Hi, Your response to my post was EXTREMELY useful. For georegistering I used the site:
http://www.runningmap.com which lets one click on a a Yahoo! map and get lat/long in lower right. As my data is in UTM I converted the lat/long to UTM using: http://home.hiwaay.net/~taylorc/toolbox/geography/geoutm.html Then using locator() in R I picked out the corresponding points on the R image that I had created using your code. That allowed me to calculate the inputs for GridTopology (are there any convenience functions to assist in this? -- I found it a bit error prone) (I actually noticed that there is a lat/long in Yahoo! map URLs so it might be that if the maps are a known width and height that one could georegister them based only on that.) At any rate I seem to be having some success with this. Thanks !!! On 3/4/07, Michael Sumner <[EMAIL PROTECTED]> wrote: > > Is there any R software that create an image from Yahoo maps together > > with points of known UTM coordinates (or lat/long marked? Note that my > > region of interest is not covered in sufficient detail by Google maps. > > It actually does not have to be Yahoo maps as long as it has > > sufficient coverage of my region but that's the one I have found with > > coverage of my region. The scale I am interested in is a city block. > > Thanks. > > rgdal in combination with sp. You'll need to georegister the image, as > the yahoo maps interface is unlikely to provide a format supporting this > metadata, but you can do that easily with two reference coordinates. > > The workflow might go like this (untested). You could continue the > query on R-Sig-Geo. Hope this helps. > > There are freely available tools for reading image data directly from > Yahoo, Google and Virtual Earth in georeferenced > versions, but I've only used them via commercial GIS. If you can > describe the map you want I'd be interested in > exploring that option. > > Cheers, Mike. > > library(rgdal) > im <- readGDAL("yahooMap.png") ## a file saved from Yahoo Maps > > ## create index for RGB colours > col <- SGDF2PCT(im) ## im is a spatialGridDataFrame with 3 bands > > im$ind <- col$idx ## add the colour index to the data frame > > image(im, "ind", col = col$ct) > > ## BUT you won't be able to plot UTM on this yet > > ## you'll need to create a new GridTopology with appropriate > cellcentre.offset and cellsize > > ogt <- getGridTopology(im) > . . . find values for offsets and pixel size > > gt <- GridTopology(cc.offset, csize, [EMAIL PROTECTED]) > > ## recreate the object > im <- SpatialGridDataFrame(gt, [EMAIL PROTECTED]) > > image(im, "idx", col = col$ct) > points(utm.x, utm.y) ## etc. > > ______________________________________________ > R-help@stat.math.ethz.ch mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code. > ______________________________________________ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.