Abstract: Face recognition in the infrared spectrum has attracted a lot of interest in recent years. Many of the techniques used in infrared are based on their visible counterpart, especially linear techniques like PCA and LDA. In this work, we introduce a probabilistic Bayesian framework for face recognition in the infrared spectrum. In the infrared spectrum, variations can occur between face images of the same individual due to pose, metabolic, time changes, etc. Bayesian approaches permit to reduce intrapersonal variation, thus making them very interesting for infrared face recognition. This framework is compared with classical linear techniques. Non linear techniques we developed recently for infrared face recognition are also presented and compared to the Bayesian face recognition framework. A new approach for infrared face extraction based on SVM is introduced. Experimental results show that the Bayesian technique is promising and lead to interesting results in the infrared spectrum when a sufficient number of face images is used in an intrapersonal learning process.
Abstract: The objective of this study was to determine the accuracy to estimation fetal weight by Johnson-s method and compares it with actual birth weight. The sample group was 126 infants delivered in Dan KhunThot hospital from January March 2012. Fetal weight was estimated by measuring fundal height according to Johnson-s method. The information was collected by studying historical delivery records and then analyzed by using the statistics of frequency, percentage, mean, and standard deviation. Finally, the difference was analyzed by a paired t-test.The results showed had an average birth weight was 3093.57 ± 391.03 g (mean ± SD) and 3,455 ± 454.55 g average estimated fetal weight by Johnson-s method higher than average actual birth weight was 384.09 grams. When classifying the infants according to birth weight found that low birth weight ( 4000 g) actual birth weight was more than estimated fetal weight. The difference was found between actual birth weight and estimation fetal weight of the minimum weight in high birth weight ( > 4000 g) , the appropriate birth weight (2500-3999g) and low birth weight (
Abstract: This paper presents a study of laminar to turbulent transition on a profile specifically designed for wind turbine blades, the DU91-W2-250, which belongs to a class of wind turbine dedicated airfoils, developed by Delft University of Technology. A comparison between the experimental behavior of the airfoil studied at Delft wind tunnel and the numerical predictions of the commercial CFD solver ANSYS FLUENT® has been performed. The prediction capabilities of the Spalart-Allmaras turbulence model and of the γ-θ Transitional model have been tested. A sensitivity analysis of the numerical results to the spatial domain discretization has also been performed using four different computational grids, which have been created using the mesher GAMBIT®. The comparison between experimental measurements and CFD results have allowed to determine the importance of the numerical prediction of the laminar to turbulent transition, in order not to overestimate airfoil friction drag due to a fully turbulent-regime flow computation.
Abstract: In many applications there is a broad variety of
information relevant to a focal “object" of interest, and the fusion of such heterogeneous data types is desirable for classification and
categorization. While these various data types can sometimes be treated as orthogonal (such as the hull number, superstructure color,
and speed of an oil tanker), there are instances where the inference and the correlation between quantities can provide improved fusion
capabilities (such as the height, weight, and gender of a person). A
service-oriented architecture has been designed and prototyped to
support the fusion of information for such “object-centric" situations.
It is modular, scalable, and flexible, and designed to support new data sources, fusion algorithms, and computational resources without affecting existing services. The architecture is designed to simplify
the incorporation of legacy systems, support exact and probabilistic entity disambiguation, recognize and utilize multiple types of
uncertainties, and minimize network bandwidth requirements.
Abstract: In this study, the problem of discriminating between interictal epileptic and non- epileptic pathological EEG cases, which present episodic loss of consciousness, investigated. We verify the accuracy of the feature extraction method of autocross-correlated coefficients which extracted and studied in previous study. For this purpose we used in one hand a suitable constructed artificial supervised LVQ1 neural network and in other a cross-correlation technique. To enforce the above verification we used a statistical procedure which based on a chi- square control. The classification and the statistical results showed that the proposed feature extraction is a significant accurate method for diagnostic discrimination cases between interictal and non-interictal EEG events and specifically the classification procedure showed that the LVQ neural method is superior than the cross-correlation one.
Abstract: The purpose of this research aims to discover the
knowledge for analysis student motivation behavior on e-Learning
based on Data Mining Techniques, in case of the Information
Technology for Communication and Learning Course at Suan
Sunandha Rajabhat University. The data mining techniques was
applied in this research including association rules, classification
techniques. The results showed that using data mining technique can
indicate the important variables that influence the student motivation
behavior on e-Learning.
Abstract: The growing interest on national heritage
preservation has led to intensive efforts on digital documentation of
cultural heritage knowledge. Encapsulated within this effort is the
focus on ontology development that will help facilitate the
organization and retrieval of the knowledge. Ontologies surrounding
cultural heritage domain are related to archives, museum and library
information such as archaeology, artifacts, paintings, etc. The growth
in number and size of ontologies indicates the well acceptance of its
semantic enrichment in many emerging applications. Nowadays,
there are many heritage information systems available for access.
