[gdal-dev] Converting from 16 bit to 8 bit images (best
practices?)
Nikolaos Hatzopoulos
nhatzop at gmail.com
Thu Jun 9 19:49:05 EDT 2011
for modis truecolor images?
--Nikos
On Thu, Jun 9, 2011 at 4:44 PM, Brian Case <rush at winkey.org> wrote:
> Jonathan
>
> I don't know about worldview, but with modis I wound up using multiple
> ranges
>
> 0:0,750:110,1500:160,3000:210,4750:240,6375:255
>
> see <lut> in the vrt tutorial
>
> Brian
>
>
>
>
>
>
> On Thu, 2011-06-09 at 12:39 -0700, Jonathan Greenberg wrote:
> > Brian:
> >
> >
> > Thanks! Great solution, definitely saved me some time. The main
> > issue (in terms of doing this entirely within GDAL) is getting the
> > right scaling to make the image look good. I was able to figure out
> > the correct scale range within ENVI, and then apply the scales as you
> > suggested below. Nikos' suggestion to use rgb2pct makes me wonder if
> > there is some way to figure out the optimal scale (which, AFAIK, is
> > part of how rgb2pct is functioning) to apply to this image.
> >
> >
> > Incidentally, what I'm trying to do is load Worldview-2 data onto an
> > iPad 2 for use with the GISRoam mobile mapping app. Worldview-2 is
> > 16-bit per band, 8 band data, with a very high dynamic range (we have
> > snow in our image, so we have lots of VERY bright pixels which throw
> > off the adjustment if using -scale without any parameters). 99%
> > (100%?) of mobile mapping applications need 3-band data, usually with
> > only 8 bits per band.
> >
> >
> > Pushing forward, but if someone knows how to figure out an optimal set
> > of scale values to use within GDAL, please post the solution.
> >
> >
> > --j
> >
> > On Wed, Jun 8, 2011 at 9:45 PM, Brian Case <rush at winkey.org> wrote:
> > Jonathan
> >
> > you can do scaling on individual bands.
> >
> > gdal_translate -b 1 -scale 0 5000 -ot Byte -of VRT infile
> > outfile1.vrt
> > gdal_translate -b 2 -scale 0 4000 -ot Byte -of VRT infile
> > outfile2.vrt
> > gdal_translate -b 3 -scale 0 3000 -ot Byte -of VRT infile
> > outfile3.vrt
> >
> > gdalbuildvrt -separate output.vrt outfile1.vrt outfile2.vrt
> > outfile3.vrt
> >
> > gdal_translate -of gtiff output.vrt output.tif
> >
> >
> >
> > for more complicated scaling you may need to hand write a vrt
> >
> > http://www.gdal.org/gdal_vrttut.html
> >
> > Brian
> >
> >
> > On Wed, 2011-06-08 at 19:31 -0700, Jonathan Greenberg wrote:
> > > Nikos and GDALers:
> > >
> > >
> > > This is CLOSE to what I'm looking to do, but I'm having a
> > hard time
> > > getting it working properly. Here's my input dataset
> > gdalinfo dump:
> > >
> > >
> > > ***
> > >
> > >
> > > Driver: ENVI/ENVI .hdr Labelled
> > > Files:
> > 10SEP10191223-M2AS-052377832030_01_P003_ps_gs_bl_tc.envi
> > >
> > 10SEP10191223-M2AS-052377832030_01_P003_ps_gs_bl_tc.hdr
> > > Size is 42255, 51712
> > > Coordinate System is:
> > > GEOGCS["WGS 84",
> > > DATUM["WGS_1984",
> > > SPHEROID["WGS 84",6378137,298.257223563,
> > > AUTHORITY["EPSG","7030"]],
> > > TOWGS84[0,0,0,0,0,0,0],
> > > AUTHORITY["EPSG","6326"]],
> > > PRIMEM["Greenwich",0,
> > > AUTHORITY["EPSG","8901"]],
> > > UNIT["degree",0.0174532925199433,
> > > AUTHORITY["EPSG","9108"]],
> > > AUTHORITY["EPSG","4326"]]
> > > Origin = (-120.066479999999999,39.271554000000002)
> > > Pixel Size = (0.000004500000005,-0.000004500000005)
> > > Image Structure Metadata:
> > > INTERLEAVE=BAND
> > > Corner Coordinates:
> > > Upper Left (-120.0664800, 39.2715540) (120d 3'59.33"W,
> > > 39d16'17.59"N)
> > > Lower Left (-120.0664800, 39.0388500) (120d 3'59.33"W, 39d
> > > 2'19.86"N)
> > > Upper Right (-119.8763325, 39.2715540) (119d52'34.80"W,
> > > 39d16'17.59"N)
> > > Lower Right (-119.8763325, 39.0388500) (119d52'34.80"W, 39d
> > > 2'19.86"N)
> > > Center (-119.9714062, 39.1552020) (119d58'17.06"W, 39d
> > > 9'18.73"N)
> > > Band 1 Block=42255x1 Type=UInt16, ColorInterp=Undefined
> > > Band 2 Block=42255x1 Type=UInt16, ColorInterp=Undefined
> > > Band 3 Block=42255x1 Type=UInt16, ColorInterp=Undefined
> > >
> > >
> > > ***
> > >
> > >
> > > I need to end up with a Byte TIF image with good color
> > balance from
> > > this input. Running rgb2pct.py on this dataset just
> > outputted a
> > > "blank" (grey) image. Running gdal_translate -of GTiff -ot
> > Byte
> > > -scale [min] [max] worked somewhat, but its near impossible
> > to get
> > > this looking right, since the scaling I want to do differs
> > from band
> > > to band (e.g. if there any way to use -scale to adjust each
> > band
> > > separately)? Alternatively, how would I get rgb2pct.py
> > working on the
> > > above image correctly? Thanks!
