Ernst Jaarsma wrote: > I recently needed a frequency filter for Aster-gdem-data. I am aware > that it is possible to smooth the data by averaging, but a frequency > filter would be more accurate and would give more options. > > So my questions are: * Has this been considered? * Would it be > possible (regarding calculation time / memory resources)?* If not > considered, and possible; Is it good idea for other people as > well?
GRASS has i.fft and i.ifft. However, in 6.x, the only operation which can be performed upon the FFT data is a binary mask; a given component can be included or excluded, but not scaled. In 7.0, the frequency coefficients are saved as raster maps, so they can be manipulated with e.g. r.mapcalc. The main constraint is that both i.fft and i.ifft require the data to fit into memory. Eliminating this requirement boils down to transposing a large array efficiently. -- Glynn Clements <[email protected]> _______________________________________________ grass-dev mailing list [email protected] http://lists.osgeo.org/mailman/listinfo/grass-dev
