[GRASS-stats] Re: [GRASS-user] Testing i.pca ~ prcomp(), m.eigensystem ~ princomp()

Nikos Alexandris nikos.alexandris at felis.uni-freiburg.de
Tue Mar 31 11:07:29 EDT 2009


Agustin:   
> >> Nikos, I'm probably missing a message (I do not see which is
> >> exactly the difference between the spot and the modis example),
> >>     

Nikos:
> > Hi Agus. The difference is:
> > Using SPOT (range up to max. 255):
> > i.pca == prcomp(x, center=TRUE, scale=FALSE)
> > Using MODIS (range varries between bands, up to max. ~5500):
[...]

Markus M:
> Sure? Valid range: -100 - 16000 for MODIS e.g. Surface Reflectance 8-Day 
> L3 Global 500m.

I am referring to the range of the specific MODIS bands that I've used,
that is:
# band 2
r.info -r mod_b2

min=0
max=5504

# band 6
r.info -r mod_b6

min=93
max=5488

# band 7
r.info -r mod_b7

min=0
max=4133


> As I would use the max possible for SPOT (255) I would also use the max 
> possible for other sensors, not the observed max, otherwise statistics 
> that don't do proper normalization on the fly are not comparable between 
> different sensors.

> But any PCA should do normalization beforehand, not sure what R or i.pca 
> are doing.

That's the question _now_ (if I am right with my tests): why isn't
normalisation (=data centering in this case) performed with the MODIS
bands I use?


> Please excuse my random comments :-)

Markus, your comments are required :-)
Cheers, Nikos



More information about the grass-stats mailing list