A low-cost image processing system on a 32-bit RISC 
microcomputer 
Mulder N.J., H.R. Kostwinder and LJ. Blesius 
ITC, Image Processing Lab 
P.O. Box 6, 7500 AA Enschede 
The Netherlands 
Abstract 
A recent proof-of-concept system addresses the problem of the use of one 
workstation for the digital treatment of small format photographs, analog maps and 
satellite data. It is called Alexander in recognition of A. von Humboldt, one of the 
fathers of modern physical geography. Improvements in technology, especially in 
the areas of reduced instruction set computers (RISC) and video frame grabbers, 
have led to a breakthrough in the price/performance ratio. Alexander runs on the 
Archimedes, a 32-bit RISC microcomputer. Depending on the configuration, the 
hardware of the described system costs in the order of US$ 2500 to US$ 5000 
including the 32-bit RISC processor. This sort of development means that ITC 
offers all students in remote sensing programmes access to a network of powerful 
workstations and a much higher likelihood of using digital equipment in their home 
countries than in the past when the costs were too high. 
KEYWORDS: Alexander, Archimedes, education, GIS, network, remote sensing, RISC, user 
interface 
1. Introduction 
Remote sensing techniques and geographic 
information systems are receiving an increa 
sing attention, not only in the earth science 
community, but also among other groups. 
This is reflected in recent efforts to promote 
these methods with special consideration of 
development in Third World countries, 
environmental monitoring, as well as urban 
and regional planning. 
Numerous satellites already provide data in a 
variety of spatial and spectral resolutions. 
Meteosat. NOAA, Landsat and SPOT images 
are used for climatologic or environmental 
purposes in an operational environment. They 
will be supplemented by new generations of 
satellites, such as ERS1, which will provide 
new sources of information. The digital format 
of the data has fostered the use of computers 
for processing the images. Digital image 
processing techniques have evolved gradually 
from statistical classification to structural pat 
tern recognition. 
Parallel to the developments in remote sen 
sing, geographic information systems have 
become a major tool. The integration of 
remote sensing and GIS will improve image 
processing efficiency and is being investi 
gated at various research levels. 
One of ITC’s biggest assets is the expertise of 
the staff in various application fields. The 
educational programmes provide not just 
technology transfer but also knowledge trans 
fer. Most ITC students come from developing 
Paper presented at the ISPRS Commission II Symposium in Dresden on 8-12 September 1990