[GRASS-user] Clustering process in i.segment

Markus Metz markus.metz.giswork at gmail.com
Mon May 7 23:22:22 PDT 2018


On Mon, May 7, 2018 at 4:03 PM, Giuseppe Esposito <peppegeo84 at gmail.com>
wrote:
>
> Dear GRASS GIS experts,
>
> I’m approaching segmentation in GRASS GIS and some doubts arised when
reading the manual of the i.segment module (
https://grass.osgeo.org/grass74/manuals/i.segment.html). For this reason, I
would like to ask if someone may provide me further explanations about the
following point.
>
> I understood that the mean shift image segmentation consists of 2 steps:
anisotrophic filtering and clustering. It is non immediatelly clear to me
how the second step (clustering) is working and I can't find help from the
manual. I am not understanding how this process is performed in the
i.segment module, as well as if I can set it somehow. Do someone can help
me with some explanation or reference about the clustering process in
i.segment?

The clustering is the same like in r.clump: it identifies connected
components by grouping neighboring cells together that are similar to each
other with regard to hr (spectral bandwidth). Each group of cells becomes
an object in the final output.

Markus M

>
> Thanks in advance for your help.
>
> Giuseppe Esposito
>
>
>
> _______________________________________________
> grass-user mailing list
> grass-user at lists.osgeo.org
> https://lists.osgeo.org/mailman/listinfo/grass-user
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/grass-user/attachments/20180508/000ab762/attachment.html>


More information about the grass-user mailing list