Urban density algorithms

[James Darrell McCauley]

Piero Boccardo - Italy (DORICE at polito.it) writes to 
grassu-list at moon.cecer.army.mil on  8 March 1995:
>I'm trying to find out an algorithm capable to slice urban areas in
>different density sets using Tm data. Has anyone any idea about that?

I'd suggest looking at i.gensigset and i.smap (Sequential MAP, see
appended refs).

I assume that you're talking about population densities - if
that's the case, I would be inclined to take a more naive approach
of calculating a texture map (one of Haralick's features; see i.texture
in GRASS) of just one band and seeing what kind of correlations this 
gives me, but then again I may be doing things the hard way.

The kind folks on the IMAGRS list may be able to provide an
answer or at least some leads to previous work.
-- 
James Darrell McCauley, PhD        http://soils.ecn.purdue.edu/~mccauley/
Dept of Agricultural Engineering   mccauley at ecn.purdue.edu
Purdue University            new-> tel: 317.494.1198 fax: 317.496.1115

@Article{ bouman94,
  author =       "Charles Bouman and Michael Shapiro",
  title =        "A Multiscale Random Field Model for {B}ayesian Image
                  Segmentation",
  journal =      "IEEE Transactions on Image Processing",
  year =         1994,
  volume =       3,
  number =       2,
  pages =        "162--177",
  month =        mar
}

@InProceedings{ bouman92,
  author =       "Charles Bouman and Michael Shapiro",
  title =        "Multispectral Image Segmentation using a Multiscale
                  Image Model",
  pages =        "III-565--III-568",
  booktitle = "Proc. of IEEE Int'l Conf. on Acoust., Speech and Sig. Proc.",
  year =         "1992",
  organization = "IEEE",
  address =      "San Francisco, " # CA,
  month =        "March 23--26"
}