Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > sci.image.processing > #4025 > unrolled thread

Gra Level Co-occurence Matrix

Back to article view | Back to sci.image.processing

Contents

  Gra Level Co-occurence Matrix Hitendra Kumar <hitendrak565@gmail.com> - 2015-09-05 01:29 -0700
    Re: Gra Level Co-occurence Matrix Dale <dale@dalekelly.org> - 2015-09-16 17:04 -0400

#4025 — Gra Level Co-occurence Matrix

I am trying to find out how statistics derived from GLCM vary with different pixel distance.
For this I am using an image with short and discontinuous vertical stripes.
Following are the steps:
1. consider one row -> obtain GLCM -> calculate contrast, correlation and energy statistics.
2. loop through all the rows.
3. change pixel distance and repeat 1 & 2.

The plots of correlation and energy have a exponential decay trend.
I am not able to figure out:
1. Why do they decay exponentially? 
2. In case of correlation, I should get a peak whenever the pixel distance is equal to stripe widths. Why is this not evident in plots?

[toc] | [next] | [standalone]

#4030

On 09/05/2015 04:29 AM, Hitendra Kumar wrote:
> I am trying to find out how statistics derived from GLCM vary with different pixel distance.
> For this I am using an image with short and discontinuous vertical stripes.
> Following are the steps:
> 1. consider one row -> obtain GLCM -> calculate contrast, correlation and energy statistics.
> 2. loop through all the rows.
> 3. change pixel distance and repeat 1 & 2.
>
> The plots of correlation and energy have a exponential decay trend.
> I am not able to figure out:
> 1. Why do they decay exponentially?
> 2. In case of correlation, I should get a peak whenever the pixel distance is equal to stripe widths. Why is this not evident in plots?
>

gamma is an exponential function? what does a linear gamma do?

-- 
Dale
http://www.dalekelly.org

[toc] | [prev] | [standalone]

Back to top | Article view | sci.image.processing

csiph-web