[Geoserver-users] WCS request parameters

djud · January 8, 2009, 3:12pm

Hi,

I'm using a WCS request to download a portion of a coverage stored in
geoserver.

There are two WCS options that can be used to define the returned image
size.

WIDTH / HEIGHT
or
RESX / RESY

As an example (given a constant BBOX), If I use WIDTH = 150 pixels and
HEIGHT=250 pixels (corresponding to X and Y resolutions of 100 m/pixel), the
dimensions of the returned image are correct.

However, if I specify RESX=100 and RESY=100, the returned image has
dimensions: w=250, h=150 (i.e. opposite to the previous).

Both images cover the correct BBOX region but the second image is sampled
incorrectly.

Is the a WCS bug in geoserver or is my understanding of the parameters
incorrect?

Any help appreciated,
Dave

--
View this message in context: http://www.nabble.com/WCS-request-parameters-tp21353711p21353711.html
Sent from the GeoServer - User mailing list archive at Nabble.com.

Alessio_Fabiani · January 8, 2009, 3:36pm

Which CRS are you requesting? The meaning of resX and resY parameters
depend on the Coordinate Reference System in which the BBOX is
expressed.

-------------------------------------------------------
Eng. Alessio Fabiani
Vice-President /CTO GeoSolutions S.A.S.
Via Carignoni 51
55041 Camaiore (LU)
Italy

phone: +39 0584 980933
fax: +39 0584 983027
mob: +39 349 8227000

http://www.geo-solutions.it

-------------------------------------------------------

On Thu, Jan 8, 2009 at 4:12 PM, djud <djud@anonymised.com> wrote:

Hi,

I'm using a WCS request to download a portion of a coverage stored in
geoserver.

There are two WCS options that can be used to define the returned image
size.

WIDTH / HEIGHT
or
RESX / RESY

As an example (given a constant BBOX), If I use WIDTH = 150 pixels and
HEIGHT=250 pixels (corresponding to X and Y resolutions of 100 m/pixel), the
dimensions of the returned image are correct.

However, if I specify RESX=100 and RESY=100, the returned image has
dimensions: w=250, h=150 (i.e. opposite to the previous).

Both images cover the correct BBOX region but the second image is sampled
incorrectly.

Is the a WCS bug in geoserver or is my understanding of the parameters
incorrect?

Any help appreciated,
Dave

--
View this message in context: http://www.nabble.com/WCS-request-parameters-tp21353711p21353711.html
Sent from the GeoServer - User mailing list archive at Nabble.com.

------------------------------------------------------------------------------
Check out the new SourceForge.net Marketplace.
It is the best place to buy or sell services for
just about anything Open Source.
http://p.sf.net/sfu/Xq1LFB
_______________________________________________
Geoserver-users mailing list
Geoserver-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/geoserver-users

djud · January 8, 2009, 4:12pm

It's CRS EPSG:3031. Antarctic Polar Stereographic.

--
View this message in context: http://www.nabble.com/WCS-request-parameters-tp21353711p21354845.html
Sent from the GeoServer - User mailing list archive at Nabble.com.

Alessio_Fabiani · January 8, 2009, 4:23pm

Is it expressed in meters?

-------------------------------------------------------
Eng. Alessio Fabiani
Vice-President /CTO GeoSolutions S.A.S.
Via Carignoni 51
55041 Camaiore (LU)
Italy

phone: +39 0584 980933
fax: +39 0584 983027
mob: +39 349 8227000

http://www.geo-solutions.it

-------------------------------------------------------

On Thu, Jan 8, 2009 at 5:12 PM, djud <djud@anonymised.com> wrote:

It's CRS EPSG:3031. Antarctic Polar Stereographic.

--
View this message in context: http://www.nabble.com/WCS-request-parameters-tp21353711p21354845.html
Sent from the GeoServer - User mailing list archive at Nabble.com.

------------------------------------------------------------------------------
Check out the new SourceForge.net Marketplace.
It is the best place to buy or sell services for
just about anything Open Source.
http://p.sf.net/sfu/Xq1LFB
_______________________________________________
Geoserver-users mailing list
Geoserver-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/geoserver-users

djud · January 8, 2009, 4:31pm

Yes, in metres.

