ISPRS Commission III, Vol.34, Part 3A ,,Photogrammetric Computer Vision“, Graz, 2002 
PRECISE ORIENTATION OF SPOT PANCHROMATIC IMAGES WITH TIE POINTS TO 
A SAR IMAGE 
J. A. Goncalves?, I. Dowman" 
? Science Faculty - University of Porto, Observatório Astronómico — Alameda Monte da Virgem 
: 4430-146 V.N. Gaia, Portugal - jgoncalv(2)oa.fc.up.pt 
Dept. of Geomatic Engineering, University College London, Gower Street, London, WC1E 6BT UK - 
idowman@ge.ucl.ac.uk 
KEY WORDS: 
Image orientation, stereoscopic, multisensor, SAR, SPOT, parallax 
ABSTRACT: 
The extraction of spatial data from satellite imagery requires that precise sensor models are used to orientate images. This is 
particularly important with images acquired by linear sensors, such as SPOT, frequently pointed with large incidence angles. Image 
orientation with pixel accuracy requires that accurate ground control points are used. The acquisition of ground control is especially 
difficult in remote areas, where satellite images are important data sources for the production and updating of topographic maps. 
This paper describes a method for the orientation of SPOT panchromatic images making use of tie points with a SAR image. SAR 
and SPOT images compose stereo-pairs with a good stereo intersection, from which heights can be derived. For the SAR-SPOT tie- 
points, approximate heights are derived from parallaxes and can be corrected using altimetric ground control points. The method 
described makes use of the fact that the SAR image orientation can be derived from the orbit and SAR processing parameters, 
provided with SAR images, and does not require ground control points. 
The paper reports on a study carried out with SPOT and Radarsat images from Portugal. Accuracy assessments were done with 
digital cartography and field surveyed GPS data. It was possible to conclude that using a SAR image, accurate SPOT image 
orientation can be achieved requiring only very few altimetric control points. 
1. INTRODUCTION 
1.1 Satellite image orientation 
Satellite images are important data sources for the production 
and update of topographic maps at medium scales, such as 
1:50,000. This is especially important in remote and 
undeveloped regions. Optical images acquired by linear sensors, 
such as SPOT, provide significant amounts of topographic 
detail for that scale (Gugan and Dowman, 1988). Stereopairs, 
acquired in two different orbits, with variable across-track 
pointing angles, provide the height data required to plot 
planimetric detail in 3D and to generate digital elevation models 
(DEMs). Orthoimages can then be produced and act by 
themselves as map products or can be used to provide 
planimetric data. Sub-pixel accuracy can be achieved in the 2D 
and 3D data extracted from SPOT (Westin, 1990, Gugan and 
Dowman, 1988). 
Both the generation of orthoimages and the extraction of 3D 
data require that precise sensor models are used (Olander, 
1998). Equations must be set up, using sensor specific 
parameters, to relate ground and image coordinates. For a given 
sensor, equations can be written according to the physical 
process of image formation, expressing the position of a point 
on the image space (x,y) as a function of its ground coordinates 
XYZ). 
v= f(T. Zid 4) (1) 
yz UL ZA vA.) 
Usually the geocentric Cartesian terrestrial System (CTS) is 
used to exepress ground coordinates. A set of n parameters 
(4,,...,4,) that characterize a particular image (the exterior 
orientation parameters) are also involved in the sensor 
equations. The knowledge of precise exterior orientation 
parameters for a given image of a particular sensor is an 
important requirement in order to fully exploit the resolution of 
the sensor. Using an appropriate number of ground control 
points (GCPs) the exterior orientation parameters can be 
determined, after which the image is said to be oriented. Image 
orientation is required for the topographic applications of 
satellite images, such as ortho-rectification and height 
extraction. In order to maximize the potential of the image 
resolution, the accuracy of GCPs should be better than the 
image resolution. In the case of SPOT panchromatic imagery, 
with a pixel size of 10 m, ground control should be acquired by 
field survey or from topographic maps at least of scale 
1:25,000, to ensure that accuracy. 
Remote areas of the world, which are the ones where 
topographic mapping from satellite images can be more useful, 
are normally poorly mapped. Field surveys can be extremely 
expensive or not possible to carry out. In such situations GCPs 
will not be available with appropriate accuracy to extract spatial 
data from images to their highest potential. 
The requirement of GCPs for image orientation can be 
overridden if exterior orientation parameters are collected by 
onboard equipment (direct georeferencing). Positioning 
systems, such as GPS (Global Positioning System) and DORIS 
(Doppler Orbitography and Radiopositioning Integrated by 
Satellite) can be used to determine satellite trajectory with better 
accuracy than image resolution. That is the case of SPOTA, 
positioned by DORIS, with 1 m accuracy (CNES, 2000). The 
attitude angles of optical sensors, which are also exterior 
orientation parameters, are determined by navigation equipment 
but not as accurately as the orbit. Pointing errors on the ground 
are, in the case of SPOT, of the order of 500 m (CNES, 2000). 
A - 125