I got a set of i.segment output objects, and would like to classify the pixels within the regions.
how can I do that?
best wishes
–
Miltinho - mcr@rc.unesp.br
Laboratório de Ecologia Espacial e Conservação - LEEC
Depto de Ecologia - UNESP - Rio Claro
Av. 24A, 1515- Bela Vista
13506-900 Rio Claro, SP, Brasil
I got a set of i.segment output objects, and would like to classify the
pixels within the regions.
Check out the v.class.ml addon [1].
For a general procedure with modules in GRASS core see the script attached to this message [2]
And the web page that Martin Landa elaborated in czech based on that script [3] which you can run through Google translate to get it in your desired language [4].
For a general procedure with modules in GRASS core see the script attached to this message [2]
And the web page that Martin Landa elaborated in czech based on that script [3] which you can run through Google translate to get it in your desired language [4].
Miltinho - mcr@rc.unesp.br
Laboratório de Ecologia Espacial e Conservação - LEEC
Depto de Ecologia - UNESP - Rio Claro
Av. 24A, 1515- Bela Vista
13506-900 Rio Claro, SP, Brasil
--
Miltinho - mcr@rc.unesp.br
Laboratório de Ecologia Espacial e Conservação - LEEC
Depto de Ecologia - UNESP - Rio Claro
Av. 24A, 1515- Bela Vista
13506-900 Rio Claro, SP, Brasil
how can I use the objects output from i.segment in the steps to classify the image using i.maxlik?
I tried to find a step-by-step classification using the raster objects from i.segment by no luck.
best wishes
milton
···
2014-06-26 18:34 GMT-03:00 Helmut Kudrnovsky <hellik@web.de>:
But anyway, I’m not sure, I can copy and paste it there but it is just
bunch of links, nothing nice.
IMHO nice links to nice howto’s are often lost in MLs and are some kind of
more persistent in wikis.
Miltinho - mcr@rc.unesp.br
Laboratório de Ecologia Espacial e Conservação - LEEC
Depto de Ecologia - UNESP - Rio Claro
Av. 24A, 1515- Bela Vista
13506-900 Rio Claro, SP, Brasil
how can I use the objects output from i.segment in the steps to classify
the image using i.maxlik?
I tried to find a step-by-step classification using the raster objects
from i.segment by no luck.
If you really want to use i.maxlik, you have to create as many raster layers as you want to have variables characterizing your objects. Then you select a series of these objects as training areas and then you can run i.gensig + i.maxlik.
But I think that the machine learning approach, using Pietros addon-modules, is probably more efficient.