Abstract: A driver identification system using finger-vein technology and an artificial neural network is presented in this paper. The principle of the proposed system is based on the function of near infra-red finger-vein patterns for biometric authentication. Finger-vein patterns are required by transmitting near infra-red through a finger and capturing the image with an infra-red CCD camera. The algorithm of the proposed system consists of a combination of feature extraction using Radon transform and classification using the neural network technique. The Radon transform can concentrate the information of an image in a few high-valued coefficients in the transformed domain. The neural networks are used to develop the training and testing modules. The artificial neural network techniques using radial basis function network and probabilistic neural network are proposed to develop a driver identification system. The experimental results indicated the proposed system performs well for personal identification. The average identification rate of PNN network is over 99.2%. The details of the image processing technique and the characteristic of system are also described in this paper.

Abstract: In this paper we present a method for improving the generalization performance of a radial basis function (RBF) neural network. The method uses a statistical linear regression technique which is based on the orthogonal least squares (OLS) algorithm. We first discuss a modified way to determine the center and width of the hidden layer neurons. Then, substituting a QR algorithm for the traditional Gram-Schmidt algorithm, we find the connected weight of the hidden layer neurons. Cross-validation is utilized to determine the stop training criterion. The generalization performance of the network is further improved using a bootstrap technique. Finally, the solution method is used to solve a simulation and a real problem. The results demonstrate the improved generalization performance of our algorithm over the existing methods.

Abstract: The problem of air pollution is a frequently recurring situation and its management has social and economic considerable effects. Given the interaction of the numerous factors involved in the raising of the atmospheric pollution rates, it should be considered that the relation between the intensity of emission produced by the polluting source and the resulting pollution is not immediate. The aim of this study was to realise and to compare two support decision system (neural networks and multivariate regression model) that, correlating the air quality data with the meteorological information, are able to predict the critical pollution events. The development of a back-propagation neural network is presented to predict the daily PM10 concentration 1, 2 and 3 days early. The measurements obtained by the territorial monitoring stations are one of the primary data sources; the forecasting of the major weather parameters available on the website and the forecasting of the Saharan dust obtained by the “Centro Nacional de Supercomputacion” website, satellite images and back trajectories analysis are used for the weather input data. The results obtained with the neural network were compared with those obtained by a multivariate linear regression model for 1 and 2 days forecasting. The relative root mean square error for both methods shows that the artificial neural networks (ANN) gives more accurate results than the multivariate linear regression model mostly for 1 day forecasting; moreover, the regression model used, in spite of ANN, failed when it had to fit spiked high values of PM10 concentration.

Abstract: A reliable prognostic model is very useful for industries to forecast equipment behaviors. The aim of this research is to verify the effectiveness of the neuro-fuzzy model in predicting the health condition of bearings. Simulation and an experiment have been carried out to verify the model, with results showing that the neuro-fuzzy model is a reliable and robust forecasting tool, and more accurate than a radial basis function network. In the experiment, vibration data collected from the equipment is used to predict the future condition.

Abstract: Novel direct adaptive robust state and output feedback controllers are presented for the output tracking control of a class of nonlinear systems with unknown system dynamics and disturbances. Both controllers employ a variable-structure radial basis function (RBF) network that can determine its structure dynamically to approximate unknown system dynamics. Radial basis functions are added or removed online in order to achieve the desired tracking accuracy and prevent to network redundancy. The raised-cosine RBF is employed to enable fast and efficient training and output evaluation of the RBF network. The direct adaptive robust output feedback controller is constructed by utilizing a high-gain observer to estimate the tracking error for the controller implementation. The closed-loop systems driven by the variable neural direct adaptive robust controllers are actually switched systems.

Abstract: Attractors and energy spectrum of neural structures Communication signal processing Complex-valued magnetic resonance images Complex-valued neural network Complex-valued recurrent neural networks Complex-valued symmetric radial basis function network Complex-valued time delay neural networks Flexible blind signal separation Global stability analysis Image reconstruction Magnetic resonance spectroscopy Model of the quantum harmonic oscillator Quantum neural networks Quaternionic neural networks Qubit neural networks.

