Abstract: Classification of Persian printed numeral characters
has been considered and a proposed system has been introduced. In
representation stage, for the first time in Persian optical character
recognition, extended moment invariants has been utilized as
characters image descriptor. In classification stage, four different
classifiers namely minimum mean distance, nearest neighbor rule,
multi layer perceptron, and fuzzy min-max neural network has been
used, which first and second are traditional nonparametric statistical
classifier. Third is a well-known neural network and forth is a kind of
fuzzy neural network that is based on utilizing hyperbox fuzzy sets.
Set of different experiments has been done and variety of results has
been presented. The results showed that extended moment invariants
are qualified as features to classify Persian printed numeral
characters.
Abstract: In this paper, a new face recognition method based on
PCA (principal Component Analysis), LDA (Linear Discriminant
Analysis) and neural networks is proposed. This method consists of
four steps: i) Preprocessing, ii) Dimension reduction using PCA, iii)
feature extraction using LDA and iv) classification using neural
network. Combination of PCA and LDA is used for improving the
capability of LDA when a few samples of images are available and
neural classifier is used to reduce number misclassification caused by
not-linearly separable classes. The proposed method was tested on
Yale face database. Experimental results on this database
demonstrated the effectiveness of the proposed method for face
recognition with less misclassification in comparison with previous
methods.
Abstract: This paper presents a new growing neural network for
cluster analysis and market segmentation, which optimizes the size
and structure of clusters by iteratively checking them for multivariate
normality. We combine the recently published SGNN approach [8]
with the basic principle underlying the Gaussian-means algorithm
[13] and the Mardia test for multivariate normality [18, 19]. The new
approach distinguishes from existing ones by its holistic design and
its great autonomy regarding the clustering process as a whole. Its
performance is demonstrated by means of synthetic 2D data and by
real lifestyle survey data usable for market segmentation.
Abstract: Nowadays, precipitation prediction is required for proper planning and management of water resources. Prediction with neural network models has received increasing interest in various research and application domains. However, it is difficult to determine the best neural network architecture for prediction since it is not immediately obvious how many input or hidden nodes are used in the model. In this paper, neural network model is used as a forecasting tool. The major aim is to evaluate a suitable neural network model for monthly precipitation mapping of Myanmar. Using 3-layerd neural network models, 100 cases are tested by changing the number of input and hidden nodes from 1 to 10 nodes, respectively, and only one outputnode used. The optimum model with the suitable number of nodes is selected in accordance with the minimum forecast error. In measuring network performance using Root Mean Square Error (RMSE), experimental results significantly show that 3 inputs-10 hiddens-1 output architecture model gives the best prediction result for monthly precipitation in Myanmar.
Abstract: This paper proposes a novel approach to the question of lithofacies classification based on an assessment of the uncertainty in the classification results. The proposed approach has multiple neural networks (NN), and interval neutrosophic sets (INS) are used to classify the input well log data into outputs of multiple classes of lithofacies. A pair of n-class neural networks are used to predict n-degree of truth memberships and n-degree of false memberships. Indeterminacy memberships or uncertainties in the predictions are estimated using a multidimensional interpolation method. These three memberships form the INS used to support the confidence in results of multiclass classification. Based on the experimental data, our approach improves the classification performance as compared to an existing technique applied only to the truth membership. In addition, our approach has the capability to provide a measure of uncertainty in the problem of multiclass classification.
Abstract: In this paper, an automatic determination algorithm for nuclear magnetic resonance (NMR) spectra of the metabolites in the living body by magnetic resonance spectroscopy (MRS) without human intervention or complicated calculations is presented. In such method, the problem of NMR spectrum determination is transformed into the determination of the parameters of a mathematical model of the NMR signal. To calculate these parameters efficiently, a new model called modified Hopfield neural network is designed. The main achievement of this paper over the work in literature [30] is that the speed of the modified Hopfield neural network is accelerated. This is done by applying cross correlation in the frequency domain between the input values and the input weights. The modified Hopfield neural network can accomplish complex dignals perfectly with out any additinal computation steps. This is a valuable advantage as NMR signals are complex-valued. In addition, a technique called “modified sequential extension of section (MSES)" that takes into account the damping rate of the NMR signal is developed to be faster than that presented in [30]. Simulation results show that the calculation precision of the spectrum improves when MSES is used along with the neural network. Furthermore, MSES is found to reduce the local minimum problem in Hopfield neural networks. Moreover, the performance of the proposed method is evaluated and there is no effect on the performance of calculations when using the modified Hopfield neural networks.
