World Academy of Science, Engineering and Technology

International Journal of Electronics and Communication Engineering

Vol:6, No:4, 2012

Object tracking system using Camshift, Meanshift

and Kalman filter
Afef SALHI and Ameni YENGUI JAMMOUSSI

Abstract—This paper presents a implementation of an object in the earlier researches. Selecting the right feature plays a
tracking system in a video sequence. This object tracking is an critical role in the tracking object in video sequence. Feature
important task in many vision applications. The main steps in video selection is closely related to the object representation. The
analysis are two: detection of interesting moving objects and tracking
of such objects from frame to frame. In a similar vein, most tracking object motion analysis usually focuses on simple characteris-
algorithms use pre-specified methods for preprocessing. In our work, tics such as texture, color, shape, geometry, etc. In general,
we have implemented several object tracking algorithms (Meanshift, many tracking algorithms use a combination of these features.
Camshift, Kalman filter) with different preprocessing methods. Then, In general, these features are selected manually by the user,
Open Science Index, Electronics and Communication Engineering Vol:6, No:4, 2012 waset.org/Publication/15036

we have evaluated the performance of these algorithms for different depending on the application domain. However, the problem of
video sequences. The obtained results have shown good performances
according to the degree of applicability and evaluation criteria. automatic feature selection has received considerable attention
in pattern recognition, namely the detection of objects to
Keywords—Tracking, Meanshift, Camshift, Kalman filter, Evalua- achieve tracking in video sequences. Tracking is often the first
tion.
step in an analysis of activities, interactions and relationships
between objects of interest. Many algorithms of object tracking
I. I NTRODUCTION have been proposed and developed. Tracking is the estimation
and analysis of trajectories of an object in the plane of the
Object tracking is an important research topic in computer
image by moving through a sequence of images or video.
vision. The rapid increase in numbers of great powered
These include for example when the main existing algorithms
computers, the avaibility of high quality and high precision
in the literature as the block-matching, KLT, the Kalman filter
and cheap video cameras. Consequently, the use of object
[6], [5] , Meanshift [2], [7] and Camshift [11], [9], [10].
tracking (person, face, hand, glass, car) is in the tasks of
The rest of paper is organized as follow: section 2 presents
automobile driver assistance, vehicle navigation, robotics,
the object tracking system. In section 3, we describe the
human-computer interaction, video surveillance, biometric,
principle methods of preprocessing, to detect the exact position
video games and industrial automation and security. Most of
of the object in the images of the sequence analyzed and the
these applications require reliable object tracking techniques
experimental results and shows the performance of our object
that meet in real-time constraints. The object tracking in
tracking in a video sequence. In section 4 summarizes our
a video sequence can be defined the dynamic entities that
work and outlines potential further improvements.
constantly change under the influence of several factors. In
the physical world, there are five parameters to be considered:
changes of the appearance, illumination, scale, object, and II. O BJECT TRACKING SYSTEM
movement of the object (fast or slow). The variation of one To ensure good organization the progress of work, we used
of these parameters can influence the results of our tracking the benefits of modular design in our algorithm implemented
algorithm (the change of illumination, the change of scale, using MATLAB. The goal of an object tracking is to gen-
the change in appearance, the change of motion of the object erate the trajectory of an object over time by discovering
and the change of the object). A huge number of tracking its exact position in every frame of the video sequence. we
methods which have been proposed in earlier researches. In have implemented several object tracking algorithms (Mean-
a tracking video sequence, an object can defined as anything shift, Camshift, Kalman filter) with different preprocessing
that is interesting for analysis. For example, people walking methods. The algorithm for object tracking is composed of
on a road, planes in the air, cars in the road, and hand or three modules: selection object module in the first frame of
a face in motion, etc. In recent researches, the form and the sequence, the module of Meanshift algorithm and the
appearance representations are classified into three families module of Camshift algorithm. The selection module selects
as the representation by points, representation by bounding the position of the object in the first frame. It consists of
boxes and the object silhouettes and contour [4], [16]. These extracting the module initialization parameters that are moving
methods for object tracking in video sequence and the feature through the position, size, width, length of the search window
selection for tracking are applied for many tracking algorithms of the object in the first frame of the sequence. The step of
object tracking system are shown in figure 1.
Computers Imaging Electronics and Systems Group(CIELS)
Research unit on Intelligent Control, design and Optimisation of complex
Systems (ICOS) A. Meanshift
University of Sfax, Sfax Engineering School, BP W, 3038 Sfax, Tunisia
Phone: (216-74) 274.088, Fax. (216-74) 275.595. The meanshift algorithm is a non-parametric method [13],
Emails:[email protected]. [2], [7]. It provides accurate localization and efficient matching

International Scholarly and Scientific Research & Innovation 6(4) 2012 421 ISNI:0000000091950263
 World Academy of Science, Engineering and Technology
International Journal of Electronics and Communication Engineering
Vol:6, No:4, 2012

video sequence zero (1) represents the area occupied by the shape of the frame.