--
View this message in context: http://www.nabble.com/WCS-request-parameters-tp21353711p21355266.html
Sent from the GeoServer - User mailing list archive at Nabble.com.

Alessio_Fabiani · January 8, 2009, 4:43pm

Mmm, quite strange then ... can you send the whole GetCoverage
requests in both cases?

The meaning of the resX and resY params should be resY = BBOX.w /
width and resX = BBOX.h / height

-------------------------------------------------------
Eng. Alessio Fabiani
Vice-President /CTO GeoSolutions S.A.S.
Via Carignoni 51
55041 Camaiore (LU)
Italy

phone: +39 0584 980933
fax: +39 0584 983027
mob: +39 349 8227000

http://www.geo-solutions.it

-------------------------------------------------------

On Thu, Jan 8, 2009 at 5:31 PM, djud <djud@anonymised.com> wrote:

Yes, in metres.

--
View this message in context: http://www.nabble.com/WCS-request-parameters-tp21353711p21355266.html
Sent from the GeoServer - User mailing list archive at Nabble.com.

------------------------------------------------------------------------------
Check out the new SourceForge.net Marketplace.
It is the best place to buy or sell services for
just about anything Open Source.
http://p.sf.net/sfu/Xq1LFB
_______________________________________________
Geoserver-users mailing list
Geoserver-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/geoserver-users

djud · January 8, 2009, 5:02pm

I execute the HTTP request using a hidden form. The two cases (downloadWH and
downloadXY) are as follows:

Thanks for the help,
Dave
--
View this message in context: http://www.nabble.com/WCS-request-parameters-tp21353711p21355960.html
Sent from the GeoServer - User mailing list archive at Nabble.com.

Alessio_Fabiani · January 8, 2009, 5:55pm

Hello,

actually I don't understand exactly what you would like to do but
taking a look at your forms I get a bit different values for RESX and
RESY in order to have the same final resolution, in particular they
should be

RESX = (6470158,7 - 4547761,2) / 424 = 4533,9563679245283018867924528302

RESY = (1233777,5 - 114770) / 246 = 4548,8109756097560975609756097561

is it right? Can you also send if possible the output images obtined?

-------------------------------------------------------
Eng. Alessio Fabiani
Vice-President /CTO GeoSolutions S.A.S.
Via Carignoni 51
55041 Camaiore (LU)
Italy

phone: +39 0584 980933
fax: +39 0584 983027
mob: +39 349 8227000

http://www.geo-solutions.it

-------------------------------------------------------

On Thu, Jan 8, 2009 at 6:02 PM, djud <djud@anonymised.com> wrote:

I execute the HTTP request using a hidden form. The two cases (downloadWH and
downloadXY) are as follows:

<form name="downloadWH" action="http://localhost/geoserver/wcs">
  <input type="hidden" name="service" value="WCS">
  <input type="hidden" name="version" value="1.0.0">
  <input type="hidden" name="request" value="GetCoverage">
  <input type="hidden" name="coverage" value="GST:int_pol.edt">
  <input type="hidden" name="crs" value="EPSG:3031">
  <input type="hidden" name="bbox"
value="-6470158.7,114770,-4547761.2,1233777.5">
  <input type="hidden" name="width" value="246">
  <input type="hidden" name="height" value="424">
  <input type="hidden" name="format" value="GeoTIFF">
</form>

<form name="downloadXY" action="http://localhost/geoserver/wcs">
  <input type="hidden" name="service" value="WCS">
  <input type="hidden" name="version" value="1.0.0">
  <input type="hidden" name="request" value="GetCoverage">
  <input type="hidden" name="coverage" value="GST:int_pol.edt">
  <input type="hidden" name="crs" value="EPSG:3031">
  <input type="hidden" name="bbox"
value="-6470158.7,114770,-4547761.2,1233777.5">
  <input type="hidden" name="resx" value="4531.13905">
  <input type="hidden" name="resy" value="4531.13905">
  <input type="hidden" name="format" value="GeoTIFF">
</form>

Thanks for the help,
Dave
--
View this message in context: http://www.nabble.com/WCS-request-parameters-tp21353711p21355960.html
Sent from the GeoServer - User mailing list archive at Nabble.com.

