Posted by
Mike Esterman on
Dec 11, 2012; 8:20pm
URL: http://imagej.273.s1.nabble.com/texture-analysis-tp5001039p5001121.html
One way to measure texture is measure the standard deviation of the mean
gray value. Look at the attached images a color swatch converted to
grayscale with essentially no texture and the second a leaf grayscale image
that you can see obvious texture.
The Mean grayscale for the swatch is 67.9 Std Dev 2.46 very little change
in grayscale across the image
The mean grayscale for the leaf is 84.2 Std Dev 14.5 giving a value that
relates to roughness of the surface which results in variation in the light
reflection and absorbance across the surface.
Another measure that may be implemented in a plug-in I haven't looked
recently is the fractal dimension.
Hope this helps.
Mike Esterman
-----Original Message-----
From: ImageJ Interest Group [mailto:
[hidden email]] On Behalf Of
Karsten
Sent: Tuesday, December 11, 2012 6:29 AM
To:
[hidden email]
Subject: Re: texture analysis
I think you should still be a bit more specific, as you might understand
from the silence around here!
At least the term texture in combination with biological tissue let me
conjecture that you are neither an expert in texture, image
analysis/processing nor in biological tissues. Dependent on the observer of
tissue there is or there is no texture!
Am 09.12.2012 um 10:50 schrieb akash verma:
> Thanks Sir for replying
> I want to do texture analysis and segmentation in biological tissues.
>
> On Thu, Dec 6, 2012 at 8:12 PM, Karsten <
[hidden email]> wrote:
>
>> Hi,
>> texture analysis and texture segmentation is a quite broad field.
>> Possibly you should slightly specify what you would like to do.
>>
>> Regards
>>
>> Am 05.12.2012 um 05:36 schrieb akash verma:
>>
>>> hello
>>>
>>> Can anybody tell me how we do texture analysis and segmentation in
>>> Image
>> j
>>> . Since i am new user of image J .
>>> Thanks in advance
>>>
>>> --
>>> ImageJ mailing list:
http://imagej.nih.gov/ij/list.html>>
>> Karsten
>>
>> --
>> ImageJ mailing list:
http://imagej.nih.gov/ij/list.html>>
>
> --
> ImageJ mailing list:
http://imagej.nih.gov/ij/list.htmlKarsten
[hidden email]
--
ImageJ mailing list:
http://imagej.nih.gov/ij/list.html--
ImageJ mailing list:
http://imagej.nih.gov/ij/list.html