Among others is community-based e-museum designed to support the
digital cultural heritage preservation. This work extends previous
effort of developing the Traditional Malay Textile (TMT) Knowledge
Model where the model is designed with the intention of auxiliary
mapping with CIDOC CRM. Due to its internal constraints, the
model needs to be transformed in advance. This paper addresses the
issue by reviewing the previous harmonization works with CIDOC
CRM as exemplars in refining the facets in the model particularly
involving TMT-Artifact class. The result is an extensible model
which could lead to a common view for automated mapping with
CIDOC CRM. Hence, it promotes integration and exchange of
textile information especially batik-related between communities in
e-museum applications.
Abstract: In aircraft applications, according to the nature of
electrical equipment its location may be in unpressurized area or very
close to the engine; thus, the environmental conditions may change
from atmospheric pressure to less than 100 mbar, and the temperature
may be higher than the ambient one as in most real working
conditions of electrical equipment. Then, the classical Paschen curve
has to be replotted since these parameters may affect the discharge
ignition voltage. In this paper, we firstly investigate the domain of
validity of two corrective expressions on the Paschen-s law found in
the literature, in case of changing the air environment and known as
Peek and Dunbar corrections. Results show that these corrections are
no longer valid for combined variation of temperature and pressure.
After that, a new empirical expression for breakdown voltage is
proposed and is validated in the case of combined variations of
temperature and pressure.
Abstract: This paper presents the analysis of similarity between local decisions, in the process of alphanumeric hand-prints classification. From the analysis of local characteristics of handprinted numerals and characters, extracted by a zoning method, the set of classification decisions is obtained and the similarity among them is investigated. For this purpose the Similarity Index is used, which is an estimator of similarity between classifiers, based on the analysis of agreements between their decisions. The experimental tests, carried out using numerals and characters from the CEDAR and ETL database, respectively, show to what extent different parts of the patterns provide similar classification decisions.
Abstract: In the present study, a support vector machine (SVM) learning approach to character recognition is proposed. Simple
feature detectors, similar to those found in the human visual system, were used in the SVM classifier. Alphabetic characters were rotated
to 8 different angles and using the proposed cognitive model, all characters were recognized with 100% accuracy and specificity.
These same results were found in psychiatric studies of human character recognition.
Abstract: Emotions are related with learning processes and
physiological signals can be used to detect them for the
personalization of learning resources and to control the pace of
instruction. A model of relevant emotions has been developed, where
specific combinations of emotions and cognition processes are
connected and integrated with the concept of 'flow', in order to
improve learning. The cardiac pulse is a reliable signal that carries
useful information about the subject-s emotional condition; it is
detected using a classroom chair adapted with non invasive EMFi
sensor and an acquisition system that generates a ballistocardiogram
(BCG), the signal is processed by an algorithm to obtain
characteristics that match a specific emotional condition. The
complete chair system is presented in this work, along with a
framework for the personalization of learning resources.
Abstract: This article simulates the wind generator set which has
two fault bearing collar rail destruction and the gear box oil leak fault.
The electric current signal which produced by the generator, We use
Empirical Mode Decomposition (EMD) as well as Fast Fourier
Transform (FFT) obtains the frequency range-s signal figure and
characteristic value. The last step is use a kind of Artificial Neural
Network (ANN) classifies which determination fault signal's type and
reason. The ANN purpose of the automatic identification wind
generator set fault..
Abstract: To make the modulation classification system more suitable for signals in a wide range of signal to noise rate (SNR), a novel method of designing combined classifier based on fuzzy neural network (FNN) is presented in this paper. The method employs fuzzy neural network classifiers and interclass distance (ICD) to improve recognition reliability. Experimental results show that the proposed combined classifier has high recognition rate with large variation range of SNR (success rates are over 99.9% when SNR is not lower than 5dB).
Abstract: This paper presents an analytical method to solve
governing consolidation parabolic partial differential equation (PDE)
for inelastic porous Medium (soil) with consideration of variation of
equation coefficient under cyclic loading. Since under cyclic loads,
soil skeleton parameters change, this would introduce variable
coefficient of parabolic PDE. Classical theory would not rationalize
consolidation phenomenon in such condition. In this research, a
method based on time space mapping to a virtual time space along
with superimposing rule is employed to solve consolidation of
inelastic soils in cyclic condition. Changes of consolidation
coefficient applied in solution by modification of loading and
unloading duration by introducing virtual time. Mapping function is
calculated based on consolidation partial differential equation results.
Based on superimposing rule a set of continuous static loads in
specified times used instead of cyclic load. A set of laboratory
consolidation tests under cyclic load along with numerical
calculations were performed in order to verify the presented method.
Numerical solution and laboratory tests results showed accuracy of
presented method.
Abstract: The lecture represents significant advances in
understanding of the transfer processes mechanism in turbulent
separated flows. Based upon experimental data suggesting the
governing role of generated local pressure gradient that takes place in
the immediate vicinity of the wall in separated flow as a result of
intense instantaneous accelerations induced by large-scale vortex
flow structures similarity laws for mean velocity and temperature and
spectral characteristics and heat and mass transfer law for turbulent
separated flows have been developed. These laws are confirmed by
available experimental data. The results obtained were employed for
analysis of heat and mass transfer in some very complex processes
occurring in technological applications such as impinging jets, heat
transfer of cylinders in cross flow and in tube banks, packed beds
where processes manifest distinct properties which allow them to be
classified under turbulent separated flows. Many facts have got an
explanation for the first time.
