Note: This document is for an older version of GRASS GIS that is outdated. You should upgrade, and read the current manual page.

GRASS logo

NAME

i.landsat.rgb - Performs auto-balancing of colors for LANDSAT images.

KEYWORDS

raster, imagery, colors

SYNOPSIS

i.landsat.rgb
i.landsat.rgb help
i.landsat.rgb [-fprs] red=string green=string blue=string [strength=float] [--verbose] [--quiet]

Flags:

-f
Extend colors to full range of data on each channel
-p
Preserve relative colors, adjust brightness only
-r
Reset to standard color range
-s
Process bands serially (default: run in parallel)
--verbose
Verbose module output
--quiet
Quiet module output

Parameters:

red=string
LANDSAT red channel
green=string
LANDSAT green channel
blue=string
LANDSAT blue channel
strength=float
Cropping intensity (upper brightness level)
Options: 0-100
Default: 98

DESCRIPTION

i.landsat.rgb auto-balances and enhances the color channels of a LANDSAT RGB image to provide a more natural color mixture. Only the color table of each image band is modified, the base data remains untouched.

The module works by calculating a histogram for each color channel and removing an adjustable amount of outliers from either end before recalibrating the color scale with r.colors.

It will work with any 8-bit RGB imagery set and the script is easily modified to work with other datasets of greater band-depth.

NOTES

Depending on the image, it may or may not be advantageous to use the -p flag to preserve the relative color scaling. You will have to experiment with the different options to find a setting that works best for your particular imagery.

The strength option should generally be set in the 90-99 range. The lower the number, the more saturated the image becomes. It represents the percentage cut-off for the top end of the color histogram curve. The lower end is fixed at 2% of the area under the curve.

For quicker execution of this module on large images you can achieve largely similar results by switching to a coarser resolution before the running of the module (using g.region) and then back to the original resolution afterwards.

By default the module will process all three R,G,B bands in parallel. To run serially use the -s flag. Parallel mode requires three times the memory while running, but takes one third of the time to complete.

EXAMPLE

North Carolina sample dataset example:
g.region rast=lsat7_2002_10 -p
d.rgb blue=lsat7_2002_10 green=lsat7_2002_20 red=lsat7_2002_30
d.barscale at=55.312500,93.125000 bcolor=white tcolor=black -t

i.landsat.rgb blue=lsat7_2002_10 green=lsat7_2002_20 red=lsat7_2002_30 strength=95
d.rgb blue=lsat7_2002_10 green=lsat7_2002_20 red=lsat7_2002_30
d.barscale at=55.312500,93.125000 bcolor=white tcolor=black -t
Uncorrected RGB composite of Landsat ETM channels B/G/R
Uncorrected RGB composite of Landsat ETM channels B/G/R

Color corrected RGB composite of Landsat ETM channels B/G/R
Color corrected RGB composite of Landsat ETM channels B/G/R

TODO

The strength option requires further refinement.

SEE ALSO

d.rgb, g.region, i.oif, r.colors, r.composite, r.univar,

AUTHORS

Markus Neteler, Trento, Italy
M. Hamish Bowman, Dept. Marine Science, Otago University, New Zealand

Last changed: $Date: 2012-06-18 01:16:05 -0700 (Mon, 18 Jun 2012) $


Main index - imagery index - Full index

© 2003-2014 GRASS Development Team