on, 
me 
put 
lew 
ion 
atic 
ally 
vith 
nts 
lain 
the 
atic 
ling 
Ilite 
ual 
sis, 
ose 
s of 
> is 
iine 
of 
lite 
of 
are 
ata 
ght 
be 
  
combined with minimum of topographical or thematic 
reference elements to make it readable and be able to 
analyze the contents of the "space-map". 
Considering the two families of maps, where satellite 
imagery plays a role, i.e. general (or topographic) maps and 
thematic maps, different problems and specific methods are 
detected for combining the available information. In order to 
consider these methods, depending on the problems that 
arise, the following aspects will be discussed: 
Geometric adjustment of the satellite information and 
conventional information, Assembly and cutting of the 
image along the map edges, Radiometric processing of the 
image to form the map background, Adding the 
conventional information to the satellite data 
2.1 Image map production 
The production of image map has three major parts (table 
24): 
Table 2.1 Production line steps in Image mapping 
  
  
  
  
Input Processing Output 
Digital Imagery Geometric General image 
correction maps 
Digital Elevation Radiometric Thematic 
Model (DEM) correction (classified) image 
maps 
Image orientation Mosaicking Thematic image 
parameters maps 
  
  
  
  
  
2.2 Geometric corrections 
Geometric corrections concerns operations such as 
Geometrical preprocessing, Geometrical processing, 
Geodetic referencing (considering GPCs, and GPS 
measurements), Establishing transformation model 
(Polynomials, others), Resampling 
2.3 Radiometric correction 
Radiometric correction concerns Radiometric 
preprocessing, Defective pixels and lines replacement, 
Destriping, Clouds substitution, Haze correction, Noise 
smoothing, Slope correction, 
Radiometric Enhancement ( Contrast enhancement, Colour 
enhancement, Edge enhancement ) 
2.4 Quality control of Image maps 
With respect to the process involved in satellite image 
mapping, it will be easily understood that three elements are 
effecting the overall quality of final result: firstly the quality 
of input data such as image (Discussed on chapters 2 and 
7), Digital Elevation Model and ground control 
points(discussed in chapter 6), secondly the method, 
equipment and finally operator skill and care applied during 
the processes such as rectification, mosaicking, screening, 
775 
reproduction, etc.. Apparently, reaching high quality satellite 
image maps is impossible unless all elements 
3. DIGITAL ELEVATION MEDEL, GENERATION 
AND ASSESSMENT 
A Digital Elevation Model (DEM) is an ordered array of 
numbers that represents the spatial distribution of terrain 
characteristics. Different methods of data acquisition for 
DTM is rewied, namely; sampling (selective, progressive, 
composite, automatic) DTM from aerial photography, more 
attention is paid to automatic DTM generation using image 
correlation or image maching techniques. 
3.1 DEM data structure 
DEM is used to present the terrain in the form of surfaces 
that can, mathematically or numerically, be defined. They 
are broadly classified into models which are based on 
structuring the points into some specific order, taking into 
account their spatial relationships, and models which are 
based on fitting mathematical functions into the elevation 
data [Ackermann, 1978]. 
DEM is classified into the grid structure (regular, 
semiregular, irregular), TIN data structure, surface patch 
quadtree, mathematical representation (polynomials, 
others). 
3.2 DEM generation 
A DEM system consist of 3 basic components: Data 
acquisition, processing (pre, main, post processing), 
storage and retrival. the method of data acquisition is 
adapted to specification of output, constraints and terrain 
characteristics. 
3.2.1 Data acquisition. 
Data acquisition can be performed directly or indirectly. 
Indirect (photogrammetric) data acquisition concerns 
sampling (selective, prograssive, composite) [Makarovic, 
1976], image matching (correlation) technics. 
3.3 Image Matching Concepts 
One of the most significant applications of image matching 
is modelling of the terrain relief (automated DEM 
generation). Automatic measurement of the parallaxes in 
stereo images (image disparities) by software, is in fact, the 
essential base of this method. Because the amount of 
image information involved is huge, the process has to be 
time efficient. Moreover, to attain sufficient quality, the 
accuracy and reliability of image matching are of dominant 
significance. 
The principle behind image matching and height 
determination relies on a matching points being found in two 
images. A mathematical relationship exists between the 
parallax, distance between matching points in the images 
due to the different view positions, and the height of the 
actual terrain at the matched position. 
International Archives of Photogrammetry and Remote Sensing. Vol. XXXI, Part B4. Vienna 1996