M 00 = I(x, y) (1)
preprocessing block
x y

The moments of order one (M 10,M 01) are calculated by

Block prediction
Kalman filter
equations (2).
 
M 10 = x × I(x, y); M 01 = y × I(x, y) (2)
Block tracking x y x y
(Selection, Meanshift, Camshift)
We find the centre of mass of the object by means of the
moments of zero order and one, (xc, yc) is expressed by the
No Correction block
Kalman Filter
equation (3).
M 10 M 01
xc = ; yc = (3)
Yes M 00 M 00
Tracking results
B. Camshift
Fig. 1. flow chart of the object tracking system. The principle of the CamShift algorithm is given in [12],
Open Science Index, Electronics and Communication Engineering Vol:6, No:4, 2012 waset.org/Publication/15036

[9], [10], [11], is based on the principles of the algorithm

Meanshift. Camshift is able to handle the dynamic distribution
without expensive exhaustive search. The size of the window by adjusting the size of the search window for the next frame
of search is fixed. It is an iterative process, that is to say, first based on the moment zero of the current distribution of images.
compute the meanshift value for the current point position, In contrast to the algorithm Meanshift who is conceived for
then move the point to its meanshift value as the new position, the static distributions, Camshift is conceived for a dynamic
then compute the meanshift until it fulfill certain condition. For evolution of the distributions. It adjusts the size of searching
an frame, we use the distribution of the levels of grey which window by invariant moments. This allows the algorithm to
gives the description of the shape and we are going to converge anticipate the movement of objects to quickly track the object
on the centre of mass of the object calculated by means of in the next frame. Even during the fast movements of an object,
moments. The flow chart of meanshift in figure 2 described Camshift is still capable of tracking well. It occurs when
the steps of the algorithm. The number of iterations of the objects in video sequences are tracked and the object moves
convergence of the algorithm is obtained when the subject is such that the size and location of the change in probability
followed within the image sequence [9]. distribution changes in time. The initial search window was
determined by a detection algorithm or software dedicated
Image in HSV to video processing. The CamShift algorithm calls upon the
MeanShift one to calculate the target centre in the probability
Select the variable H distribution image, but also the orientation of the principal
of the image axis and dimensions of the probability distribution. Defining
the first and second moments for x and y. These parameters
Choose the size of
the search window are given from the first and second moments, are defined by
equations (4, 5 et 6).
Choose the initial location 
of the search window M 20 = x2 × I(x, y) (4)
x y

Find center of mass
of the object M 02 = y 2 × I(x, y) (5)
x y
Center search window 
Compute the parameters at the center of mass M 11 = x × y × I(x, y) (6)
x y
Yes Convergence? No The orientation of the major axis and the scale of the distribu-
tion are determined by finding an equivalent rectangle that has
the same moments as those measured from the 2D probability
Fig. 2. flow chart of Meanshift algorithm.
distribution image. The orientation is defined by the equation
(7).
1) Calculation of parameters:: The inputs of this module 2×( M 11
M 00 −xc×yc)
arctan( ( M 20 −x 2 )−( M 02 −y 2 ) )
are the initial position, width and length of the initial search
θ= M 00 c M 00 c
(7)
window of the object. Let I(x, y) the image of ”2D”, (x, y) 2
is the position of the object in the image. The zero-order The first two eigenvalues (the length and width of the proba-
moment is given by equation (1), an order moments spatial and bility distribution) are calculated in closed form as follows in
the center of mass are given by equations (2) and (3). These the equations (8), (9) and (10).
parameters are the output of Meanshift algorithm module and M 20
the input of Camshift algorithm module. The moment of order a= − x2c (8)
M 00

International Scholarly and Scientific Research & Innovation 6(4) 2012 422 ISNI:0000000091950263
 World Academy of Science, Engineering and Technology
International Journal of Electronics and Communication Engineering
Vol:6, No:4, 2012

