Provided by: pktools_2.6.7.6+ds-2build2_amd64 
NAME
pkfilter - program to filter raster images
SYNOPSIS
pkfilter -i input -o output [-f filter | -srf file [-srf file]... | -fwhm value [-fwhm
value]... ] [options] [advanced options]
DESCRIPTION
pkfilter implements spatial and spectral filtering for raster data. In the spatial domain
(X, Y), the filter typically involves a rectangular convolution kernel (moving window).
To avoid image shifting, the size of the window should be odd (3, 5, 7, ...). You can set
the window sizes in X and Y directions separately with the options -dx and -dy. A
circular kernel (disc) is applied if option -circ is set. An overview of the supported
filters (option -f|--filter) is given below. You can create customized filters by
defining your own filter taps (multiplicative elements of the filter kernel) via an ascii
file (option -tap). In the spectral/temporal domain (Z) you can filter multi-band raster
inputs. The kernel filter size can be set with the option -dz (use odd values only).
FILTERS IN SPATIAL (DX, DY) AND SPECTRAL/TEMPORAL (DZ) DOMAIN
The number of output bands equals number of input bands
Filter Description
dilate morphological dilation
erode morphological erosion
close morpholigical closing
(dilate+erode)
open morpholigical opening
(erode+dilate)
smoothnodata values smooth nodata values (set
--nodata option!)
Example: "Smooth" (interpolate) nodata in spectral/temporal domain (-dz 1), using a linear
interpolation
pkfilter -i input.tif -o smoothed.tif -dz 1 -f smoothnodata -interp linear
Example: Filter input.tif in spatial domain with morphological dilation filter with kernel
size 3x3.
pkfilter -i input.tif -o dilated.tif -dx 3 -dy 3 -f dilate
In case of moving window, the number of output bands equals number of input bands. In
case dz=1, the single output band is calculated as the result of the statistical function
applied to all bands.
Filter Description
nvalid report number of valid (not
nodata) values in window
median perform a median filter in
spatial (dx, dy) or
spectral/temporal (dz) domain
var calculate variance in window
min calculate minimum in window
max calculate maximum in window
sum calculate sum in window
mean calculate mean in window
stdev calculate standard deviation in
window
savgolay Savitzky-Golay filter (check
examples page!)
percentile calculate percentile value in
window
proportion calculate proportion in window
Example: Median filter in spatial domain
pkfilter -i input.tif -o median.tif -dx 3 -dy 3 -f median
Example: Calculate statistical variance in spectral/temporal domain (single output band)
pkfilter -i input.tif -o var.tif -dz 1 -f var
WAVELET FILTERS
The number of output bands equals number of input bands
Filter Description
dwt discrete wavelet transform
dwti discrete inverse wavelet
transform
dwt_cut discrete wavelet + inverse
transform, using threshold
option to cut percentile of
coefficients
Example: Calculate discrete wavelet in spatial domain
pkfilter -i lena.tif -o lena_dwt.tif -f dwt
Example: Calculate discrete wavelet in spectral/temporal domain
pkfilter -i timeseries.tif -o dwt.tif -f dwt -dz 1
The number of output bands equals number of input bands
Filter Description
dwt_cut_from discrete wavelet + inverse
transform, setting all high
frequence coefficients to zero
(scale >= threshold)
Example: Calculate low frequency time series based on discrete wavelet + inverse transform
in spectral/temporal domain, retaining only coefficients until scale 3.
pkfilter -i timeseries.tif -o lowfrequency.tif -f dwt_cut_from -dz 1 -t 4
FILTERS IN SPATIAL DOMAIN ONLY (DX, DY > 1 AND ODD).
The number of output bands equals number of input bands.
Filter Description
mrf Markov random field
ismin pixel is minimum?
ismax pixel is maximum?
shift perform a pixel shift in spatial
window
scramble scramble pixels in a spatial
window
mode (majority voting) perform a majority voring (set
--class option)
sobelx horizontal edge detection
sobely vertical edge detection
sobelxy diagonal edge detection (NE-SW)
sobelyx diagonal edge detection (NW-SE)
countid count digital numbers in window
order rank pixels in order
density calculated the density
homog central pixel must be identical
to all other pixels within
window
heterog central pixel must be different
than all other pixels within
window
sauvola Sauvola's thresholding method
Example: Sobel edge detection in horizontal direction
pkfilter -i lena.tif -o sobelx.tif -f solbelx -dx 5 -dy 5
OPTIONS
-i filename, --input filename
input image
-o filename, --output filename
output classification image
-f filter, --filter filter
filter function (nvalid, median, var, min, max, sum, mean, dilate, erode, close,
open, homog (central pixel must be identical to all other pixels within window),
heterog (central pixel must be different than all other pixels within window),
sobelx (horizontal edge detection), sobely (vertical edge detection), sobelxy
(diagonal edge detection NE-SW),sobelyx (diagonal edge detection NW-SE), density,
countid, mode (majority voting, only for classes), smoothnodata (smooth nodata
values only) values, ismin, ismax, order (rank pixels in order), stdev, mrf, dwt,
dwti, dwt_cut, dwt_cut_from, scramble, shift, savgolay, percentile, proportion)
-dx value, --dx value
filter kernel size in x, use odd values only
-dy value, --dy value
filter kernel size in y, use odd values only
-dz value, --dz value
filter kernel size in z (band or spectral dimension), must be odd (example: 3).
