My understanding is that the fdim library is orientated towards vector data filling a space, as opposed to a raster type image. Importantly, I utilise the GRASS library interface with the GRASS GIS to import raster images that are spatially referenced into R. Hence methods such as box-counting and Minkowsky sausage are perhaps more applicable, and certainly found more often in the image-analysis/landscape ecology literature than the information or correlation dimension.
I have managed to come up with a primitive implementation of the box-counting algorithm in the meantime (i'm pleased with it although I am a mediocre programmer), and my next step is an adjacency measure - to tell me whether I have one big patch or lots of small patches.
I'm working from a few references, but one of the most useful is:
L da F Costa & R M Cesar 2001 Shape Analysis and Classification: Theory and Practice. CRC Press. Boca Raton
Thanks for replying
Rohan Sadler
Manuel Castej�n Limas wrote:
Dear Rohan, Have a look at the fdim library, it may be of interest to you as far as fractal dimension (or box counting if you prefer) is concerned.Best wishes, Manuel Castejon ______________________________________________ [EMAIL PROTECTED] mailing list http://www.stat.math.ethz.ch/mailman/listinfo/r-help
-- Ecosystems Research Group (ERGO) School of Plant Biology (Botany), Faculty of Natural & Agricultural Sciences, The University of Western Australia, 35 Stirling Highway, Crawley WA 6009, Australia Ph: +61 8 9380 7914 Fax: +61 8 9380 7925 email: [EMAIL PROTECTED] ERGO's web site:<http://www.botany.uwa.edu.au/ergo> ______________________________________________ [EMAIL PROTECTED] mailing list http://www.stat.math.ethz.ch/mailman/listinfo/r-help
