Abstract: In this paper three different approaches for person
verification and identification, i.e. by means of fingerprints, face and
voice recognition, are studied. Face recognition uses parts-based
representation methods and a manifold learning approach. The
assessment criterion is recognition accuracy. The techniques under
investigation are: a) Local Non-negative Matrix Factorization
(LNMF); b) Independent Components Analysis (ICA); c) NMF with
sparse constraints (NMFsc); d) Locality Preserving Projections
(Laplacianfaces). Fingerprint detection was approached by classical
minutiae (small graphical patterns) matching through image
segmentation by using a structural approach and a neural network as
decision block. As to voice / speaker recognition, melodic cepstral
and delta delta mel cepstral analysis were used as main methods, in
order to construct a supervised speaker-dependent voice recognition
system. The final decision (e.g. “accept-reject" for a verification
task) is taken by using a majority voting technique applied to the
three biometrics. The preliminary results, obtained for medium
databases of fingerprints, faces and voice recordings, indicate the
feasibility of our study and an overall recognition precision (about
92%) permitting the utilization of our system for a future complex
biometric card.
Abstract: Over the past decades, automatic face recognition has become a highly active research area, mainly due to the countless application possibilities in both the private as well as the public sector. Numerous algorithms have been proposed in the literature to cope with the problem of face recognition, nevertheless, a group of methods commonly referred to as appearance based have emerged as the dominant solution to the face recognition problem. Many comparative studies concerned with the performance of appearance based methods have already been presented in the literature, not rarely with inconclusive and often with contradictory results. No consent has been reached within the scientific community regarding the relative ranking of the efficiency of appearance based methods for the face recognition task, let alone regarding their susceptibility to appearance changes induced by various environmental factors. To tackle these open issues, this paper assess the performance of the three dominant appearance based methods: principal component analysis, linear discriminant analysis and independent component analysis, and compares them on equal footing (i.e., with the same preprocessing procedure, with optimized parameters for the best possible performance, etc.) in face verification experiments on the publicly available XM2VTS database. In addition to the comparative analysis on the XM2VTS database, ten degraded versions of the database are also employed in the experiments to evaluate the susceptibility of the appearance based methods on various image degradations which can occur in "real-life" operating conditions. Our experimental results suggest that linear discriminant analysis ensures the most consistent verification rates across the tested databases.
Abstract: In this paper, we proposed a method to classify each
type of natural rock texture. Our goal is to classify 26 classes of rock
textures. First, we extract five features of each class by using
principle component analysis combining with the use of applied
spatial frequency measurement. Next, the effective node number of
neural network was tested. We used the most effective neural
network in classification process. The results from this system yield
quite high in recognition rate. It is shown that high recognition rate
can be achieved in separation of 26 stone classes.
Abstract: A model of user behaviour based automated planning
is introduced in this work. The behaviour of users of web interactive
systems can be described in term of a planning domain encapsulating
the timed actions patterns representing the intended user profile. The
user behaviour recognition is then posed as a planning problem
where the goal is to parse a given sequence of user logs of the
observed activities while reaching a final state.
A general technique for transforming a timed finite state automata
description of the behaviour into a numerical parameter planning
model is introduced.
Experimental results show that the performance of a planning
based behaviour model is effective and scalable for real world
applications. A major advantage of the planning based approach is to
represent in a single automated reasoning framework problems of
plan recognitions, plan synthesis and plan optimisation.
Abstract: We report in this paper the model adopted by our
system of continuous speech recognition in Arab language SySRA
and the results obtained until now. This system uses the database
Arabdic-10 which is a corpus of word for the Arab language and
which was manually segmented. Phonetic decoding is represented
by an expert system where the knowledge base is translated in the
form of production rules. This expert system transforms a vocal
signal into a phonetic lattice. The higher level of the system takes
care of the recognition of the lattice thus obtained by deferring it in
the form of written sentences (orthographical Form). This level
contains initially the lexical analyzer which is not other than the
module of recognition. We subjected this analyzer to a set of
spectrograms obtained by dictating a score of sentences in Arab
language. The rate of recognition of these sentences is about 70%
which is, to our knowledge, the best result for the recognition of the
Arab language. The test set consists of twenty sentences from four
speakers not having taken part in the training.
