Detection and classification of buried dielectric anomalies usingneural networks-further results

The development of a neural network-based detection and classification system for use with buried dielectric anomalies is the main focus of this paper. Several methods of data representation are developed to study their effects on the trainability and generalization capabilities of the neural networks. The method of Karhonen-Loeve (KL) transform is used to extract energy dependent features and to reduce the dimensionality of the weight space of the original data set. To extract the shape-dependent features of the data, another data preprocessing method known as Zernike moments is also studied for its use in the detector/classifier system. The effects of different neural network paradigms, architectural variations, and selection of proper training data on detection and classification rates are studied. Simulation results for nylon and wood targets indicate superior performance when compared to conventional schemes     

An Efficient Internet Traffic Classification System Using Deep Learning for IoT

Internet of Things (IoT) defines a network of devices connected to the internet and sharing a massive amount of data between each other and a central location. These IoT devices are connected to a network therefore prone to attacks. Various management tasks and network operations such as security, intrusion detection, Quality-of-Service provisioning, performance monitoring, resource provisioning, and traffic engineering require traffic classification. Due to the ineffectiveness of traditional classification schemes, such as port-based and payload-based methods, researchers proposed machine learning-based traffic classification systems based on shallow neural networks. Furthermore, machine learning-based models incline to misclassify internet traffic due to improper feature selection. In this research, an efficient multilayer deep learning based classification system is presented to overcome these challenges that can classify internet traffic. To examine the performance of the proposed technique, Moore-dataset is used for training the classifier. The proposed scheme takes the pre-processed data and extracts the flow features using a deep neural network (DNN). In particular, the maximum entropy classifier is used to classify the internet traffic. The experimental results show that the proposed hybrid deep learning algorithm is effective and achieved high accuracy for internet traffic classification, i.e., 99.23%. Furthermore, the proposed algorithm achieved the highest accuracy compared to the support vector machine (SVM) based classification technique and k-nearest neighbours (KNNs) based classification technique.     

Rule Extraction from Trained Artifical Neural Networks

The neural network rule extraction problem is aimed at obtaining rules from an arbitrarily trained artificial neural network. Recently there have been several approaches to rule extraction. Approaches to rule extraction implement a priori knowledge of data or rule requirements into neural networks before the rules are extracted. Although this may lead to a simplified final phase of acquitting the rules from particular type of neural networks, it limits the methodologies for general-purpose use. This article approaches the neural network rule extraction problem in its essential and general form. Preference is given to multilayer perceptron networks (MLP networks) due to their universal approximation capabilities. The article establishes general theoretical grounds for rule extraction from trained artificial neural networks and further focuses on the problem of crisp rule extraction. The problem of crisp rule extraction from trained MLP networks is first approached on theoretical level. Present ed theoretical results state conditions guaranteeing equivalence between classification by an MLP network and crisp logical formalism. Based on the theoretical results an algorithm for crisp rule extraction, independent of training strategy, is proposed. The rule extraction algorithm can be used even in cases where the theoretical conditions are not strictly satisfied; by offering an approximate classification. An introduced rule extraction algorithm is experimentally demonstrated.

     

Neural Networks in Applied Statistics

Artificial neural networks are computer algorithms or computer programs derived in part from attempts to model the activity of nerve cells. They have been applied to pattern recognition, classification, and optimization problems in the physical and chemical sciences, as well as in other fields. We introduce the principles of the multilayer feedforward network that is among the most commonly used neural networks in practical problems. The relevance of neural network models for the applied statistician is considered using a time series prediction problem as an example. The multilayer feedforward neural network uses a nonlinear function of the predictors to obtain predictions for future time series values. We illustrate the considerations involved in specifying a neural network model and evaluate the accuracy of neural network models relative to the accuracy obtained using other computer-intensive, nonmodel-based techniques.     

Control chart pattern recognition using learning vector quantization networks

Pattern recognition systems using neural networks for discriminating between different types of control chart patterns are discussed. A class of pattern recognizers based on the Learning Vector Quantization (LVQ) network is described. A procedure to increase the classification accuracy and decrease the learning time for LVQ networks is presented. The results of control chart pattern recognition experiments using both existing LVQ networks and an LVQ network implementing the proposed procedure are given.     