M 11
b=2×( − x c × yc ) (9) The measurement model is defined by equation (17).
M 00
zk = H × sk + vk (17)
M 02
c= − yc2 (10)
M 00 With H (18) is the measurement matrix:
⎛ ⎞
The length and width of the distribution around the centroid 1 0 0 0 0 0
is then given by equations (11) and (12). ⎜0 1 0 0 0 0⎟
 H=⎜ ⎝0 0 1 0 0 0⎠
⎟ (18)

(a + c) + b2 + (a − c)2 0 0 0 1 0 0
l= (11)
2 Noise process ”wk−1 ” and as ”vk ” are assumed independent
  of state vectors and measurement and normal distributions and
((a + c) − b2 + (a − c)2 white that are presented by equations (19) et (20):
w= (12)
2
p(w) ∼ N (0, Q) (19)
C. Block Kalman filter p(v) ∼ N (0, R) (20)
The Kalman filter estimates the position of the object in
Open Science Index, Electronics and Communication Engineering Vol:6, No:4, 2012 waset.org/Publication/15036

The process noise is of the form (21):

each frame of the sequence [6], [5]. The input parameters ⎛ ⎞
of the Kalman filter, respectively, the position of the object 1
⎜1⎟
in the image at time k, the size of the object and the width ⎜ ⎟
⎜1⎟
and length of the search window of the object which vary wk−1 = ⎜ ⎟ (21)
⎜1⎟
due to the mobility of the object during the sequence. These ⎜ ⎟
⎝1⎠
parameters represent the state vector and measurement vector
of the Kalman filter. In general, the estimation of parameters 1
followed with a Kalman filter is a process that requires the The measurement noise is presented by the matrix of dimen-
following steps: sion (4 × 1) (22):
- The measure is to take the tracking parameters calculated ⎛ ⎞
0.1
by the algorithm Camshift. ⎜0.1⎟
- The estimate, which updates the position of the object. vk = ⎜⎝0⎠
⎟ (22)
- The prediction, which calculates the position of the object 0
in the next frame.
The variable parameters of the Kalman filter are the state So the covariance of process noise and measurement are
vector and measurement vector: The state vector is composed deducted from wk−1 and vk by the matrices (23) et (24):
⎛ ⎞
of the initial position, width and length of the search window 1 0 0 0 0 0
and the center of mass of the object (xc, yc) at time tk . This ⎜0 1 0
⎜ 0 0 0⎟ ⎟
vector is presented by equation (13). ⎜0 0 1 0 0 0⎟ ⎟
Q=⎜ (23)
⎜0 0 0
⎜ 1 0 0⎟ ⎟
sk = (xk , yk , Wk , Lk , xc , yc ) (13) ⎝0 0 0 0 1 0⎠
The measurement vector of the Kalman filter is composed of 0 0 0 0 0 1
the initial position, length and width of the search window of ⎛ ⎞
the object at time tk. This vector is given by equation (14). 0.1 0 0 0
⎜0 0.1 0 0⎟
zk = (xk , yk , Wk , Lk ) R=⎜
⎝0
⎟ (24)
(14) 0 0 0⎠
0 0 0 0
1) Process to estimate : The Kalman filter estimates the
state s a discrete process, this state is modeled by the linear 2) The equations for updating: Finally, the output equations
equation (15). for the two blocks of prediction and correction of Kalman filter
sk = A × sk−1 + wk−1 (15) are:
- Equations for predicting (25), (26):
With A (16) is the transition matrix, wk is the noise process
and dt is the difference between the two moments k and k −
ŝk = A × ŝk−1 (25)
1(dt = 1).
⎛ ⎞ Pk− = A × Pk−1 × A + Q T
(26)
1 0 dt 0 0 0
⎜0 1 0 dt 0 0⎟
⎜ ⎟ - Correction Equations (27), (28) and (29):
⎜0 0 1 0 0 0 ⎟
A=⎜ ⎜ ⎟ (16)
⎟
⎜0 0 0 1 0 0 ⎟ Kk = Pk− × H T × (H × Pk− × H T + R)−1 (27)
⎝0 0 0 0 1 0 ⎠
0 0 0 0 0 1 ŝk = ŝ− −
k + Kk × (zk − H × ŝk ) (28)

