[GRASS-user] GRASS to ID Nuclear Facilities?

[Markus Metz]

On Fri, Apr 26, 2013 at 10:50 PM, Melissa Hanham <mhanham at miis.edu> wrote:
> Re-posted from "announcement" list.
>
> Hi GRASS Developers and Users!
>
> I work for the James Martin Center for Nonproliferation Studies (CNS)
> in Monterey, CA. One of our biggest challenges is identifying
> undeclared nuclear facilities. CNS has experienced imagery analysts,
> and we can identify various signatures of different types of
> facilities, but we'd like to try automating the early stages of
> imagery analysis by using GIS information derived from images to
> predict the likelihood that a nuclear facility may exist in the area
> -- essentially triage the areas which are most likely to host a
> facility.
>
> We feel pretty confident that we can develop known and new signatures
> of different types of nuclear facilities, but don't have the in-house
> capability on object-based image analysis using GRASS GIS.
>
> My questions are:
>
> 1) Is this a good use of GRASS?
Given known and new signatures of different types of nuclear
facilities, it should be easy to automatically identify objects that
match these signatures using existing GRASS functionality.

Maybe of interest is also a GRASS module for object recognition,
i.segment. It should be possible to filter the objects identified by
i.segment with known signatures.

> 2) Where can I find a developer?

On the GRASS developer mailing list:
http://lists.osgeo.org/mailman/listinfo/grass-dev

and on the osgeo service provider site:
http://www.osgeo.org/search_profile

Markus M

>
> Kind regards,
> Melissa
>
> --
> Melissa Hanham
> Research Associate
> +1.831.647.6507
> James Martin Center for Nonproliferation Studies
> Combating the spread of weapons of mass destruction with training & analysis
> _______________________________________________
> grass-user mailing list
> grass-user at lists.osgeo.org
> http://lists.osgeo.org/mailman/listinfo/grass-user