C C T V B A S E D SENSING T E C H N I Q U E S F O R 
A D A P T I V E C O N T R O L OF 
TRAFFIC SIGNALS 
IN M U L T I P R O C E S S O R A R C H I T E C T U R E S 
S U B M I T T E D FOR T H E D E G R E E OF M A S T E R OF P H I L O S O P H Y 
IN E L E C T R O N I C A N D T E L E C O M M U N I C A T I O N E N G I N E E R I N G 
0 2 . 
6 
UNIVERSITY OF MORATUWA UM Thesis coll. 
SRI LANKA 
APRIL 2002
 7 ( 0 0 / 0 , 
University of Moratuwa 
79891 
D e c l a r a t i o n 
T h e w o r k p r e s e n t e d in t h i s t h e s i s h a s n o t b e e n s u b m i t t e d for t h e f u l f i l l m e n t o f a n y o t h e r 
d e g r e e . 
G . G . D . N i s h a n t h a Prof . J . A . K . S . J a y a s i n g h e 
( C a n d i d a t e ) ( S u p e r v i s o r ) 
Abstract 
As the problem of urban traffic congestion spreads, there is a pressing need for 
the introduction of advanced technology and equipment to improve the state of 
the art of traffic control. In this context, the techniques used for sensing traffic 
flow information plays a vital role, ensuring accurate real t ime controllabil i ty. 
Closed Circuit Television (CCTV) has gained popularity because of its ability to 
provide diverse information on relatively large regions leading to opportuni t ies 
for performing substantially more complex tasks than conventional detect ion 
schemes . By processing these video images, traffic parameters such as speed, 
traffic composi t ion, queue length etc. can be extracted. In addition C C T V 
images can be further processed for other useful information such as detection of 
vehicle shapes, vehicle types, occurrence of traffic violations and vehicle 
identification numbers etc. 
To introduce C C T V based vehicle detection for automation of road traffic 
control in Sri Lanka involves an unaffordable investment cost to purchase 
foreign technology and equipment unless some locally developed system is 
introduced. 
This thesis presents complete design of a C C T V system, which involves 
mul t i tude of design and implementat ional aspects. It involves deve lopment of an 
image grabber, a remote communicat ion interface and detection a lgor i thms. In 
addition to visual moni tor ing of remote road traffic scenes, the designed system 
is capable of assessing many different traffic parameters , which can be used for 
adapt ive control of road traffic. 
T h e underlined project is an attempt to seek the possibility of introducing image-
based traffic sensing technology to Sri Lanka. Significant attention has given to 
reduce the cost of development throughout the project to ensure that the concept 
of C C T V can be realized in practice in Sri Lanka for road traffic control. . 
T o 
Eve ry body 
W h o he lped mc 
Acknowledgements 
First I thank with deep grati tude to my project supervisor, Dr. J A K S Jayasinghe, 
for the excellent guidance and inspiration he has provided m e through out this 
work. He provided a lot of encouragement and help for which I will be forever 
grateful. 
1 also thank Dr. (Mrs.) Deleeka Dias, Head/Electronics and Prof. (Mrs . ) I. J. 
Dayawansa for their kind attention and valuable suggest ions offered. T h e work 
carried out by them at official levels, towards the continuation of this project is 
deeply appreciated. 
My thanks also go to Dr. Amal S. Kumarage and Dr. J .M.S.J. Bandera for their 
valuable advises. 
1 specially thank the administration of road development Authori ty ( R D A ) and 
Road Construct ion and development company ( R C D C ) for al locating funds for 
this research project. I would like to personally thank Mr. G A M Sumanasekara 
(DGM/Mecan ica l , RCDC) for his kind cooperation offered for the upliftment of 
this project. I also thank Mr. Santha Kumara G a m e g e and Mr. Wasantha 
Ranjewwa for their s incere cooperation offered throughout the project. 
I am extremely grateful to my colleague Hemantha Kodikara Arachchi , wi thout 
whose valuable support and inspirations I could have never achieved this goal . 
Thanks are due to all my good friends in the research lab Ranaweera , 
Chandana , a n d N u w a n for helping mc in many ways to make this project a 
success. 1 also thank the two final year students Dewasurendra and Waravvita for 
spending their valuable t ime to participate on Held visits and helping m e in many 
other ways . 
I also thank Mr. Jayantha Perera and Mr. Senadeera for their help offerd to m e 
throughout this project. Mr. Thushara and Mr. D D Sumanapala are also 
remembered for spending their t ime after working hours to keep the research lab 
open. 
