But the landsat images are of different dates and in some case different years.
I found some mosaicing debates here: http://www.faqs.org/faqs/sci/Satellite-Imagery-FAQ/part3/section-14.html 2 schools of thought appear: 1. Mosaic image then classify 2. Classify individual images then mosaic I am following the great grass developer Michael Shapiro's (is he the grass developer?) advice on option 2. On Wed, Jan 21, 2009 at 7:16 PM, Markus Neteler <[email protected]> wrote: > On Wed, Jan 7, 2009 at 4:04 AM, maning sambale > <[email protected]> wrote: >> Hi, >> I have a series of classified rasters from different landsat images >> (various dates). I want to moasic/stitch them to a one raster. The >> problem is there are subtle differences in the classified output. Any >> advise in mosaicing them? > > (I am afraid that it is easier to first mosaic (e.g., with OSSIM), then > do the classification...) > > You could vectorize the raster areas and then try some SQL > magic to match close polygons. Also v.clean to remove tiny > areas from the map if needed. > > Markus > -- cheers, maning ------------------------------------------------------ "Freedom is still the most radical idea of all" -N.Branden wiki: http://esambale.wikispaces.com/ blog: http://epsg4253.wordpress.com/ ------------------------------------------------------ _______________________________________________ grass-user mailing list [email protected] http://lists.osgeo.org/mailman/listinfo/grass-user
