You are using an outdated browser that does not fully support the intranda viewer.
As a result, some pages may not be displayed correctly.

We recommend you use one of the following browsers:

Full text

Remote sensing for resources development and environmental management
Damen, M. C. J.

Symposium on Remote Sensing for Resources Development and Environmental Management / Enschede / August 1986
3 automatic
Land cover. Proc.
sensing applied
SC. 1984; ESA
L - 140
j 4: 631 - 637
cultural crops
M. Veugen 1984.
SEE Trans.
)f tidal patterns
igery of the sea.
:orm Noordwijk
ig. OE-7: 15-20
!. Attema 1982.
:. Geosci. and
etability -
dimensions and
Aerospace and
•icultural area
it observations
:s; b. with 30
Developing tools for digital radar image data evaluation
G.Domik & F.Leberl
VEX CEL Corporation, Boulder, Colo., USA
Research Center Graz, Austria
Abstract: The refinement of radar image analysis methods has led to a need for a systems
approach to radar image processing software. Developments stimulated through satellite radar
are combined with; standard image processing techniques to create a user environment to
manipulate and analyze airborne and satellite radar images. One aim is to create radar products
for the user from the original data to enhance the ease of understanding the contents. The
results are called secondary image products and derive from the original digital images. Another
aim is to support interactive SAR image analysis. Software methods permit use of a digital
height model to create ortho images, synthetic images, stereo-ortho images, radar maps or color
combinations of different component products. Efforts are ongoing to integrate individual tools
into a combined hardware/software environment for interactive radar image analysis.
Post processing of digital synthetic aperture radar (SAR) images is
usually meaningful. One reason is the unreliable geometry due to the
kinematic nature of the imaging process and the effect of topography.
Thus the radar image has to be rectified. Another reason is the
complex interaction of terrain and microwaves as expressed in the gray
tones of the individual SAR image pixels. What data and tools would
be necessary to exploit the full content of digital radar images?
One may think of adding information not contained in the image itself,
e.g. map data, terrain slope or heights, to enhance the interpretability
of the image for different applications. Height data might be used to
give a general idea about the topography in form of a perspective view,
an illuminated height model, or even better through a radar image
simulation. Such presentations of a digital elevation model (DEM)
can be combined with the radar image grey values or stand alone.
After rectification other methods can be used to merge information
with the image.
Essential tools found necessary for processing of single and
overlapping SAR images are set up of single images or stereo models,
integration of topographic and thematic maps, simulation and
rectification of radar images, general purpose image processing as well
as archiving and retrieval routines, such as specific tape read
programs and image organization. The following will discuss some
essential SAR post-processing tools and their application.
Software can be divided into radar-specific and non-radar software.
There is of course
(a) General image processing.
(j) Radar image rectification;
(k) Slope-effect reduction;
(l) Mosaicking;
(m) Height extraction from stereo images;
(n) Feature extraction from single and stereo images.
The future should be open to the development of specific multi-sensor
and time-series/monitoring concepts.
2.1 General Image Processing
Obviously a full range of general purpose functions needs to include
display and save functions; grey value enhancement; color coding in
red, green and blue (RGB) or intensity, hue and saturation (IHS);
rotation and mirroring; pixel edit functions; general purpose filters
and edge operators; statistical processing; a graphics subset.
2.2 Archiving/Retrieval
One needs tape read, write and test programs; must be able to
produce hardcopies on film or on a printer (symbol overstriking,
matrix printing); to manage logfiles; to generally organize, name,
store, recall, window images.
2.3 DEM Manipulations
It is useful to digitize points and linear features; have vector to raster
conversion; be able to create axonometric and contour line plots,
coordinate transformations; statistics; to illuminate the DEM; have
functions to edit, rotate and mirror DEMs.
The non-radar software includes mainly the handling of input and
output data and is not specific to SAR-application. It may, however,
not be available in general purpose image processing systems such as:
(b) Archiving/Retrieval;
(c) DEM manipulations.
The radar software should include several specific capabilities such as
(d) Ground control management;
(e) Single image setup;
(f) Stereo-viewing of SAR soft copies;
(g) Stereo model setup;
(h) Radar image simulation;
(i) Stereo ortho image production;
2.4 Ground Control Management
This is not specifically radar relevant. It permits one to enter map
points into the computer, either from a keyboard or by manual
digitizing. In addition one needs to name, edit, organize, store and
retrieve the ground control data such as points (GCP).
2.5 Single Image Setup
This represents functions to input available parameters of the sensor
platform, and of the sensor itself. Examples include: flight
recordings and processing parameters, GCPs have to be accepted and
used for further calculations, e.g. a resection in space. One can
relate a map to an image supporting this by automated map-to-image
correspondence techniques.
2.6 Stereo-Viewing of SAR Soft Copy Images
A digital stereo viewer commonly consists of a display or pair of
displays presenting 2 image segments, and a pair of movable cursors.
This needs to be complemented by software to maintain a parallax-free
stereo impression.