Abstract: The presented work is motivated by a french law regarding nuclear waste management. In order to avoid the limitation coming with the usage of the existing scenario codes, as COSI, VISION or FAMILY, the Core Library for Advance Scenario Simulation (CLASS) is being develop. CLASS is an open source tool, which allows any user to simulate an electronuclear scenario. The main CLASS asset, is the possibility to include any type of reactor, even a complitely new concept, through the generation of its ACSII evolution database. In the present article, the CLASS working basis will be presented as well as a simple exemple in order to show his potentiel. In the considered exemple, the effect of the transmutation will be assessed on Minor Actinide Inventory produced by PWR reactors.
Abstract: Airbag deployment has been known to be responsible
for huge death, incidental injuries and broken bones due to low crash
severity and wrong deployment decisions. Therefore, the authorities
and industries have been looking for more innovative and intelligent
products to be realized for future enhancements in the vehicle safety
systems (VSSs). Although the VSSs technologies have advanced
considerably, they still face challenges such as how to avoid
unnecessary and untimely airbag deployments that can be hazardous
and fatal. Currently, most of the existing airbag systems deploy
without regard to occupant size and position. As such, this paper will
focus on the occupant and crash sensing performances due to frontal
collisions for the new breed of so called smart airbag systems. It
intends to provide a thorough discussion relating to the occupancy
detection, occupant size classification, occupant off-position
detection to determine safe distance zone for airbag deployment,
crash-severity analysis and airbag decision algorithms via a computer
modeling. The proposed system model consists of three main
modules namely, occupant sensing, crash severity analysis and
decision fusion. The occupant sensing system module utilizes the
weight sensor to determine occupancy, classify the occupant size,
and determine occupant off-position condition to compute safe
distance for airbag deployment. The crash severity analysis module is
used to generate relevant information pertinent to airbag deployment
decision. Outputs from these two modules are fused to the decision
module for correct and efficient airbag deployment action. Computer
modeling work is carried out using Simulink, Stateflow,
SimMechanics and Virtual Reality toolboxes.
Abstract: The paper describes a self supervised parallel self organizing neural network (PSONN) architecture for true color image segmentation. The proposed architecture is a parallel extension of the standard single self organizing neural network architecture (SONN) and comprises an input (source) layer of image information, three single self organizing neural network architectures for segmentation of the different primary color components in a color image scene and one final output (sink) layer for fusion of the segmented color component images. Responses to the different shades of color components are induced in each of the three single network architectures (meant for component level processing) by applying a multilevel version of the characteristic activation function, which maps the input color information into different shades of color components, thereby yielding a processed component color image segmented on the basis of the different shades of component colors. The number of target classes in the segmented image corresponds to the number of levels in the multilevel activation function. Since the multilevel version of the activation function exhibits several subnormal responses to the input color image scene information, the system errors of the three component network architectures are computed from some subnormal linear index of fuzziness of the component color image scenes at the individual level. Several multilevel activation functions are employed for segmentation of the input color image scene using the proposed network architecture. Results of the application of the multilevel activation functions to the PSONN architecture are reported on three real life true color images. The results are substantiated empirically with the correlation coefficients between the segmented images and the original images.
Abstract: In the hardening energy context, the transport sector
which constitutes a large worldwide energy demand has to be
improving for decrease energy demand and global warming impacts.
In a controversial situation where subsists an increasing demand for
long-distance and high-speed travels, high-speed trains offer many
advantages, as consuming significantly less energy than road or air
transports.
At the project phase of new rail infrastructures, it is nowadays
important to characterize accurately the energy that will be induced
by its operation phase, in addition to other more classical criteria as
construction costs and travel time.
Current literature consumption models used to estimate railways
operation phase are obsolete or not enough accurate for taking into
account the newest train or railways technologies.
In this paper, an updated model of consumption for high-speed is
proposed, based on experimental data obtained from full-scale tests
performed on a new high-speed line. The assessment of the model
is achieved by identifying train parameters and measured power
consumptions for more than one hundred train routes. Perspectives
are then discussed to use this updated model for accurately assess
the energy impact of future railway infrastructures.
Abstract: In the framework of adaptive parametric modelling of images, we propose in this paper a new technique based on the Chandrasekhar fast adaptive filter for texture characterization. An Auto-Regressive (AR) linear model of texture is obtained by scanning the image row by row and modelling this data with an adaptive Chandrasekhar linear filter. The characterization efficiency of the obtained model is compared with the model adapted with the Least Mean Square (LMS) 2-D adaptive algorithm and with the cooccurrence method features. The comparison criteria is based on the computation of a characterization degree using the ratio of "betweenclass" variances with respect to "within-class" variances of the estimated coefficients. Extensive experiments show that the coefficients estimated by the use of Chandrasekhar adaptive filter give better results in texture discrimination than those estimated by other algorithms, even in a noisy context.