Abstract: Recent scientific investigations indicate that
multimodal biometrics overcome the technical limitations of
unimodal biometrics, making them ideally suited for everyday life
applications that require a reliable authentication system. However,
for a successful adoption of multimodal biometrics, such systems
would require large heterogeneous datasets with complex multimodal
fusion and privacy schemes spanning various distributed
environments. From experimental investigations of current
multimodal systems, this paper reports the various issues related to
speed, error-recovery and privacy that impede the diffusion of such
systems in real-life. This calls for a robust mechanism that caters to
the desired real-time performance, robust fusion schemes,
interoperability and adaptable privacy policies.
The main objective of this paper is to present a framework that
addresses the abovementioned issues by leveraging on the
heterogeneous resource sharing capacities of Grid services and the
efficient machine learning capabilities of artificial neural networks
(ANN). Hence, this paper proposes a Grid-based neural network
framework for adopting multimodal biometrics with the view of
overcoming the barriers of performance, privacy and risk issues that
are associated with shared heterogeneous multimodal data centres.
The framework combines the concept of Grid services for reliable
brokering and privacy policy management of shared biometric
resources along with a momentum back propagation ANN (MBPANN)
model of machine learning for efficient multimodal fusion and
authentication schemes. Real-life applications would be able to adopt
the proposed framework to cater to the varying business requirements
and user privacies for a successful diffusion of multimodal
biometrics in various day-to-day transactions.
Abstract: Problem solving has traditionally been one of the principal research areas for artificial intelligence. Yet, although artificial intelligence reasoning techniques have been employed in several product support systems, the benefit of integrating product support, knowledge engineering, and problem solving, is still unclear. This paper studies the synergy of these areas and proposes a knowledge engineering framework that integrates product support systems and artificial intelligence techniques. The framework includes four spaces; the data, problem, hypothesis, and solution ones. The data space incorporates the knowledge needed for structured reasoning to take place, the problem space contains representations of problems, and the hypothesis space utilizes a multimodal reasoning approach to produce appropriate solutions in the form of virtual documents. The solution space is used as the gateway between the system and the user. The proposed framework enables the development of product support systems in terms of smaller, more manageable steps while the combination of different reasoning techniques provides a way to overcome the lack of documentation resources.
Abstract: Detection of human emotions has many potential applications. One of application is to quantify attentiveness audience in order evaluate acoustic quality in concern hall. The subjective audio preference that based on from audience is used. To obtain fairness evaluation of acoustic quality, the research proposed system for multimodal emotion detection; one modality based on brain signals that measured using electroencephalogram (EEG) and the second modality is sequences of facial images. In the experiment, an audio signal was customized which consist of normal and disorder sounds. Furthermore, an audio signal was played in order to stimulate positive/negative emotion feedback of volunteers. EEG signal from temporal lobes, i.e. T3 and T4 was used to measured brain response and sequence of facial image was used to monitoring facial expression during volunteer hearing audio signal. On EEG signal, feature was extracted from change information in brain wave, particularly in alpha and beta wave. Feature of facial expression was extracted based on analysis of motion images. We implement an advance optical flow method to detect the most active facial muscle form normal to other emotion expression that represented in vector flow maps. The reduce problem on detection of emotion state, vector flow maps are transformed into compass mapping that represents major directions and velocities of facial movement. The results showed that the power of beta wave is increasing when disorder sound stimulation was given, however for each volunteer was giving different emotion feedback. Based on features derived from facial face images, an optical flow compass mapping was promising to use as additional information to make decision about emotion feedback.
Abstract: This paper presents an optimal design of linear phase
digital high pass finite impulse response (FIR) filter using Improved
Particle Swarm Optimization (IPSO). In the design process, the filter
length, pass band and stop band frequencies, feasible pass band and
stop band ripple sizes are specified. FIR filter design is a multi-modal
optimization problem. An iterative method is introduced to find the
optimal solution of FIR filter design problem. Evolutionary
algorithms like real code genetic algorithm (RGA), particle swarm
optimization (PSO), improved particle swarm optimization (IPSO)
have been used in this work for the design of linear phase high pass
FIR filter. IPSO is an improved PSO that proposes a new definition
for the velocity vector and swarm updating and hence the solution
quality is improved. A comparison of simulation results reveals the
optimization efficacy of the algorithm over the prevailing
optimization techniques for the solution of the multimodal, nondifferentiable,
highly non-linear, and constrained FIR filter design
problems.
Abstract: In this paper, we propose novel algorithmic models
based on information fusion and feature transformation in crossmodal
subspace for different types of residue features extracted from
several intra-frame and inter-frame pixel sub-blocks in video
sequences for detecting digital video tampering or forgery. An
evaluation of proposed residue features – the noise residue features
and the quantization features, their transformation in cross-modal
subspace, and their multimodal fusion, for emulated copy-move
tamper scenario shows a significant improvement in tamper detection
accuracy as compared to single mode features without transformation
in cross-modal subspace.
