AUTOMATED VISUAL TRAFFIC MONITORING AND SURVEILLANCE THROUGH A 
NETWORK OF DISTRIBUTED UNITS 
A. Koutsia 3 , T. Semertzidis 3 , K. Dimitropoulos 3 , N. Grammalidis 3 and K. Georgouleas b 
d Informatics and Telematics Institute, Centre for Research and Technology Hellas, 1st km Thermi-Panorama road, 
57001 Thessaloniki, Greece - (koutsia, theosem, dimitrop, ngramm)@iti.gr 
b MARAC Electronics, 165 Marias Kiouri & Tripoleos Str, 188 63, Piraeus, Greece - georgouleas@marac.gr 
Commission III, WG III/5 
KEY WORDS: Computer Vision, Visual Analysis, Fusion, Location Based Services, Calibration, Change Detection, Matching 
ABSTRACT: 
This work aims to present an intelligent system for tracking moving targets (such as vehicles, persons etc) based on a network of 
distributed autonomous units that capture and process images from one or more pre-calibrated visual sensors. The proposed system, 
which has been developed within the framework of TRAVIS (TRAffic VISual monitoring) project, is flexible, scalable and can be 
applied in a broad field of applications. Two different pilot installations have been installed for initial evaluation and testing, one for 
traffic control of aircraft parking areas and one for tunnels at highways. Various computer vision techniques which were 
implemented and tested during the development of the project, are described and analysed. Multiple background extraction and data 
fusion algorithms are comparatively evaluated. 
1. INTRODUCTION 
1.1 Relative Work 
Traffic control and monitoring using video sensors has drawn 
increasing attention recently due to the significant advances in 
the field of computer vision. Many commercial and research 
systems use video processing, aiming to solve specific problems 
in road traffic monitoring (Kastrinaki, 2003). An efficient 
application for monitoring and surveillance from multiple 
cameras is the Reading People Tracker (Le Bouffant, 2002), 
which was later used as a base for the development of a system 
called AVITRACK, which monitors airplane servicing 
operations (Thirde, 2006). Furthermore, in the FP5 
INTERVUSE project, an artificial vision network-based system 
was developed to monitor the ground traffic at airports 
(Pavlidou, 2005). The system uses the Autoscope® Solo Wide 
Area Video Vehicle Detection System which has been 
successfully deployed worldwide for monitoring and 
controlling road traffic (Michalopoulos, 1991). 
1.2 Motivation and Aims 
Robust and accurate detection and tracking of moving objects 
has always been a complex problem. Especially in the case of 
outdoor video surveillance systems, the visual tracking problem 
is particularly challenging due to illumination or background 
changes, occlusions problems etc. The aim of the TRAVIS 
project was to determine whether the recent changes in the field 
of Computer Vision can help overcome these problems and 
develop a robust traffic surveillance application. The final 
system is easily adjustable and parameterised, in order to be 
suitable for diverse applications related to target tracking. Two 
prototypes have been installed each for a different application: • 
• Traffic control of aircraft parking areas (APRON). This 
application focuses more on the graphical display of the ground 
situation at the APRON. The system calculates the position, 
velocity and direction of the targets and it classifies them 
according to their type (car, man, long vehicle etc). Alerts are 
displayed for dangerous situations, such as speeding. This 
information can be accessible by the respective employees, 
even if they are situated in a distant area, with no direct eye- 
contact to the APRON. A pilot installation of this system took 
place at “Macedonia” airport of Thessaloniki, Greece. 
• Traffic control of tunnels at highways. The focus of this 
application is on the collection of traffic statistics, such as speed 
and traffic loads per lane. It can also identify dangerous 
situations, such as objects falling, animals or traffic jams. These 
results can be sent to traffic surveillance centres or used to 
activate road signs/waming lights. This prototype was installed 
at a highway tunnel at Piraeus Harbour, Athens, Greece. 
2. SYSTEM ARCHITECTURE 
The proposed system consists of a scalable network of 
autonomous tracking units (ATUs) that use cameras to capture 
images, detect moving objects and provide results to a central 
sensor data fusion server (SDF). The SDF server is responsible 
for tracking and visualizing moving objects in the scene as well 
as collecting statistics and providing alerts for dangerous 
situations. The system provides a choice between two modes, 
each supporting a different data fusion technique. Grid mode 
separates the ground plane into cells and fuses neighbouring 
observations while map fusion mode warps greyscale images of 
foreground objects in order to fuse them. 
The topology of the ATUs network varies in each application 
depending on the existing infrastructure, géomorphologie facts 
and bandwidth and cost limitations. The network architecture is 
based on a wired or wireless TCP/IP connection as illustrated in 
Figure 1. These topologies can be combined to produce a 
hybrid network of ATUs. Depending on the available network 
bandwidth, images captured from specific video sensors may