基于径向基函数神经网络的滚动轴承故障模式的识别

径向基函数（RBF）神经网络是一种3层前馈性神经网络，它具有较强的函数逼近能力和分类能力。鉴于径向基函数神经网络的优点，在对滚动轴承振动信号特征分析的基础上，提出了采用时序方法对其建立AR模型，利用AR模型参数建立径向基函数神经网络，并用该网络对滚动轴承的故障模式进行了识别。理论和试验证明了该方法的有效性，且具有较高的识别精度。     

Adaptive resonance theory-based neural algorithms for manufacturing process quality control

The demand for quality products in industry is continuously increasing. To produce products with consistent quality, manufacturing systems need to be closely monitored for any unnatural deviation in the state of the process. Neural networks are potential tools that can be used to improve the analysis of manufacturing processes. Indeed, neural networks have been applied successfully for detecting groups of predictable unnatural patterns in the quality measurements of manufacturing processes. The feasibility of using Adaptive Resonance Theory (ART) to implement an automatic on-line quality control method is investigated. The aim is to analyse the performance of the ART neural network as a means for recognizing any structural change in the state of the process when predictable unnatural patterns are not available for training. To reach such a goal, a simplified ART neural algorithm is discussed then studied by means of extensive Monte Carlo simulation. Comparisons between the performances of the proposed neural approach and those of well-known SPC charts are also presented. Results prove that the proposed neural network is a useful alternative to the existing control schemes.     

A Novel Combinational Convolutional Neural Network for Automatic Food-Ingredient Classification

With the development of deep learning and Convolutional Neural Networks (CNNs), the accuracy of automatic food recognition based on visual data have significantly improved. Some research studies have shown that the deeper the model is, the higher the accuracy is. However, very deep neural networks would be affected by the overfitting problem and also consume huge computing resources. In this paper, a new classification scheme is proposed for automatic food-ingredient recognition based on deep learning. We construct an up-to-date combinational convolutional neural network (CBNet) with a subnet merging technique. Firstly, two different neural networks are utilized for learning interested features. Then, a well-designed feature fusion component aggregates the features from subnetworks, further extracting richer and more precise features for image classification. In order to learn more complementary features, the corresponding fusion strategies are also proposed, including auxiliary classifiers and hyperparameters setting. Finally, CBNet based on the well-known VGGNet, ResNet and DenseNet is evaluated on a dataset including 41 major categories of food ingredients and 100 images for each category. Theoretical analysis and experimental results demonstrate that CBNet achieves promising accuracy for multi-class classification and improves the performance of convolutional neural networks.     

ART1神经网络在基于CBR概念设计中的应用

提出一种机构编码方法,可以分别对机构的运动功能属性和功能质量属性进行二进制编码,在此基础上,深入探讨了ART1神经网络在机构实例的分类和决策中的应用并对其算法进行了改进．与其它神经网络相比,ART1网络的优点在于可以实现机构动态分类和决策,设计者可根据自己的意图,通过调节警戒参数得到满意或最优的设计方案．实例表明,该方法合理可行．     

Neural networks applied to diffraction-pattern sampling

While diffraction-pattern sampling has been widely applied in the classification of patterns, still its usage has been limited somewhat by the need to devise rather sophisticated algorithms. In this paper we describe sorting or classification of a variety of patterns with commercially available neural-network software together with the ring-wedge photodetector to supply optical transform data for the input neurons. With this combination of neural networks and diffraction-pattern sampling it is no longer necessary to write specialized software. The training and testing methodology is carried out for this new system, and excellent results are obtained for sorting thumbprints. In sorting thumbprints the neural network can be trained for orientation-independent or wide-scale size-independent classifications by use of ring-only or wedge-only input neurons, respectively. Separate experiments are described for the sorting of particulates. Again, these are cases in which writing appropriate software based on diffraction theory would be extremely difficult. Two interesting novel neural networks are obtained: one is for real-time control of a submicrometer colloidal suspension of CdS, and the second is for concentration measurements of 2.02-μm polyvinyltoulene spheres in methyl alcohol. Widespread new applications are predicted for this hybrid system that combines diffraction-pattern sampling and the neural network.     