International Scholarly and Scientific Research & Innovation 6(4) 2012 423 ISNI:0000000091950263
 World Academy of Science, Engineering and Technology
International Journal of Electronics and Communication Engineering
Vol:6, No:4, 2012

Pk = Pk− − Kk × H × Pk− (29)

The expression of the estimation error is given by equation 30.
We applied this equation 30 for all parameters of state vectors
Video Redcup: Frame 2 Video Stforge: Frame 53
(x, y, W, L, xc , yc ) algorithm implemented in our tracking Video Forman: Frame 2

of an object in a video sequence (Camshift algorithm and the Fig. 3. Preprocessing results of 3 sequences video by the histogram method.
algorithm of the Kalman filter):
Ex = xKalman − xCamshif t (30)
The results of error calculation for estimating the parameters Video Forman: Frame 27 Frame 63 Frame 93

of our monitoring algorithm (Camshift and Kalman filter) we

obtain the test sequence ”Foreman” are given in table I.
TABLE I
T HE RESULTS OF ERROR CALCULATION OF THE PARAMETERS OF STATE Video Redcup: Frame 2 Frame 45 Frame 99
VECTORS CALCULATED .
Open Science Index, Electronics and Communication Engineering Vol:6, No:4, 2012 waset.org/Publication/15036

Video Stforge: Frame 53 Frame 54 Frame 63

N d’images E(x) E(y) E(W) E(L) E(xc ) E(yc )

2 17 -10 -5 -8 -0.157 -0.0162 Fig. 4. Tracking results of 3 sequences video by the Histogram methods.
10 14 -12 -3 -4 -0.8106 -0.7257
15 0 1 -3 -3 -0.4723 -0.1218 Trajectory in pixels from the centre of the glass with Camshift algorithm.

25 2 0 11 -15 0.5163 -0.9389

35 1 0 11 -16 0.1011 -0.2007 (xc, yc) with Camshift algorithm
Trajectory of centre of the glass X: 139.4
Y: 146.9
Final position of centre of glass

45 1 1 11 -22 -0.0832 0.1103

55 0 -17 11 -25 -0.5864 0.0067 X: 112.1
Y: 111.3 X: 162.6
Y: 106.2

65 1 4 11 -27 -0.0258 -0.1601

Initial position of centre of glass

75 0 6 11 -28 -0.7492 0.1237

85 1 -1 11 -34 -0.389 -0.2474 50 100 150
X (pixels)
200 250 300

99 1 1 11 -32 0.1073 0.2374 Trajectory from the centre of the glass with Kalman filter.

yc) with Kalman filter

ctory of the centre of the glass X: 140.5 Final position of centre of the glass
Y: 147.8

III. E XPERIMENTAL RESULTS

X: 114.4
X: 163.2
Y: 111
Y: 107.1

We implemented the proposed algorithm in MATLAB, and Initial position of centre of the glass

50 100 150 200 250 300

tested it on five sequence video. The tests video sequences was X (pixels)

a public in Internet, which can be downloaded from [22], [21],

Fig. 5. Trajectory in pixels from the centre of the glass.
[20], [19]. This video has been adopted by many researchers
to test their algorithms, because of its capacity in simulating
various tracking conditions, including illumination changes, Table II shows the experimental results obtained with the
pose variations, occlusions, and distraction. Since ensemble tests sequences by the proposed method preprocessing of
tracking is a general framework for tracking objects, several histogram. Thus precision and execution time ratio were
object tracking algorithms (Meanshift, Camshift, Kalman fil- obtained comparing to those for the tests video sequence. The
ter), and a pre-specified methods for preprocessing (method of experimental results show that the proposed method tracks
histogram color [9], method of background subtraction [1], [9], the face and the glass more reliably and more accurately but
and method of detection skin color [18], [17]) are evaluated the car information is lost thereby leading to failure tracking.
to achieve a good tracking performance. An adaptive first I remedied the problems tracking a moving car using the
pre-processing the histogram color was tested to the video
sequences. The algorithm tracks the object as it moves from TABLE II
T HE PRECISION VALUES CALCULATED FOR THE HISTOGRAM METHOD .
one frame to the next one in approximately 0.15 second on
a 2.2GHz PC in Matlab. The first expriment is on a video
sequence of glass in hand man with 100 frames of spatial
resolution 320 × 240. The tracking target is the moving is the video frames Precision(%) execution time(s)
glass. The second experiment is on a sequence with 100 frames Redcup 100 99 24.408675
Foreman 100 96 15.938718
of spatial resolution 144 × 176. In this video, we will track a PETS 20011 100 44 118.523104
motion of face. To save space, we only show the experimental PETS 20012 100 4 145.519240
results preprocessing and tracking object in three sequence an Stforge 10 3 10.104903
by the proposed method in Figure 3, 4. As shown in Figure
5, the trajectories of the glass for the three algorithms tracking second method of background subtraction. TableIII shows
by the Camshift algorithm and the estimated trajet location of measurements of precision calculated and execution time for
the Kalman filter in the frame plane. test sequences of our tracking algorithm using as preprocessing
method of background subtraction. The Figure 6 show tracking