Finally my thank goes to a l l w ho helped me to make this venture a success . 
iii 
Contents 
Abstract 
Dedication 
Acknowledgements 
Contents 
List of Abbreviations 
List of Figures 
List of Tables 
Chapter 1 Introduction 
1.1. The congestion problem 
1.2. Solutions to the congestion problem 
1.3. C C T V based traffic detection 
1.3.1 .The importance of CCTV techniques 
1.4. Scope of the project 
Chapter 2 Vehicular traffic parameters 
2 .1 . Traffic parameter definitions 
2.1.1 Quantity measure 
2.1.2 Quality assessment measure 
2.1.3 Movement measure 
2.1.4 Composition and Classification measure 
Chapter 3 Vehicle detectors 
3.1 Vehicle detection technologies 
3.1.1. U Urasonic Detectors 
3.1.2. Microwave Detectors 
3.1.3. Acoustic Arrays Detectors 
3.1.4. Magnetic Detectors 
3.1.5. Infrared (Passive and active) Detectors 
3.1.6. Inductive Loop Detectors 
3.1.7. Video Image Processing VIP Detectors 
3.2. The relative cost of detectors 
3.3. Advantages of VIP detectors 
C h a p t e r 4 V i d e o i m a g e p r o c e s s i n g t e c h n o l o g y < ^ ; • > V 1 5 
4 . 1 . CCTV system topology 15 
4 .1 .1 . Composition of a CCTV system 15 
4.1.1.1. CCTV cameras 16 
4.1.1.2 Camera Mounting and traffic viewing consideration 17 
4.1.1.3. Camera orientation and field of view 19 
4.1.2. Control Hardware 20 
4.1.3. Control software 21 
4.1.4. Processing software 21 
4.2. Integration of a CCTV system 22 
C h a p t e r 5 T h e I m a g e G r a b b e r d e s i g n 2 5 
5.1 . System Architecture 25 
5.1.1. Sensor Interface Unit (SIU) 25 
5.1.2. Digitizing and Storage Unit (DSU) 26 
5.1.2.1. Clock generation 27 
5.1.2.2. Signal conditioning 27 
5.1.2.2.1 Conditioning for sync separation 27 
5.1.2.2.2 Conditioning at A/D input 28 
5.1.2.3 SYNC separation 29 
5.1.2.3.1 Horizontal S Y N C separation 29 
5.1.2.3.2. Vertical SYNC separation 29 
5.1.2.4. Control Unit 31 
5.1.2.5. Finite State Machine (FSM) 31 
5.1.2.5.1. FSM State transition diagram 31 
5.1.2.5.2. State transition table 33 
5.1.2.5.3. FSM State function and output equation 34 
5.1.2.5.4. Modes of Operation 35 
5.1.2.5.5. Operation in the Capture Mode (MODE_C) 35 
5.1.2.5.6. Data Transfer Mode (MODE J T ) 36 
5.1.2.6. Static RAM Storage 36 
5.1.2.6.1. RAM Organization 37 
5.1.2.6.1. RAM control logic 38 
5.1.2.6.3. RAM Read/Write operation and buss control 39 
5.1.3. Microprocessors Unit (MPU) 41 
5.1.3.1. MPU architecture 41 
5.1.3.2. STE interface 42 
5.1.3.2.1. STE signal definitions 43 
5.3. 
