r.pi.import - GRASS GIS manual

NAME

r.pi.import - Import and generation of patch raster data

KEYWORDS

SYNOPSIS

r.pi.import

r.pi.import --help

r.pi.import [-a] input=string raster=name output=name keyval=integer id_col=integer val_col=integer [title="phrase"] [--overwrite] [--help] [--verbose] [--quiet] [--ui]

Flags:

-a: Set for 8 cell-neighbors. 4 cell-neighbors are default
--overwrite: Allow output files to overwrite existing files
--help: Print usage summary
--verbose: Verbose module output
--quiet: Quiet module output
--ui: Force launching GUI dialog

Parameters:

input=string [required]: Name of the input ASCII-file
raster=name [required]: Name of input raster map
output=name [required]: Name for output raster map
keyval=integer [required]: Category value of the patches
id_col=integer [required]: Number of the column with patch IDs
val_col=integer [required]: Number of the column with patch values
title="phrase": Title for resultant raster map

DESCRIPTION
NOTES
EXAMPLE
SEE ALSO
AUTHORS

DESCRIPTION

Import and generation of patch raster data based on individual patch based raster data.

NOTES

...

EXAMPLE

An example for the North Carolina sample dataset: In order to run r.pi.import we need an exported patch index raster:

r.pi.index input=landclass96 output=landclass96_forestclass5_area keyval=5 method=area

export this resulting map:

r.pi.export input=landclass96_forestclass5_area output=patch_area_out values=patch_area_values id_raster=forestclass5_ID stats=average,variance,min

modify it with R or just import the file again and assign the percentage coverage to each fragment. You need the patch_area_values file and the previously used input file forestclass96 raster (important: the same patch coverage is mandatory otherwise patch ID in the text file and raster are not congruent!):

r.pi.import input=patch_area_values raster=landclass96 output=imported_values keyval=5 id_col=1 val_col=2

if you want to export the patch values to R and do e.g. a linear regression of two patch values and import them again in GRASS, do:
apply r.pi.export with two indices (A and B), in R, do:

resid.AB <- resid(lm(A[,3]~B[,3])) #write residuals of a linear regression
df.resid.AB <- data.frame(A[,1],resid.AB) #merge patch IDs and resid into same data frame
write.table(df.resid.AB,"resid.for.GRASS",row.names=F,col.names=F)

exit R and run in GRASS:

r.pi.import input=resid.for.GRASS raster=landclass96 output=resid.AB keyval=5 id_col=1 val_col=2

AUTHORS

Programming: Elshad Shirinov
Scientific concept: Dr. Martin Wegmann
Department of Remote Sensing
Remote Sensing and Biodiversity Unit
University of Wuerzburg, Germany

Port to GRASS GIS 7: Markus Metz

SOURCE CODE

Available at: r.pi.import source code (history)