Geocoding Processor is intended for precision geometric correction of SAR images use by an accurate SAR
sensor model based on the satellite position, survey features, and image formation parameters.
Sophisticated algorithms are used for detailed positioning of each pixel on output image.
The standard processing procedures are supported: georeferening, geocoding, and orthorectification.
The inputs of Geocoding Processor are the spaceborne radar imagery presented in single look complex CEOS format
for the georeferencing procedure and georeferenced CEOS format for the geocoding/orthorectification once.
Processor imports automatically both the imagery and the CEOS headers information into the internal environment.
Orbit correction option for input imagery using three-dimensional GCP points is supported. Control point could be set
by user directly on the imagery, edited, saved into file, and opened lately. If DEM for processed scene is available
the terrain distortion could be removed.
The resulted image has the user definable pixel size in degrees or meters. An output geolocation accuracy is
limited only by the satellite and GCPs positions accuracy due of the mathematically precise sensor modeling is
Geo Processor is intended to transform source radar image (SAR image) from
coordinate system time – slant range into another one. In Geo Processor there
are three transformation methods:
Georeferencing — image transformation
to the time – ground range coordinate system;
Geocoding — image transformation
to the geodetic coordinate system without help of digital elevation model;
Orthorectification — image
transformation to the geodetic coordinate system using digital elevation
Radar images (SAR images) are usually presented in the following coordinate systems:
Antenna coordinate system (ACS)
is two-dimension matrix with azimuth (space platform flight path direction)
and slant range (the direction of viewing line direction from SAR to target)
Ground reference coordinate
system (GFS) is two-dimension matrix with azimuth (space platform flight path
direction) and ground range (the direction of viewing line direction from SAR
ground track point to target) axes.
To perform of target recognition it is recommended to process images acquired by
different sensors. For this purpose it is necessary to ensure space compatibility
of these images, i.e. the images from different sensors
must be presented in common basic coordinate system. It may be one of sensor's
coordinate system, geodetic or one of cartographic projection.
Space referencing is divided into two types:
Geometry calibration (for every
image pixel in antenna coordinate system "time — slant range" the ground coordinates
in some cartographic projection are given, calibration includes only coordinates
calculation without image transformation);
Geometry correction (geometry
correction means image transformation to basic coordinate system, includes geometry
calibration and geometry image transformation to new grid, corresponding to
parameters of given basic coordinate system, results of geometry calibration
and given scale).
Orbit parameters adjustment.
Orbit parameters adjustment is divided into two stages:
1. Along track adjustment;
2. Cross track adjustment.
Orbit adjustment along track is provided by iterations. Orbit adjustment cross
track is produced by differential method.
Radar image in the orbital coordinates system and coordinates of Ground Control
Points (GCP) are used to orbit correction. Besides time of survey, slant range
and platform coordinates are known for each image pixel.
Flight path information is corrected thereby that coordinates of identified
points at image should be situated close to GCP position greatly after geocoding.
Georeferencing — the output of this operation
is image representation in ground coordinate system "time — ground range".
The raster of this image is formed to provide given pixel sizes on Earth surface.
Geocoding — image pixel coordinates transformation
from antenna SAR system to one of cartographic projection. The geocoded image
is oriented in North/South direction. Pixel sizes are determined corresponding
to given projection (degrees or meters), scale is chosen by operator. Precision
geocoding — revision of image cartographic coordinates with using of
ground control points. The coordinates accuracy of ground control points must
be better than accuracy of ballist calculating coordinates.
Orthorectification — image pixel coordinates
transformation from antenna SAR system to one of cartographic projection with
using of Digital Elevation Model (DEM). Digital Elevation Model — it is two-dimensional
matrix where each element has three coordinates (latitude, longitude and height).
Amplitude of each element is proportional to the height of corresponding ground
point above Earth geoid.
At the picture the orthorectified (transformed
into the plane projection) radar image, combined with digital terrain
map is shown. Image was taken over Jukovsky site, Moscow region.