5.1.3.3. MPU device address 44 ' 
5.1.4. Transmission Unit (TXU) 45 
5.1.5. Power Supply Unit (PSU) 46 
System Performance 46 
Data Transfer Routines 47 
5.3.1. Data link control 47 
5.3.2. Data Flow Control 48 
5.3.3. Provision built in to the MPU for remote data access 49 
5.3.3.1. Method of increasing field transmission rate 50 
5.3.3.2. Implementation of data transmission routines 51 
5.3.3.2.1. Image Capturing command 53 
5.3.3.2.2. Full Image transfer command 55 
5.3.3.2.3. Window transfer command 57 
5.3.3.2.4. Line transfer command 59 
5.3.3.2.5. Window store command and.Line store 
command 
59 
5.3.3.2.6. Stored data transfer command 60 
5.3.3.3. Transmission control at host PC 60 
5.3.3.4 The user interface 61 
C h a p t e r 6 V e h i c l e d e t e c t i o n 6 4 
6 .1 . General image processing problems applied to traffic monitoring 64 
6.2. Solutions to the general problems of image processing applied to 65 
traffic monitoring 
6.2.1. Avoiding a bandwidth extensive transmission facility 66 
6.2.1.1. Processing at the camera site itself 66 
6.2.1.2. Reduce the amount data to a manageable rate 66 
6.2.1.3. Introduce offline-processing 68 
6.2.1.4. Elimination of high speed processing requirements 68 
6.2.1.5. Noise reduction 68 
6.2.1.6. Correction to ambient light changes 69 
6.2.1.7. Avoiding problems due to neighboring vehicles 69 
6.3. A Review of research towards traffic monitoring using video images 70 
6.4. Methodologies Of Vehicle Detection 74 
6.4.1. Back ground frame differencing 74 
6.4.1.1. Manipulation of back ground image 75 
6.4.2. Inter Frame Differencing 77 
6.4.3. Vehicle Detection at selected windows 77 
6.4.3.1. Traffic parameter extraction 78 
6.4.4. Tracking objects by feature matching 80 
Chapter 7 Detection algorithms implemented on the designed system 82 
7.1 . System limitations 82 
7.1.1. Image Grabber provision 83 
7.2. Algorithm development for vehicle detection 84 
7.2.1. The detection approach 84 
7.2.2. Sampling rate requirements 84 
7.2.3. Effect of vehicle speed 84 
7.2.4. Detection Algorithm 85 
7.2.4.1.Notation 85 
7.2.4.2. Processing steps 86 
7.2.5. Calculation of traffic parameters 91 
7.3. Interpretation of estimated traffic parameters 92 
7.4. Functionality review 92 
Chapter 8 Offline processing of traffic video 94 
8.1. Site selection 94 
8.2. Results of offline processing 95 
Chapter 9 Conclusion 96 
References 98 
VII 
Abbreviations 
A T C S Advanced Traffic Control Systems 
C C D Charge Coupled Device 
CCTV Closed Circuit Television 
IVHS Intelligent Vehicles and Highway Systems 
MPEG Motion Picture Expert Group 
PATH- Partners for Advanced transit and highway 
T S C Traffic Control System 
U S B Universal Serial Bus 
VIP Video Image Processing 
List of figures 
Fig. 4.1 C C T V based traffic monitoring and intelligent traffic control 16 
Fig 4.2 Camera mounting parameters 18 
Fig 4.3 Line of site detection geometry 19 
Fig 5.1 The image grabber architecture 25 
Fig 5.2 Simplified block diagram of DSU 26 
Fig 5.3 • C L K Circuit 27 
Fig 5.4 Gain/Offset control amplifier at SYNC separation 27 
Fig 5.5 ' Gain/Offset control amplifier at A/D conversion 28 
Fig 5.6 Circuit arrangement at Hsync separation 29 
Fig 5.7.a Circuit arrangement for Vsync separation 30 
Fig 5.7.b Selected waveforms 30 
Fig 5.8 State diagram of the FSM 32 
Fig 5.9 Memory map for field 1 36 
Fig 5.10 SRAM organization 37 
Fig 5.1 1 RAM access and bus control signals 39 
Fig 5.12 RAM write timing - 40 
Fig 5.13 RAM read timing 41 
Fig 5.14 MPU architecture 42 
Fig 5.15 MPU - D S U interface 42 
Fig 5.16.a MPU I/O map 44 
Fig 5.16.D MPU memory map 44 
Fig 5.17.a Null modem connection map 45 
Fig 5.17.D Through modem connection map 45 
Fig 5.18 Detailed functional block diagram of DSU 46 
Fig 5.20.a Data read subroutine 49 
Fig 5.20.b Dats write subroutine 49 
Fig 5.21 MPU main routine 53 
Fig 5.22 Image Capturing routine 54 
Fig 5.23 Transmitted image format 55 
Fig 5.24 Full image transfer routine 56 
ix 
Fig 5.25 Window transfer routine 
Fig. 5.26 The user interface 
Fig. 5.27 Picture examples - different spatial resolutions 
Fig. 5.28 Picture examples - different bit resolutions 
List of tables 
Table 4.1 Comparison of upstream and downstream viewing 21 
Table 5.1 Signal description of FSM 32 
Table 5.2 State transition table for FSM design 33 
Table 5.3 FSM state dependent functions 34 
Table 5.4 Mode of operation 35 
Table 5.5 ' SRAM address counter control 38 
Table 5.6 SRAM chip access control 38 
Table 5.7 . Address map for DSU access 43 
Table 5.8 STE signal definitions 43 
Table 8.1 Comparison between actual and delected vehicle positions 95 
X I (