[GRASSLIST:1624] LANDSAT-7 ETM+ IMAGERY

Friends,

I would like to import L7 imagery into GRASS. I am most interested in a
command or script of commands that interrogates the header files and i
mports the bands noted.

I use r.in.gdal to import L5 imagery, works pretty well. I notice it does not
attempt to utilize the georeferencing information included in the header.
Does anyone have a scripting solution? I assume I should be able to create
transformation matrix using i.points with that metadata.

Specifically, I am interested in developing a means to automate the extraction
of AOI's to individual rasterfiles based on polygon masks, similiar to what
has been recently discussed on the ERDAS list.

I am a re-born GRASS analyst, since I re-installed Linux on my PC. The last
time I worked with GRASS was 3-4 years ago. Loved the power...hated having
multiple operating systems on one machine. I'm back, and re-learning.

TIA
--
Gerald Edward Buckmaster Jr.
Imagery Analysis Support Site & Newsletter
http://www.imagery-analyst.com
http://www.buckoptimized.com
mailto: buckmasterjunior@yahoo.com

On Fri, Mar 16, 2001 at 05:49:46PM -0700, Gerald Buckmaster wrote:

Friends,

I would like to import L7 imagery into GRASS. I am most interested in a
command or script of commands that interrogates the header files and i
mports the bands noted.

Are these BIL or BIP with a .hdr header? I thought r.in.gdal was able
to parse those header files...

I use r.in.gdal to import L5 imagery, works pretty well. I notice it does not
attempt to utilize the georeferencing information included in the header.
Does anyone have a scripting solution? I assume I should be able to create
transformation matrix using i.points with that metadata.

Should be sufficient just to change the cell_head information
(resolutions, extents, etc...). I'm guessing you're getting imagery
that is already georeferenced to UTM? I don't think something as
drastic as i.points/i.rectify should be necessary unless there's some
rotation/scaling that has to be done. Otherwise you'd probably
introduce some error...

Specifically, I am interested in developing a means to automate the extraction
of AOI's to individual rasterfiles based on polygon masks, similiar to what
has been recently discussed on the ERDAS list.

Might have to convert your polygons to raster masks, but then it should work.

I am a re-born GRASS analyst, since I re-installed Linux on my PC. The last
time I worked with GRASS was 3-4 years ago. Loved the power...hated having
multiple operating systems on one machine. I'm back, and re-learning.

--
Eric G. Miller <egm2@jps.net>

Eric,

Allow me to clarify.

I need to import FAST-L7A, which has 3 different header files,
1 for the panchomatic band, 1 for the VNIR/SWIR bands, and
1 header for the Thermal Band.

The header files end with the extension .FST not .hdr.

Actually I think I can use LPGS-LITE Version 4.1.0 to process Landsat-7 ETM+
Level1G data. Haven't tried it yet. Will summarize.

Thanks for your help.

Gerald Buckmaster
www.imagery-analyst.com

On Fri, 16 Mar 2001, Eric G. Miller wrote:

On Fri, Mar 16, 2001 at 05:49:46PM -0700, Gerald Buckmaster wrote:
> Friends,
>
> I would like to import L7 imagery into GRASS. I am most interested in a
> command or script of commands that interrogates the header files and i
> mports the bands noted.

Are these BIL or BIP with a .hdr header? I thought r.in.gdal was able
to parse those header files...

> I use r.in.gdal to import L5 imagery, works pretty well. I notice it does not
> attempt to utilize the georeferencing information included in the header.
> Does anyone have a scripting solution? I assume I should be able to create
> transformation matrix using i.points with that metadata.

Should be sufficient just to change the cell_head information
(resolutions, extents, etc...). I'm guessing you're getting imagery
that is already georeferenced to UTM? I don't think something as
drastic as i.points/i.rectify should be necessary unless there's some
rotation/scaling that has to be done. Otherwise you'd probably
introduce some error...

> Specifically, I am interested in developing a means to automate the extraction
> of AOI's to individual rasterfiles based on polygon masks, similiar to what
> has been recently discussed on the ERDAS list.

Might have to convert your polygons to raster masks, but then it should work.

> I am a re-born GRASS analyst, since I re-installed Linux on my PC. The last
> time I worked with GRASS was 3-4 years ago. Loved the power...hated having
> multiple operating systems on one machine. I'm back, and re-learning.

--
Eric G. Miller <egm2@jps.net>

--
Gerald Edward Buckmaster Jr.
Editor, Webmaster, Imagery Analyst

Imagery Analysis Support Newsletter
Imagery Analysis Support Site

http://www.imagery-analyst.com/index.shtml
http://www.buckoptimized.com/index.shtml

mailto: imagery@imagery-analyst.com
mailto: webmaster@buckoptimized.com
mailto: pi1n1x1@qwest.net