In: Wagner W., Szekely, B. (eds.): ISPRS TC VII Symposium - 100 Years ISPRS, Vienna, Austria, July 5-7, 2010, IAPRS, Vol. XXXVIII, Part 7B 
229 
In this work some basic photogrammetric processing tools, such 
as the stereo matching, 3D coordinate determination, 
interpolation algorithms to generate a regular grid DSM, ortho- 
rectification and image mosaicking were used. Complementary, 
some tools of the BLUH software package, created at the 
Institute for Photogrammetry and Geoinformation, University 
of Hannover (Jacobsen, 2000), and other programmed by the 
authors were applied to optimise the procedure efficiency. DSM 
and orthoimage data integration and analysis were carried using 
a commercial GIS software. 
2. APPLICATION AREA AND DATA ACQUISITION 
The application area has an extension of 15 km to the south of 
the city of Porto, which includes in the north part the estuary of 
the river Douro (Figure 1). The area is subject to strong effects 
of the sea (Intensive action from the ocean), which produces 
Figure 1. Map of the area with the flight plan (created with 
Google Earth) 
In the scope of the monitoring program, it was decided to 
acquire aerial digital photography twice a year, before and after 
winter. In this way it is possible to assess the effect of winter 
storms over the beaches. It was also decided to acquire 
photographs in very low tide condition in order to map as much 
as possible the intertidal area. 
forward overlap and 30% side lap. Figure 1 shows the centre 
lines of the two strips. 
The pan-chromatic images, used in the process of digital 
surface model (DSM) generation, have a dimension of 7680 by 
13824 pixels, with a pixel size 12 pm. According to the camera 
specifications the focal length is 120.000 mm, the principal 
point is at the image centre and images are virtually distortion 
free. 
The base-height ratio in digital cameras is smaller than in 
conventional analogical cameras. In the present photos a 
parallax of 1 pixel (10 cm on the ground) corresponds to a 
height difference of 34 cm, on average. 
Images were acquired by a private company, using GPS/INS 
direct georeferencing equipment (IGI Aerocontrol) and standard 
processing techniques as for any other large scale topographic 
mapping work. According to the information provided 
(Municipia, 2009), a boresight calibration was performed after 
the camera was mounted and all the relative positions of GPS, 
IMU and camera were measured with total stations. GPS 
positioning of the flights was done relative to the Portuguese 
network of GPS permanent stations (IGP, 2009). The reference 
system used was ETRS89. The data processing software was 
IGI Aero-office. According to the processing reports provided, 
exterior orientation parameters are given in UTM coordinate 
system with appropriate corrections of scale factor and meridian 
convergence. 
During the monitoring periods four flights were done. Table 1 
contains information about date and start time of the image 
acquisition. A full flight is carried in less than 10 minutes. 
Flight no. 
Date 
Start time (UTC) 
1 
14-Nov-2008 
09:25 
2 
23-Apr-2009 
08:06 
3 
18-Nov-2009 
10:42 
4 
5-May-2010 
14:11 
Table 1. Date and start time of the four flights 
Some of the times were early in the morning, which would be 
inconvenient for normal photogrammetric purposes. In the 
present case, since the area of interest is essentially sand, that 
can be an advantage because some shadow effects occur in the 
sand that facilitate the stereomatching. 
Figure 2. Example of part of the study area, in the estuary of 
river Douro, near the city of Porto 
The data was acquired with a digital camera Zeiss-Intergraph 
DMC which allows to obtain images with a ground sampling 
distance (GSD) of 0.1 meters. The 15 km of the study area were 
covered by a total of 100 photos in two strips, with a 60% 
3. ANALYSIS OF THE EXTERIOR ORIENTATION 
One of the objectives of this work is to assess if the results 
provided through the adopted methodology were appropriate for 
the main objectives of large scale mapping of coastal areas 
without field work requirements. Several assessments were 
done, some without and some with check points. 
An algorithm of space intersection incorporated in the BLUH 
software package was used. It gives the point of shortest 
distance between the straight lines defined by the co-linearity 
equations, together with the corresponding y-parallax. 
3.1 Relative orientation 
The first verification done was to see if the exterior orientation 
parameters of consecutive photos provide an accurate relative 
orientation. That is essential if stereoscopic viewing is required 
as well as if calculation of 3D coordinates. 
For selected stereopairs a total of 15 homologous points per pair 
were manually measured and the corresponding y-parallaxes