atic 
of 
for 
‘he 
ipe 
ric 
ng 
ce 
ne 
ial 
of 
tal 
ed 
y 
J 2 
% 
of 
al 
ial 
Lis 
he 
he 
er 
ill 
re 
th 
re 
er 
ys 
S 
ip 
at 
'* Ce OO 
= (PY Un 2 
P» er 
3. Development of Remote Sensing 
Methods for pattern recognition and knowledge-based image 
| interpretation are under development for remote sensing. From 
these methods a major progress for the interpretation of digital 
image data is expected. The use of neuronal networks for 
supervised classification is tested with promising signs 
(DAGM symposium at Munich, 1991). 
Radar data from ERS-1 are an interesting innovation. There 
are several problems with geometry and other sensor 
characteristics. These problems ask for some new solutions for 
classification, pattern recognition and data extraction. 
4. GIS and Digital Mapping 
The integration of photogrammetry and remote sensing into 
Geographic Information Systems is in sight. Today GIS and 
digital mapping are an operational phase in many fields. Based 
on the enormous data requirements of GIS there is a strong lift 
for photogrammetry and remote sensing. This tendency will 
lead to integrated systems, bringing the tasks of 
photogrammetry, remote sensing, GIS and digital mapping 
together. Hybrid GIS for common processing of raster and 
vector data already gives the possibility to use orthophotos for 
landscape documentation in GIS. 
5. Education and Research 
Meanwhile there are not only expensive high end DPS, but 
also low end systems with sufficient performance. These 
systems are ideal for educational use. In the last years the 
tendency of photogrammetric development shows clearly that 
the education of photogrammeters in DPS has become very 
important. 
COMMISSION III 
(Priv.-Doz. Dr.-Ing. D. Fritsch) 
1. Introduction 
Further developments of mathematical methods for 
photogrammetric applications in the quadrennium 1988 - 1992 
are closely related with the progress in information 
technology. While in the previous quadrennium 1984 - 1987 
digital photogrammetry was initiated, we now observe a 
consolidation and deepening on a broad basis. In particular, the 
design of digital photogrammetric systems demands for 
comprehensive solutions, which can only be provided by 
intensive research. 
In the recent past only fragments of digital photogrammetry 
were available. These fragments were completed and coded; 
today they are already operational to be used in 
photogrammetric workstations. In order to name some 
important modules, not only the digital orthophoto and image 
matching approaches are advanced in design and operation but 
also the extraction of topological primitives in the form of 
nodes and edges to update databases of Geographic 
Information Systems (GIS). 
The problem of photogrammetric data acquisition has been 
reduced considerably using the Global Positioning System 
(GPS) for the determination of the exterior orientation 
parameters of an aerial camera. Thus, aerial triangulation can 
be carried out without or with only a few control points when 
integrating digital image matching techniques for point 
transfer. 
37 
Further use of photogrammetric data in GIS was also strongly 
influenced. The strategies which have been worked out will 
not only capture and structure vectorial data accordingto some 
object definitions and geometric-topological models but use 
the same data model for image data. However, some 
deficiencies still remain: the integration of photogrammetric 
data and methods has broadened the spectrum of hybrid GIS 
but we miss elements in image processing e.g. object 
identification, hybrid data structures, efficient data 
compression and editing. 
This very short review will indicate the change of 
photogrammetry today, which is related to the digital era. The 
same result came out from the scientific events of the working 
groups to be explained in more detail in the following. With 
regard to some references the developments in digital 
photogrammetry will only be sketched; the interested reader 
can find further hints in the references listed below. 
The research work of Commission III is initiated and 
coordinated by its working groups (WG). In the quadrennium 
1988 - 1992 the following WG's were set up: 
WG IVII: Design and Algorithmic Aspects of Digital 
Photogrammetric Systems (Chairmen: H. Ebner, 
Munich/I. Dowman, London) 
WG II/VI: Educational Aspects in Mathematical Analysis of 
Data (Chairmen: L. Mussio, Milan/T. Bouloucos, 
Enschede) 
WG III/1: Geographic Information Theory (Chairmen: M. 
Molenaar, Wageningen/R. Groot, Ottawa) 
WG III/2: Object Reconstruction and Location by Image 
Analysis (Chairmen: W. Foerstner, Bonn/R. Haralick, 
Seattle) 
WG III/3: Thematic Information Extraction from Digital 
Images (Chairmen: T. Schenk, Columbus/B.S. Schulz, 
Frankfurt) 
WGIII/4: Knowledge Based Systems (Chairmen: N.J. Mulder, 
Enschede/T. Sarjakoski, Helsinki) 
Initiated and organized by the german chairmen the following 
workshops and symposia should be mentioned: 
Workshop WG IV/III organized in cooperation with WG V/3 in 
London (February 1990) 
Workshop WG II/III in Boulder (March 1991) 
Workshop WG IVIII in Munich (September 1991, H. 
Ebner/D. Fritsch/C. Heipke, 1991) 
Workshop WG IIU/2 in Seattle (October 1990) 
Workshop WG III/2 in Bonn (March 1992, W. Foerstner/S. 
Ruwiedel, 1992) 
2. Development of Photogrammetry 
Within the developments of photogrammetry data acquisition 
has been supplemented by GPS observations for the 
parameters of the exterior camera orientation. In this context, 
e.g., the contribution given by P. Friess (1990) shows, that the 
accuracy of these parameters can be derived within the cm 
level. He points out an operational mode when the satellite 
constellation does allow kinematic positioning. 
Examples for close range photogrammetry are dealt with in the 
work of T. Voegtle (1989) and P. Krzystek (1990): While in 
the first case street traffic data are captured by electro optical 
sensors, the latter one demonstrates algorithms for an optical 
positioning system to be used in highly accurate engineering 
surveys. Herewith photogrammetry supplements classical 
surveying methods (Commission V ISPRS). The synergy of 
photogrammetry and surveying leads to hybrid theodolites and 
autonomous systems to be maneuvred and managed by remote 
control.