Abstract: This article presents the developments of efficient
algorithms for tablet copies comparison. Image recognition has
specialized use in digital systems such as medical imaging,
computer vision, defense, communication etc. Comparison between
two images that look indistinguishable is a formidable task. Two
images taken from different sources might look identical but due to
different digitizing properties they are not. Whereas small variation
in image information such as cropping, rotation, and slight
photometric alteration are unsuitable for based matching
techniques. In this paper we introduce different matching
algorithms designed to facilitate, for art centers, identifying real
painting images from fake ones. Different vision algorithms for
local image features are implemented using MATLAB. In this
framework a Table Comparison Computer Tool “TCCT" is
designed to facilitate our research. The TCCT is a Graphical Unit
Interface (GUI) tool used to identify images by its shapes and
objects. Parameter of vision system is fully accessible to user
through this graphical unit interface. And then for matching, it
applies different description technique that can identify exact
figures of objects.
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: In this paper we address the issue of classifying the fluorescent intensity of a sample in Indirect Immuno-Fluorescence (IIF). Since IIF is a subjective, semi-quantitative test in its very nature, we discuss a strategy to reliably label the image data set by using the diagnoses performed by different physicians. Then, we discuss image pre-processing, feature extraction and selection. Finally, we propose two ANN-based classifiers that can separate intrinsically dubious samples and whose error tolerance can be flexibly set. Measured performance shows error rates less than 1%, which candidates the method to be used in daily medical practice either to perform pre-selection of cases to be examined, or to act as a second reader.
Abstract: We introduce an algorithm based on the
morphological shared-weight neural network. Being nonlinear and
translation-invariant, the MSNN can be used to create better
generalization during face recognition. Feature extraction is
performed on grayscale images using hit-miss transforms that are
independent of gray-level shifts. The output is then learned by
interacting with the classification process. The feature extraction and
classification networks are trained together, allowing the MSNN to
simultaneously learn feature extraction and classification for a face.
For evaluation, we test for robustness under variations in gray levels
and noise while varying the network-s configuration to optimize
recognition efficiency and processing time. Results show that the
MSNN performs better for grayscale image pattern classification
than ordinary neural networks.
Abstract: This research aimed to find out the determining
factors for ISO 14001 EMS implementation among SMEs in
Malaysia from the Resource based view. A cross-sectional approach
using survey was conducted. A research model been proposed which
comprises of ISO 14001 EMS implementation as the criterion
variable while physical capital resources (i.e. environmental
performance tracking and organizational infrastructures), human
capital resources (i.e. top management commitment and support,
training and education, employee empowerment and teamwork) and
organizational capital resources (i.e. recognition and reward,
organizational culture and organizational communication) as the
explanatory variables. The research findings show that only
environmental performance tracking, top management commitment
and support and organizational culture are found to be positively and
significantly associated with ISO 14001 EMS implementation. It is
expected that this research will shed new knowledge and provide a
base for future studies about the role played by firm-s internal
resources.
Abstract: In this paper we propose an NLP-based method for
Ontology Population from texts and apply it to semi automatic
instantiate a Generic Knowledge Base (Generic Domain Ontology) in
the risk management domain. The approach is semi-automatic and
uses a domain expert intervention for validation. The proposed
approach relies on a set of Instances Recognition Rules based on
syntactic structures, and on the predicative power of verbs in the
instantiation process. It is not domain dependent since it heavily
relies on linguistic knowledge.
A description of an experiment performed on a part of the
ontology of the PRIMA1 project (supported by the European
community) is given. A first validation of the method is done by
populating this ontology with Chemical Fact Sheets from
Environmental Protection Agency2. The results of this experiment
complete the paper and support the hypothesis that relying on the
predicative power of verbs in the instantiation process improves the
performance.
