[ITK] Determine surface area of certain region in image
Dženan Zukić
dzenanz at gmail.com
Wed Aug 10 11:19:28 EDT 2016
There are nine 1s in the LabelImage, which means that your object with
label 1 is 9 pixels big. Content of the 3DImage is irrelevant. If the
example LabelImage content you provided does not reflect what you want, you
need to come up with a better LabelImage (a different, better
segmentation), derived from 3DImage.
On Wed, Aug 10, 2016 at 11:11 AM, Matthias B <matthias.b at hotmail.com> wrote:
> Thank you for the code
>
> sp[0] and sp[1] is both 1 in my code. I don't get how I would get 12 in
> your example.
> The 3DImage is also never used.
>
> The spacing is also never absolute in my situation.
> Ydata could also be like this={{3,2.5,2.1,2.5,2.2},{1.
> 4,1.2,1.3,1.1,1.1},...}
>
> ------------------------------
> From: dzenanz at gmail.com
> Date: Wed, 10 Aug 2016 09:58:32 -0400
> Subject: Re: [ITK] Determine surface area of certain region in image
> To: matthias.b at hotmail.com
> CC: dan.muel at gmail.com; community at itk.org; blowekamp at mail.nih.gov
>
>
> 3DImage of size 5x5x3, and LabelImage 5x5.
>
> Dim labelShapeStatistics = New simple.LabelShapeStatisticsImageFilter()
> labelShapeStatistics.Execute(labelImage)
> Dim perimeterValue= labelShapeStatistics.GetPerimeter(1)
> Dim pixelCount = labelShapeStatistics.GetNumberOfPixels(1) //=9 in your
> example
> Dim sp = labelImage.GetSpacing()
> Dim surfaceArea=pixelCount*sp[0]*sp[1] //sp[0]*sp[1] should be =1.25 to
> yield a total of 12 which you want
>
> HTH,
> Dženan
>
> On Wed, Aug 10, 2016 at 9:48 AM, Matthias B <matthias.b at hotmail.com>
> wrote:
>
> I think an example is the best way to solve my problem.
>
> Let us take following 3D Data with name 3DImage as itk.simple.image
> Xdata={{-4,-2,0,2,4},{-4,-2,0,2,4},{-4,-2,0,2,4},{-4,-2,0,2,4},
> {-4,-2,0,2,4}}
> Ydata={{2,2,2,2,2},{1,1,1,1,1},{0,0,0,0,0},{-1,-1,-1,-1,-1},
> {-2,-2,-2,-2,-2}}
> Zdata={{0,0,0,0,0},{0,0,0,0,0},{0,0,0,0,0},{0,0,0,0,0},{0,0,0,0,0}}
>
> and LabelImage as itk.simple.image
> LabelImage={{{0,0,0,0,0},{0,1,1,1,0},{0,1,1,1,0},{0,1,1,1,0},{0,0,0,0,0}}
>
> The surface area with label 1 should be 12.
> Which steps (C# or C++) do I need to take to solve this?
>
> Thank you guys
>
> ------------------------------
> From: dzenanz at gmail.com
> Date: Wed, 10 Aug 2016 09:27:05 -0400
> Subject: Re: [ITK] Determine surface area of certain region in image
> To: matthias.b at hotmail.com
> CC: community at itk.org; dan.muel at gmail.com; blowekamp at mail.nih.gov
>
>
> Sorry, I have confused LabelShapeStatistics
> <https://itk.org/SimpleITKDoxygen/html/classitk_1_1simple_1_1LabelShapeStatisticsImageFilter.html>
> with LabelStatistics
> <https://itk.org/Doxygen/html/classitk_1_1LabelStatisticsImageFilter.html>
> filter.
>
> With label shape statistics, there is no intensity image, only the label
> image which you are already using as input.
>
> You generally don't mix (add) label and intensity images, you just keep
> them separate and know which two are a pair. For label statistics, you
> simply set two inputs like in this example
> <https://itk.org/Wiki/ITK/Examples/ImageProcessing/LabelStatisticsImageFilter>
> .
>
> Regards
>
> On Wed, Aug 10, 2016 at 9:19 AM, Matthias B <matthias.b at hotmail.com>
> wrote:
>
> I understand, but how do I add the label on the 3D image programmatically?
> preferable in C#, but C++ is also ok.
