Sintes 2010 Caius [629734]

ISSN 2068 – 0465

PPPrrroooccceeeeeedddiiinnngggsss
ooofff ttthhheee
111444ttthhh IIInnnttteeerrrnnnaaatttiiiooonnnaaalll CCCooonnnfffeeerrreeennnccceee ooonnn
SSSyyysssttteeemmm TTThhheeeooorrryyy aaannnddd CCCooonnntttrrrooolll
(((JJJoooiiinnnttt cccooonnnfffeeerrreeennnccceee ooofff SSSIIINNNTTTEEESSS111444,,, SSSAAACCCCCCSSS111000,,, SSSIIIMMMSSSIIISSS111444)))
October 17 – 19, 2010
Sinaia, ROMANIA

Editor: Emil PETRE

Organizers:

ƒ University of Craiova, Faculty of Automation, Computers and Electronics,
Automatic Control Research Center
ƒ “Gheorghe Asachi” Technical University of Iasi, Faculty of Automatic Control
and Computer Engineering
ƒ “Dunarea de Jos” University of Galat i, Faculty of Computer Science
ƒ IPA CIFATT Craiova – Institute for Research and Engineering in Automation

Technical co-sponsored by the IEEE – CSS
Control Systems Society

1COMMITTEES

INTERNATIONAL PROGRAM COMMITTEE

Vladimir R ĂSVAN
Department of Automatic Control, Un iversity of Craiova, Romania General Chair

Chaouki T. ABDALLAH
University of New Mexico, United States Program Chair
Silviu Iulian NICULESCU
Laboratoire des signaux et systemes (L2S), Supelec, France CSS Representative

Dan POPESCU
Department of Automatic Control, University of Craiova, Romania Vice-Chairman
Vasile MANTA
Department of Computer Engineering, “Gheorghe Asachi” Technical University of Ia și, Romania Vice-Chairman
Adrian FILIPESCU
Department of Automation and Industrial Informatics,
“Dunarea de Jos” University of Gala ți, Romania Vice-Chairman

MEMBERS
Mihail Abrudean (RO) Ioan Dumitrache (RO) Sorin Olaru (FR)
Tihamer Adam (HU) Gerhard Freiling (DE) Nejat Olgac (USA)
Steve Banks (UK) Emilia Fridman (IL) Octavian Pastravanu (RO) Viorel Barbu (RO) Cornelia Gordan (RO) Emil Petre (RO)
Andrzej Bartoszewicz (PL) Eugen Iancu (RO) Dumitru Popescu (RO)
Costin Badica (RO) Mircea Ivanescu (RO) Alexander Poznyak (MX) Theodor Borangiu (RO) Robin De Keyser (BE) Radu-Emil Precup (RO)
Pierre Borne (FR) Vladimir L. Kharitonov (RU) Stefan Preitl (RO)
Paul van den Bosch (NL) Peter Kopacek (AT) Dorina Purcaru (RO) Sergiu Caraman (RO Karol Kostur (SK) Werner Purgathofer (AT)
Petru Cascaval (RO) Gheorghe La zea (RO) Sergey Ryvkin (RU)
Emil Ceanga (RO) Corneliu Lazar (RO) Bogdan Sapinski (PL) Arben Cela (FR) Rogelio Lozano (FR) Lubomir Smutny (CZ)
Ali Charara (FR) Marin Lungu (RO) Nikos Tsourdveloudis (GR)
Silviu Ciochina (RO) Gheorghe Marian (RO) Nicolae Tapus (RO) Dorian Cojocaru (RO) Qinghao Meng (CH) Erik I. Verriest (USA)
Vladimir Cretu (RO) Viorel Minzu (RO) Gabriel Vladut (RO)
Valentin Cristea (RO) Sabine Mondie (MX) Mihail Voicu (RO) Radu Dobrescu (RO) Sergiu Nedevschi (RO)
Viorel Dugan (RO) Roberto Oboe (IT)

NATIONAL ORGANIZING COMMITTEE
Eugen Bobasu (Chairman – RO) Dorin Popescu (RO)
Emil Petre (Program Editor – RO) Elvira Popescu (RO)
Adrian Burlacu (RO) Monica Roman (RO)
Daniela Cernega (RO) Dan Selisteanu (RO)
Daniela Danciu (RO) Dorin Sendrescu (RO) Augustin Ionescu (RO) Razvan Solea (RO)
Marius Marian (RO) Florina Ungureanu(RO)

