Abstract: Human perceives color in categories, which may be
identified using color name such as red, blue, etc. The categorization
is unique for each human being. However despite the individual
differences, the categorization is shared among members in society.
This allows communication among them, especially when using
color name. Sociable robot, to live coexist with human and become
part of human society, must also have the shared color
categorization, which can be achieved through learning. Many
works have been done to enable computer, as brain of robot, to learn
color categorization. Most of them rely on modeling of human color
perception and mathematical complexities. Differently, in this work,
the computer learns color categorization through interaction with
humans. This work aims at developing the innate ability of the
computer to learn the human-like color categorization. It focuses on
the representation of color categorization and how it is built and
developed without much mathematical complexity.
Abstract: This paper describes a segmentation algorithm based
on the cooperation of an optical flow estimation method with edge
detection and region growing procedures.
The proposed method has been developed as a pre-processing
stage to be used in methodologies and tools for video/image indexing
and retrieval by content. The addressed problem consists in
extracting whole objects from background for producing images of
single complete objects from videos or photos. The extracted images
are used for calculating the object visual features necessary for both
indexing and retrieval processes.
The first task of the algorithm exploits the cues from motion
analysis for moving area detection. Objects and background are then
refined using respectively edge detection and region growing
procedures. These tasks are iteratively performed until objects and
background are completely resolved.
The developed method has been applied to a variety of indoor and
outdoor scenes where objects of different type and shape are
represented on variously textured background.
Abstract: We present a non standard Euclidean vehicle
routing problem adding a level of clustering, and we revisit the use
of self-organizing maps as a tool which naturally handles such
problems. We present how they can be used as a main operator
into an evolutionary algorithm to address two conflicting
objectives of route length and distance from customers to bus stops
minimization and to deal with capacity constraints. We apply the
approach to a real-life case of combined clustering and vehicle
routing for the transportation of the 780 employees of an
enterprise. Basing upon a geographic information system we
discuss the influence of road infrastructures on the solutions
generated.
Abstract: This work presents a recursive identification algorithm. This algorithm relates to the identification of closed loop system with Variable Structure Controller. The approach suggested includes two stages. In the first stage a genetic algorithm is used to obtain the parameters of switching function which gives a control signal rich in commutations (i.e. a control signal whose spectral characteristics are closest possible to those of a white noise signal). The second stage consists in the identification of the system parameters by the instrumental variable method and using the optimal switching function parameters obtained with the genetic algorithm. In order to test the validity of this algorithm a simulation example is presented.
Abstract: CEMTool is a command style design and analyzing
package for scientific and technological algorithm and a matrix based
computation language. In this paper, we present new 2D & 3D
finite element method (FEM) packages for CEMTool. We discuss
the detailed structures and the important features of pre-processor,
solver, and post-processor of CEMTool 2D & 3D FEM packages. In
contrast to the existing MATLAB PDE Toolbox, our proposed FEM
packages can deal with the combination of the reserved words. Also,
we can control the mesh in a very effective way. With the introduction
of new mesh generation algorithm and fast solving technique, our
FEM packages can guarantee the shorter computational time than
MATLAB PDE Toolbox. Consequently, with our new FEM packages,
we can overcome some disadvantages or limitations of the existing
MATLAB PDE Toolbox.
Abstract: Image fusion aims to enhance the perception
of a scene by combining important information captured by
different sensors. Dual-Tree Complex Wavelet (DT-CWT) has been
thouroughly investigated for image fusion, since it takes advantages
of approximate shift invariance and direction selectivity. But it can
only handle limited direction information. To allow a more flexible
directional expansion for images, we propose a novel fusion scheme,
referred to as complex contourlet transform (CCT). It successfully
incorporates directional filter banks (DFB) into DT-CWT. As a result
it efficiently deal with images containing contours and textures,
whereas it retains the property of shift invariance. Experimental
results demonstrated that the method features high quality fusion
performance and can facilitate many image processing applications.
Abstract: In this contribution an innovative platform is being
presented that integrates intelligent agents and evolutionary
computation techniques in legacy e-learning environments. It
introduces the design and development of a scalable and
interoperable integration platform supporting:
I) various assessment agents for e-learning environments,
II) a specific resource retrieval agent for the provision of
additional information from Internet sources matching the
needs and profile of the specific user and
III) a genetic algorithm designed to extract efficient information
(classifying rules) based on the students- answering input
data.
The agents are implemented in order to provide intelligent
assessment services based on computational intelligence techniques
such as Bayesian Networks and Genetic Algorithms.
The proposed Genetic Algorithm (GA) is used in order to extract
efficient information (classifying rules) based on the students-
answering input data. The idea of using a GA in order to fulfil this
difficult task came from the fact that GAs have been widely used in
applications including classification of unknown data.
The utilization of new and emerging technologies like web
services allows integrating the provided services to any web based
legacy e-learning environment.