International Scholarly and Scientific Research & Innovation 6(4) 2012 424 ISNI:0000000091950263
 World Academy of Science, Engineering and Technology
International Journal of Electronics and Communication Engineering
Vol:6, No:4, 2012

process of the moving car obtained in this method. The third several methods preprocessing to detect the object in each
frame of the sequence, provides satisfactory results in the case
of the complex video sequence (illumination change, change
in appearance, scale, change of subject and object movement).
Image 53 Image 54 Image 55 Image 57

R EFERENCES
[1] A. Alper and J. Omar and S. Mubarak, “ Object Tracking: A Survey”.
ACM Computing Surveys, vol. 38, no. 45, pp. 1–45. 2006.
Image 58 Image 59 Image 60 Image 62 [2] D. Chen and F. Bai and B. Li and T. Wang, “ BEST: A Real-time
Tracking Method for Scout Robot”. IEEE/RSJ International Conference
Fig. 6. Tracking results of the moving car by the background substraction. on Intelligent Robots and Systems, no. 6, pp. 1001–1006. 2009.
[3] J. Wang and F. He and X. Zhang and G. Yun, “ Tracking Object throuth
Occuusions Using Improved Kalman Filter”. IEEE 2010, no. 6, pp. 223-
228. 2010.
TABLE III
[4] N.A Gmez, “ A Probabilistic Integrated Object Recognition and Tracking
P RECISION VALUES CALCULATED FOR THE METHOD OF BACKGROUND
Framework for Video Sequences”. Universitat Rovira I Virgili, PHD
SUBTRACTION AND EXECUTION TIME .
thesis, Espagne, 2009.
[5] X. Sun and H. Yao, and S. Zhang, “ A Refined Particle Filter Method for
video frames Precision (%) execution time(s) Contour Tracking”. Visual Communications and Image Processing, no. 8,
Open Science Index, Electronics and Communication Engineering Vol:6, No:4, 2012 waset.org/Publication/15036

PETS 20011 100 99 79.931015 pp. 1–8. 2010.

