Abstract: Skin color can provide a useful and robust cue
for human-related image analysis, such as face detection,
pornographic image filtering, hand detection and tracking,
people retrieval in databases and Internet, etc. The major
problem of such kinds of skin color detection algorithms is
that it is time consuming and hence cannot be applied to a real
time system. To overcome this problem, we introduce a new
fast technique for skin detection which can be applied in a real
time system. In this technique, instead of testing each image
pixel to label it as skin or non-skin (as in classic techniques),
we skip a set of pixels. The reason of the skipping process is
the high probability that neighbors of the skin color pixels are
also skin pixels, especially in adult images and vise versa. The
proposed method can rapidly detect skin and non-skin color
pixels, which in turn dramatically reduce the CPU time
required for the protection process. Since many fast detection
techniques are based on image resizing, we apply our
proposed pixel skipping technique with image resizing to
obtain better results. The performance evaluation of the
proposed skipping and hybrid techniques in terms of the
measured CPU time is presented. Experimental results
demonstrate that the proposed methods achieve better result
than the relevant classic method.
Abstract: The automatic classification of non stationary signals is an important practical goal in several domains. An essential classification task is to allocate the incoming signal to a group associated with the kind of physical phenomena producing it. In this paper, we present a modular system composed by three blocs: 1) Representation, 2) Dimensionality reduction and 3) Classification. The originality of our work consists in the use of a new wavelet called "Ben wavelet" in the representation stage. For the dimensionality reduction, we propose a new algorithm based on the random projection and the principal component analysis.
Abstract: The volume of XML data exchange is explosively
increasing, and the need for efficient mechanisms of XML data
management is vital. Many XML storage models have been proposed
for storing XML DTD-independent documents in relational database
systems. Benchmarking is the best way to highlight pros and cons of
different approaches. In this study, we use a common benchmarking
scheme, known as XMark to compare the most cited and newly
proposed DTD-independent methods in terms of logical reads,
physical I/O, CPU time and duration. We show the effect of Label
Path, extracting values and storing in another table and type of join
needed for each method-s query answering.
Abstract: The goal of steganography is to avoid drawing
suspicion to the transmission of a hidden message. If suspicion is
raised, steganography may fail. The success of steganography
depends on the secrecy of the action. If steganography is detected,
the system will fail but data security depends on the robustness of the
applied algorithm. In this paper, we propose a novel plausible
deniability scheme in steganography by using a diversionary message
and encrypt it with a DES-based algorithm. Then, we compress the
secret message and encrypt it by the receiver-s public key along with
the stego key and embed both messages in a carrier using an
embedding algorithm. It will be demonstrated how this method can
support plausible deniability and is robust against steganalysis.
Abstract: Mining frequent tree patterns have many useful
applications in XML mining, bioinformatics, network routing, etc.
Most of the frequent subtree mining algorithms (i.e. FREQT,
TreeMiner and CMTreeMiner) use anti-monotone property in the
phase of candidate subtree generation. However, none of these
algorithms have verified the correctness of this property in tree
structured data. In this research it is shown that anti-monotonicity
does not generally hold, when using weighed support in tree pattern
discovery. As a result, tree mining algorithms that are based on this
property would probably miss some of the valid frequent subtree
patterns in a collection of trees. In this paper, we investigate the
correctness of anti-monotone property for the problem of weighted
frequent subtree mining. In addition we propose W3-Miner, a new
algorithm for full extraction of frequent subtrees. The experimental
results confirm that W3-Miner finds some frequent subtrees that the
previously proposed algorithms are not able to discover.
Abstract: Due to new distributed database applications such as
huge deductive database systems, the search complexity is constantly
increasing and we need better algorithms to speedup traditional
relational database queries. An optimal dynamic programming
method for such high dimensional queries has the big disadvantage of
its exponential order and thus we are interested in semi-optimal but
faster approaches. In this work we present a multi-agent based
mechanism to meet this demand and also compare the result with
some commonly used query optimization algorithms.
Abstract: This paper proposes an easy-to-use instruction hiding
method to protect software from malicious reverse engineering
attacks. Given a source program (original) to be protected, the
proposed method (1) takes its modified version (fake) as an input,
(2) differences in assembly code instructions between original and
fake are analyzed, and, (3) self-modification routines are introduced
so that fake instructions become correct (i.e., original instructions)
before they are executed and that they go back to fake ones after
they are executed. The proposed method can add a certain amount
of security to a program since the fake instructions in the resultant
program confuse attackers and it requires significant effort to discover
and remove all the fake instructions and self-modification routines.
Also, this method is easy to use (with little effort) because all a user
(who uses the proposed method) has to do is to prepare a fake source
code by modifying the original source code.
Abstract: Design Patterns have gained more and more
acceptances since their emerging in software development world last
decade and become another de facto standard of essential knowledge
for Object-Oriented Programming developers nowadays.
Their target usage, from the beginning, was for regular computers,
so, minimizing power consumption had never been a concern.
