Fill raster regions by interpolation from edges.


gdal_fillnodata [--help] [--help-general] [-q] [-md <max_distance>]
           [-si <smoothing_iterations>] [-o <name>=<value> [<name>=<value> ...]]
           [-mask <filename>] [-interp {inv_dist,nearest}] [-b <band>]
           [-of <gdal_format>] [-co <name>=<value>]
           <src_file> <dst_file>


gdal_fillnodata fills selection regions (usually nodata areas) by interpolating from valid pixels around the edges of the area.

Additional details on the algorithm are available in the GDALFillNodata() docs.


gdal_fillnodata is a Python utility, and is only available if GDAL Python bindings are available.


Show this help message and exit


Gives a brief usage message for the generic GDAL commandline options and exit.


The script runs in quiet mode. The progress monitor is suppressed and routine messages are not displayed.

-md <max_distance>

The maximum distance (in pixels) that the algorithm will search out for values to interpolate. The default is 100 pixels.

-si <smooth_iterations>

The number of 3x3 average filter smoothing iterations to run after the interpolation to dampen artifacts. The default is zero smoothing iterations.

-o <name>=<value>

Specify a special argument to the algorithm. Currently none are supported.

-b <band>

The band to operate on, by default the first band is operated on.

-mask <filename>

Use the first band of the specified file as a validity mask (zero is invalid, non-zero is valid).

-of <format>

Select the output format. The default is GTiff -- GeoTIFF File Format. Use the short format name.

-interp {inv_dist,nearest}

Added in version 3.9.

By default, pixels are interpolated using an inverse distance weighting (inv_dist). It is also possible to choose a nearest neighbour (nearest) strategy.


The source raster file used to identify target pixels. Only one band is used.


The new file to create with the interpolated result.