6Mihai George RADUCU, Mircea NITULESCU: Control Algorithms for a Self Reconfiguring
Robotic System ………………………..………………………………………………………………..………………………….
438
Gabriel R ĂDULESCU, Nicolae PARASCHIV: Using a Virtualization Te chniques – Based Platform
for Advanced Studies on Operatin g Systems ………………………………………….………………………………………….
444
Vladimir R ĂSVAN, Dan POPESCU, Daniela DANCIU: Monotone and slope restricted
nonlinearities – a PIO II case study …………………………………………………………………………………………………….
449
Ionuț Cristian RE ȘCEANU, George-Cristian C ĂLUGĂRU, Cristina Floriana RE ȘCEANU:
Smith Predictor Structure Experiments fo r a Quanser Servo Motor ………………..…………………………….….
455
Cristina Floriana RE ȘCEANU: Control of Legged Robot with Locked Joint ……….……….…………….. 461
Pedro RODRIGUEZ-AYERBE, Sorin OLARU: Disturbance model in explicit control laws ………….. 467
Monica ROMAN: Modeling and Simulation of a Baker’s Yeast Fed-batch Bioprocess .…..……………. 473
Maria SANTA, Octavian CUIBUS and Tiberiu LETIA: Train Scheduling with Delay Time Petri
Nets and Genetic Algorithms ……………………………………………………………………………………………………………….
479
Mihnea SCAFE Ș, Costin B ĂDICĂ: Complex negotiations in multi-agent systems ………………………. 485
Adriana SCARLAT, Iulian MUNTEANU, Anto neta Iuliana BRATCU and Emil CEANGA:
Improved power optimization method for squirre l-cage-inductiongenerator-based wind energy
conversion systems …………………………………………………………………………….…..……………………………..

491
Daniel SCHERZER: An Overview of Temporal Coherence Methods in Real-Time Rendering ………… 497
Mohamed SEHILI, Dan ISTRATE, Jérôme BOUDY: Primary Investigation of Sound Recognition
for a domotic application using Support Vect or Machines ……………………………….…..……………………….
503
Dorin SENDRESCU, Constantin MARIN, Emil PETRE: Weighted Moments Based Identification
of DC Motor ………………………..…………………………………………………….………..……….…………………………………….
507
Adriana SERBENCU, Adrian Emanoil SE RBENCU, Daniela Cristina CERNEGA: Evolutionary
Strategies for Sliding Mode Contro ller Parameters …………………………………………………………………………….
513
Cristian SMOCHINA, Vasile MANTA, Giovanna BISES, Radu ROGOJANU: Automatic cell
nuclei detection in tissue sections fr om colorectal cancer …………………………………………………………………
519
Veaceslav SPINU, Mircea LAZAR and Paul van den BOSCH: On low complexity model
predictive control of DC/DC co nverters ……………………………………..……………………………………………………….
525
Andrei STAN, Lucian PANDURU and Florina UNGUREANU: Architectural Support for
Subroutine Execution Time Monitoring in Em bedded Microprocessors ……….………………………………..
531
Roxana ST ĂNICĂ, Emil PETRE: Control the Packets Transmission Using Quality of Service
Protocol ……………………………………………………………….…………….…….…..…………………………………………………..
535
Florin STINGA, Dan MITOIU, Ionut NISIPEANU, Andreea SOIMU: Hybrid Control Scheme
Implemented on a Programmable Logical Contro ller …………………………………………..………………………
541
Cosmin STOICA SPAHIU: Questions generation management system for e-learning ………………
546
Caius SULIMAN, Cristina CRUCERU, G. MACESANU and Florin MOLDOVEANU: Person
Tracking in Video Surveillance System s Using Kalman Filtering ………………………………………………………..
550
Grigore VASILIU, Ionut MIHALCEA, Serghei RADJABOV, Adriana FILIPESCU: Intelligent
Trajectory Tracking in Sliding Mode Based Wh eeled Mobile Robot Control ……………………………..………
556
Matei VINATORU: Dynamic Stability for Hydropower Plant Systems ………………….…..…………….. 562
Gabriel VLADUT, Monica MATEESCU, Lian a Simona SBIRNA, Sebastian SBIRNA:
Mathematical model proposed for simulating the influence on air quality of the three ash dumps that affect Craiova, Romania ………………………………………………………………..………………………………………….

