[R-sig-Geo] More questions on setting up ENVI headers...
Roger Bivand
Roger.Bivand at nhh.no
Fri Jul 10 09:57:34 CEST 2009
On Thu, 9 Jul 2009, Jonathan Greenberg wrote:
> R-sig-geo'ers:
>
> I am using the following code to create an ENVI header:
>
> driver="ENVI"
> d.drv <- new("GDALDriver", driver)
> tds.out <- new("GDALTransientDataset", driver = d.drv, rows = dims[2],cols
> = dims[1], bands = bands, type = type)
> gt <- c(offset[1] - 0.5 * cellsize[1], cellsize[1], 0.0,offset[2] +
> (dims[2] -0.5) * cellsize[2], 0.0, -cellsize[2])
> print(gt)
> .Call("RGDAL_SetGeoTransform", tds.out, gt, PACKAGE = "rgdal")
> .Call("RGDAL_SetProject", tds.out, file_projection, PACKAGE = "rgdal")
> fn <- tempfile()
> saveDataset(tds.out, fn)
> GDAL.close(tds.out)
>
> gt prints out to be:
> [1] 736801.7 30.0 0.0 4447309.0 0.0 -30.0
>
> The issue I'm having is that the output header, while containing the
> correct projection, is not saving out the pixel sizes, an upper left
> coordinates (e.g. I'm not getting RGDAL_SetGeoTransform to work correctly).
It is tricky, partly because in places the order of the coordinates is
reversed - sometimes (columns, rows) for R-facing functions, often (rows,
columns) for GDAL-facing code). I suspect that this is what is confusing
here. Try to follow the code in writeGDAL() through called functions for
examples.
Roger
> Here's the text of the output header:
>
> ENVI
> description = {
> /var/folders/-Y/-Yv+caBK2RaJhU+8ZQAOk++++TQ/-Tmp-//RtmpX7PvqQ/file4431b782}
> samples = 1172
> lines = 2810
> bands = 7
> header offset = 0
> file type = ENVI Standard
> data type = 4
> interleave = bsq
> byte order = 0
> map info = {UTM, 1, 1, 0, 0, 1, 1, 10, North,North America 1983}
> band names = {
> Band 1,
> Band 2,
> Band 3,
> Band 4,
> Band 5,
> Band 6,
> Band 7}
>
> Thoughts?
>
> --j
>
>
--
Roger Bivand
Economic Geography Section, Department of Economics, Norwegian School of
Economics and Business Administration, Helleveien 30, N-5045 Bergen,
Norway. voice: +47 55 95 93 55; fax +47 55 95 95 43
e-mail: Roger.Bivand at nhh.no
More information about the R-sig-Geo
mailing list