On Mon, Jun 17, 2013 at 3:36 PM, Andrea Aime
<andrea.a...@geo-solutions.it>wrote:
> On Mon, Jun 17, 2013 at 7:38 PM, Chris Holmes <chol...@opengeo.org> wrote:
>
>> One thing I have been mulling over lately is if we might be able to make
>> it easier to go from a single image to an image mosiac / image pyramid.
>>
>> Like from a user's perspective they don't really feel like distinct
>> 'stores', they feel more like advanced configuration options for an image.
>> Like ideally I'd like to be able to upload an big image, and then maybe
>> have it pyramided in GeoServer for me, instead of creating it and
>> uploading. I guess you guys have built some stuff to do that with the
>> importer. And maybe that's just the way we present it.
>>
>
> That's something we did not build. Our version of the importer allows
> single files and directories of tiffs to be imported (as list of tiff
> stores) and allows the admin to add overviews and inner tiling. There is no
> special handling of mosaics and pyramids.
>
>
Ah, ok, you just do them as inner tiles and overviews in the tiff directly.
Gotcha.
>
>> The use case I've been thinking about a decent bit is MapStory. Most
>> everything is time based, so we need image mosiacs to handle them - we want
>> one layer with different time slices. It's all crowd sourced / user
>> contributed. And one of our big use cases is taking scanned maps and
>> rectifying them, using the warper built for nypl (see
>> http://warper.mapstory.org). What we'd ideally like to do is have people
>> be able to upload more scanned maps to an existing layer.
>>
>> Now I think we'll be fine - we'll just create everything as an image
>> mosiac to start, even if it's just one layer, so we can add more in the
>> future. But from a user or even programmatic client perspective it'd be
>> nice if GeoServer handled that abstraction. You just have an image layer,
>> and you could choose to add mosiacs to it, or add pyramids to it. Or have
>> it generate pyramids, etc.
>>
>> I realize this is a big topic, just curious on current thinking, if it's
>> a goal to unite image stuff in some way. Or if that's just unrealistic. Or
>> if there's better routes to achieve it.
>>
>
> I don't think it's unrealistic, on the contrary, makes sense, seems a
> reasonable option to allow the creation of a mosaic instead of a single
> tiff store out of a single file, but we don't have anything ready to
> support it. Pyramid out of a single file or a bunch of files would make
> sense too, but we'd need steps to build the pyramid out of the file, it's
> not something you add to later (at least, that's not easy, plus pyramids at
> the moment have no support for dimensions as far as I know, the datasets
> we're used to deal with tend to have a small/trivial spatial extent and a
> very large temporal/dimensional one, so using pyramid on it makes no sense).
>
>
Cool, yeah, I always get a bit lost on pyramids vs. mosiacs vs. tiling
within tiffs. MapStory before too long we probably will have a dataset with
large spatial extent and large temporal one, but short term is definitely
large temporal and probably smaller spatial. So we'll be able to get along
fine with just doing geotiff overviews and inner tiling I imagine.
Thanks for the thoughts Andrea.
C
> Cheers
> Andrea
>
> --
> ==
> Our support, Your Success! Visit http://opensdi.geo-solutions.it for more
> information.
> ==
>
> Ing. Andrea Aime
> @geowolf
> Technical Lead
>
> GeoSolutions S.A.S.
> Via Poggio alle Viti 1187
> 55054 Massarosa (LU)
> Italy
> phone: +39 0584 962313
> fax: +39 0584 1660272
> mob: +39 339 8844549
>
> http://www.geo-solutions.it
> http://twitter.com/geosolutions_it
>
> -------------------------------------------------------
>
------------------------------------------------------------------------------
This SF.net email is sponsored by Windows:
Build for Windows Store.
http://p.sf.net/sfu/windows-dev2dev
_______________________________________________
Geoserver-devel mailing list
Geoserver-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/geoserver-devel