Abstract: The Internet telephony employs a new type of Internet communication on which a mutual communication is realized by establishing sessions. Session Initiation Protocol (SIP) is used to establish sessions between end-users. For unreliable transmission (UDP), SIP message should be retransmitted when it is lost. The retransmissions increase a load of the SIP signaling network, and sometimes lead to performance degradation when a network is overloaded. The paper proposes an overload control for a SIP signaling network to protect from a performance degradation. Introducing two thresholds in a queue of a SIP proxy server, the SIP proxy server detects a congestion. Once congestion is detected, a SIP signaling network restricts to make new calls. The proposed overload control is evaluated using the network simulator (ns-2). With simulation results, the paper shows the proposed overload control works well.
Abstract: The anti-lock braking systems installed on vehicles
for safe and effective braking, are high-order nonlinear and timevariant.
Using fuzzy logic controllers increase efficiency of such
systems, but impose a high computational complexity as well. The
main concept introduced by this paper is reducing computational
complexity of fuzzy controllers by deploying problem-solution data
structure. Unlike conventional methods that are based on
calculations, this approach is based on data oriented modeling.
Abstract: MultiProtocol Label Switching (MPLS) is an
emerging technology that aims to address many of the existing issues
associated with packet forwarding in today-s Internetworking
environment. It provides a method of forwarding packets at a high
rate of speed by combining the speed and performance of Layer 2
with the scalability and IP intelligence of Layer 3. In a traditional IP
(Internet Protocol) routing network, a router analyzes the destination
IP address contained in the packet header. The router independently
determines the next hop for the packet using the destination IP
address and the interior gateway protocol. This process is repeated at
each hop to deliver the packet to its final destination. In contrast, in
the MPLS forwarding paradigm routers on the edge of the network
(label edge routers) attach labels to packets based on the forwarding
Equivalence class (FEC). Packets are then forwarded through the
MPLS domain, based on their associated FECs , through swapping
the labels by routers in the core of the network called label switch
routers. The act of simply swapping the label instead of referencing
the IP header of the packet in the routing table at each hop provides
a more efficient manner of forwarding packets, which in turn allows
the opportunity for traffic to be forwarded at tremendous speeds and
to have granular control over the path taken by a packet. This paper
deals with the process of MPLS forwarding mechanism,
implementation of MPLS datapath , and test results showing the
performance comparison of MPLS and IP routing. The discussion
will focus primarily on MPLS IP packet networks – by far the
most common application of MPLS today.
Abstract: Lossless compression schemes with secure
transmission play a key role in telemedicine applications that helps in
accurate diagnosis and research. Traditional cryptographic algorithms
for data security are not fast enough to process vast amount of data.
Hence a novel Secured lossless compression approach proposed in
this paper is based on reversible integer wavelet transform, EZW
algorithm, new modified runlength coding for character
representation and selective bit scrambling. The use of the lifting
scheme allows generating truly lossless integer-to-integer wavelet
transforms. Images are compressed/decompressed by well-known
EZW algorithm. The proposed modified runlength coding greatly
improves the compression performance and also increases the
security level. This work employs scrambling method which is fast,
simple to implement and it provides security. Lossless compression
ratios and distortion performance of this proposed method are found
to be better than other lossless techniques.
Abstract: In this paper, a framework is presented trying to make
the most secure web system out of the available generic and web
security technology which can be used as a guideline for
organizations building their web sites. The framework is designed to
provide necessary security services, to address the known security
threats, and to provide some cover to other security problems
especially unknown threats. The requirements for the design are
discussed which guided us to the design of secure web system. The
designed security framework is then simulated and various quality of
service (QoS) metrics are calculated to measure the performance of
this system.
Abstract: The mobile users with Laptops need to have an
efficient access to i.e. their home personal data or to the Internet from
any place in the world, regardless of their location or point of
attachment, especially while roaming outside the home subnet. An
efficient interpretation of packet losses problem that is encountered
from this roaming is to the centric of all aspects in this work, to be
over-highlighted. The main previous works, such as BER-systems,
Amigos, and ns-2 implementation that are considered to be in
conjunction with that problem under study are reviewed and
discussed. Their drawbacks and limitations, of stopping only at
monitoring, and not to provide an actual solution for eliminating or
even restricting these losses, are mentioned. Besides that, the
framework around which we built a Triple-R sequence as a costeffective
solution to eliminate the packet losses and bridge the gap
between subnets, an area that until now has been largely neglected, is
presented. The results show that, in addition to the high bit error rate
of wireless mobile networks, mainly the low efficiency of mobile-IP
registration procedure is a direct cause of these packet losses.
Furthermore, the output of packet losses interpretation resulted an
illustrated triangle of the registration process. This triangle should be
further researched and analyzed in our future work.
Abstract: Among neural models the Support Vector Machine
(SVM) solutions are attracting increasing attention, mostly because
they eliminate certain crucial questions involved by neural network
construction. The main drawback of standard SVM is its high
computational complexity, therefore recently a new technique, the
Least Squares SVM (LS–SVM) has been introduced. In this paper we
present an extended view of the Least Squares Support Vector
Regression (LS–SVR), which enables us to develop new
formulations and algorithms to this regression technique. Based on
manipulating the linear equation set -which embodies all information
about the regression in the learning process- some new methods are
introduced to simplify the formulations, speed up the calculations
and/or provide better results.