Abstract: Current OCR technology does not allow to
accurately recognizing small text images, such as those found
in web images. Our goal is to investigate new approaches to
recognize very low resolution text images containing antialiased
character shapes.
This paper presents a preliminary study on the variability of
such characters and the feasibility to discriminate them by
using geometrical features. In a first stage we analyze the
distribution of these features. In a second stage we present a
study on the discriminative power for recognizing isolated
characters, using various rendering methods and font
properties. Finally we present interesting results of our
evaluation tests leading to our conclusion and future focus.
Abstract: The ability of the brain to organize information and generate the functional structures we use to act, think and communicate, is a common and easily observable natural phenomenon. In object-oriented analysis, these structures are represented by objects. Objects have been extensively studied and documented, but the process that creates them is not understood. In this work, a new class of discrete, deterministic, dissipative, host-guest dynamical systems is introduced. The new systems have extraordinary self-organizing properties. They can host information representing other physical systems and generate the same functional structures as the brain does. A simple mathematical model is proposed. The new systems are easy to simulate by computer, and measurements needed to confirm the assumptions are abundant and readily available. Experimental results presented here confirm the findings. Applications are many, but among the most immediate are object-oriented engineering, image and voice recognition, search engines, and Neuroscience.
Abstract: Although face recognition seems as an easy task for
human, automatic face recognition is a much more challenging task
due to variations in time, illumination and pose. In this paper, the
influence of time-lapse on visible and thermal images is examined.
Orthogonal moment invariants are used as a feature extractor to
analyze the effect of time-lapse on thermal and visible images and the
results are compared with conventional Principal Component
Analysis (PCA). A new triangle square ratio criterion is employed
instead of Euclidean distance to enhance the performance of nearest
neighbor classifier. The results of this study indicate that the ideal
feature vectors can be represented with high discrimination power
due to the global characteristic of orthogonal moment invariants.
Moreover, the effect of time-lapse has been decreasing and enhancing
the accuracy of face recognition considerably in comparison with
PCA. Furthermore, our experimental results based on moment
invariant and triangle square ratio criterion show that the proposed
approach achieves on average 13.6% higher in recognition rate than
PCA.
Abstract: As in today's semiconductor industries test costs can make up to 50 percent of the total production costs, an efficient test error detection becomes more and more important. In this paper, we present a new machine learning approach to test error detection that should provide a faster recognition of test system faults as well as an improved test error recall. The key idea is to learn a classifier ensemble, detecting typical test error patterns in wafer test results immediately after finishing these tests. Since test error detection has not yet been discussed in the machine learning community, we define central problem-relevant terms and provide an analysis of important domain properties. Finally, we present comparative studies reflecting the failure detection performance of three individual classifiers and three ensemble methods based upon them. As base classifiers we chose a decision tree learner, a support vector machine and a Bayesian network, while the compared ensemble methods were simple and weighted majority vote as well as stacking. For the evaluation, we used cross validation and a specially designed practical simulation. By implementing our approach in a semiconductor test department for the observation of two products, we proofed its practical applicability.
Abstract: An automatic speech recognition system for the
formal Arabic language is needed. The Quran is the most formal
spoken book in Arabic, it is spoken all over the world. In this
research, an automatic speech recognizer for Quranic based speakerindependent
was developed and tested. The system was developed
based on the tri-phone Hidden Markov Model and Maximum
Likelihood Linear Regression (MLLR). The MLLR computes a set
of transformations which reduces the mismatch between an initial
model set and the adaptation data. It uses the regression class tree, as
well as, estimates a set of linear transformations for the mean and
variance parameters of a Gaussian mixture HMM system. The 30th
Chapter of the Quran, with five of the most famous readers of the
Quran, was used for the training and testing of the data. The chapter
includes about 2000 distinct words. The advantages of using the
Quranic verses as the database in this developed recognizer are the
uniqueness of the words and the high level of orderliness between
verses. The level of accuracy from the tested data ranged 68 to 85%.