However, in this decade, demands of more complicated software for
running on mobile devices has grown rapidly as the much higher
performance portable gadgets have been supplied to the market
continuously. To get along with time to market that is business
reason, the section of software development for power conscious,
battery, devices has shifted itself from using specific low-level
languages to higher level ones. Currently, complicated software
running on mobile devices are often developed by high level
languages those support OOP concepts. These cause the trend of
embracing Design Patterns to mobile world.
However, using Design Patterns directly in software development
for power conscious systems is not recommended because they were
not originally designed for such environment. This paper
demonstrates the adapted Design Pattern for power limitation system.
Because there are numerous original design patterns, it is not possible
to mention the whole at once. So, this paper focuses only in creating
Energy Conscious version of existing regular "Builder Pattern" to be
appropriated for developing low power consumption software.
Abstract: In this paper, we present a novel approach to accurately
detect text regions including shop name in signboard images with
complex background for mobile system applications. The proposed
method is based on the combination of text detection using edge
profile and region segmentation using fuzzy c-means method. In the
first step, we perform an elaborate canny edge operator to extract all
possible object edges. Then, edge profile analysis with vertical and
horizontal direction is performed on these edge pixels to detect
potential text region existing shop name in a signboard. The edge
profile and geometrical characteristics of each object contour are
carefully examined to construct candidate text regions and classify the
main text region from background. Finally, the fuzzy c-means
algorithm is performed to segment and detected binarize text region.
Experimental results show that our proposed method is robust in text
detection with respect to different character size and color and can
provide reliable text binarization result.
Abstract: Ontologies and tagging systems are two different ways to organize the knowledge present in the current Web. In this paper we propose a simple method to model folksonomies, as tagging systems, with ontologies. We show the scalability of the method using real data sets. The modeling method is composed of a generic ontology that represents any folksonomy and an algorithm to transform the information contained in folksonomies to the generic ontology. The method allows representing folksonomies at any instant of time.
Abstract: In this paper, a comparative study of application of
supervised and unsupervised learning algorithms on illumination
invariant face recognition has been carried out. The supervised
learning has been carried out with the help of using a bi-layered
artificial neural network having one input, two hidden and one output
layer. The gradient descent with momentum and adaptive learning
rate back propagation learning algorithm has been used to implement
the supervised learning in a way that both the inputs and
corresponding outputs are provided at the time of training the
network, thus here is an inherent clustering and optimized learning of
weights which provide us with efficient results.. The unsupervised
learning has been implemented with the help of a modified
Counterpropagation network. The Counterpropagation network
involves the process of clustering followed by application of Outstar
rule to obtain the recognized face. The face recognition system has
been developed for recognizing faces which have varying
illumination intensities, where the database images vary in lighting
with respect to angle of illumination with horizontal and vertical
planes. The supervised and unsupervised learning algorithms have
been implemented and have been tested exhaustively, with and
without application of histogram equalization to get efficient results.
Abstract: Network on a chip (NoC) has been proposed as a viable solution to counter the inefficiency of buses in the current VLSI on-chip interconnects. However, as the silicon chip accommodates more transistors, the probability of transient faults is increasing, making fault tolerance a key concern in scaling chips. In packet based communication on a chip, transient failures can corrupt the data packet and hence, undermine the accuracy of data communication. In this paper, we present a comparative analysis of transient fault tolerant techniques including end-to-end, node-by-node, and stochastic communication based on flooding principle.
Abstract: RoboCup Rescue simulation as a large-scale Multi
agent system (MAS) is one of the challenging environments for
keeping coordination between agents to achieve the objectives
despite sensing and communication limitations. The dynamicity of
the environment and intensive dependency between actions of
different kinds of agents make the problem more complex. This point
encouraged us to use learning-based methods to adapt our decision
making to different situations. Our approach is utilizing
reinforcement leaning. Using learning in rescue simulation is one of
the current ways which has been the subject of several researches in
recent years. In this paper we present an innovative learning method
implemented for Police Force (PF) Agent. This method can cope
with the main difficulties that exist in other learning approaches.
Different methods used in the literature have been examined. Their
drawbacks and possible improvements have led us to the method
proposed in this paper which is fast and accurate. The Brain
Emotional Learning Based Intelligent Controller (BELBIC) is our
solution for learning in this environment. BELBIC is a
physiologically motivated approach based on a computational model
of amygdale and limbic system. The paper presents the results
obtained by the proposed approach, showing the power of BELBIC
as a decision making tool in complex and dynamic situation.
