Image (Re-)projections and Merging - Open Source Geospatial Tools: Applications in Earth Observation - page 102

different resampling methods. As in the previous experiment, the figures on the right
correspond to the pixel values of the center row of the synthetic images.
Similar conclusions as for the upsampling can be drawn. Some of the contrast
has been lost for the cubicspline and average resampling methods. Both the methods
cubic and lanczos result in overshoots. Due to the kernel-based method, you can
expect some artifacts at the image borders. For the example shown here, this is
particularly striking in Fig. 8.3 g for the average resampling method (the lower and
right edges of the image). Both the methods mode and near do not introduce new
pixel values and, based on the synthetic image, seem to well preserve the original
image. However, this can be misleading as shown for the lena image. You must be
aware that for these methods, there is an issue with the pixel locations.

8.3 gdalwarp

gdalwarp is a multi-purpose command line utility. It combines several operations:
reproject, warp and mosaic raster data. It can transform (reproject) images from a
source to a target coordinate system. As with the other GDAL command line utilities,
multiple formats for the source and target coordinate systems are supported as listed
in Sect. 3.1 . Raw images (without a defined coordinate system) can be warped using
a set of GCPs. This set should reside within the image, which can be achieved
with gdal_edit.py , gdal_translate or by editing a VRT (virtual raster)
(see Chap. 11 ) . Writing to an existing image will not change its bounding box unless
the -overwrite option is selected.

Usage: gdalwarp [--help-general] [--formats] [-s_srs srs_def]
ₒ

[-t_srs srs_def] [-to"NAME=VALUE"] [-order n | -tps |

ₒ

-rpc | -geoloc] [-et err_threshold] [-refine_gcps

ₒ

tolerance [minimum_gcps]] [-te xmin ymin xmax ymax] [-tr

ₒ

xres yres] [-tap] [-ts width height] [-wo "NAME=VALUE"]

ₒ

[-ot Byte/Int16/...] [-wt Byte/Int16] [-srcnodata "value

ₒ [value...]"] [-dstnodata "value [value...]"] -dstalpha

ₒ [-r resampling_method] [-wm memory_in_mb] [-multi] [-q]

ₒ [-cutline datasource] [-cl layer] [-cwhere expression]

ₒ [-csql statement] [-cblend dist_in_pixels]

ₒ [-crop_to_cutline] [-of format] [-co "NAME=VALUE"]*

ₒ [-overwrite] srcfile* dstfile

-s_srs srs def
source spatial reference set. The coordinate systems that can be passed are any-
thing supported by the OGRSpatialReference.SetFromUserInput() call, which
includes EPSG PCS and GCSes (ie. EPSG:4296), PROJ.4 declarations (as above),
or the name of a .prf file containing well known text.
-t_srs srs_def
target spatial reference set. The coordinate systems that can be passed are anything