[Insight-users] Memory issue

Arunachalam Kana Kana.Arunachalam at fh-wels.at
Fri Nov 26 04:51:14 EST 2010


Hi Sergio Vera,

Thank you for your response. The image resolution is 3µm. The extent is approx 3300*3300*1300.
I will have a look at the recommendations in the link.

As you said Hessian calculation might be a problem. I was thinking of breaking the image into smaller region of interest
and calculate the hessian, then apply my function in the region and store the result in segmented image(equal to size of input image) in the respective region.
This was just an idea ... Is this possible ? Will this lead to the changes in the hessian calculation at the end which is shared with next region?

Thank you,
Regards,
Kana

From: Sergio Vera [mailto:sergio.vera at alma3d.com]
Sent: 26 November 2010 09:34
To: Arunachalam Kana
Cc: insight-users at itk.org
Subject: Re: [Insight-users] Memory issue

Without knowing the exact image resolution/size, I would expect difficulties with the hessian at least: on a hessian image each pixel is a 3x3 double matrix... just multiply...and add intermediate images inbetween the pipeline, other procesess..

You may certainly experiment some memory problems. Check this thread for tips related to memory and speed optimizations within ITK: http://old.nabble.com/Guidelines-for-speed-and-memory-footprint-optimization-td30178329.html

On Thu, Nov 25, 2010 at 5:54 PM, Arunachalam Kana <Kana.Arunachalam at fh-wels.at<mailto:Kana.Arunachalam at fh-wels.at>> wrote:

Hi ITK Users,



Situation: I have 32Gb unsigned short image and I use a 64bit machine, with linux, and my RAM size is 64Gb. I would like to apply the following itk algorithm:

Step1: Original image -> Anisotropic diffusion image filter -> smoothed image



Step 2: Own filter: I calculate the hessian for the input image. I use the hessian image and the input image to segment and produce an output image.

Smoothed image -> hessianrecurssivegaussianimagefilter-> smoothed image + hessian image ->segmentation function -> segmented image



Step 3: Segmented image -> 3dbinarythinning -> thinned image.



After every step the output image will be save to local hard disk.



Questions:

1.       Will there be any memory problem in application of anisotropic diffusion filter, hessian filter and 3dbinarythinning?

2.       If the answer is yes to the above question. How can i solve it?

Thank you,

Regards,

Kana Arunachalam Kannappan

Research Associate

FH OÖ Forschungs & Entwicklungs GmbH

Stelzhamer Strasse 23,

4600 Wels,

Austria.

Phone: +43 (0)7242 72811 -4420

kana.arunachalam at fh-wels.at<mailto:kana.arunachalam at fh-wels.at>

www.fh-ooe.at<http://www.fh-ooe.at>; www.3dct.at<http://www.3dct.at>



_____________________________________
Powered by 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.html

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://www.itk.org/mailman/listinfo/insight-users



--
Sergio Vera

 Alma IT Systems
 C/ Vilana, 4B, 4º 1ª
 08022 Barcelona
 T. (+34) 932 380 592
 www.alma3d.com<http://www.alma3d.com>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.itk.org/pipermail/insight-users/attachments/20101126/d23c6d8f/attachment.htm>


More information about the Insight-users mailing list