[Gwyddion-users] image derivatives or pyramid analysis

[Nicolas Delorme]

Hi,
We are studying surfaces formed by pyramids.
We want to statistically analyse their numbers and height (the shape is
the same for all the pyramids).
The masking procedure do not work since some of the pyramids are in contact.
One solution would be to derivate the topographic image (but I don't
know if gwyddion can do it) and to count the obtained maxima and look at
their relative height...
Is there a way to obtain a derivative of an image with Gwyddion?
If you have another solution for the pyramids analysis it will be great.
Best regards,
Nicolas
--
------------------------------------------------------------------------------------------------------------------

*Nicolas DELORME*
Maître de conférences (associate professor)
Institut Molécules et Matériaux du Mans (IMMM)
UMR CNRS 6283
Université du Maine Avenue Olivier Messiaen,72085 Le Mans Cedex 9 – France
Tél : [phone]
Fax : [phone]
website:http://perso.univ-lemans.fr/~bardeau/IMMM-PEC/surfpropspec/indexsurface.html
Licence Professionnelle Ecoconstruction et Energie :
http://sciences.univ-lemans.fr/Licence-Pro-Eco-construction-et
Facebook :https://www.facebook.com/licence.ecoconstruction

[Stefania Carapezzi]

Hi.
I don't know in Gwyddion, but you could derivate a surface with Origin or
Matlab. I would say there should be troubles with the tops because a
pyramidal surface should have discontinuities in 1st derivative just there.
I thought over and could suggest some other solutions:
1) the kind of surface you described it seems to me could be assimilated to
a mounded surface. If it is applicable, then you can extract in a
statistical way at least the interdistance between pyrmids by fitting the
Height Height correlation function by mounded ideal function. You
can compute the HHCF by Gwyddion and then export and fit by Origin or
Matlab. here are some references:
Zhao et al. Characterization of Amorphous and Crystalline Rough
Surface Principles and Applications mounded surface, page 43
(this page is accesible online from googlebooks)
Pelliccione, Lu Evolution of Thin Film Morphology; Modeling and
Simulations, chapter 4
2) you can "mark grains by threshold" using height as a threshold. in this
case you can separate the highest parts of your surface. If you extract the
mask, you should get a black/white image, with top-nearby regions
black. You can study it as a distribution of "grains" or "particles". If
you save it and use ImageJ the interdistance between your "grains". you
can play with the concept and study the next neighbour distance
distribution, if it is of some interest. I don't know if it could be done
also entirely by Gwyddion.
As regards the heights, you can extract the data from the height
distribution (from "Calculate 1D functions"), but this is strictly related
on how much your features are regular and isomorphic.
Hope to have been of some use.
Best regards

Stefania Carapezzi
Post-Doc
Dipartimento di Fisica e Astronomia
Alma Mater Studiorum - Università di Bologna
V.le Berti Pichat 6/2
40127, Bologna Italy
[phone]
***@unibo.it


On Fri, Jun 12, 2015 at 9:28 AM, Nicolas Delorme <

Post by Nicolas Delorme
Hi,
We are studying surfaces formed by pyramids.
We want to statistically analyse their numbers and height (the shape is
the same for all the pyramids).
The masking procedure do not work since some of the pyramids are in contact.
One solution would be to derivate the topographic image (but I don't know
if gwyddion can do it) and to count the obtained maxima and look at their
relative height...
Is there a way to obtain a derivative of an image with Gwyddion?
If you have another solution for the pyramids analysis it will be great.
Best regards,
Nicolas
--
------------------------------------------------------------------------------------------------------------------
*Nicolas DELORME*
Maître de conférences (associate professor)
Institut Molécules et Matériaux du Mans (IMMM)
UMR CNRS 6283
Université du Maine Avenue Olivier Messiaen,72085 Le Mans Cedex 9 – France
Tél : [phone]
Fax : [phone]
http://perso.univ-lemans.fr/~bardeau/IMMM-PEC/surfpropspec/indexsurface.html
http://sciences.univ-lemans.fr/Licence-Pro-Eco-construction-et
Facebook : https://www.facebook.com/licence.ecoconstruction
------------------------------------------------------------------------------
_______________________________________________
Gwyddion-users mailing list
https://lists.sourceforge.net/lists/listinfo/gwyddion-users

[David Nečas (Yeti)]

Post by Nicolas Delorme
The masking procedure do not work since some of the pyramids are in contact.

The waterhsed should fare better. There are actually two completely
different watershed algorithms, one denoted ‚Mark by Segmentation‘ but I
should still to improve it to make it more useful to topographical data.

Post by Nicolas Delorme
One solution would be to derivate the topographic image (but I don't
know if gwyddion can do it) and to count the obtained maxima and
look at their relative height...
Is there a way to obtain a derivative of an image with Gwyddion?

Functions in Presentation → Gradient calculate the classical directional
gradients (Sobel, Prewitt), the corresponding functions in Presentation
→ Edge Detection calculate the total gradient. Note these functions
create presentations as the output was not intended for further
analysis, but you still extract them...

Function Integral Transforms → Local Slope transforms the image to the
total derviative using the simple symmetrical derivative operator.

Regards,

Yeti


------------------------------------------------------------------------------