568
Index of authors ……………………………………………………………………………………………………………………………….. 572
List of reviewers ……………………………………………………………………………………………………………………………….. 575


Abstract — In this paper we have developed a Simulink based
model for monitoring contacts in a video surveillance sequence.
To correctly identify a contact in a surveillance video, we have
used the Lucas-Kanade optical flow algorithm. The position
and the behavior of the correctly detected contact was
monitored with the help of the traditional Kalman filter. Here
we compare the results obtained from the optical flow with the
ones obtained from the Kalman filter, and we show the correct
functionality of the Kalman filter based tracking. The tests
were performed using video data taken with the help of a fix
camera. The tested algo-rithm has shown promising results.
I. I NTRODUCTION
HE problem of using vision to track and understand the
behavior of humans is a very important one. The main
applications that it has are in the areas concerning human-robot interaction [6], robot learning, and video surveillance.
Here we try to focus our attention on video surveillance
systems. A high level of security in public places is an extremely complex challenge. A number of technologies can be applied to various aspects of security, including biometric systems, screening systems, and video surveillance systems. Nowadays video surveillance systems act as large-scale
video recorders, analog or digital. These systems serve two
main purposes: to provide a human operator with images to detect and react to potential threats and recording for future investigative purposes.
From the perspective of real-time detection, it is well
known that the human’s visual attention drops below acceptable levels even if that operator is a trained one in the task of visual monitoring. Video analysis technologies can
be applied to develop smart surveillance systems that can aid
the operator in the detection and in the investigatory tasks.
For surveillance applications, the tracking problem is a
fundamental component. In video surveillance one of the most used method for tracking contacts is the particle filter [7][10][11][13]. Another well known method in the research community is the use of the traditional Kalman filter [9]. In
Manuscript received August 31, 2010. This paper was supported by the
Sectoral Operational Programme Human Resources Development (SOP
HRD), fi-nanced from the European Social Fund and by the Romanian
Government under the contract number POSDRU/6/1.5/S/6.
Suliman C. is with the Automatics Department, “ Transylvania ”
University of Brasov, Brasov, Romania (e-mail: caius.suliman@unitbv.ro).
Cruceru C. is with the Automatics Department, “ Transylvania ”
University of Brasov, Brasov, Romania (e-mail: cri.cruceru@yahoo.com).
Macesanu G. is with the Automatics Department, “ Transylvania ”
University of Brasov, Brasov, Romania (e-mail: gigel.macesanu@unitbv.ro)
Moldoveanu F. is with the Automatics Department, “ Transylvania ”
University of Brasov, Brasov, Romania (e-mail: moldof@unitbv.ro). many cases the use of this type of filter is sufficient. This is
due to the controlled indoor and outdoor environments that are used in the studies.
Many papers in the literature detail methods that track
single persons only [5][10], but there are also many authors that describe different methods for the detection and tracking of multiple persons [2][3][4][11]. Most of these
methods involve as testing grounds indoor environments
[1][3][7][13] as well as outdoor environments [2][4][7][9], where these methods are applied to track groups.
The objective of this paper is the development of a video
surveillance system capable of tracking multiple persons in an outdoor environment. In Section II we describe the structure of the proposed video surveillance system. In Section III we present the method used for contact detection and the method used for the extraction of useful data from
the video feed. Section IV describes the Kalman filter
algorithm applied in our case. In Section V and VI we present the results obtained from the Simulink model’s
simulation, the conclusions drawn from this study and the
possible future developments.
II.S
URVEILLANCE SYSTEM STRUCTURE
In this paper we examine the feasibility of using the
optical flow algorithm in conjunction with the Kalman filter algorithm [9][12] for tracking multiple contacts in a
surveillance scene. In order to create an algorithm that is
able to track a contact in a scene, three different, large-scale
task must be accomplished (see Fig. 1). First the algorithm
needs to take an incoming surveillance video signal and segment it into a stream of frames where contacts are distinguished from the background of the scene. The next step is the tracking of the contacts throughout the video sequence. Finally, the resulting track must be processed in
order to analyze the contact’s behavior.
For the segmentation process of the incoming video
signal, the optical flow algorithm developed by Lucas and
Kanade was used [8]. The optical flow algorithm
approximates the movement of the contact in the current
frame as referenced to the previous frame. By determining the motion of objects, one can distinguish between the contact and the background of the scene. After careful
tuning and processing, the output of the segmentation
process is passed to the Kalman filter algorithm for further processing. The Kalman filter is a recursive, adaptive filter that operates Person Tracking in Video Surveillance
Systems Using Kalman Filtering
Suliman C., Cruceru C., Macesanu G. and Moldoveanu F., Member, IEEE
T
550