PETS 20012 133 88,72 93.119679 [6] D. Comaniciu, and V. Ramesh, and P. Meer , “ Kernel-Based Object
Stforge 10 10 7.476155 Tracking”. IEEE Transactions on Pattern Analysis and Machine Intelli-
gence, vol. 25 no. 14, pp. 564–577. 2003.
[7] L. Ido, and L. Michael, and R. Ehud, “ Mean Shift tracking with multiple
proposed method preprocessing is the detection of human reference color histograms”. Computer Vision and Image Understanding,
no. 9, pp. 400–408. 2010.
skin color by RGB space. This method is what determines [8] G. bradski, and R. Gary, and L.M. Olga, “ Computer Vision Face Tracking
that an image contains faces or not and if any is to return For Use in a Perceptual User Interface”. Intel Technology Journal Q2’98,
their size and position in the image. The detection is based vol. 10, no. 15, pp. 1–15. 1998.
[9] G. Bradski, and T. Ogiuchi, and M. Higashikubo, “ Visual Tracking
on the RGB color space. Detection in this case requires no Algorithm using Pixel-Pair Feature”. International Conference on Pattern
skin model and no conversion of color space, it checks if Recognition, no. 4, pp. 1808–1811. 2010.
just the triple (R, G, B) is a skin color or no-skin color. [10] Y. Ruiguo, and Z. Xinrong, “ The Design and Implementation of Face
Tracking in Real Time Multimedia Recording System”. IEEE Transaction,
The proposed method successfully tracks the face over the no. 3, pp. 1–3. 2009.
whole sequence of Forman. The precision value and the time [11] E. David, and B. Erich, and K. Daniel, and S. Anselm, “ Fast and Robust
execution are respectively 97,05%, 17.021 secondes. Table Camshift Tracking”. IEEE Transaction, no. 8, pp. 1–8. 2010.
[12] S. Avidan, “ Support vector tracking”. In IEEE Conference on Computer
IV lists the execution time corresponding to the different Vision and Pattern Recognition, no. 8, pp.184–191. 2001.
preprocessing used with our algorithm object tracking for all [13] S. Franois, B. and R.J. Alexandre, “ Camshift Tracker Design Experi-
video sequences analyzed. We note that the two pre-processing ments”. IMSC, no. 11, pp. 1–11. 2004.
[14] L. Zhang, and D. Zhang, “ Robust Object Tracking Using Joint Color
TABLE IV Texture Histogram”. International Journal of Pattern Recognition and
T HE EXECUTION TIME FOR DIFFERENT METHOD PRE - PROCESSING . Artificial Intelligence, vol. 23 no. 19, pp. 1245–1263. 2009.
[15] X. Song, and R. Nevatia, “ Camshift Tracker Design Experiments with
Intel OpenCV and Sai”. International Conference on Pattern Recogni-
tion,IEEE Computer Society, no. 4, pp. 1–4. 2004.
[16] S. Afzal, and N. Vaswani, and A. Tannenbaum ,A. and Yezzi, “ Object
Sequence frames Histogram Background skin color Tracking: A Survey”. IEEE, no. 6, pp. 4244–4249. 2009.
Video [17] A. Iyad, and H. Mahmoud, “ Smart Human Face Detection System”.
execution time time time International Journal of Computers, vol. 5, no. 8, pp. 210–221. 2011.
Foreman 100 18.558961 (s) × 17.021(s) [18] C. Abbas, and C. Joan, and C. Kevin, and M.K. Paul “ A Skin Tone
Redcup 100 21.113581 (s) × × Detection Algorithm for an Adaptive Approach to Steganography”. , no.
PETS 20011 100 118.523104(s) 79.931015(s) × 23, pp. 1–23. 2011.
PETS 20012 133 145.519240 (s) 93.119679(s) × [19] http://developer.com/technology/itj.
[20] http://ftp.pets.rdg.ac.uk/PETS2001/DATASET1/TESTING/.
[21] http://developer.com/technology/itj.
methods (background subtraction and detection of skin color) [22] http://ftp.pets.rdg.ac.uk/PETS2001/DATASET1/TESTING/.
are the fastest and the calculation of a histogram is the slowest.

IV. C ONCLUSION
This paper focuses on simultaneous tracking of multiple
object (a human,a face, a hand, a glass, a car). During the test
sequences generated with different methods of pre-processing,
we can conclude that the tracking of objects differs from one
object to another and several parameters can affect the results
of tracking. Experimental results show that our algorithm
(Camshift and the Kalman filter) is superior in terms of
precision, reliability and execution time, compared the various
methods presented in the literature. In particular, the use of

International Scholarly and Scientific Research & Innovation 6(4) 2012 425 ISNI:0000000091950263
 World Academy of Science, Engineering and Technology
International Journal of Electronics and Communication Engineering
Vol:6, No:4, 2012
SALHI Afef was born in Sfax, Tunisia in 1985. He received a degree
in electrical engineering and Master degree in Control and
Computer Engineer- ing from National School of Engineers of Sfax in
2009 and 2011. He is currently pursuing a thesis at the same university
in Laboratory of Electronics and Information Technologies (L.E.T.I). His
main research interests include Codesign software Hard- ware high level
of MPSoCs (FPGA, DSP, GPP) for real time application, image and
video processing,multimedia application, medical application, system
tracking.was born in Sfax, Tunisia in 1985. He received a degree in electrical
engineering and Master degree in Control and Computer Engineering from
National School of Engineers of Sfax in 2009 and 2011. He is currently
pursuing a thesis at the same university in Laboratory of Electronics and
Information Technologies (L.E.T.I). His main research interests include
Code- sign software Hardware high level of MPSoCs (FPGA, DSP, GPP)
for real time application, image and video processing, multimedia
application, medical application, system tracking.
Open Science Index, Electronics and Communication Engineering Vol:6, No:4, 2012 waset.org/Publication/15036

International Scholarly and Scientific Research & Innovation 6(4) 2012 426 ISNI:0000000091950263