------------------------------------------------------------------------------
Check out the new SourceForge.net Marketplace.
It is the best place to buy or sell services for
just about anything Open Source.
http://p.sf.net/sfu/Xq1LFB
_______________________________________________
Geoserver-users mailing list
Geoserver-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/geoserver-users

djud · January 9, 2009, 9:50am

I've attached the two WCS files and a dump of the tiff headers.

Yeah, ideally I'd like a resolution of 4531.13905 but, because the width and
height have to be whole integers, you can't get the exact resolution you
want; so 4548.8 and 4533.9 are acceptable.

If you look at the tiff headers though, I don't even get 4548.8 and 4533.9.
Because of the raster type (PixelIsPoint) and probably because it's a tiled
tiff, the WCS returns resolutions of 4676.1 and 4660.8. This again is
acceptable.

The problem is that if I specify the resolutions instead of the width and
height the WCS should return an image that's close to 246x424 but it doesn't
- it returns an image of 424x246 (w=h, h=w) and the resolutions are
consequently really wrong (2713.0 and 8033.3).

Looks like some values are being switched in the code - X has become Y and Y
has become X.

Hope that helps, cheers,
Dave

http://www.nabble.com/file/p21369210/int_pol.edt.wh.tif int_pol.edt.wh.tif
http://www.nabble.com/file/p21369210/int_pol.edt.xy.tif int_pol.edt.xy.tif
http://www.nabble.com/file/p21369210/int_pol.edt.wh.tif.dump
int_pol.edt.wh.tif.dump
http://www.nabble.com/file/p21369210/int_pol.edt.xy.tif.dump
int_pol.edt.xy.tif.dump
--
View this message in context: http://www.nabble.com/WCS-request-parameters-tp21353711p21369210.html
Sent from the GeoServer - User mailing list archive at Nabble.com.

Alessio_Fabiani · January 9, 2009, 10:07am

Ah ok, I see ... we will inspect the code if some error are present in
the wh computation given the xy resolutions.

Many thanks, I'll let you know as soon as possible.

-------------------------------------------------------
Eng. Alessio Fabiani
Vice-President /CTO GeoSolutions S.A.S.
Via Carignoni 51
55041 Camaiore (LU)
Italy

phone: +39 0584 980933
fax: +39 0584 983027
mob: +39 349 8227000

http://www.geo-solutions.it

-------------------------------------------------------

On Fri, Jan 9, 2009 at 10:50 AM, djud <djud@anonymised.com> wrote:

I've attached the two WCS files and a dump of the tiff headers.

Yeah, ideally I'd like a resolution of 4531.13905 but, because the width and
height have to be whole integers, you can't get the exact resolution you
want; so 4548.8 and 4533.9 are acceptable.

If you look at the tiff headers though, I don't even get 4548.8 and 4533.9.
Because of the raster type (PixelIsPoint) and probably because it's a tiled
tiff, the WCS returns resolutions of 4676.1 and 4660.8. This again is
acceptable.

The problem is that if I specify the resolutions instead of the width and
height the WCS should return an image that's close to 246x424 but it doesn't
- it returns an image of 424x246 (w=h, h=w) and the resolutions are
consequently really wrong (2713.0 and 8033.3).

Looks like some values are being switched in the code - X has become Y and Y
has become X.

Hope that helps, cheers,
Dave

http://www.nabble.com/file/p21369210/int_pol.edt.wh.tif int_pol.edt.wh.tif
http://www.nabble.com/file/p21369210/int_pol.edt.xy.tif int_pol.edt.xy.tif
http://www.nabble.com/file/p21369210/int_pol.edt.wh.tif.dump
int_pol.edt.wh.tif.dump
http://www.nabble.com/file/p21369210/int_pol.edt.xy.tif.dump
int_pol.edt.xy.tif.dump
--
View this message in context: http://www.nabble.com/WCS-request-parameters-tp21353711p21369210.html
Sent from the GeoServer - User mailing list archive at Nabble.com.

------------------------------------------------------------------------------
Check out the new SourceForge.net Marketplace.
It is the best place to buy or sell services for
just about anything Open Source.
http://p.sf.net/sfu/Xq1LFB
_______________________________________________
Geoserver-users mailing list
Geoserver-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/geoserver-users