in the state space. It is well known for its ability to track
objects in a timely and accurate manner. The tracking
algorithm developed in this paper is able to process multiple
contacts.
Fig. 1. The surveillance system structure.
III. P ERSON DETECTION AND POSITION ESTIMATION
In Fig. 2 we will present the main component parts of the
Optical Flow Analysis block.
Fig. 2. The optical flow analysis block.
A. Optical flow analysis
One of the important blocks presented in the above
scheme is the so called optical flow analysis block. The
main purpose of this block is to determine the existence of
possible contacts in the incoming video signal and process
them in such manner that the Kalman filter will be able to
estimate their position with minimal error.
In the following we will describe the functionality for
each component of the optical flow analysis block.
1) Segmentation
In our case, the term segmentation is used to describe the
process through which a video signal passes to become a series of binary images. At the output of this sub-block, each
of the resulting binary images will contain black and white
areas. The black areas correspond to the portion of the frame
where no motion was detected, and the white areas
correspond to the portion of the frame where motion was detected.
Because we use the optical flow to detect motion, the
video signal needs to be converted to the intensity color
space. To estimate the optical flow between two images we
use the algorithm developed by Lucas and Kanade [8]. In our case this algorithm is used to compute the optical flow between the current frame and the previous one. This is one
of the tunable parameters used in our experiments. The
method proposed by Lucas and Kanade woks by assuming
that optical flow is constant in an
mmu window (mask),
with 1!m . This window is centered on the pixel. The
numbering of the pixels begins from 1 to n, where 2mn .
From this we will have the following set of equations:
»»»
¼ș
«««
¬ă

»¼ș
«¬ă
»»»»»
¼ș
«««««
¬ă
321
2 21 1
ttt
yn xny xy x
III
vu
I II II I
 (1)
In the above equations we can see that we have only two
unknowns ( u and v). The solution for the above system of
equations will have the following form:
n iIIII
I IIII I
vu
tiyitixi
yi yixiyixi xi…1,22
»
¼ș
«
¬ă
¦¦
»»
¼ș
««
¬ă
¦ ¦¦¦ »¼ș
«¬ă (2)
From the above we can say that the optical flow can be
obtained by computing all of the image derivatives for all
the values of i.
Another important tunable parameter used in the optical
flow estimation is the threshold for noise reduction. It is
used for eliminating the small movements between frames.
The higher its value, the less small movements impact the
optical flow computing. Our experiments pointed out that the optimal value for the smoothness factor is 0.005.
Before the processed video signal exits the segmentation
sub-block it is compared with a certain threshold to keep
only what interests us from the video feed.
Fig. 3. The segmentation process.
2) Median Filtering
One of the biggest problems that optical flow has it’s that
it is very sensitive to changes in illumination or to the
quality of the video. This sensitivity conduces in erroneous
blobs appearing in individual frames. If these blobs are
551

