[GRASS-dev] On i.histo.match (Re: On (Landsat) imagery naming patterns)

[Nikos Alexandris]

Nikos Alexandris:

> >> Greetings to the dev-list.
> >> I successfully tested <i.histo.match>, by feeding it with 12 Landsat
> >> scenes (working on identical bands, different Path/Row tiles) in one go. 
> >> It works and I really enjoy it.

Nikos A:

> > Could it be that histo-matching works nicely only on 0-255 ranges but
> > doesn't play nice in ranges 0-1.0? Integers only?

Markus Metz wrote:
 
> Yes, integers only. The i.histo.match script forces input raster
> values to integers. But it might not be restricted to the 0-255 range.

(
Oh how nice it would be to do the tranformation internally :-)
)
 
> > For example, when feeding images ranging from 0-1.0 (all Landsat bands 3)
> > with the optional parameter "max=1", many of the histo-matched images are
> > "flatten" to 1.  I will try to "recode" 0-1.0 images to 0-255 and repeat
> > the histo-matching process.
 
> For rescaling to 0-1000, you could try something like
> r.mapcalc "bandX_int = round(1000 * bandX)"

Nice!  But I wonder why round-ing is required?  I want to jump back afterwards 
(i.e. divide by 1000 the histo-matched images).  Rounding does not hurt much, 
does it?

Thank you Markus, Nikos
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 230 bytes
Desc: This is a digitally signed message part.
URL: <http://lists.osgeo.org/pipermail/grass-dev/attachments/20130220/3c2b2fde/attachment.pgp>