Filters: Variance

Hello,

Here one question about Process/Filter/Variance. I am trying to
understand what is behind the Variance filter. I make a text image with
the following matrix:

1 1 1
1 2 1
1 1 1

Then I run Variance filter on the image, with radiue = 2. The I got an
image with all 9 pixel value = 0.95238.

My understanding is that it would take a 3x3 matrix, calculate variance
of the 9 elements (which is 0.1111), then put it back to the center
position of the 3x3 matrix, without changing value of the elements at
the edge, i.e.
1 1 1
1 0.11 1
1 1 1
       
Is my understanding correct? Could anybody tell me how ImageJ calculate
this Variance filter?

Any comments will be appreciated.

ZQ

[jmutterer]

You can check how the variance filter is implemented by looking at the
following pluginfilter code (RankFilters.java) available at :

http://rsb.info.nih.gov/ij/developer/source/ij/plugin/filter/RankFilters.java.html

Jerome


On 11/16/06, ZQ@Bend <[hidden email]> wrote:

> Hello,
>
> Here one question about Process/Filter/Variance. I am trying to
> understand what is behind the Variance filter. I make a text image with
> the following matrix:
>
> 1       1       1
> 1       2       1
> 1       1       1
>
> Then I run Variance filter on the image, with radiue = 2. The I got an
> image with all 9 pixel value = 0.95238.
>
> My understanding is that it would take a 3x3 matrix, calculate variance
> of the 9 elements (which is 0.1111), then put it back to the center
> position of the 3x3 matrix, without changing value of the elements at
> the edge, i.e.
> 1       1       1
> 1       0.11    1
> 1       1       1
>
> Is my understanding correct? Could anybody tell me how ImageJ calculate
> this Variance filter?
>
> Any comments will be appreciated.
>
> ZQ
>

[M. Nur Heriawan]

Dear list,

I am a new user of ImageJ. I would like to use ImageJ
to analyze the sequence of images resulted from X-ray
CT scanner.

My question: Is it possible to draw the profile of CT
value or attenuation coefficient instead of (just)
grayscale value using ImageJ? Are there any plugins
available to address this purpose?

Thank you very much for kind attention.

Best regards,
 

---------------
M. Nur Heriawan
Lab. of Applied Geosciences & Tech.  
Kumamoto Univ., Japan


 
____________________________________________________________________________________
Sponsored Link

Mortgage rates near 39yr lows.
$420k for $1,399/mo. Calculate new payment!
www.LowerMyBills.com/lre

[Zhenyu Shou]

H Nur:

 You have to calibrate your film from gray value to CT number/Hausfeld
number/attentuation coefficient. The function is located in
Analyze/Calibrate..You have to scan film (films) with known Hausfeld
number and scan them into ImageJ. And then you read the gray value and
establish the one-on-one relationship between measured gray value and
known Hausfeld values and save it in a global calibration file (or look-up
table). Please check the documentation for that.


----------------------------------------------------------------------
Zhenyu Shou, Ph.D
Radiation Oncology Physicist
RadAmerica, LLC
Franklin Square Radiation Oncology Center
9105 Franklin Square Drive, Suite 100
Baltimore, MD 21237
Phone: [phone] (office), 408-306-8801 (cell)
Fax:      [phone]
Email: [hidden email]
--------------------------------------------------------------------------



"M. Nur Heriawan" <[hidden email]>
Sent by: ImageJ Interest Group <[hidden email]>
11/16/2006 05:17 AM
Please respond to
ImageJ Interest Group <[hidden email]>


To
[hidden email]
cc

Subject
Analysis of CT image






Dear list,

I am a new user of ImageJ. I would like to use ImageJ
to analyze the sequence of images resulted from X-ray
CT scanner.

My question: Is it possible to draw the profile of CT
value or attenuation coefficient instead of (just)
grayscale value using ImageJ? Are there any plugins
available to address this purpose?

Thank you very much for kind attention.

Best regards,
 

---------------
M. Nur Heriawan
Lab. of Applied Geosciences & Tech.
Kumamoto Univ., Japan


 
____________________________________________________________________________________
Sponsored Link

Mortgage rates near 39yr lows.
$420k for $1,399/mo. Calculate new payment!
www.LowerMyBills.com/lre

[Robert Martin-3]

Hi Nur,

I had a similar question awhile ago.  What I did find that you could  
do is 1) use the line tool and 2) change the line width tool  
(edit>>Options>>LIne width). Then you can find the profile of a  
widthed line to your choicing.  (Cntl-K for doing the profile).

Good Lunk
Robert Martin
Carleton University


On Nov 16, 2006, at 5:17 AM, M. Nur Heriawan wrote:

> Dear list,
>
> I am a new user of ImageJ. I would like to use ImageJ
> to analyze the sequence of images resulted from X-ray
> CT scanner.
>
> My question: Is it possible to draw the profile of CT
> value or attenuation coefficient instead of (just)
> grayscale value using ImageJ? Are there any plugins
> available to address this purpose?
>
> Thank you very much for kind attention.
>
> Best regards,
>
>
> ---------------
> M. Nur Heriawan
> Lab. of Applied Geosciences & Tech.
> Kumamoto Univ., Japan
>
>
>
> ______________________________________________________________________
> ______________
> Sponsored Link
>
> Mortgage rates near 39yr lows.
> $420k for $1,399/mo. Calculate new payment!
> www.LowerMyBills.com/lre

[M. Nur Heriawan]

Sincerely thank to Dr. Zhenyu and Robert Martin for
your comments regarding my CT image. Previously I
opened the CT image in *.BMP of 8-bit format using
ImageJ. That is why when I made profile, the gray
value always showed around 240-245. This really not
represented the CT value.

Actually when I checked again, I have the original CT
image in *.IMG, then I import (not open) it by raw and
set the format to be 16-bit signed with image size of
2048 x 2048. When I make a profile, the gray value
automatically corresponds to the CT value. Therefore
no need to make the calibration.  

I have one request...may someone address me to the
references either books or papers that discussed about
relationship between CT value (or attenuation
coefficient) with (the density of) geological
material. Are there any standards for this
relationship?

Last question...is it possible to produce the contour
lines for the image (in my case is CT value) by using
ImageJ?

I appreciate any comments and advices from you all.

Thank you.


--- "M. Nur Heriawan" <[hidden email]> wrote:

> Dear list,
>
> I am a new user of ImageJ. I would like to use
> ImageJ
> to analyze the sequence of images resulted from
> X-ray
> CT scanner.
>


---------------
M. Nur Heriawan
Lab. of Applied Geosciences & Tech.  
Kumamoto Univ., Japan
http://www.civil.kumamoto-u.ac.jp/tansa/index.shtml


 
____________________________________________________________________________________
Sponsored Link

Don't quit your job - take classes online
www.Classesusa.com