24 
INTEGRATION OF PHOTOGRAMMETRIC AND GEOGRAPHIC DATABASES 
Peggy Agouris 
Anthony Stefanidis 
Department of Spatial Information Science and Engineering 
and the National Center for Geographic Information Science and Analysis 
University of Maine 
5711 Boardman Hall, Rm. 348 
Orono, ME 04469-5711 
Tel: (207) 581 2180, Fax: (207) 581 2206, e-mail: {peggy, tony} @spatial.maine.edu 
Commission IV, Working Group 1 
KEY WORDS: Integration, GIS, Digital Images, Image Analysis Methods. 
ABSTRACT 
This paper addresses the role of digital photogrammetry within the current trend towards integrated databases, comprising 
photos and maps in digital format, relevant additional information in raster or vector format (e.g. extracted features and 
DEM), as well as a set of interrelating operations (e.g. feature extraction algorithms, georeferencing). The relationship 
between photogrammetry and integrated geographic systems is bi-directional. From digital images, through georeferencing, 
object extraction and DEM generation, information for geodatabase layers is obtained. At the same time, information from 
geographic databases is used to guide photogrammetric operations, especially for updating purposes (e.g. using previously 
available DEMS for deriving approximate conjugate locations, or using vector feature information for updating). In this paper 
we present the state-of-the-art in relevant digital photogrammetric research issues, and focus in particular on automatic 
orientations, DTM and orthoimage generation, and man-made object extraction. We evaluate how the current form of these 
algorithms and corresponding research activities meet the needs of such integrated environments, in terms of accuracy, 
efficiency, and productivity, and we identify research trends and needs resulting from this integration. 
1. INTRODUCTION 
The qualitative and quantitative geoinformation contained in 
spatial databases like maps, topographic databases, and GIS 
in general, is of vital importance to a large, multifaceted 
array of applications. During the last few years, mainly due 
to the increased awareness of the economical consequences 
and importance of organized planning and spatial 
information management, the number of users, and 
consequently the demand for such data, is steadily 
increasing. The spatial information user community has 
expanded well beyond its traditional limits, even making the 
transition towards non-professional users, with novel forms 
of spatial information (e.g. mapping data on CD-ROM) 
being nowadays available to the general public. 
In addition to the increased demand for spatial 
geoinformation, the accelerated rate of change in modern 
environments makes the frequent updating of spatial 
databases rather imperative to ensure their continuous 
validity. Photogrammetry surpasses existing alternative 
methods for geodata collection in terms of accuracies over 
involved costs, offering unmatched potential for fast 
collection of large amounts of accurate spatial information. 
The great advancements of the last decade in the field of 
digital photogrammetry have further solidified the role of 
24 
photogrammetry as data collection methodology, allowing 
the bypassing of the need for expensive, dedicated 
instruments and trained personnel. We are currently at the 
stage where automation through the use of digital imagery 
and suitable image analysis techniques is making the 
transition from research and development to production, 
with softcopy workstations appearing set to substitute 
analytical instruments as the choice of the practitioners. 
However, while research and development activities in 
digital photogrammetry so far mainly concentrated on 
improving the performance of photogrammetric operations 
through automation (with undoubtedly remarkable results), 
another great advantage of digital photogrammetry remains 
rather unexplored: the compatibility of digital imagery with 
other geographic databases in terms of storage formats, 
analysis operations, and media, which permits the full 
integration of digital imagery in spatial geoinformation 
systems. 
The realization of the potential for the integration of digital 
imagery in GIS can be traced back to the beginning of the 
decade [Ehlers et al., 1989; Ehlers et al., 1991; Dobson, 
1993; Gahegan, 1994; Ehlers et al, 1994]. The initial 
concepts though were focusing on remote sensing rather 
than photogrammetry, as reliable digital imagery was 
associated solely with remote sensing applications. 
However, advancements in digital photogrammetry allow us 
to reevaluate this concept. In this paper we will discuss the 
International Archives of Photogrammetry and Remote Sensing. Vol. XXXI, Part B4. Vienna 1996 
major 
aerial : 
of-the- 
Geogr 
storin 
qualite 
data. 
examp 
and th 
1991]. 
positic 
Regar 
impor 
inforn 
curren 
integr: 
The si 
where 
realm 
to the 
emplo 
throug 
is sub: 
not f 
excepi 
georef 
inforn 
directi 
7 
Fi 
Systei 
With 
prohil 
analo, 
openi 
that tl 
only 
withit 
of im 
these 
releve 
integr 
inforr