Message Boards Message Boards

GeoRaster should be a thing

A GeoRaster object isn't a thing, but I can't think of anything more generally useful.

A GeoRaster object would bundle a raster — either an Image or a 2D Array — with the information needed to geo-locate it: a projection definition and either a LL origin coordinate and cell size, or LL and UR coordinates, in the specified projection (or in lat/long if no projection is specified).

A GeoRaster is basically the internal Mathematica representation of a GeoTIFF, and importing a GeoTIFF would, by default, produce a GeoRaster object.

Most importantly, GeoRaster should function as a GeoGraphics primitive, in that GeoGraphics[{myGeoRaster}, GeoProjection->xxx, etc.] should result in the raster being re-projected into the GeoGraphics projection and overlaid on the background (the same way the background images themselves are reprojected — presumably the internal code for reprojection is already present).

GeoRasters could be uploaded to the Wolfram Data Repository so that anyone can and grab and display raster geographic data with a minimum of fuss. For example, in my field there are loads of biodiversity mapping projects that all essentially produce this kind of output.

Would this be useful to other people?

Perhaps this in the works for v13. Any Wolfram insiders have insight?

POSTED BY: Gareth Russell
3 Replies

Ah yes, so I see! Add me to the list then. A small difference is that I am suggesting an object, analogous to GeoVector, rather than a special plotting function. (Although I see that a GeoVector needs to be wrapped in GeoMarker to become a GeoGraphics primitive.)

POSTED BY: Gareth Russell

Got it. Yes, better.

POSTED BY: Vincent Virgilio
Reply to this discussion
Community posts can be styled and formatted using the Markdown syntax.
Reply Preview
Attachments
Remove
or Discard

Group Abstract Group Abstract