Abstract: In this paper, a novel scheme is proposed for Ownership Identification and Color Image Authentication by deploying Cryptography & Digital Watermarking. The color image is first transformed from RGB to YST color space exclusively designed for watermarking. Followed by color space transformation, each channel is divided into 4×4 non-overlapping blocks with selection of central 2×2 sub-blocks. Depending upon the channel selected two to three LSBs of each central 2×2 sub-block are set to zero to hold the ownership, authentication and recovery information. The size & position of sub-block is important for correct localization, enhanced security & fast computation. As YS ÔèÑ T so it is suitable to embed the recovery information apart from the ownership and authentication information, therefore 4×4 block of T channel along with ownership information is then deployed by SHA160 to compute the content based hash that is unique and invulnerable to birthday attack or hash collision instead of using MD5 that may raise the condition i.e. H(m)=H(m'). For recovery, intensity mean of 4x4 block of each channel is computed and encoded upto eight bits. For watermark embedding, key based mapping of blocks is performed using 2DTorus Automorphism. Our scheme is oblivious, generates highly imperceptible images with correct localization of tampering within reasonable time and has the ability to recover the original work with probability of near one.
Abstract: A low bit rate still image compression scheme by
compressing the indices of Vector Quantization (VQ) and generating
residual codebook is proposed. The indices of VQ are compressed by
exploiting correlation among image blocks, which reduces the bit per
index. A residual codebook similar to VQ codebook is generated that
represents the distortion produced in VQ. Using this residual
codebook the distortion in the reconstructed image is removed,
thereby increasing the image quality. Our scheme combines these two
methods. Experimental results on standard image Lena show that our
scheme can give a reconstructed image with a PSNR value of 31.6 db
at 0.396 bits per pixel. Our scheme is also faster than the existing VQ
variants.
Abstract: The aim of this paper is to rank the impact of Object
Oriented(OO) metrics in fault prediction modeling using Artificial
Neural Networks(ANNs). Past studies on empirical validation of
object oriented metrics as fault predictors using ANNs have focused
on the predictive quality of neural networks versus standard
statistical techniques. In this empirical study we turn our attention to
the capability of ANNs in ranking the impact of these explanatory
metrics on fault proneness. In ANNs data analysis approach, there is
no clear method of ranking the impact of individual metrics. Five
ANN based techniques are studied which rank object oriented
metrics in predicting fault proneness of classes. These techniques are
i) overall connection weights method ii) Garson-s method iii) The
partial derivatives methods iv) The Input Perturb method v) the
classical stepwise methods. We develop and evaluate different
prediction models based on the ranking of the metrics by the
individual techniques. The models based on overall connection
weights and partial derivatives methods have been found to be most
accurate.
Abstract: In this paper we have proposed three and two
stage still gray scale image compressor based on BTC. In our
schemes, we have employed a combination of four techniques
to reduce the bit rate. They are quad tree segmentation, bit
plane omission, bit plane coding using 32 visual patterns and
interpolative bit plane coding. The experimental results show
that the proposed schemes achieve an average bit rate of 0.46
bits per pixel (bpp) for standard gray scale images with an
average PSNR value of 30.25, which is better than the results
from the exiting similar methods based on BTC.
Abstract: In this paper, a novel scheme is proposed for ownership identification and authentication using color images by deploying Cryptography and Digital Watermarking as underlaying technologies. The former is used to compute the contents based hash and the latter to embed the watermark. The host image that will claim to be the rightful owner is first transformed from RGB to YST color space exclusively designed for watermarking based applications. Geometrically YS ÔèÑ T and T channel corresponds to the chrominance component of color image, therefore suitable for embedding the watermark. The T channel is divided into 4×4 nonoverlapping blocks. The size of block is important for enhanced localization, security and low computation. Each block along with ownership information is then deployed by SHA160, a one way hash function to compute the content based hash, which is always unique and resistant against birthday attack instead of using MD5 that may raise the condition i.e. H(m)=H(m'). The watermark payload varies from block to block and computed by the variance factorα . The quality of watermarked images is quite high both subjectively and objectively. Our scheme is blind, computationally fast and exactly locates the tampered region.
Abstract: Recently, grid computing has been widely focused on
the science, industry, and business fields, which are required a vast
amount of computing. Grid computing is to provide the environment
that many nodes (i.e., many computers) are connected with each
other through a local/global network and it is available for many
users. In the environment, to achieve data processing among nodes
for any applications, each node executes mutual authentication by
using certificates which published from the Certificate Authority
(for short, CA). However, if a failure or fault has occurred in the
CA, any new certificates cannot be published from the CA. As
a result, a new node cannot participate in the gird environment.
In this paper, an off-the-shelf scheme for dependable grid systems
using virtualization techniques is proposed and its implementation is
verified. The proposed approach using the virtualization techniques
is to restart an application, e.g., the CA, if it has failed. The system
can tolerate a failure or fault if it has occurred in the CA. Since
the proposed scheme is implemented at the application level easily,
the cost of its implementation by the system builder hardly takes
compared it with other methods. Simulation results show that the
CA in the system can recover from its failure or fault.
Abstract: We study how the outcome of evolutionary dynamics on
graphs depends on a randomness on the graph structure. We gradually
change the underlying graph from completely regular (e.g. a square lattice) to completely random. We find that the fixation probability increases as the randomness increases; nevertheless, the increase is
not significant and thus the fixation probability could be estimated by the known formulas for underlying regular graphs.