>
> Thank you for your help
> ------------------------------
> From: dzenanz at gmail.com
> Date: Wed, 10 Aug 2016 09:14:35 -0400
> Subject: Re: [ITK] Determine surface area of certain region in image
> To: matthias.b at hotmail.com
> CC: blowekamp at mail.nih.gov; dan.muel at gmail.com; community at itk.org
>
>
> The 3D image where the label will be applied is need for statistics such
> as mean, min and max values. If you don't care about those, you can supply
> the same image as both label and original (intensity) image. Or you can
> supply an empty (all black) image there.
>
> HTH,
> Dženan
>
> On Wed, Aug 10, 2016 at 8:47 AM, Matthias B <matthias.b at hotmail.com>
> wrote:
>
> Thank you for answering Dženan,
>
> ok, the perimeter attribute is what I need.
>
> Problem is that I don't see a way where I can
> 1. Choose a label image to define the region
> 2. Choose the 3D Image where the label will be aplied
>
> At this moment it looks for me that the label image already contains the
> 3D data, which confuses me.
>
> ------------------------------
> From: dzenanz at gmail.com
> Date: Wed, 10 Aug 2016 08:32:53 -0400
> Subject: Re: [ITK] Determine surface area of certain region in image
> To: matthias.b at hotmail.com
> CC: blowekamp at mail.nih.gov; community at itk.org
>
>
> Hi Matthias,
>
> perimeter should be surface area in 3D (which you can divide by 2 to
> eliminate back side if you have a very flat object). In 2D image, it is the
> circumference. In 2D image, you can get object's surface area by counting
> how many pixels it has and then multiplying that by the size of each pixel
> (spacing[0]*spacing[1]).
>
> Regards,
> Dženan
>
> On Wed, Aug 10, 2016 at 2:40 AM, Matthias B <matthias.b at hotmail.com>
> wrote:
>
> Thank you for answering!
>
> At this moment I get the perimeter in pixel units (not the surface area of
> my X,Y-data) of my labeled image.
> I have also really no idea how to use my 3D data with a labeled image. I
> see no way how to add 3D data to a LabelMap for example.
>
> This is what I have at the moment.
> Dim labelShapeStatistics = New simple.LabelShapeStatisticsImageFilter()
> labelShapeStatistics.Execute(labelImage)
> Dim perimeterValue= labelShapeStatistics.GetPerimeter(1)
>
> Thanks for any help, Matthias
>
>
>
> ------------------------------
> From: blowekamp at mail.nih.gov
> To: matthias.b at hotmail.com
> CC: community at itk.org
> Subject: Re: [ITK] Determine surface area of certain region in image
> Date: Tue, 9 Aug 2016 15:58:56 +0000
>
>
> Hello,
>
> You should look into the label map framework [1]. In particular the
> LabelImageToShapeLabelMapFilter[2] and the Perimeter attribute [3] [4].
>
> HTH,
> Brad
>
> [1] http://hdl.handle.net/1926/584
> [2] https://itk.org/Doxygen/html/classitk_1_1LabelImageToSha
> peLabelMapFilter.html
> [3] https://itk.org/Doxygen/html/classitk_1_1ShapeLabelObjec
> t.html#aff1209e925293c15775520ef89e97afb
> [4] http://hdl.handle.net/10380/3342
>
> On Aug 9, 2016, at 11:50 AM, Matthias B <matthias.b at hotmail.com> wrote:
>
> Hello ITK-community,
>
> At the moment I'm working with 3D Images with following structure:
> 100x80x3. Every pixel contains a X,Y,Z value.
> I want to calculate the surface area of a certain region that I have
> defined with a boolean 2D-Image (100x80).
> The surface area should only be calculated with the X,Y-data.
>
> In Matlab I have experience with this by using the function bwboundaries
> on the boolean image. This function determines the indices of the the
> boundaries in clockwise direction. Next I take the corresponding X,Y
> values of the indices so I can calculate the surface area with the function
> polyarea.
>
> Is this possbile with the ITK-library?
>
> Thank you for your help guys.
>
>
> <https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail>
> Virusvrij. www.avast.com
> <https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail>
> _______________________________________________
> Community mailing list
> Community at itk.org
> http://public.kitware.com/mailman/listinfo/community
>
>
>
> _______________________________________________
> Community mailing list
> Community at itk.org
> http://public.kitware.com/mailman/listinfo/community
>
>
>
>
> <https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail> Virusvrij.
> www.avast.com
> <https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail>
> <#m_-24046198973060639_m_876262044166110828_m_1512104202195024141_m_4192781686007042514_DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2>
>
>
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/community/attachments/20160810/81510076/attachment.html>
More information about the Community
mailing list