Abstract: A numerical investigation on a technique for choosing an optimal shape parameter is proposed. Radial basis functions (RBFs) and their derivatives are used as interpolants in the asymmetric collocation radial basis method, for solving systems of partial differential equations. The shape parameter c in RBFs plays a major role in obtaining high quality solutions for boundary value problems. As c is a user defined value, inexperienced users may compromise the quality of the solution, often a problem of this meshless method. Here we propose a statistical technique to choose the shape parameter in radial basis functions. We use a cross-validation technique suggested by Rippa Adv Comput Math 11 (1999), 193–210 for interpolation problems to find a cost function Cost(c) that ideally has the same behavior as an error function. If that is the case, the parameter c that minimizes the cost function will be an optimal shape parameter, in the sense that it minimizes the error function. The form of the cost and error functions are analized for several examples, and for most cases the two functions have a similar behavior. The technique produced very accurate results, even with a small number of points and irregular grids. © 2009 Wiley Periodicals, Inc. Numer Methods Partial Differential Eq, 2010

Abstract: A major problem in medical science is attaining the correct diagnosis of disease in precedence of its treatment. This paper presents the diagnosis of thyroid disorders using Artificial Neural Networks (ANNs). The feed-forward neural network has been trained using three ANN algorithms; the Back propagation algorithm (BPA), the Radial Basis Function (RBF) Networks and the Learning Vector Quantization (LVQ) Networks. The networks are simulated using MATLAB and their performance is assessed in terms of factors like accuracy of diagnosis and training time. The performance comparison helps to find out the best model for diagnosis of thyroid disorders.

Abstract: Being a tool that assigns optical parameters used in interactive visualization, transfer functions (TF) have important effects on the quality of volume rendered medical images. Unfortunately, finding accurate TFs is a tedious and time consuming task because of the trade off between using extensive search spaces and fulfilling the physician's expectations with interactive data exploration tools and interfaces. By addressing this problem, we introduce a semi-automatic method for initial generation of TFs. The proposed method uses a self generating hierarchical radial basis function network to determine the lobes of a volume histogram stack (VHS) which is introduced as a new domain by aligning the histograms of slices of a image series. The new self generating hierarchical design strategy allows the recognition of suppressed lobes corresponding to suppressed tissues and representation of the overlapping regions which are parts of the lobes but can not be represented by the Gaussian bases in VHS. Moreover, approximation with a minimum set of basis functions provides the possibility of selecting and adjusting suitable units to optimize the TF. Applications on different CT/MR data sets show enhanced rendering quality and reduced optimization time in abdominal studies.

Abstract: The abrasive water jet machining process, a material removal process, uses a high velocity jet of water and an abrasive particle mixture. The estimation of appropriate values of the process parameters is an essential step toward an effective process performance. This has led to the development of numerous mathematical and empirical models. However, the complexity of the process confines the use of these models for limited operating conditions; e.g., some of these models are valid for special material combinations while others are based on the selection of only the most critical variables such as pump pressure, traverse rate, abrasive mass flow rate and others that affect the process. Furthermore, these models may not be generalized to other operating conditions. In this respect, a neural network approach has been proposed in this paper. Two neural network approaches, backpropagation and radial basis function networks, are proposed. The results from these two neural network approaches are compared with that from the linear and non-linear regression models. The neural networks provide a better estimation of the parameters for the abrasive water jet machining process.

Abstract: In this paper, we deal with the problem of time series prediction from a given set of input/output data. This problem consists of the prediction of future values based on past and/or present data. We present a new method for prediction of time series data using radial basis functions. This approach is based on a new efficient method of clustering of the centers of the radial basis function neural network; it uses the error committed in every cluster using the real output of the radial basis function neural network trying to concentrate more clusters in those input regions where the error is bigger and move the clusters instead of just the input values of the I/O data. This method of clustering, improves the performance of the time series prediction system obtained, compared with other methods derived from traditional algorithms.