large, that means that they are approaching the average size
Fig. 4. Median filtering.
of the contact that we are trying to track, they can create
problems for successful morphological operations.
The main reason for choosing the median filter is that
most of these abnormalities, the erroneous blobs, appear in singular frames and they do not appear again for several more frames. The median filter is used to decrease the effect
of these abnormalities while still maintaining the
information of the correctly detected contacts (see Fig. 4).
3) Morphological Operations
The morphological operation will process the video signal
coming from the output of the median filtering sub-block in
such way that all erroneous blobs residing in the image are eliminated and all and only the correctly detected blob is
maintained and classified as a real contact. The main morphological operations used by us in this study are the
erosion and dilatation . Optimal erosion is achieved when
the structuring element keeps at least the remnant of a blob
for all correct contacts. If we use a sub-optimal structuring element for the erosion and dilatation operations, a valid
contact could be lost completely, or an erroneous blob could
be tracked. Both these errors can produce significant barriers to optimal dilatation. Optimal dilation is obtained when the
structuring element merges all remnants of a single blob into
one contact. If a sub-optimal structuring element is used for
dilation, one contact could be viewed as multiple contacts or multiple contacts could be viewed as one contact. After an
optimal structuring element for erosion was determined,
each frame was eroded using the chosen structuring element.
Determination of the optimal structuring element for dilation
was similar to that of erosion. Each frame was dilated with a square structuring element.
An infinite number of possibilities exist for size and shape
of structuring elements. Depending on the data used, the size and shape of the optimal structuring element could vary
significantly.
Fig. 5. Morphological operations. After the median filtering the erosion operation removed
great part of the remaining erroneous blobs residing in the image, thus deciding that they were not contacts. Further
dilation has created a solid blob out of the area where
motion was detected, and this blob will be tracked as a
contact (see Fig. 5).
4) Blob Analysis
The main functionality of the blob analysis sub-block is to
determine the minimum size of a blob and the maximum
number of blobs that will be used in the Kalman position
estimation step. By setting the minimum blob size we obtain
a new level of protection against abnormalities by specifying
a minimum size that a blob must have in order to be correctly tracked. Thus, any blob that doesn’t fulfill this
condition will not be tracked.
The other tunable parameter of the blob analysis sub-
block, the maximum number of blobs, is used to set the
number of Kalman filters to be used in the tracking process.
In our case this parameter was set to 1.
B. Kalman Filtering
Filtering is a very used method in engineering and
embedded systems. A good filtering algorithm can reduce the noise from signals while retaining the useful information. It estimates the states of linear systems. This
type of filter works very well in practice and that is why it is
often implemented in embedded control system and because we need an accurate estimate of the process variables. The
discrete Kalman filter is characterized by both a process
model and a measurement equation.
The process model is characterized by the assumption that
the present state,
kx, can be related to the past state, 1kx,
as follows:
k kk k w x x ) 1 , (2)
wherekw is assumed to be a discrete, white, zero-mean
process noise with known covariance matrix, kQ;k)
represents the state transition matrix which determines the
relationship between the present state and the previous one.
In our case we try to estimate the current state of an
contact based on its last known state. Here, the state vector consists of a two-dimensional position expressed in Cartesian coordinates, a two-dimensional velocity and a
two-dimensional acceleration. By considering a constant
acceleration, the state transiti on matrix can be determined
from the basic kinematic equations as follows:
2
1 1 121ta tv s sk k k k   , (3)
ta v vk k k 1 1 , (4)
1 k ka a , (5)
where s is defined to be the contact’s position, v is its
velocity, a is the contact’s acceleration and t is the
sampling period. In a matrix form, the above equations can
be written as:
552

