Ciao Martin,
thanks for the info.

I have a few more questions, since I would not mind at all fixing the
few issues that Andrea is having and then join your effort despite the
little time I have fofr this.

1>Did you envision the possibility to do alpha blending between
coverages that overlap?
2>Did you envision the possibility to control the background for the mosaic?
3>Did you consider the usage of overviews?

On Dec 28, 2007 5:22 PM, Martin Desruisseaux
<[EMAIL PROTECTED]> wrote:
> Simone Giannecchini a écrit :
> > How do you assemble them if you are not using the mosaic?
>
> Step 1
> ------
> Create the target BufferedImage (initially empty). This is the only image
> created, and it is created with exactly the final size as specified in
> ImageReadParam, nothing more.
>
> Step 2
> ------
> For each tile, invoke the following:
>
>  * ImageReadParam.setSourceRegion(Rectangle)
>    With only the part of the tile we want to read
>
>  * ImageReadParam.setDestination(image)
>    where 'image' is the BufferedImage created at step 1.
>
>  * ImageReadParam.setDestinationOffset(Point)
>    with the coordinate where to write the image.
>

Believe it or not we where discussing with daniele about adding
support for these two methods in imageio-ext just before christmas!!!
Well it seems that we've got to that in the end! :-)



>
> Step 3
> ------
> Invoke ImageReader.read(param) for each tile. You can verify that it returns
> exactly the same BufferedImage than the one specified at step 2. No new
> BufferedImage was created.
>
>
>
> Aftermath
> ---------
> Exactly *ONE* and only one image is created, only the final destination image
> and nothing else. No temporary BufferedImage created for each tile. No copy
> needed using JAI "Mosaic" operation. Actually no operation at all, neither JAI
> or whatever - the pixels are exactly where they should be straight after
> ImageRead.read(...) call, nothing more, nothing less.
>
>
> References
> ----------
> http://java.sun.com/javase/6/docs/api/javax/imageio/ImageReadParam.html#setDestination(java.awt.image.BufferedImage)
> http://java.sun.com/javase/6/docs/api/javax/imageio/IIOParam.html#setDestinationOffset(java.awt.Point)
>
>
>
> > Are you designin with in mind using deferred loading via JAI ImageRead
> > or directly the ImageRead as you showed me for SeaGis?
>
> This work is done at javax.imageio.ImageReader level, so JAI is not involved 
> at
> all at this level. The MosaicImageReader can be wrapped in a JAI "ImageRead"
> operation like any other ImageReader operation, at user choice. If the
> "ImageRead" operation requires only a Tile, it should be cascaded through
> MosaicImageReader straight to the appropriate image to read.
>
>
>         Martin
>

4>What kind of index do you envision to use/support for the mosaic?

I probably have other questions but I will fire them way after the nye
break, after looking at the code (is it still in coveraegio
unsupported?)


Ciao,
Simone.

-- 
-------------------------------------------------------
Eng. Simone Giannecchini
President /CEO GeoSolutions S.A.S.
Via Carignoni 51
55041  Camaiore (LU)
Italy

phone: +39 0584983027
fax:      +39 0584983027
mob:    +39 333 8128928


http://www.geo-solutions.it

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

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Geotools-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/geotools-devel

Reply via email to