Abstract: The number of users and their network utilization will enumerate the traffic of the network. The accurate and timely estimation of network traffic is increasingly becoming important in achieving guaranteed Quality of Service (QoS) in a wireless network. The better QoS can be maintained in the network by admission control, inter or intra network handovers by knowing the network traffic in advance. Here wireless network traffic is modeled as a nonlinear and nonstationary time series. In this framework, network traffic is predicted using neural network and statistical methods. The results of both the methods are compared on different time scales or time granularity. The Neural Network(NN) architectures used in this study are Recurrent Radial Basis Function Network (RRBFN) and Echo state network (ESN).The statistical model used here in this work is Fractional Auto Regressive Integrated Moving Average (FARIMA) model. The traffic prediction accuracy of neural network and statistical models are in the range of 96.4% to 98.3% and 78.5% to 80.2% respectively.

Abstract: This study proposes RBF Network hybrid learning with Particle Swarm Optimization (PSO) for better convergence, error rates and classification results. In conventional RBF Network structure, different layers perform different tasks. Hence, it is useful to split the optimization process of hidden layer and output layer of the network accordingly. RBF Network hybrid learning involves two phases. The first phase is a structure identification, in which unsupervised learning is exploited to determine the RBF centers and widths. This is done by executing different algorithms such as k-mean clustering and standard derivation respectively. The second phase is parameters estimation, in which supervised learning is implemented to establish the connections weights between the hidden layer and the output layer. This is done by performing different algorithms such as Least Mean Squares (LMS) and gradient based methods. The incorporation of PSO in RBF Network hybrid learning is accomplished by optimizing the centers, the widths and the weights of RBF Network. The results for training, testing and validation of five datasets (XOR, Balloon, Cancer, Iris and Ionosphere) illustrates the effectiveness of PSO in enhancing RBF Network learning compared to conventional Backpropogation.

Abstract: Detecting all kinds of intrusions efficiently is significant to network security. Radial basis function (RBF) neural network is a kind of feed forward neural network, which is widely employed as a real-time pattern classification. In RBF neural network, the center of radial basis function, the variance of radial basis of function and the weight have to be chosen. If they are not appropriately chosen, the RBF neural network may degrade validity and accuracy of modeling. Particle swarm optimization algorithm (PSO) is a member of the wide category of swarm intelligence methods to solve non-linear programming problems. PSO has proved to be competitive with genetic algorithm (GA) in parameter optimization. So PSO is used to optimize the RBF neural network parameters in this work. Therefore, the novel combination method based on RBF neural network and PSO (PSO-RBFNN) is adapted to network intrusion detection. The experimental results show that the proposed model is superior to the conventional RBF neural network.

Abstract: A Pharmaceutical formulation is composed of several formulation factors and process variables. Quantitative model based pharmaceutical formulation involves establishing mathematical relations between the formulation variables and the resulting responses, and optimizing the formulation conditions. In a formulation system involving several objectives, the desirable formulation conditions for one property may not always be desirable for other characteristics, thus leading to the problem of conflicting objectives. Therefore, efficient modeling and optimization techniques are needed to devise an optimal formulation system. In this work, a novel method based on radial basis function network (RBFN) is proposed for modeling and optimization of pharmaceutical formulations involving several objectives. This method has the advantage that it automatically configures the RBFN using a hierarchically self organizing learning algorithm while establishing the network parameters. This method is evaluated by using a trapidil formulation system as a test bed and compared with that of a response surface method (RSM) based on multiple regression. The simulation results demonstrate the better performance of the proposed RBFN method for modeling and optimization of pharmaceutical formulations over the regression based RSM technique.

Abstract: The standard approach to the solution of the radial basis function interpo- lation problem has been recognized as an ill-conditioned problem for many years. This is especially true when infinitely smooth basic functions such as multiquadrics or Gaussians are used with extreme values of their associated shape parameters. Various approaches have been described to deal with this phenomenon. These tech- niques include applying specialized preconditioners to the system matrix, changing the basis of the approximation space or using techniques from complex analysis. In this paper we present a preconditioning technique based on residual iteration of an approximate moving least squares quasi-interpolant that can be interpreted as a change of basis. In the limit our algorithm will produce the perfectly conditioned cardinal basis of the underlying radial basis function approximation space. Although our method is motivated by radial basis function interpolation problems, it can also be adapted for similar problems when the solution of a linear system is involved such as collocation methods for solving differential equations.

