[GRASSLIST:2823] Re: image correction tools

[Markus Neteler]

On Tue, Mar 02, 2004 at 04:24:15PM +0100, werchowyna at pf.pl wrote:
> thank you for your help Markus
> 
> one thing more: how do I correct the effect of uneven lighting conditions in
> aerial photos?
> (some areas are brighter like if there was a spot-light or a flash cast on
> it and some are unnaturaly dark)

That should be minimized by the (below) suggested terrain correction
which calculates the illumination based on the DEM and then normalizes
the brightness differences to make the photos appear as it were in
flat terrain.

Markus
 
> Regards
> Maciek
> 
> ----- Original Message -----
> From: "Markus Neteler" <neteler at itc.it>
> To: <GRASSLIST at baylor.edu>
> Sent: Tuesday, March 02, 2004 10:17 AM
> Subject: [GRASSLIST:2799] Re: image correction tools
> 
> 
> > On Mon, Mar 01, 2004 at 11:49:12AM +0100, werchowyna at pf.pl wrote:
> > > hello Everyone!
> > >
> > > this my first post to GRASSLIST
> > >
> > > since it's not strictly GRASSy you may answer directly to me
> > > werchowyna at pf.pl
> > >
> > > I'm looking for a Grass module/standalone software (both Linux and
> Windos)
> > > for radiometric (solar, atmospheric, topographic) correction of aerial
> and
> > > satellite images  of subalpine and alpine areas
> > >
> > > open source/freeware/cheap preferably
> > >
> > > can someone help?
> > > any ideas where to ask?
> >
> > There is an implementation of "6S" in GRASS:
> > http://grass.itc.it/grass_addons.html
> >  -> i.atcorr: 6S atmospheric correction program (Christo Zietsman, South
> Africa)
> >
> > Concerning terrain correction we are suggesting a simple method
> > with r.mapcalc etc in our book (http://mpa.itc.it/grassbook/).
> > Basically it is using the terrain geometry (slope, aspect), sun position
> > and the calculation of a normalized map. The upcoming second edition of
> > the book will describe a slightly improved method. It's not very
> > difficult.
> > An older version is here (my MSc thesis, 1999):
> >
> http://www.geog.uni-hannover.de/users/neteler/itc_irst/atmos_sma/9_atmos.htm
> l
> > (click onto the right buttons to proceed).
> >
> > You should be sure to have an elevation model with a resolution
> > which corresponds to the resolution of the aerial or satellite image.
> >
> > Best regards
> >
> >  Markus