By flattening, do you mean level? Those can work on the masked area,
and then you can use the binary mask to "reclaim" the original image
that was not under the mask and combine it with the leveled masked area.
Is that what you are trying to do?
Best regards
--aryeh
On 21/06/2021 11:29, Nicolas Delorme wrote:
Thank you Aryeh,
the problem is to apply a flattening method (needed to calculate the
roughness) to the resulting image since the flattening function seems
to take into account all the points in the image.
Best regards
------------------------------------------------------------------------
*Prof. Nicolas Delorme*
Physics and Material Science
Institut des Molécules et Matériaux du Mans - UMR CNRS 6283
IUT Le Mans
Le Mans Université
Avenue Olivier Messiaen, 72085 - LE MANS Cedex 09 - FRANCE
Tél. +33 (0)2 43 83 27 44
Publications list
<https://scholar.google.fr/citations?user=UQviEeYAAAAJ&hl=fr&oi=ao>
Le 21/06/2021 à 10:18, Aryeh Weiss a écrit :
Do you want to mask your feature and then create a new image with
just that feature?
I attached a set of images that do that -- you make your mask ,
extract it, and then and it with the original using the arithmetic
operations tool.
Best regards
--aryeh
On 21/06/2021 09:47, Nicolas Delorme wrote:
Thank you for your answer,
my image is less complicated since i have only one fiber per image.
At this point, my only solution was the one you gave : use of 1D
roughness function. But as you said the values are quite dispersed
and it is difficult to make a statistical comparison between samples.
I wonder if there is an option allowing to extract the data under
the mask and make a new image with it (like a the crop option but
without the need to make a rectangular image).
Best regards
------------------------------------------------------------------------
*Prof. Nicolas Delorme*
Physics and Material Science
Institut des Molécules et Matériaux du Mans - UMR CNRS 6283
IUT Le Mans
Le Mans Université
Avenue Olivier Messiaen, 72085 - LE MANS Cedex 09 - FRANCE
Tél. +33 (0)2 43 83 27 44
Publications list
<https://scholar.google.fr/citations?user=UQviEeYAAAAJ&hl=fr&oi=ao>
Le 18/06/2021 à 19:46, David Nečas (Yeti) a écrit :
On Fri, Jun 18, 2021 at 05:33:42PM +0200, Nicolas Delorme wrote:
I would like to measure the roughness of fibers.
For this I have to flatten the "fiber zone" of the image.
By using the mask option in Gwyddion i did not succeed in flattening only the
fiber zone.
If the image looks anything like
http://gwyddion.net/download/test/fibres.gwy
then it is difficult – and possibly ill-defined. With multiple fibres
there is no common background; each defines locally the overall form
which needs to be subtracted. So usual background subtraction does not
work.
It seems the dumb method gives reasonable results: Just use the 1D
Roughness tool, select a profile along the top of the fibre (with small
or even zero cutoff) and read the roughness. Repeat for multiple fibres
and possibly multiple parallel profiles along each – there will be a
large variation. If you have to, use Straighten Path to straighten
curved fibres. Do not use too short segments.
I would still call not the process measurement, maybe rough estimation…
Regards,
Yeti
_______________________________________________
Gwyddion-users mailing list
Gwyddion-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/gwyddion-users
_______________________________________________
Gwyddion-users mailing list
Gwyddion-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/gwyddion-users
--
Aryeh Weiss
Faculty of Engineering
Bar Ilan University
Ramat Gan 52900 Israel
Ph: 972-3-5317638
FAX: 972-3-7384051
--
Aryeh Weiss
Faculty of Engineering
Bar Ilan University
Ramat Gan 52900 Israel
Ph: 972-3-5317638
FAX: 972-3-7384051
_______________________________________________
Gwyddion-users mailing list
Gwyddion-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/gwyddion-users