Abstract: In this paper two models using a functional network
were employed to solving classification problem. Functional networks
are generalized neural networks, which permit the specification of
their initial topology using knowledge about the problem at hand. In
this case, and after analyzing the available data and their relations, we
systematically discuss a numerical analysis method used for
functional network, and apply two functional network models to
solving XOR problem. The XOR problem that cannot be solved with
two-layered neural network can be solved by two-layered functional
network, which reveals a potent computational power of functional
networks, and the performance of the proposed model was validated
using classification problems.
Abstract: With the advance of information technology in the
new era the applications of Internet to access data resources has
steadily increased and huge amount of data have become accessible
in various forms. Obviously, the network providers and agencies,
look after to prevent electronic attacks that may be harmful or may
be related to terrorist applications. Thus, these have facilitated the
authorities to under take a variety of methods to protect the special
regions from harmful data. One of the most important approaches is
to use firewall in the network facilities. The main objectives of
firewalls are to stop the transfer of suspicious packets in several
ways. However because of its blind packet stopping, high process
power requirements and expensive prices some of the providers are
reluctant to use the firewall. In this paper we proposed a method to
find a discriminate function to distinguish between usual packets and
harmful ones by the statistical processing on the network router logs.
By discriminating these data, an administrator may take an approach
action against the user. This method is very fast and can be used
simply in adjacent with the Internet routers.
Abstract: Ensemble learning algorithms such as AdaBoost and
Bagging have been in active research and shown improvements in
classification results for several benchmarking data sets with mainly
decision trees as their base classifiers. In this paper we experiment to
apply these Meta learning techniques with classifiers such as random
forests, neural networks and support vector machines. The data sets
are from MAGIC, a Cherenkov telescope experiment. The task is to
classify gamma signals from overwhelmingly hadron and muon
signals representing a rare class classification problem. We compare
the individual classifiers with their ensemble counterparts and
discuss the results. WEKA a wonderful tool for machine learning has
been used for making the experiments.
Abstract: Support Vector Machine (SVM) is a statistical
learning tool that was initially developed by Vapnik in 1979 and later
developed to a more complex concept of structural risk minimization
(SRM). SVM is playing an increasing role in applications to
detection problems in various engineering problems, notably in
statistical signal processing, pattern recognition, image analysis, and
communication systems. In this paper, SVM was applied to the
detection of SAR (synthetic aperture radar) images in the presence of
partially developed speckle noise. The simulation was done for single
look and multi-look speckle models to give a complete overlook and
insight to the new proposed model of the SVM-based detector. The
structure of the SVM was derived and applied to real SAR images
and its performance in terms of the mean square error (MSE) metric
was calculated. We showed that the SVM-detected SAR images have
a very low MSE and are of good quality. The quality of the
processed speckled images improved for the multi-look model.
Furthermore, the contrast of the SVM detected images was higher
than that of the original non-noisy images, indicating that the SVM
approach increased the distance between the pixel reflectivity levels
(the detection hypotheses) in the original images.
Abstract: This paper presents a new color face image database
for benchmarking of automatic face detection algorithms and human
skin segmentation techniques. It is named the VT-AAST image
database, and is divided into four parts. Part one is a set of 286 color
photographs that include a total of 1027 faces in the original format
given by our digital cameras, offering a wide range of difference in
orientation, pose, environment, illumination, facial expression and
race. Part two contains the same set in a different file format. The
third part is a set of corresponding image files that contain human
colored skin regions resulting from a manual segmentation
procedure. The fourth part of the database has the same regions
converted into grayscale. The database is available on-line for
noncommercial use. In this paper, descriptions of the database
development, organization, format as well as information needed for
benchmarking of algorithms are depicted in detail.
Abstract: In this paper a novel scheme for watermarking digital
audio during its compression to MPEG-1 Layer III format is
proposed. For this purpose we slightly modify some of the selected
MDCT coefficients, which are used during MPEG audio
compression procedure. Due to the possibility of modifying different
MDCT coefficients, there will be different choices for embedding the
watermark into audio data, considering robustness and transparency
factors. Our proposed method uses a genetic algorithm to select the
best coefficients to embed the watermark. This genetic selection is
done according to the parameters that are extracted from the
perceptual content of the audio to optimize the robustness and
transparency of the watermark. On the other hand the watermark
security is increased due to the random nature of the genetic
selection. The information of the selected MDCT coefficients that
carry the watermark bits, are saves in a database for future extraction
of the watermark. The proposed method is suitable for online MP3
stores to pursue illegal copies of musical artworks. Experimental
results show that the detection ratio of the watermarks at the bitrate
of 128kbps remains above 90% while the inaudibility of the
watermark is preserved.