Abstract: The Artificial immune systems algorithms are Meta
heuristic optimization method, which are used for clustering and
pattern recognition applications are abundantly. These algorithms in
multimodal optimization problems are more efficient than genetic
algorithms. A major drawback in these algorithms is their slow
convergence to global optimum and their weak stability can be
considered in various running of these algorithms. In this paper,
improved Artificial Immune System Algorithm is introduced for the
first time to overcome its problems of artificial immune system. That
use of the small size of a local search around the memory antibodies
is used for improving the algorithm efficiently. The credibility of the
proposed approach is evaluated by simulations, and it is shown that
the proposed approach achieves better results can be achieved
compared to the standard artificial immune system algorithms
Abstract: In this paper, we present an approach for soccer video
edition using a multimodal annotation. We propose to associate with
each video sequence of a soccer match a textual document to be used
for further exploitation like search, browsing and abstract edition.
The textual document contains video meta data, match meta data, and
match data. This document, generated automatically while the video
is analyzed, segmented and classified, can be enriched semi
automatically according to the user type and/or a specialized
recommendation system.
Abstract: This paper describes a platform that faces the main
research areas for e-learning educational contents. Reusability tackles
the possibility to use contents in different courses reducing costs and
exploiting available data from repositories. In our approach the
production of educational material is based on templates to reuse
learning objects. In terms of interoperability the main challenge lays
on reaching the audience through different platforms. E-learning
solution must track social consumption evolution where nowadays
lots of multimedia contents are accessed through the social networks.
Our work faces it by implementing a platform for generation of
multimedia presentations focused on the new paradigm related to
social media. The system produces videos-courses on top of web
standard SMIL (Synchronized Multimedia Integration Language)
ready to be published and shared. Regarding interfaces it is
mandatory to satisfy user needs and ease communication. To
overcome it the platform deploys virtual teachers that provide natural
interfaces while multimodal features remove barriers to pupils with
disabilities.
Abstract: Optimization is often a critical issue for most system
design problems. Evolutionary Algorithms are population-based,
stochastic search techniques, widely used as efficient global
optimizers. However, finding optimal solution to complex high
dimensional, multimodal problems often require highly
computationally expensive function evaluations and hence are
practically prohibitive. The Dynamic Approximate Fitness based
Hybrid EA (DAFHEA) model presented in our earlier work [14]
reduced computation time by controlled use of meta-models to
partially replace the actual function evaluation by approximate
function evaluation. However, the underlying assumption in
DAFHEA is that the training samples for the meta-model are
generated from a single uniform model. Situations like model
formation involving variable input dimensions and noisy data
certainly can not be covered by this assumption. In this paper we
present an enhanced version of DAFHEA that incorporates a
multiple-model based learning approach for the SVM approximator.
DAFHEA-II (the enhanced version of the DAFHEA framework) also
overcomes the high computational expense involved with additional
clustering requirements of the original DAFHEA framework. The
proposed framework has been tested on several benchmark functions
and the empirical results illustrate the advantages of the proposed
technique.
Abstract: In this paper, a new hybrid of genetic algorithm (GA)
and simulated annealing (SA), referred to as GSA, is presented. In
this algorithm, SA is incorporated into GA to escape from local
optima. The concept of hierarchical parallel GA is employed to
parallelize GSA for the optimization of multimodal functions. In
addition, multi-niche crowding is used to maintain the diversity in
the population of the parallel GSA (PGSA). The performance of the
proposed algorithms is evaluated against a standard set of multimodal
benchmark functions. The multi-niche crowding PGSA and normal
PGSA show some remarkable improvement in comparison with the
conventional parallel genetic algorithm and the breeder genetic
algorithm (BGA).
Abstract: The dramatic increasing of sea-freight container
transportations and the developing trends for using containers in the
multimodal handling systems through the sea, rail, road and land in
nowadays market cause general managers of container terminals to
face challenges such as increasing demand, competitive situation,
new investments and expansion of new activities and need to use new
methods to fulfil effective operations both along quayside and within
the yard. Among these issues, minimizing the turnaround time of
vessels is considered to be the first aim of every container port
system. Regarding the complex structure of container ports, this
paper presents a simulation model that calculates the number of
trucks needed in the Iranian Shahid Rajaee Container Port for
handling containers between the berth and the yard. In this research,
some important criteria such as vessel turnaround time, gantry crane
utilization and truck utilization have been considered. By analyzing
the results of the model, it has been shown that increasing the number
of trucks to 66 units has a significant effect on the performance
indices of the port and can increase the capacity of loading and
unloading up to 10.8%.