Abstract: As a vital activity for companies, new product
development (NPD) is also a very risky process due to the high
uncertainty degree encountered at every development stage and the
inevitable dependence on how previous steps are successfully
accomplished. Hence, there is an apparent need to evaluate new
product initiatives systematically and make accurate decisions under
uncertainty. Another major concern is the time pressure to launch a
significant number of new products to preserve and increase the
competitive power of the company. In this work, we propose an
integrated decision-making framework based on neural networks and
fuzzy logic to make appropriate decisions and accelerate the
evaluation process. We are especially interested in the two initial
stages where new product ideas are selected (go/no go decision) and
the implementation order of the corresponding projects are
determined. We show that this two-staged intelligent approach allows
practitioners to roughly and quickly separate good and bad product
ideas by making use of previous experiences, and then, analyze a
more shortened list rigorously.
Abstract: Designing and implementing intelligent systems has become a crucial factor for the innovation and development of better products of space technologies. A neural network is a parallel system, capable of resolving paradigms that linear computing cannot. Field programmable gate array (FPGA) is a digital device that owns reprogrammable properties and robust flexibility. For the neural network based instrument prototype in real time application, conventional specific VLSI neural chip design suffers the limitation in time and cost. With low precision artificial neural network design, FPGAs have higher speed and smaller size for real time application than the VLSI and DSP chips. So, many researchers have made great efforts on the realization of neural network (NN) using FPGA technique. In this paper, an introduction of ANN and FPGA technique are briefly shown. Also, Hardware Description Language (VHDL) code has been proposed to implement ANNs as well as to present simulation results with floating point arithmetic. Synthesis results for ANN controller are developed using Precision RTL. Proposed VHDL implementation creates a flexible, fast method and high degree of parallelism for implementing ANN. The implementation of multi-layer NN using lookup table LUT reduces the resource utilization for implementation and time for execution.
Abstract: In this paper, we propose a new modular approach called neuroglial consisting of two neural networks slow and fast which emulates a biological reality recently discovered. The implementation is based on complex multi-time scale systems; validation is performed on the model of the asynchronous machine. We applied the geometric approach based on the Gerschgorin circles for the decoupling of fast and slow variables, and the method of singular perturbations for the development of reductions models.
This new architecture allows for smaller networks with less complexity and better performance in terms of mean square error and convergence than the single network model.
Abstract: This paper investigates the problem of exponential stability for a class of uncertain discrete-time stochastic neural network with time-varying delays. By constructing a suitable Lyapunov-Krasovskii functional, combining the stochastic stability theory, the free-weighting matrix method, a delay-dependent exponential stability criteria is obtained in term of LMIs. Compared with some previous results, the new conditions obtain in this paper are less conservative. Finally, two numerical examples are exploited to show the usefulness of the results derived.
Abstract: Investigating language acquisition is one of the most
challenging problems in the area of studying language. Syllable
learning as a level of language acquisition has a considerable
significance since it plays an important role in language acquisition.
Because of impossibility of studying language acquisition directly
with children, especially in its developmental phases, computer
models will be useful in examining language acquisition. In this
paper a computer model of early language learning for syllable
learning is proposed. It is guided by a conceptual model of syllable
learning which is named Directions Into Velocities of Articulators
model (DIVA). The computer model uses simple associational and
reinforcement learning rules within neural network architecture
which are inspired by neuroscience. Our simulation results verify the
ability of the proposed computer model in producing phonemes
during babbling and early speech. Also, it provides a framework for
examining the neural basis of language learning and communication
disorders.
Abstract: The electric power industry is currently undergoing an unprecedented reform. One of the most exciting and potentially profitable recent developments is increasing usage of artificial intelligence techniques. The intention of this paper is to give an overview of using neural network (NN) techniques in power systems. According to the growth rate of NNs application in some power system subjects, this paper introduce a brief overview in fault diagnosis, security assessment, load forecasting, economic dispatch and harmonic analyzing. Advantages and disadvantages of using NNs in above mentioned subjects and the main challenges in these fields have been explained, too.
Abstract: This paper derives some new sufficient conditions for
the stability of a class of neutral-type neural networks with discrete
time delays by employing a suitable Lyapunov functional. The
obtained conditions can be easily verified as they can be expressed
in terms of the network parameters only. It is shown that the results
presented in this paper for neutral-type delayed neural networks establish
a new set of stability criteria, and therefore can be considered
as the alternative results to the previously published literature results.
A numerical example is also given to demonstrate the applicability
of our proposed stability criterion.
Abstract: Prediction of highly non linear behavior of suspended
sediment flow in rivers has prime importance in the field of water
resources engineering. In this study the predictive performance of
two Artificial Neural Networks (ANNs) namely, the Radial Basis
Function (RBF) Network and the Multi Layer Feed Forward (MLFF)
Network have been compared. Time series data of daily suspended
sediment discharge and water discharge at Pari River was used for
training and testing the networks. A number of statistical parameters
i.e. root mean square error (RMSE), mean absolute error (MAE),
coefficient of efficiency (CE) and coefficient of determination (R2)
were used for performance evaluation of the models. Both the models
produced satisfactory results and showed a good agreement between
the predicted and observed data. The RBF network model provided
slightly better results than the MLFF network model in predicting
suspended sediment discharge.
