so .,.. probably at the moment the best solution is to reproject the raster off 
line with some client tools (gdal?) and have two versions loaded in geoserver

Am i right?

________________________________
Inizio: andrea.a...@gmail.com [andrea.a...@gmail.com] per conto di Andrea Aime 
[andrea.a...@geo-solutions.it]
Inviato: venerdì 6 gennaio 2012 10.59
Fine: Pierluigi Santin Fintel
Cc: geoserver-users@lists.sourceforge.net
Oggetto: Re: [Geoserver-users] raster reprojection

On Thu, Jan 5, 2012 at 4:28 PM, Pierluigi Santin Fintel 
<p.san...@fintelengineering.it<mailto:p.san...@fintelengineering.it>> wrote:
Hi all!
hope to find someone that could help me.
I have a big raster (geotiff) georeferenced in EPSG:3003

I've loaded it to geoserver everything is fine if I query the layer in WMS 
using 3003, but if I query the same layer using epsg 900913, the quality of the 
reprojection is very poor. I've adapted the proj parameters to correcly 
reproject 3003 to 900913 for my local area but while this works perfectly for 
vector data the result is instead very bad for rasters.

Am i missing something?

I cannot be sure without looking at your data, but certainly there are a couple 
of
things missing to get optimal quality in raster reprojection:
- be able to select the overview policy. It can be set via code, but it's not 
exposed
  anywhere in GeoServer. At the moment the code uses the "closest overview"
  policy, which is a compromise between speed and quality.
  There is another policy that always chooses the overview with the higher 
resolution
  than the request, that one provides better results at the expense of 
performance
- take into account the reprojection expansion. The Google projection expands
  pixels in the south to north direction, Italy is north enough that the 
expansion
  starts to become visible. Long story short, to get good results we should read
  more pixels from the source than what we have in the final output image,
  but right now we don't. We'd need an estimator that can determine how much
  more of the input we should read (different projections expand the pixels
  in different directions, and by different amounts depending on where on the
  earth you are using them).

Internally in my company we discussed these problems a few times already,
but so far failed to raise the funding necessary to turn the ideas into actual 
code

Cheers
Andrea

--
-------------------------------------------------------
Ing. Andrea Aime
GeoSolutions S.A.S.
Tech lead

Via Poggio alle Viti 1187
55054  Massarosa (LU)
Italy

phone: +39 0584 962313
fax:      +39 0584 962313
mob:    +39 339 8844549

http://www.geo-solutions.it
http://geo-solutions.blogspot.com/
http://www.youtube.com/user/GeoSolutionsIT
http://www.linkedin.com/in/andreaaime
http://twitter.com/geowolf

-------------------------------------------------------
------------------------------------------------------------------------------
Ridiculously easy VDI. With Citrix VDI-in-a-Box, you don't need a complex
infrastructure or vast IT resources to deliver seamless, secure access to
virtual desktops. With this all-in-one solution, easily deploy virtual 
desktops for less than the cost of PCs and save 60% on VDI infrastructure 
costs. Try it free! http://p.sf.net/sfu/Citrix-VDIinabox
_______________________________________________
Geoserver-users mailing list
Geoserver-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/geoserver-users

Reply via email to