[Qgis-user] Principal Component Analysis ?

[Nikos Alexandris]

* image <laurent.celati at gmail.com> [2017-06-23 08:03:42 -0700]:

>Good afternoon,

Hi Laurent,


>i'm working on windows with several opensource tools (qgis, otb, grass...),
>
>I generated several OTB texture indices. Now, i want to evaluate the
>information 's redundancy thanks to a Principal Component Analysis.
>
>=> Is it possible to do that with some opensource tools (otb? qgis? grass?)

For GRASS, please have a look at
https://grass.osgeo.org/grass73/manuals/i.pca.html and a related wiki
page (needs some cleaning, yet useful--includes R examples):
https://grasswiki.osgeo.org/wiki/Principal_Components_Analysis.


>Moreover, I would like set a mask AND set a spectral subset (in order to
>ignore some indices bands which seems "strange" and not relevant to
>introduce into the process).

You can perform PCA selectively into as many input (spectral, in your
case) dimensions as you need to.  And, the MASK concept in GRASS will
operate only inside the current region (speak extent) and the unmasked
cells.


>Could you throw light for me?

If the wiki is not helping, please ask in grass-user.

Good luck, Nikos


>In advance, thank you very much for your help.
>
>Kind regards.