»»»»»»»»
¼ș
««««««««
¬ă
»»»»»»»
¼ș
«««««««
¬ă

»»»»»»»»
¼ș
««««««««
¬ă
1,1,1,1,1,1,
,,,,,,
1 000000 100001 010000 101005.00101005.00101
kykxkykxkykx
kykxkykxkykx
aavvss
aavvss
. (6)
Here, the subscripts x and y refer to the direction of the
contact’s position, velocity and acceleration in the two-
dimensional plane. The value of the sampling period is set to 1. From the above equation the state transition matrix,
k),
is:
»»»»»»»
¼ș
«««««««
¬ă
)
1 000000 100001 010000 101005.00101005.00101k . (7)
The measurement equation is defined as:
k kk k v xH z  , (8)
where kz represents the measurement vector, kv is
assumed to be a discrete, white, zero-mean process noise
with known covariance matrix, kR. The matrix kH
describes the relationship between the measurement vector,
kz, and the state vector, kx. Given the fact that the state
vector is of length six and the measurement vector is of length two, the matrix
kH must be of length six by two:
»¼ș
«¬ă 5.00101005.00101
kH . (9)
From the process model and measurement equation it
results that the Kalman filter attempts to improve the prior
state estimate using the incoming measurement which has
been corrupted by noise. This improvement can be achieved
by linearly blending the prior state estimate, 1ˆkx, with the
noisy measurement, kz, in:
)ˆ ( ˆ ˆ  kk k k k k xHzK x x . (10)
Here 
kxˆ means the a-priori estimate; kK is known as the
blending factor. The minimum mean squared error of the estimate is obtained when the blending factor assumes the
value of the Kalman gain:
1) (   kT
k kkT
k k k R HPH HP K , (11)
where kP is known as the state covariance matrix. Generally
the state covariance matrix is a diagonal matrix. The state covariance matrix is determined from the a-priori state
covariance matrix as follows:
 k kk k PHKI P ) ( . (12) After the Kalman gain has been computed, and the state
and state error covariance matrices have been updated, the
Kalman filter makes projections for the next value of k.
These projections will be used as the a-priori estimates
during processing of the next frame of data.
kk k x x ˆ ˆ1) 
 , (13)
kT
kkk k Q P P )) 
1 . (14)
The above equations are the projection equations for the
state estimate and for the state covariance matrix.
The main role of the Kalman filtering block is to assign a
position estimation filter to each of the measurements
entering the system from the optical flow analysis block.
Therefore we have a filter for each of the detected contacts.
For an easy implementation of the Kalman filter in
Simulink, we wrote an embedded Matlab function. This
method is often used when the function that needs to be implemented is more easily to express in Matlab’s symbolic
language than in Simulink’s graphical language.
Fig. 6. The Kalman filter algorithm.
IV. P OST PROCESSING
The last block that we will discuss is the post
processing/video output block. This block was used to
process the output from the optical flow analysis block and
the output of the Kalman filtering block.
The post processing block is composed of three video
output sub-blocks. The first sub-block is used only to view
the original video signal.
The second sub-block is used to visualize the resulting
signal from the optical flow analysis block and to allow the
user to be sure of the correct functionality of the optical flow
analysis block. This sub-block is in direct connection with the blob analysis sub-block, block that produces the
coordinates for a bounding box. This bounding box is a
rectangle drawn around each correctly detected blob. The user is able to watch in real-time which contact in the video
553

feed is being sent to the Kalman filtering block. If rectangles
are not surrounding the correctly detected contacts in an
image, this thing means that the optical flow analysis bloc is
not working properly. Figure 7a presents the resulting output of the optical flow video viewer sub-block. We
present only four frames taken at 17.5 FPS of each other. It
can be clearly seen that a contact was detected and a bounding box was correctly superimposed on the contact.
The same thing happens in figure 7b where two contacts
have been detected. The last sub-block discussed in this section is the Kalman filtering video viewer. This sub-block
is in direct connection with the Kalman filtering block. This
block produces at its output a matrix containing the position of the detected contact. The output is used by the Kalman
filtering video viewer to draw markers in the video. These
markers are represented here by red filled circles. The user is
able to see in real-time if the detected contact is correctly
tracked by the Kalman filter. If a marker doesn’t follow consistently a contact we can say that the Kalman filtering block isn’t working properly. Figure 8a presents four frames
resulting from the Kalman filtering video viewer, taken like
in the previous case, at 17.5 FPS of each other. We can clearly see that the marker is correctly tracking the detected contact thus confirming that the Kalman filter is working properly. The Kalman filter is even capable to track the contact that is leaving at some time the visual field of the
camera and then correctly reassign the marker to the contact
that reenters in the visual field. Figure 8b presents the
tracking process for two correctly detected contacts.
V.C
ONCLUSIONS AND FUTURE WORK
There are two main factors that affect the problem of
tracking: the accuracy to distinguish between contacts
passing through the scene and the speed to process the video
feed in real-time. In this paper we have shown that with the
help of the optical flow and Kalman filter algorithms it is possible to detect and track a person passing through a
scene.
The video signal used in our experiments it is provided by
a Linksys WVC200 PTZ IP video camera at a resolution of
240×320. The entire experiment was conducted using an
Intel Core2Duo T9300 computer with 4 GB of RAM.
From the optical flow analysis used in our research we
have deduced that there is an inevitable trade-off between the accuracy and speed of processing. To accurately
distinguish a contact that passes through a scene, the
computational time of the optical flow algorithm must be increased. If this increase of the processing time is too large, the algorithm will not operate in real-time.
(a)
(b)
Fig. 7. Output sample from the optical flow video viewer.
(a)
554