Abstract: The learning strategy of the radial basis function network (RBFN) commonly uses a hybrid learning process to identify the structure and then proceed to search the model parameters, which is a time-consuming procedure. We proposed an evolutionary way to automatically configure the structure of RBFN and search the optimal parameters of the network. The strategy can effectively identify an appropriate structure of the network by the orthogonal least squares algorithm and then systematically search the optimal locations of centres and the widths of their corresponding kernel function by the genetic algorithm. The proposed strategy of auto-configuring RBFN is first testified in predicting the future values of the chaotic Mackey-Glass time series. The results demonstrate the superiority, on both effectiveness and efficiency, of the proposed strategy in predicting the chaotic time series. We then further investigate the model's suitability and reliability in flood forecast. The Lan-Young River in north-east Taiwan is used as a case study, where the hourly river flow of 23 flood events caused by typhoons or storms is used to train and validate the neural networks. The back propagation neural network (BPNN) is also performed for the purpose of comparison. The results demonstrate that the proposed RBFN has much better performance than the BPNN. The RBFN not only provides an efficient way to model the rainfall-runoff process but also gives reliable and precise one-hour and two-hour ahead flood forecasts. Copyright © 2009 John Wiley & Sons, Ltd.

Abstract: Accurate weather predictions are important for planning our dayto-day activities. In recent years, a large literature has evolved on the use of artificial neural networks (ANNs) in many forecasting applications. Neural networks are particularly appealing because of their ability to model an unspecified non-linear relationship between weather variables. This paper evaluates three neural networks architectures with different training techniques, in this context: the popular multilayer perceptron (MLP), the radial basis function network (RBF) and feed forward neural networks which were trained by differential evolution algorithm. Different testing and training scenarios are presented. Those scenarios are designed to obtain the most suitable one for weather predication at different neural network architectures. Simulation results for each scenario demonstrate the effectives of both neural network architectures and its associated training algorithm.

Abstract: This paper presents a segmentation-based handwriting recognizer and the performance that it achieves on the numerical fields extracted from a large single-writer historical collection. Our recognizer has the particularity that it uses morphing during training: random elastic deformations are applied to fabricate synthetic training character patterns yielding an improved final recognition performance. Two different digit recognizers are evaluated, a multilayer perceptron (MLP) and radial basis function network (RBF), by plugging them into the same left-to-right Viterbi search framework with a tree organization of there cognition lexicon. We also compare with the performance obtained when no dictionary is used to constrain the recognition results.

Abstract: Marble quality classification is an important procedure generally performed by human experts. However, using human experts for classification is error prone and subjective. Therefore, automatic and computerized methods are needed in order to obtain reproducible and objective results. Although several methods are proposed for this purpose, we demonstrate that their performance is limited when dealing with diverse datasets containing a large number of quality groups. In this work, we test several feature sets and neural network topologies to obtain a better classification performance. During these tests, it is observed that different feature sets represent different subgroup(s) in a quality group rather than representing the whole group. Therefore, our approach is to use these features in a cascaded manner in which a quality group is classified by classifying all of its subgroups. We first realize this approach by using a two-stage cascaded network. Then, we design a hierarchical radial basis function network (HRBFN) in which correctly classified marble samples are taken out of the dataset and a different feature extraction method is applied to the remaining samples at each network level. The HRBFN system produces successful results for industrial applications and facilitates the desirable property of implementation in a quasi real-time manner.

Abstract: Radial basis function neural network (RBFNN) is widely used in nonlinear function approximation. One of the key issues in RBFNN modeling is to improve the approximation ability with samples as few as possible, so as to limit the network’s complexity. To solve this problem, a gradient-based sequential RBFNN modeling method is proposed. This method can utilize the gradient information of the present model to expand the sample set and refine the model sequentially, so as to improve the approximation accuracy effectively. Two mathematical examples and one practical problem are tested to verify the efficiency of this method.