[R] finding contours in a matrix

Roger Bivand Roger.Bivand at nhh.no
Wed Jun 18 18:35:36 CEST 2003


On Wed, 18 Jun 2003, Douglas Beare wrote:

> Hi,
> I have matrix of sea bottom depths that I am plotting in R with the function
> 'image'.
> I am particularly interested in the 200m depth contour (I'm using 'distance
> from this feature' as a covariate in a model) and would like to extract the
> data at evenly 
> spaced points along it.  I can easily superimpose a line at 200m using the
> function 'contour'.  What I
> want to know is: can I actually get the data used to plot this line? If I
> just select the relevant data depth[depth==200] I
> get a very few observations, but if I say select depth[depth > 180 & depth <
> 220] I get a "band" of 
> unevenly space data along the contour which I don't want either.

This is a different question to the "can I retrieve a contour line from 
contour()" - it presupposes that we have the line, but asks which cells of 
the array are intersected by the line, is that right? It could also be 
used to interrogate any other image/raster layer for the same area, not 
just depth. Polygon clipping is available in the gcplib package, but that 
may be overkill. If the (contour) line(s) can be converted into sequences 
of points closer together than the raster resolution, it should be 
possible to identify the raster cells they belong to (row and column 
numbers). If this was GIS data, it might be easier to do the overlay 
operation there, especially if you need a buffer around the line.

Roger

> Regards,
> Doug Beare
> Fisheries Research Services,
> Marine Laboratory,
> Victoria Road,
> Torry,
> Aberdeen, UK.
> Tel. 44 (0) 1224 295314
> 
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://www.stat.math.ethz.ch/mailman/listinfo/r-help
> 

-- 
Roger Bivand
Economic Geography Section, Department of Economics, Norwegian School of
Economics and Business Administration, Breiviksveien 40, N-5045 Bergen,
Norway. voice: +47 55 95 93 55; fax +47 55 95 93 93
e-mail: Roger.Bivand at nhh.no




More information about the R-help mailing list