> > The message is displayed twice in my debugger. Well, even if I'm using
>> > two
>> > or one buffer, and when I comment the buffer line, it does not appear
>> > anymore.
>> >
>> > Thanks for helping me.
>>
>> MONOCHROME2 and getSamplesPerPixel really means you only have one
>> component (=grayscale) per pixel, do not try to tell Qt there are 3
>> using the RGB16 format. Depending on the value from GetPixelFormat()
>> you may need a 16 bits QtImage or 8bits QtImage (or even 32bits).
>>
>> Check you example such as ConvertPIL.py / ConvertNumpy.py /
>> ConvertMPL.py, to see how gdcm can be used to pass the raw byte buffer
>> to 3rd party library.
>>
>> HTH
>> --
>> Mathieu
>
> ________________________________
> Découvrez toutes les possibilités de communication avec vos proches



-- 
Mathieu