Abstract: Rapid advancement in computing technology brings
computers and humans to be seamlessly integrated in future. The
emergence of smartphone has driven computing era towards
ubiquitous and pervasive computing. Recognizing human activity has
garnered a lot of interest and has raised significant researches-
concerns in identifying contextual information useful to human
activity recognition. Not only unobtrusive to users in daily life,
smartphone has embedded built-in sensors that capable to sense
contextual information of its users supported with wide range
capability of network connections. In this paper, we will discuss the
classification algorithms used in smartphone-based human activity.
Existing technologies pertaining to smartphone-based researches in
human activity recognition will be highlighted and discussed. Our
paper will also present our findings and opinions to formulate
improvement ideas in current researches- trends. Understanding
research trends will enable researchers to have clearer research
direction and common vision on latest smartphone-based human
activity recognition area.
Abstract: This study examines perception of environmental
approach in small and medium-sized enterprises (SMEs) – the
process by which firms integrate environmental concern into
business. Based on a review of the literature, the paper synthesizes
focus on environmental issues with the reflection in a case study in
the Czech Republic. Two themes of corporate environmentalism are
discussed – corporate environmental orientation and corporate
stances toward environmental concerns. It provides theoretical
material on greening organizational culture that is helpful in
understanding the response of contemporary business to
environmental problems. We integrate theoretical predictions with
empirical findings confronted with reality. Scales to measure these
themes are tested in a survey of managers in 229 Czech firms. We
used the process of in-depth questioning. The research question was
derived and answered in the context of the corresponding literature
and conducted research. A case study showed us that environmental
approach is variety different (depending on the size of the firm) in
SMEs sector. The results of the empirical mapping demonstrate
Czech company’s approach to environment and define the problem
areas and pinpoint the main limitation in the expansion of
environmental aspects. We contribute to the debate for recognition of
the particular role of environmental issues in business reality.
Abstract: The feature extraction method(s) used to recognize
hand-printed characters play an important role in ICR applications.
In order to achieve high recognition rate for a recognition system, the
choice of a feature that suits for the given script is certainly an
important task. Even if a new feature required to be designed for a
given script, it is essential to know the recognition ability of the
existing features for that script. Devanagari script is being used in
various Indian languages besides Hindi the mother tongue of majority
of Indians. This research examines a variety of feature extraction
approaches, which have been used in various ICR/OCR applications,
in context to Devanagari hand-printed script. The study is conducted
theoretically and experimentally on more that 10 feature extraction
methods. The various feature extraction methods have been evaluated
on Devanagari hand-printed database comprising more than 25000
characters belonging to 43 alphabets. The recognition ability of the
features have been evaluated using three classifiers i.e. k-NN, MLP
and SVM.
Abstract: Achievement motivation is believed to promote
giftedness attracting people to invest in many programs to adopt
gifted students providing them with challenging activities.
Intellectual giftedness is founded on the fluid intelligence and
extends to more specific abilities through the growth and inputs from
the achievement motivation. Acknowledging the roles played by the
motivation in the development of giftedness leads to an effective
nurturing of gifted individuals. However, no study has investigated
the direct and indirect effects of the achievement motivation and
fluid intelligence on intellectual giftedness. Thus, this study
investigated the contribution of motivation factors to giftedness
development by conducting tests of fluid intelligence using Cattell
Culture Fair Test (CCFT) and analytical abilities using culture
reduced test items covering problem solving, pattern recognition,
audio-logic, audio-matrices, and artificial language, and self report
questionnaire for the motivational factors. A number of 180 highscoring
students were selected using CCFT from a leading university
in Malaysia. Structural equation modeling was employed using Amos
V.16 to determine the direct and indirect effects of achievement
motivation factors (self confidence, success, perseverance,
competition, autonomy, responsibility, ambition, and locus of
control) on the intellectual giftedness. The findings showed that the
hypothesized model fitted the data, supporting the model postulates
and showed significant and strong direct and indirect effects of the
motivation and fluid intelligence on the intellectual giftedness.