GRASSers,
I am not getting all the colors displaying on the GRASS display
monitor. Colormode float only seems to enhance the colors but not
provide any more colors than does fixed mode.
The images I am displaying are scanned photographs that we are
using v.digit to map on top of. The colr file in the database shows
256 categories each with its own unique rgb mix, and yet when I run
d.colortable many of the 256 colors are the same.
When the image is displayed similar colors seem to be blended into
one and this washes out some of the detail needed for mapping.
Is this a problem with the display monitor (d.mon)? Is it realy this
limited in its color display capabilities? Or am I just missing
something here?
The photographs are scanned, saved as tiff, converted to
sunraster using xv (24 to 8 bit), and imported into GRASS using
r.in.sunrast. Both the tiff and sunraster files show all the colors
when displayed using xv.
Thanks!
Morris Trimmer
University Computing
Washington & Lee University
Lexington, VA. 24450
trimmer@wlu.edu
(703) 463-8846
Hi Morris,
this is a feature of Grass that has caused reams of email already. You
can find a summary of the discussions on this list in my Grass Beginners'
Manual, available by anon ftp from scanner.frw.uva.nl:pub. Briefly, with
Grass in default condition, you will see only 6 different greys in an
image that has 256 grey values - I've never checked how many of 256
colors are actually visible. The d.colormode float command will release
additional colors for display, but only AFTER you d.erase and redisplay
your image. Hope this helps and that Grass developers will find a way to
emulate the display capabilities of Xv...
P. Martijn van Leusen
Institute of Prehistory, University of Amsterdam martijn@scanner.frw.uva.nl
On Fri, 5 Aug 1994, Morris Trimmer wrote:
GRASSers,
I am not getting all the colors displaying on the GRASS display
monitor. Colormode float only seems to enhance the colors but not
provide any more colors than does fixed mode.
The images I am displaying are scanned photographs that we are
using v.digit to map on top of. The colr file in the database shows
256 categories each with its own unique rgb mix, and yet when I run
d.colortable many of the 256 colors are the same.
When the image is displayed similar colors seem to be blended into
one and this washes out some of the detail needed for mapping.
Is this a problem with the display monitor (d.mon)? Is it realy this
limited in its color display capabilities? Or am I just missing
something here?
The photographs are scanned, saved as tiff, converted to
sunraster using xv (24 to 8 bit), and imported into GRASS using
r.in.sunrast. Both the tiff and sunraster files show all the colors
when displayed using xv.
Thanks!
Morris Trimmer
University Computing
Washington & Lee University
Lexington, VA. 24450
trimmer@wlu.edu
(703) 463-8846
Morris hello,
I had the exact same problem when I tried to import TM data using first
xv then r.in.sunrast. Instead of displaying 256 levels in the histogram,
GRASS is saying I have only max 31 grey levels. I sent a message last
week regarding this problem, but nobody seemed to know what to do. I am
still at a loss. Let's keep on this!
Sorry I can't help you!
Jonathan
Hi Martijn,
Itried the exact procedure you suggest in your message to Morris, but
there is no change. The histogram continues to show that in TM band 4,
for example, only 31 levels are present.
Jonathan
On Fri, 5 Aug 1994 martijn@scanner.frw.uva.nl wrote:
Hi Morris,
this is a feature of Grass that has caused reams of email already. You
can find a summary of the discussions on this list in my Grass Beginners'
Manual, available by anon ftp from scanner.frw.uva.nl:pub. Briefly, with
Grass in default condition, you will see only 6 different greys in an
image that has 256 grey values - I've never checked how many of 256
colors are actually visible. The d.colormode float command will release
additional colors for display, but only AFTER you d.erase and redisplay
your image. Hope this helps and that Grass developers will find a way to
emulate the display capabilities of Xv...P. Martijn van Leusen
Institute of Prehistory, University of Amsterdam martijn@scanner.frw.uva.nlOn Fri, 5 Aug 1994, Morris Trimmer wrote:
> GRASSers,
> I am not getting all the colors displaying on the GRASS display
> monitor. Colormode float only seems to enhance the colors but not
> provide any more colors than does fixed mode.
> The images I am displaying are scanned photographs that we are
> using v.digit to map on top of. The colr file in the database shows
> 256 categories each with its own unique rgb mix, and yet when I run
> d.colortable many of the 256 colors are the same.
> When the image is displayed similar colors seem to be blended into
> one and this washes out some of the detail needed for mapping.
> Is this a problem with the display monitor (d.mon)? Is it realy this
> limited in its color display capabilities? Or am I just missing
> something here?
> The photographs are scanned, saved as tiff, converted to
> sunraster using xv (24 to 8 bit), and imported into GRASS using
> r.in.sunrast. Both the tiff and sunraster files show all the colors
> when displayed using xv.
> Thanks!
> Morris Trimmer
> University Computing
> Washington & Lee University
> Lexington, VA. 24450
> trimmer@wlu.edu
> (703) 463-8846
>
One other reason for Grass images to contain less greyscale values than
they should is that they really do. That is, you either never had 256
different values in the image or you lost them during import/conversion
procedures. The first thing to do is to query your Grass image using
"r.cats" to find out if there really are only 31 values left. If this
proves true, and you're POSITIVE that the original image contained more
different values than that, you must have lost them somewhere. This means
you have to check every step along the way. Hopefully you will discover
that you inadvertently resampled the image - the `save at original size'
checkbox in Xv comes to mind as a plausible cause.
P. Martijn van Leusen
Institute of Prehistory, University of Amsterdam martijn@scanner.frw.uva.nl
Thanks for the reply martijn,
I checked the dat using "r.cats". It appears that all the bands have been
reduced somehow. In xV, I notice that I can put a check in the box
labelled "save at normal size". Do I want to do that? Also, when I try to
save the images in the greyscale mode, I fail as the images appear to be
saved automatically and unchangeably in the 32 color mode. As far as I
can tell, it is at this juncture that my problems arise. Why won't xV
allow me to save to the greyscale mode?
Thanks for your patience!
Jonathan
On Sun, 7 Aug 1994 martijn@scanner.frw.uva.nl wrote:
One other reason for Grass images to contain less greyscale values than
they should is that they really do. That is, you either never had 256
different values in the image or you lost them during import/conversion
procedures. The first thing to do is to query your Grass image using
"r.cats" to find out if there really are only 31 values left. If this
proves true, and you're POSITIVE that the original image contained more
different values than that, you must have lost them somewhere. This means
you have to check every step along the way. Hopefully you will discover
that you inadvertently resampled the image - the `save at original size'
checkbox in Xv comes to mind as a plausible cause.P. Martijn van Leusen
Institute of Prehistory, University of Amsterdam martijn@scanner.frw.uva.nl