Abstract: Single biometric modality recognition is not able to meet the high performance supplies in most cases with its application become more and more broadly. Multimodal biometrics identification represents an emerging trend recently. This paper investigates a novel algorithm based on fusion of both fingerprint and fingervein biometrics. For both biometric recognition, we employ the Monogenic Local Binary Pattern (MonoLBP). This operator integrate the orginal LBP (Local Binary Pattern ) with both other rotation invariant measures: local phase and local surface type. Experimental results confirm that a weighted sum based proposed fusion achieves excellent identification performances opposite unimodal biometric systems. The AUC of proposed approach based on combining the two modalities has very close to unity (0.93).
Abstract: A wide spectrum of systems require reliable
personal recognition schemes to either confirm or determine the
identity of an individual person. This paper considers multimodal
biometric system and their applicability to access control,
authentication and security applications. Strategies for feature
extraction and sensor fusion are considered and contrasted. Issues
related to performance assessment, deployment and standardization
are discussed. Finally future directions of biometric systems
development are discussed.
Abstract: Currently electronic slide (e-slide) is one of the most common styles in educational presentation. Unfortunately, the utilization of e-slide for the visually impaired is uncommon since they are unable to see the content of such e-slides which are usually composed of text, images and animation. This paper proposes a model for presenting e-slide in multimodal presentation i.e. using conventional slide concurrent with voicing, in both languages Malay and English. At the design level, live multimedia presentation concept is used, while at the implementation level several components are used. The text content of each slide is extracted using COM component, Microsoft Speech API for voicing the text in English language and the text in Malay language is voiced using dictionary approach. To support the accessibility, an auditory user interface is provided as an additional feature. A prototype of such model named as VSlide has been developed and introduced.
Abstract: The paper presents a multimodal approach for biometric authentication, based on multiple classifiers. The proposed solution uses a post-classification biometric fusion method in which the biometric data classifiers outputs are combined in order to improve the overall biometric system performance by decreasing the classification error rates. The paper shows also the biometric recognition task improvement by means of a carefully feature selection, as much as not all of the feature vectors components support the accuracy improvement.
Abstract: Evolutionary Algorithms are population-based,
stochastic search techniques, widely used as efficient global
optimizers. However, many real life optimization problems often
require finding optimal solution to complex high dimensional,
multimodal problems involving computationally very expensive
fitness function evaluations. Use of evolutionary algorithms in such
problem domains is thus practically prohibitive. An attractive
alternative is to build meta models or use an approximation of the
actual fitness functions to be evaluated. These meta models are order
of magnitude cheaper to evaluate compared to the actual function
evaluation. Many regression and interpolation tools are available to
build such meta models. This paper briefly discusses the
architectures and use of such meta-modeling tools in an evolutionary
optimization context. We further present two evolutionary algorithm
frameworks which involve use of meta models for fitness function
evaluation. The first framework, namely the Dynamic Approximate
Fitness based Hybrid EA (DAFHEA) model [14] reduces
computation time by controlled use of meta-models (in this case
approximate model generated by Support Vector Machine
regression) to partially replace the actual function evaluation by
approximate function evaluation. However, the underlying
assumption in DAFHEA is that the training samples for the metamodel
are generated from a single uniform model. This does not take
into account uncertain scenarios involving noisy fitness functions.
The second model, DAFHEA-II, an enhanced version of the original
DAFHEA framework, incorporates a multiple-model based learning
approach for the support vector machine approximator to handle
noisy functions [15]. Empirical results obtained by evaluating the
frameworks using several benchmark functions demonstrate their
efficiency
Abstract: Biometric techniques are gaining importance for
personal authentication and identification as compared to the
traditional authentication methods. Biometric templates are
vulnerable to variety of attacks due to their inherent nature. When a
person-s biometric is compromised his identity is lost. In contrast to
password, biometric is not revocable. Therefore, providing security
to the stored biometric template is very crucial. Crypto biometric
systems are authentication systems, which blends the idea of
cryptography and biometrics. Fuzzy vault is a proven crypto
biometric construct which is used to secure the biometric templates.
However fuzzy vault suffer from certain limitations like nonrevocability,
cross matching. Security of the fuzzy vault is affected
by the non-uniform nature of the biometric data. Fuzzy vault when
hardened with password overcomes these limitations. Password
provides an additional layer of security and enhances user privacy.
Retina has certain advantages over other biometric traits. Retinal
scans are used in high-end security applications like access control to
areas or rooms in military installations, power plants, and other high
risk security areas. This work applies the idea of fuzzy vault for
retinal biometric template. Multimodal biometric system
performance is well compared to single modal biometric systems.
The proposed multi modal biometric fuzzy vault includes combined
feature points from retina and fingerprint. The combined vault is
hardened with user password for achieving high level of security.
The security of the combined vault is measured using min-entropy.
The proposed password hardened multi biometric fuzzy vault is
robust towards stored biometric template attacks.