Abstract: In this paper, an ultrasonic technique is proposed to
predict oil content in a fresh palm fruit. This is accomplished by
measuring the attenuation based on ultrasonic transmission mode.
Several palm fruit samples with known oil content by Soxhlet
extraction (ISO9001:2008) were tested with our ultrasonic
measurement. Amplitude attenuation data results for all palm samples
were collected. The Feedforward Neural Networks (FNNs) are
applied to predict the oil content for the samples. The Root Mean
Square Error (RMSE) and Mean Absolute Error (MAE) of the FNN
model for predicting oil content percentage are 7.6186 and 5.2287
with the correlation coefficient (R) of 0.9193.
Abstract: The counting process of cell colonies is always a long
and laborious process that is dependent on the judgment and ability
of the operator. The judgment of the operator in counting can vary in
relation to fatigue. Moreover, since this activity is time consuming it
can limit the usable number of dishes for each experiment. For these
purposes, it is necessary that an automatic system of cell colony
counting is used. This article introduces a new automatic system of
counting based on the elaboration of the digital images of cellular
colonies grown on petri dishes. This system is mainly based on the
algorithms of region-growing for the recognition of the regions of
interest (ROI) in the image and a Sanger neural net for the
characterization of such regions. The better final classification is
supplied from a Feed-Forward Neural Net (FF-NN) and confronted
with the K-Nearest Neighbour (K-NN) and a Linear Discriminative
Function (LDF). The preliminary results are shown.
Abstract: In this paper we present an off line system for the
recognition of the handwritten numeric chains. Our work is divided
in two big parts. The first part is the realization of a recognition
system of the isolated handwritten digits. In this case the study is
based mainly on the evaluation of neural network performances,
trained with the gradient back propagation algorithm. The used
parameters to form the input vector of the neural network are
extracted on the binary images of the digits by several methods: the
distribution sequence, the Barr features and the centred moments of
the different projections and profiles. The second part is the
extension of our system for the reading of the handwritten numeric
chains constituted of a variable number of digits. The vertical
projection is used to segment the numeric chain at isolated digits and
every digit (or segment) will be presented separately to the entry of
the system achieved in the first part (recognition system of the
isolated handwritten digits). The result of the recognition of the
numeric chain will be displayed at the exit of the global system.
Abstract: Software project effort estimation is frequently seen
as complex and expensive for individual software engineers.
Software production is in a crisis. It suffers from excessive costs.
Software production is often out of control. It has been suggested that
software production is out of control because we do not measure.
You cannot control what you cannot measure. During last decade, a
number of researches on cost estimation have been conducted. The
metric-set selection has a vital role in software cost estimation
studies; its importance has been ignored especially in neural network
based studies. In this study we have explored the reasons of those
disappointing results and implemented different neural network
models using augmented new metrics. The results obtained are
compared with previous studies using traditional metrics. To be able
to make comparisons, two types of data have been used. The first
part of the data is taken from the Constructive Cost Model
(COCOMO'81) which is commonly used in previous studies and the
second part is collected according to new metrics in a leading
international company in Turkey. The accuracy of the selected
metrics and the data samples are verified using statistical techniques.
The model presented here is based on Multi-Layer Perceptron
(MLP). Another difficulty associated with the cost estimation studies
is the fact that the data collection requires time and care. To make a
more thorough use of the samples collected, k-fold, cross validation
method is also implemented. It is concluded that, as long as an
accurate and quantifiable set of metrics are defined and measured
correctly, neural networks can be applied in software cost estimation
studies with success
Abstract: In this paper, a class of generalized bi-directional associative memory (BAM) neural networks with mixed delays is investigated. On the basis of Lyapunov stability theory and contraction mapping theorem, some new sufficient conditions are established for the existence and uniqueness and globally exponential stability of equilibrium, which generalize and improve the previously known results. One example is given to show the feasibility and effectiveness of our results.
Abstract: The protection of parallel transmission lines has been a challenging task due to mutual coupling between the adjacent circuits of the line. This paper presents a novel scheme for detection and classification of faults on parallel transmission lines. The proposed approach uses combination of wavelet transform and neural network, to solve the problem. While wavelet transform is a powerful mathematical tool which can be employed as a fast and very effective means of analyzing power system transient signals, artificial neural network has a ability to classify non-linear relationship between measured signals by identifying different patterns of the associated signals. The proposed algorithm consists of time-frequency analysis of fault generated transients using wavelet transform, followed by pattern recognition using artificial neural network to identify the type of the fault. MATLAB/Simulink is used to generate fault signals and verify the correctness of the algorithm. The adaptive discrimination scheme is tested by simulating different types of fault and varying fault resistance, fault location and fault inception time, on a given power system model. The simulation results show that the proposed scheme for fault diagnosis is able to classify all the faults on the parallel transmission line rapidly and correctly.