Hi, Thanks for the link Michael. I'll have a read later.
I've done some work on comparing point distributions using a raster based approach. My Use Case was comparing the ditribution of road accidents over time. Every accident was located using a grid reference (a point given to a fixed precision), so effectively this is raster data. At the resolution of the raster I can compare/difference distribution. However, displaying the raster at the raster resolution for a large region doesn't tend to reveal much (unless the viewer zooms up close and is thus viewing for a smaller region). So, the next step I took was to generalise over a spatial scale using a distance weighted function to compare distributions. At this stage, the generalisation (interpolation) could take various forms and the difference between Zero and NoData observations is cruicial. So, I did some work on what I called Geographically Weighted Statistics. I should have written this up better, but somehow have not got around to it. The following is a URL to a Word document that drafts this work (written in the old days): http://www.geog.leeds.ac.uk/people/a.turner/publications/papers/working/ RasterBasedGWS_0.2.doc I have developed some software for doing this and provided more options for the distance based generalisation weightings. It is all Java and an old version is available via the following URL: http://www.geog.leeds.ac.uk/people/a.turner/src/andyt/java/gws/ If anyone is interested then I will try to find time to make a new release. Perhaps it would be good to fold this into GeoTools, I have wanted to, but not managed it. The Grids package on which it was based I developed for handling large raster data. It does like a swapping thing that may be like TileCache, but it operates by catching OutOfMemoryErrors. That complicates things, but... Anyway that Grids library is here: http://www.geog.leeds.ac.uk/people/a.turner/src/andyt/java/grids/ I have registered my interest in this kind of thing with OGC and I managed to go to the Coverages Domain Working Group meeting at the last Technical Committee (TC) meeting (Potsdam 2008-06). Martin, a year has passed since we met and talked about this at the OGC TC meeting in Paris! Best wishes, Andy http://www.geog.leeds.ac.uk/people/a.turner/ -----Original Message----- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Michael Bedward Sent: 12 June 2008 03:10 To: Martin Desruisseaux; [email protected] Subject: [Geotools-gt2-users] Interpolating missing data in a raster (WASRe: Status of Coverages in GeoAPI and GeoTools) Hi Martin, Your example, below, made me remember an interesting paper on a novel method of interpolating missing data in raster images which might be useful if you haven't seen it: http://sprott.physics.wisc.edu/pubs/paper276.htm Michael On Thu, Jun 12, 2008 at 12:50 AM, Martin Desruisseaux <[EMAIL PROTECTED]> wrote: > > Then everytime an evaluate method is invoked, GeoTools will tries the > bicubic interpolation first. If it get NaN, then it tried bilinear > interpolation (which requires less pixels, so less chances that one of those pixels has a NaN value). > If the later produced NaN as well, then if fallback on last resort to > nearest neighboard. This fallback mechanism is useful for coverage > like "sea surface temperature" with lot of "nodata values" (typically > clouds), and we want to interpolate only when it will not cause a lost of data. > > Martin ------------------------------------------------------------------------ - Check out the new SourceForge.net Marketplace. It's the best place to buy or sell services for just about anything Open Source. http://sourceforge.net/services/buy/index.php _______________________________________________ Geotools-gt2-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/geotools-gt2-users ------------------------------------------------------------------------- Check out the new SourceForge.net Marketplace. It's the best place to buy or sell services for just about anything Open Source. http://sourceforge.net/services/buy/index.php _______________________________________________ Geotools-gt2-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/geotools-gt2-users