An exploratory study of a neural network approach for reliability data analysis

The results of this paper show that neural networks could be a very promising tool for reliability data analysis. Identifying the underlying distribution of a set of failure data and estimating its distribution parameters are necessary in reliability engineering studies. In general, either a chi-square or a non-parametric goodness-of-fit test is used in the distribution identification process which includes the pattern interpretation of the failure data histograms. However, those procedures can guarantee neither an accurate distribution identification nor a robust parameter estimation when small data samples are available. Basically, the graphical approach of distribution fitting is a pattern recognition problem and parameter estimation is a classification problem where neural networks have been proved to be a suitable tool. This paper presents an exploratory study of a neural network approach, validated by simulated experiments, for analysing small-sample reliability data. A counter-propagation network is used in classifying normal, uniform, exponential and Weibull distributions. A back-propagation network is used in the parameter estimation of a two-parameter Weibull distribution.     

Photon-limited image classification with a feedforward neural network

An application of neural networks to the classification of photon-limited images is reported. A three-level feedforward network architecture is employed in which the input units of the network correspond to the pixels of a two-dimensional image. The network is trained in a minicomputer by the use of the backpropagation technique. The statistics of the network components are analyzed, resulting in a method by which the probability of correct classification of a given input image can be calculated. Photon-limited images of printed characters are obtained with a photon-counting camera and are classified. The experimental results are in excellent agreement with theoretical predictions.     

Image texture classification using wavelet based curve fitting and probabilistic neural network

This article describes a new approach for image texture classification based on curve fitting of wavelet domain singular values and probabilistic neural networks. Image textures are wavelet packet transformed and singular value decomposition is then employed on subband coefficient matrices after introducing non‐linearity. Lower singular values are truncated based on energy distribution to effectively classify textures in the presence of noise. The selected singular values are fitted to the exponential curve. The model parameters are estimated using population‐sample analogues method and the parameters are used for performing classification. A modified form of probabilistic neural network (PNN) called weighted PNN (WPNN) is employed for performing the classification. Compared to probabilistic neural network, WPNN includes weighting factors between pattern layer and summation layer of the PNN. Performance of the approach is compared with model based and feature based methods in terms of signal to noise ratio and classification rate. Experimental results prove that the proposed approach gives better classification rate under noisy environment. © 2007 Wiley Periodicals, Inc. Int J Imaging Syst Technol, 17, 266–275, 2007     

High impedance fault detection in power distribution networks using time?frequency transform and probabilistic neural network

An intelligent approach for high impedance fault (HIF) detection in power distribution feeders using advanced signal-processing techniques such as time-time and time-frequency transforms combined with neural network is presented. As the detection of HIFs is generally difficult by the conventional over-current relays, both time and frequency information are required to be extracted to detect and classify HIF from no fault (NF). In the proposed approach, S- and TT-transforms are used to extract time-frequency and time-time distributions of the HIF and NF signals, respectively. The features extracted using S- and TT-transforms are used to train and test the probabilistic neural network (PNN) for an accurate classification of HIF from NF. A qualitative comparison is made between the HIF classification results obtained from feed forward neural network and PNN with same features as inputs. As the combined signal-processing techniques and PNN take one cycle for HIF identification from the fault inception, the proposed approach was found to be the most suitable for HIF classification in power distribution networks with wide variations in operating conditions.     

用概率社会网络进行结构损伤位置识别

在不计测量误差情况下,神经网络能够成功地识别损伤位置及其程度,但在测量噪声影响下,神经网络的损伤识别效果则比较差,考虑到基于多变量模式分类的概率神经网络具有处理受噪声污染的测试数据的能力,本文将可能的损伤位置作为模式类,利用概率神经网络的分类能力来识别结构的损,地对两个算例,一个六层框架和一个两层框架进行数值模拟分析,并将概率神经网络与BP网络进行了比较,结果表明,概率神经网络具有更好的识别效果,是一种很有潜力的结构损伤位置识别方法。     

Ground Nephogram Recognition Algorithm Based on Selective Neural Network Ensemble

In view of the low accuracy of traditional ground nephogram recognition model, the authors put forward a k-means algorithm-acquired neural network ensemble method, which takes BP neural network ensemble model as the basis, uses k-means algorithm to choose the individual neural networks with partial diversities for integration, and builds the cloud form classification model. Through simulation experiments on ground nephogram samples, the results show that the algorithm proposed in the article can effectively improve the Classification accuracy of ground nephogram recognition in comparison with applying single BP neural network and traditional BP AdaBoost ensemble algorithm on classification of ground nephogram.     

