spatial analyses from picture

Hello, 

i think you can do that with Bio7 (OpenSource):
 
http://bio7.org

It integrates a R GUI

see: http://bio7.org/documentation/Main.html#toc-Section-4.6

and ImageJ GUI

see: http://bio7.org/documentation/Main.html#toc-Section-4.3

in one platform. 

Images, selected image data and selection coordinates (of different
selection types) can be transferred to an R workspace.

see: http://bio7.org/documentation/Main.html#toc-Subsubsection-4.3.3.1

In addition you can transfer the results of a ImageJ Particle Analysis (e.g.
centroids, diameter, e.g.) from thresholded images directly to R:

http://rsbweb.nih.gov/ij/docs/guide/146-30.html#toc-Subsection-30.2

In your case you could threshold the different colours in ImageJ, select
them (Wand Tool) and send the shapes and pixels to R for a spatial analysis.

Or if you have more than one shape (pattern) with the same signature you can
make a Particle Analysis from thresholded images and send the results to a
ROI (Region Of Interest) Manager:

http://rsbweb.nih.gov/ij/docs/guide/146-30.html#sub:ROI-Manager...

The ROI measurements (pixels) and ROI selection data from the different
shapes can be transferred with two available actions in Bio7. In addition
you can transfer the selection data of a stack of images, too.

The results can be analyzed with R e.g. with the great spatstat package:

http://www.spatstat.org/ 

or maybe the rgeos package is useful for you:

http://www.inside-r.org/packages/cran/rgeos/docs/gDistance

Best regards

Marcel




--
View this message in context: http://r-sig-geo.2731867.n2.nabble.com/spatial-analyses-from-picture-tp7586471p7586475.html
Sent from the R-sig-geo mailing list archive at Nabble.com.