Set dz>0 if 1-D filter must be used in band domain
-nodata value, --nodata value
nodata value(s) (used for smoothnodata filter)
-v level, --verbose level
verbose mode if > 0
Advanced options
-circ, --circular
circular disc kernel for dilation and erosion
-r method, --resampling-method method
Resampling method for shifting operation (near: nearest neighbour, bilinear: bi-
linear interpolation).
-co option, --co option
Creation option for output file. Multiple options can be specified.
-wt type, --wavelet type
wavelet type: daubechies,daubechies_centered, haar, haar_centered, bspline,
bspline_centered
-wf family, --wf family
wavelet family (vanishing moment, see also
http://www.gnu.org/software/gsl/manual/html_node/DWT-Initialization.html)
-class value, --class value
class value(s) to use for density, erosion, dilation, openening and closing,
thresholding
-t threshold, --threshold threshold
threshold value(s) to use for threshold filter (one for each class), or threshold
to cut for dwt_cut (use 0 to keep all) or dwt_cut_from, or sigma for shift
-tap filename, --tap filename
text file containing taps used for spatial filtering (from ul to lr). Use dimX and
dimY to specify tap dimensions in x and y. Leave empty for not using taps
-tapz value, --tapz value
taps used for spectral filtering
-pad method, --pad method
Padding method for filtering (how to handle edge effects). Choose between:
symmetric, replicate, circular, constant (pad with 0).
-fwhm value, --fwhm value
list of full width half to apply spectral filtering (-fwhm band1 -fwhm band2 ...)
-srf filename, --srf filename
list of ASCII files containing spectral response functions (two columns: wavelength
response)
-win col, --wavelengthIn col
column number of input ASCII file containing wavelengths
-wout value, --wavelengthOut value
list of wavelengths in output spectrum (-wout band1 -wout band2 ...)
-down value, --down value
down sampling factor. Use value 1 for no downsampling). Use value n>1 for
downsampling (aggregation)
-beta filename, --beta filename
ASCII file with beta for each class transition in Markov Random Field
-eps value, --eps value
error marging for linear feature
-l1, --l1
obtain longest object length for linear feature
-a1, --a1
obtain angle found for longest object length for linear feature
-a2, --a2
obtain angle found for shortest object length for linear feature
-interp type, --interp type
type of interpolation for spectral filtering (see
http://www.gnu.org/software/gsl/manual/html_node/Interpolation-Types.html)
-ot type, --otype type
Data type for output image ({Byte / Int16 / UInt16 / UInt32 / Int32 / Float32 /
Float64 / CInt16 / CInt32 / CFloat32 / CFloat64}). Empty string: inherit type from
input image
-of GDALformat, --oformat GDALformat
Output image format (see also gdal_translate(1)).
-ct filename, --ct filename
color table (file with 5 columns: id R G B ALFA (0: transparent, 255: solid)). Use
none to omit color table
EXAMPLE
Filtering in spatial domain
Filter input.tif with morphological dilation filter. Use a circular kernel (instead of
rectangular) of size 3x3.
pkfilter -i input.tif -o filter.tif -dx 3 -dy 3 -f dilate -circ
Similar to previous example, but consider only values of 255 for filtering operation.
Typical usage: dilate cloud values in input image that are flagged as 255
pkfilter -i input.tif -o filter.tif -dx 3 -dy 3 -class 255 -f dilate -circ
Filtering in spectral/temporal domain
Calculate the median value for each pixel, calculated on a moving window of width 3 (-dz
3) over all input bands. The output raster dataset will contain as many bands as the
input raster dataset.
pkfilter -i input.tif -o filter_stdev.tif -dz 3 -f median
No moving window (-dz 1). Calculate the standard deviation for each pixel, calculated on
all input bands. The output raster dataset will contain a single band only.
pkfilter -i input.tif -o filter_stdev.tif -dz 1 -f stdev
"Smooth" (interpolate) nodata in spectral/temporal domain (-dz 1), using a linear
interpolation. The following interpolation types are supported: akima (default), linear,
polynomial, cspline, cspline_periodic, akima_periodic (please check gsl ⟨http://
www.gnu.org/software/gsl/manual/html_node/Interpolation-Types.html⟩ page for more
information on the interpolation types).
pkfilter -i input.tif -o input_smoothed.tif -dz 1 -f smoothnodata -interp linear
10 February 2020 pkfilter(1)