(b)
Fig. 8. Output sample from the Kalman filtering video viewer.
(a) tracking a contact that passes through the camera’s field of view; (b) tracking two contacts that are passing through the c amera’s field of view.
The Kalman filter algorithm presented in this research was
able to correctly process a contact and to correctly assign a filter to the processed contact. After reviewing the results we
deduced that the algorithm performed quite well showing a
moderate consistency in tracking. Due the success with the
data used in our experiments, any inconsistencies in the tracking process can be traced back to the fluctuations in performance of the optical flow algorithm.
Future research in the area of surveillance systems should
be focused in two directions. First, research should be made
to determine an objective measure of performance of the
optical flow algorithm and to see if other existing algorithms are better suited to accurate, real-time processing of a video
signal. The second research should be focused in the area of
determining contact behavior, and in areas such as merging
contacts into groups or dividing groups into separate
contacts.
For a future research we will try to make a comparison
between the presented segmentation method for contact
detection and other segmentation methods .
A
CKNOWLEDGMENT
This paper is supported by the Sectoral Operational
Programme Human Resources Development (SOP HRD),
financed from the European Social Fund and by the
Romanian Government under the contract number
POSDRU/6/1.5/S/6.
REFEERENCES
[1] M. A. Ali, S. Indupalli, B. Boufama, “Tracking Multiple People for
Video Surveillance,” First Intern. Workshop on Video Processing for
Security , June 2006. [2] B. Benfold, I. Reid, “Guiding Visual Surveillance by Tracking Human
Attention,” Proc. of the 20th British Machine Vision Conf. , September
2009.
[3] L.M. Fuentes, S.A. Velastin, “From tracking to advanced surveillance,” Proc. of the Intern. Conf. on Image Processing , vol. 3,
pp. 121–124 , September 2003.
[4] C. C. Hsieh, S. S. Hsu, “A Simple and Fast Surveillance System for
Human Tracking and Behavior Analysis,” Proc. of the 3
rd Intern.
IEEE Conf. on Signal-Image Technologies and Internet-Based System ,
pp. 812–818, December 2007.
[5] F. Jean, R. Bergevin, A.B. Albu, “Body tracking in human walk from
monocular video sequences,” Proc. of the 2nd Canadian Conf. on
Computer and Robot Vision , pp. 144–151, May 2005.
[6] N. Koenig, “Toward real-time human detection and tracking in diverse
environments,” Proc. of the 6th IEEE Intern. Conf. on Development
and Learning , pp. 94–98, July 2007.
[7] S. Kong, M.K. Bhuyan, C. Sanderson, B.C. Lovell, “Tracking of
Persons for Video Surveillance of Unattended Environments,” Proc.
of the 19th Intern. Conf. on Pattern Recognition , pp. 1–4, December
2008.
[8] B. Lucas, T. Kanade, “An iterative image registration technique with
an application to stereo vision,” Proc. of the Intern. Joint Conf. on
Artificial Intelligence , pp. 674–679, 1981.
[9] W. Niu, L. Jiao, D. Han, Y. F. Wang, “Real-time multiperson tracking
in video surveillance,” Proc. of the 4th Pacific Rim Conf. on
Multimedia , vol. 2, pp. 1144–1148, December 2003.
[10] A.W. Senior, G. Potamianos, S. Chu, Z. Zhang, and A. Hampapur, “A comparison of multicamera person-tracking algorithms,” Proc. IEEE
Int. Works. Visual Surveillance , May 2006.
[11] J. Wang, Y.g Yin, H. Man, “Multiple Human Tracking Using Particle
Filter with Gaussian Process Dynamical Model,” EURASIP Journal
on Image and Video Processing , vol. 2008, Article ID 969456, 10
pages, 2008.
[12] G. Welch, G. Bishop, “An Introduction to the Kalman Filter,”,
Technical Report: TR95-041 , University of North Carolina, 2006.
[13] J. Yao, J.M. Odobez, “Multi-Camera 3d Person Tracking With
Particle Filter In A Surveillance Environment,” Proc. of the 16th
European Signal Processing Conf. , August 2008.
555