5A-5-1 
SURVEYING AND MAPPING OF URBAN STREETS BY PHOTOGRAMMETRIC TRAVERSE 
A. R. SILVA 1 , J. C. BATISTA 1 , R. A. OLIVEIRA 1 , P. O. CAMARGO 2 and J. F. C. SILVA 2 
'Cartographic Engineering Undergraduate Program 
2 Department of Cartography and Center of Mapping 
Universidade Estadual Paulista - UNESP 
19060-900 Presidente Prudente SP 
Brazil 
http://www.prudente.unesp.br/dcartog/dcartog.htm 
jfcsilva@prudente.unesp.br 
KEY WORDS: surveying, mapping, terrestrial photogrammetry, street scenes, capstone project. 
ABSTRACT 
A 1:2000 street map was made based on a topographic surveying by photogrammetric traverse technique. Photogrammetric 
traverse concatenates terrestrial stereo-pairs. The technique is suitable to map roads and streets surveyed by mobile mapping 
system terrestrial platform. Two urban blocks were surveyed by photogrammetric traverse and a street map was made using 
a collection of different equipment, hardware and software. Street images were acquired by a simple digital camera (Kodak 
DC40) mounted on a tripod which was moved forward to simulate the vehicle trip along the streets. The same thing was 
done to the antenna of the rover GPS receiver while the fixed receiver stayed on a reference station. The camera perspective 
centers were positioned by GPS surveying and the orientation angles were computed by bundle block adjustment. GPS data 
were processed by Ashtech Reliance vl.40. Digital images were processed by Kodak PhotoEnhancer vl.7 and by Zsoft 
PhotoFinish v3.0 to extract image coordinates (then transformed to photocoordinates). A bundle block adjustment (tftc), 
adapted to the particular geometry, computed the angular orientation and the coordinates of pass points and interest points. 
Autodesk AutoCAD Map vl.O was used for map edition and Golden Software Surfer v6.0 interpolated the contour lines and 
DTM. 10% significance level statistic tests show that the trend analysis and map accuracy succeeded. Accuracy resulted 
around one meter approximately. This means the map may be classified as a class B map according to Brazilian 
cartographic criteria. In this particular case, the standard error is 0.5mm in map scale. 
1 INTRODUCTION 
The scientific and technical program of this workshop 
demonstrates how developed are the mobile systems to 
acquire spatial data for mapping and GIS (Geographic 
Information Systems) purposes. However there are 
situations that modem equipment lets a practical simulation 
arises with two goals in perspective. One is to give students 
the opportunity to merge distinct techniques in order to 
make a topographic map and the other to observe and 
analyze the results of applying distinct techniques under a 
different and modem approach. 
Particularly, we have decided to proceed with simulated 
situations of collecting data on streets to make maps using 
an alternative methodology called photogrammetric 
traverse. The idea behind this project was to provide an 
understanding of the map making process to a group of 
undergraduate students (capstone project). Despite of being 
an unconventional method for topographic mapping, the 
project also brought contributions to learn the relationship 
between the stages that connect the main phases of the data 
and information when it comes to the application of 
photogrammetric traverse. 
Although simulating a mobile mapping platform acquiring 
data, the project introduced the students to practical 
problems. GPS (Global Positioning System) data and 
digital images were grabbed on urban streets then 
processed and measured to extract topographic information 
about distinct objects which were represented in a digital 
map. Besides that, the results will help the analysis of real 
application data to be produced in the months to come by a 
platform that is being constructed. 
2 STREET SURVEYING AND 
PHOTOGRAMMETRIC TRAVERSE 
Reasons for surveying the streets 
Most people live in urban areas. Along the streets facility 
companies lay their networks say water and sewer, electric 
power, telecommunication, and many others. Poles and 
trees are quite common in typical urban scenes. They do 
help but at some extent they disturb people when walking 
on the sidewalks. Pedestrians and drivers (these 
representing a large collection of different vehicles) need a 
language to communicate to each other in favor of traffic 
security, which is expressed in signs and warnings, both 
horizontal and vertical. Not only architects and 
urbanologists argue for a comfortable urban environment 
where humans can live with dignity and happiness. We 
could continue and make a long list of nice words to justify 
the need for the street and road mapping. Briefly, from a 
technical point of view, an image database and digital maps 
will help the urban administrators to reach the standards of 
a better quality of life. Particularly, mobile mapping 
systems (MMS) seem to play an important role in 
collecting street and road data for mapping and GIS 
purposes.