.TH i.pca 1grass "" "GRASS 8.3.1" "GRASS GIS User's Manual"
.SH NAME
\fI\fBi.pca\fR\fR  \- Principal components analysis (PCA) for image processing.
.SH KEYWORDS
imagery, transformation, PCA, principal components analysis
.SH SYNOPSIS
\fBi.pca\fR
.br
\fBi.pca \-\-help\fR
.br
\fBi.pca\fR [\-\fBnf\fR] \fBinput\fR=\fIname\fR[,\fIname\fR,...] \fBoutput\fR=\fIbasename\fR  [\fBrescale\fR=\fImin,max\fR]   [\fBpercent\fR=\fIinteger\fR]   [\-\-\fBoverwrite\fR]  [\-\-\fBhelp\fR]  [\-\-\fBverbose\fR]  [\-\-\fBquiet\fR]  [\-\-\fBui\fR]
.SS Flags:
.IP "\fB\-n\fR" 4m
.br
Normalize (center and scale) input maps
.br
Default: center only
.IP "\fB\-f\fR" 4m
.br
Output will be filtered input bands
.br
Apply inverse PCA after PCA
.IP "\fB\-\-overwrite\fR" 4m
.br
Allow output files to overwrite existing files
.IP "\fB\-\-help\fR" 4m
.br
Print usage summary
.IP "\fB\-\-verbose\fR" 4m
.br
Verbose module output
.IP "\fB\-\-quiet\fR" 4m
.br
Quiet module output
.IP "\fB\-\-ui\fR" 4m
.br
Force launching GUI dialog
.SS Parameters:
.IP "\fBinput\fR=\fIname[,\fIname\fR,...]\fR \fB[required]\fR" 4m
.br
Name of two or more input raster maps or imagery group
.IP "\fBoutput\fR=\fIbasename\fR \fB[required]\fR" 4m
.br
Name for output basename raster map(s)
.br
A numerical suffix will be added for each component map
.IP "\fBrescale\fR=\fImin,max\fR" 4m
.br
Rescaling range for output maps
.br
For no rescaling use 0,0
.br
Default: \fI0,255\fR
.IP "\fBpercent\fR=\fIinteger\fR" 4m
.br
Cumulative percent importance for filtering
.br
Options: \fI50\-99\fR
.br
Default: \fI99\fR
.SH DESCRIPTION
\fIi.pca\fR is an image processing program based on the
algorithm provided by Vali (1990), that processes n
(n >= 2) input raster map layers and produces n output
raster map layers containing the principal components of
the input data in decreasing order of variance
(\(dqcontrast\(dq).  The output raster map layers are assigned
names with .1, .2, ... .n suffixes. The numbers used as suffix
correspond to percent importance with .1 being the scores of the
principal component with the highest importance.
.PP
The current geographic region definition and MASK settings are
respected when reading the input raster map layers. When the rescale
option is used, the output files are rescaled to fit the min,max range.
.PP
The order of the input bands does not matter for the output maps
(PC scores), but does matter for the vectors (loadings), since each
loading refers to a specific input band.
.PP
If the output is not rescaled (\fIrescale=0,0\fR, the output raster
maps will be of type DCELL, otherwise the output raster maps will be of
type CELL.
.PP
By default, the values of the input raster maps are centered for each
map separately with \fIx \- mean\fR. With \fI\-n\fR, the input raster
maps are normalized for each map separately with \fI(x \- mean) / stddev\fR.
Normalizing is highly recommended when the input raster maps have different
units, e.g. represent different environmental parameters.
.PP
The \fI\-f\fR flag, together with the \fIpercent\fR option, can
be used to remove noise from input bands. Input bands will be
recalculated from a subset of the principal components (inverse PCA).
The subset is selected by using only the most important (highest
eigenvalue) principal components which explain together \fIpercent\fR
percent variance observed in the input bands.
.SH NOTES
Richards (1986) gives a good example of the application of principal
components analysis (PCA) to a time series of LANDSAT images of a burned
region in Australia.
.PP
Eigenvalue and eigenvector information is stored in the output maps\(cq
history files. View with \fIr.info\fR.
.SH EXAMPLE
PCA calculation using Landsat7 imagery in the North Carolina sample dataset:
.br
.nf
\fC
g.region raster=lsat7_2002_10 \-p
i.pca in=lsat7_2002_10,lsat7_2002_20,lsat7_2002_30,lsat7_2002_40,lsat7_2002_50,lsat7_2002_70 \(rs
    out=lsat7_2002_pca
r.info \-h lsat7_2002_pca.1
   Eigen values, (vectors), and [percent importance]:
   PC1   4334.35 ( 0.2824, 0.3342, 0.5092,\-0.0087, 0.5264, 0.5217) [83.04%]
   PC2    588.31 ( 0.2541, 0.1885, 0.2923,\-0.7428,\-0.5110,\-0.0403) [11.27%]
   PC3    239.22 ( 0.3801, 0.3819, 0.2681, 0.6238,\-0.4000,\-0.2980) [ 4.58%]
   PC4     32.85 ( 0.1752,\-0.0191,\-0.4053, 0.1593,\-0.4435, 0.7632) [ 0.63%]
   PC5     20.73 (\-0.6170,\-0.2514, 0.6059, 0.1734,\-0.3235, 0.2330) [ 0.40%]
   PC6      4.08 (\-0.5475, 0.8021,\-0.2282,\-0.0607,\-0.0208, 0.0252) [ 0.08%]
d.mon wx0
d.rast lsat7_2002_pca.1
# ...
d.rast lsat7_2002_pca.6
\fR
.fi
In this example, the first two PCAs (PCA1 and PCA2) already explain 94.31% of
the variance in the six input channels.
.PP
.br
Resulting PCA maps calculated from the Landsat7 imagery (NC, USA)
.SH SEE ALSO
Richards, John A.,
\fBRemote Sensing Digital Image Analysis\fR,
Springer\-Verlag, 1986.
.PP
Vali, Ali R., Personal communication,
Space Research Center, University of Texas, Austin, 1990.
.PP
\fI
i.cca,
g.gui.iclass,
i.fft,
i.ifft,
m.eigensystem,
r.covar,
r.mapcalc
\fR
.PP
\fI
Principal Components Analysis article
(GRASS Wiki)
\fR
.SH AUTHORS
David Satnik, GIS Laboratory
.PP
Major modifications for GRASS 4.1 were made by
.br
Olga Waupotitsch and
Michael Shapiro,
U.S.Army Construction Engineering
Research Laboratory
.PP
Rewritten for GRASS 6.x and major modifications by
.br
Brad Douglas
.SH SOURCE CODE
.PP
Available at:
i.pca source code
(history)
.PP
Accessed: Sunday Dec 17 17:22:45 2023
.PP
Main index |
Imagery index |
Topics index |
Keywords index |
Graphical index |
Full index
.PP
© 2003\-2023
GRASS Development Team,
GRASS GIS 8.3.1 Reference Manual