> > >
> > >
> > > --j
> > >
> > >
> > >
> > > On Wed, Jun 8, 2011 at 4:10 PM, Nikolaos Hatzopoulos
> > > <nhatzop at gmail.com> wrote:
> > > why you don't try rgb2pct
> > http://www.gdal.org/rgb2pct.html?
> > >
> > > --Nikos Hatzopoulos
> > >
> > >
> > > On Wed, Jun 8, 2011 at 2:20 PM, Jonathan Greenberg
> > > <greenberg at ucdavis.edu> wrote:
> > >
> > >
> > > Folks:
> > >
> > >
> > > I am using a piece of software which is
> > relying on an
> > > older version of GDAL that doesn't have the
> > "fix" to
> > > deal with > 8 bit geotiffs (it is trying to
> > make a
> > > jpeg overlay but can't from a 16 bit image.
> > I think
> > > this is the issue:
> > >
> > http://trac.osgeo.org/gdal/wiki/TIFF12BitJPEG), but
> > > when I do a straight convert (-ot Byte) the
> > output
> > > image looks really washed out. Any hints
> > for getting
> > > the best quality output from a 16 bit to to
> > an 8 bit
> > > conversion of a Geotiff?
> > >
> > >
> > > --j
> > >
> > > --
> > > Jonathan A. Greenberg, PhD
> > > Assistant Project Scientist
> > > Center for Spatial Technologies and Remote
> > Sensing
> > > (CSTARS)
> > > Department of Land, Air and Water Resources
> > > University of California, Davis
> > > One Shields Avenue
> > > Davis, CA 95616
> > > Phone: 415-763-5476
> > > AIM: jgrn307, MSN: jgrn307 at hotmail.com,
> > Gchat: jgrn307
> > >
> > >
> > >
> > >
> > >
> > _______________________________________________
> > > gdal-dev mailing list
> > > gdal-dev at lists.osgeo.org
> > >
> > http://lists.osgeo.org/mailman/listinfo/gdal-dev
> > >
> > >
> > >
> > >
> > > --
> > > Jonathan A. Greenberg, PhD
> > > Assistant Project Scientist
> > > Center for Spatial Technologies and Remote Sensing (CSTARS)
> > > Department of Land, Air and Water Resources
> > > University of California, Davis
> > > One Shields Avenue
> > > Davis, CA 95616
> > > Phone: 415-763-5476
> > > AIM: jgrn307, MSN: jgrn307 at hotmail.com, Gchat: jgrn307
> > >
> > >
> > > _______________________________________________
> > > gdal-dev mailing list
> > > gdal-dev at lists.osgeo.org
> > > http://lists.osgeo.org/mailman/listinfo/gdal-dev
> >
> >
> >
> >
> >
> >
> > --
> > Jonathan A. Greenberg, PhD
> > Assistant Project Scientist
> > Center for Spatial Technologies and Remote Sensing (CSTARS)
> > Department of Land, Air and Water Resources
> > University of California, Davis
> > One Shields Avenue
> > Davis, CA 95616
> > Phone: 415-763-5476
> > AIM: jgrn307, MSN: jgrn307 at hotmail.com, Gchat: jgrn307
> >
> >
> > _______________________________________________
> > gdal-dev mailing list
> > gdal-dev at lists.osgeo.org
> > http://lists.osgeo.org/mailman/listinfo/gdal-dev
>
>
> _______________________________________________
> gdal-dev mailing list
> gdal-dev at lists.osgeo.org
> http://lists.osgeo.org/mailman/listinfo/gdal-dev
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.osgeo.org/pipermail/gdal-dev/attachments/20110609/5e908a8f/attachment.html
More information about the gdal-dev
mailing list