5 replies to this topic

[GIS_noob]

I downloaded from New EarthExplorer a zip file, GLS 2000, which unzipped to quite a number of geoTIFF files and a .met file which looks like it knows how to create a multispectral color composite image from them all. However I am unable to figure out how to put it together properly.

Any tips at all about tools or processes needed to create the image will be appreciated.

Thanks

[frax]

What does the tiff files hold? what software do you have at hand?

[GIS_noob]

What does the tiff files hold? what software do you have at hand?

The tiff images are all greyscale, and seem to be seperate channels, R G B perhaps, and someother greyscal information. They come with a .met file which contains references to multi bands:
BAND1_FILE_NAME = "p009r055_7dt20000821_z18_10.tif"
BAND2_FILE_NAME = "p009r055_7dt20000821_z18_20.tif"
BAND3_FILE_NAME = "p009r055_7dt20000821_z18_30.tif"
BAND4_FILE_NAME = "p009r055_7dt20000821_z18_40.tif"
BAND5_FILE_NAME = "p009r055_7dt20000821_z18_50.tif"
BAND61_FILE_NAME = "p009r055_7dk20000821_z18_61.tif"
BAND62_FILE_NAME = "p009r055_7dk20000821_z18_62.tif"
BAND7_FILE_NAME = "p009r055_7dt20000821_z18_70.tif"
BAND8_FILE_NAME = "p009r055_7dp20000821_z18_80.tif"
GROUP = PROJECTION_PARAMETERS
REFERENCE_DATUM = "WGS84"
REFERENCE_ELLIPSOID = "WGS84"
GRID_CELL_ORIGIN = "Center"
UL_GRID_LINE_NUMBER = 1
UL_GRID_SAMPLE_NUMBER = 1
GRID_INCREMENT_UNIT = "Meters"
GRID_CELL_SIZE_PAN = 15.000
GRID_CELL_SIZE_THM = 60.000
GRID_CELL_SIZE_REF = 30.000
FALSE_NORTHING = 0
ORIENTATION = "NUP"
RESAMPLING_OPTION = "CC"
MAP_PROJECTION = "UTM"
END_GROUP = PROJECTION_PARAMETERS

I have Geographic Imager for photoshop from Avenza, GeoExpress 7 from LizardTech, and assorted freeware and demo tools (ArcGIS Explorer, MicroDEM, etc

[archimedian]

What does the tiff files hold? what software do you have at hand?

The tiff images are all greyscale, and seem to be seperate channels, R G B perhaps, and someother greyscal information. They come with a .met file which contains references to multi bands:
BAND1_FILE_NAME = "p009r055_7dt20000821_z18_10.tif"
BAND2_FILE_NAME = "p009r055_7dt20000821_z18_20.tif"
BAND3_FILE_NAME = "p009r055_7dt20000821_z18_30.tif"
BAND4_FILE_NAME = "p009r055_7dt20000821_z18_40.tif"
BAND5_FILE_NAME = "p009r055_7dt20000821_z18_50.tif"
BAND61_FILE_NAME = "p009r055_7dk20000821_z18_61.tif"
BAND62_FILE_NAME = "p009r055_7dk20000821_z18_62.tif"
BAND7_FILE_NAME = "p009r055_7dt20000821_z18_70.tif"
BAND8_FILE_NAME = "p009r055_7dp20000821_z18_80.tif"
GROUP = PROJECTION_PARAMETERS
REFERENCE_DATUM = "WGS84"
REFERENCE_ELLIPSOID = "WGS84"
GRID_CELL_ORIGIN = "Center"
UL_GRID_LINE_NUMBER = 1
UL_GRID_SAMPLE_NUMBER = 1
GRID_INCREMENT_UNIT = "Meters"
GRID_CELL_SIZE_PAN = 15.000
GRID_CELL_SIZE_THM = 60.000
GRID_CELL_SIZE_REF = 30.000
FALSE_NORTHING = 0
ORIENTATION = "NUP"
RESAMPLING_OPTION = "CC"
MAP_PROJECTION = "UTM"
END_GROUP = PROJECTION_PARAMETERS

I have Geographic Imager for photoshop from Avenza, GeoExpress 7 from LizardTech, and assorted freeware and demo tools (ArcGIS Explorer, MicroDEM, etc

It sounds like what you have is Global Land Survey 2000 data. This is composited multispectral satellite imagery which is built up out of "best available pixels" from multiple different LandSat ETM+ images captured around 2000 and intended for decadal-scale science on land-use/land-cover change.

The GLS home page is here: http://gls.umd.edu/index.html
The USGS EROS Data Center's page on the product is here: http://eros.usgs.gov...a_Available/GLS
Information on the individual LandSat ETM+ bands can be found here: http://landsathandbo...html#section6.2

Band 1 is blue, Band 2 is green, Band 3 is red, Band 4 is near infrared; these and bands 5 and 7 are 30m. Band 8 is 15m pan-chromatic (black and white). For common cartographic purposes, you wouldn't want anything other than bands 1,2,3, and possibly 8. (ArcGIS can now do a trick out of the box called "pan-sharpening", that takes the multispectral information of the color bands and fuses it with the higher-spatial-resolution of the pan-chromatic band.)

You want software that can read single-band imagery and write out a multi-band raster. Common tools in the remote sensing industry are ERDAS Imagine and ENVI, but both are very pricey. The open-source scripting language Python can do this if you have a little programming familiarity. ArcGIS has a tool to do this as well (Data Management > Raster > Raster Processing > Composite Bands) but it sounds like this app is not an option either.

I just tried downloading ArcExplorer 9.3.1 Java edition which can supposedly handle the display of GeoTIFFS, but I had no luck (it didn't recognize the data when I pointed to a GLS tile that I just downloaded. Don't know if MicroDEM can do this either. GlobalMapper can read multiple single-band TIFFS and write out a 3-band composite, but the trial version of that app won't allow you to export or save anything.

That's a whole lot of negative results - sorry, and best of luck!

[Kimi]

I will be happy to introduce you how to create a color composite files with Geographic Imager and Adobe Photoshop

I have multiple spectral data (Band 1, 2, 3, 4, 5, 7, and 8) and the goal is to create a false color composite files with the images from Band 2, 3, and 4.

1) Open three images (band 2, 3, and 4) in Adobe Photoshop - Geographic Imager. You will see that Georeference information stored in those Geotiff files.

2) Open the Channels panel from Window > Channels from Adobe Photoshop menu bar

3) From the option menu from the Channels, select "Merge Channels" option

4) In the Merge Channels option dialog window, select "RGB Color" for Mode and "3" for Channels. Click OK.

5) Now assigning each of the band to each of the color bands. I assigned Band 4 for Red, Band 3 for Green, and Band 2 for Blue.

6) Now you can see my false-color composite image in the example below


I hope this helps!

p.s.--
Geographic Imager, compatible with Adobe Photoshop CS5, will be released shortly

Best regards,
Kimi

Attached Files

•  03_MergeChannels.png   24.96K   14 downloads
•  04_MergeChannel_RGB.PNG   12.02K   13 downloads
•  05_FCC.PNG   34.16K   11 downloads
•  06_myImage_FCC.PNG   326.56K   13 downloads

[GIS_noob]

Thanks for the great information, there was just enough information to get me pointed in the correct direction. Thank you very much!