Evaluation of neural network models with generalized sensitivity analysis

A sensitivity analysis method for discovering characteristic features of the input data using neural network classification models has been devised. The sensitivity is the gradient of the neural network model response function, and because neural network models are nonlinear, the gradient depends on the point where it is evaluated. Two criteria are used for measuring the sensitivity. The first criterion calculates the sensitivity or gradient of the neural network output with respect to the average of the objects that comprise each class. The second criterion measures the average sensitivity of the class objects. The sensitivity analysis was applied to temperature-constrained cascade correlation network models and evaluated with sets of synthetic data and experimental mobility spectra. The neural network models were built using temperature-constrained cascade correlation networks (TCCCNs). A weight constraint was devised for the output units of the network models. This method implements weight decay with conjugate gradient training and yields more sensitive neural network models. Temperature-constrained hidden units furnish more sensitive network models than networks without constraints. By comparing the sensitivities of the class mean input and the mean sensitivity for all the inputs of a class, the individual input variables may be assessed for linearity. If these two sensitivities for an input variable differ by a constant factor, then that variable is modeled by a simple linear relationship. If the two sensitivities vary by a nonconstant scale factor, then the variable is modeled by higher order functions in the network. The sensitivity method was used to diagnose errors in the training data, and the test for linearity indicated a TCCCN architecture that had better predictability.     

Determination of closed form solution for acceptance sampling using ANN.

Tabled sampling schemes such as MIL-STD-105D offer limited flexibility to quality control engineers in designing sampling plans to meet specific needs. We describe a closed form solution to determine the AQL indexed single sampling plan using an artificial neural network (ANN). To determine the sample size and the acceptance number, feed-forward neural networks with sigmoid neural function are trained by a back propagation algorithm for normal, tightened, and reduced inspections. From these trained ANNs, the relevant weight and bias values are obtained. The closed form solutions to determine the sampling plans are obtained using these values. Numerical examples are provided for using these closed form solutions to determine sampling plans for normal, tightened, and reduced inspections. The proposed method does not involve table look-ups or complex calculations. Sampling plan can be determined by using this method, for any required acceptable quality level and lot size. Suggestions are provided to duplicate this idea for applying to other standard sampling table schemes.     

Artificial neural networks for the automated detection of trichloroethylene by passive Fourier transform infrared spectrometry

Artificial neural networks are applied to the automated classification of trichloroethylene (TCE) signatures from passive Fourier transform infrared remote sensing interferogram data. Through the use of three data collection methods, a combination of laboratory and field data is acquired that allows the methodology to be evaluated under a variety of infrared background conditions and in the presence of potentially interfering compounds such as sulfur hexafluoride, methyl ethyl ketone, acetone, carbon tetrachloride, and ammonia. To maximize the computational efficiency of the network optimization, experimental design techniques are employed to develop a training protocol for the network that takes into account the relationships among five variables that are related either to the network architecture or to the training process. This protocol is implemented for the case of a back-propagation neural network (BNN) and is used to develop an optimized network for the detection of TCE. The classification performance of the network is assessed by comparing both TCE detection capabilities and false detection rates to similar classification results obtained with the technique of piecewise linear discriminant analysis (PLDA). When applied to prediction data withheld from the optimization of both the BNN and PLDA algorithms, the BNN method is observed to outperform PLDA overall, with TCE detection rates in excess of 99% and false detection rates less than 0.5%.     

Chemometric classification of several olive cultivars from Trás-os-Montes region (northeast of Portugal) using artificial neural networks

This work aimed to use artificial neural networks for fruit classification according to olive cultivar, as a tool to guarantee varietal authenticity. So, 70 samples, each one containing, in general, 40 olives, belonging to the six most representative olive cultivars of Trás-os-Montes region (Cobrançosa, Cordovil, Madural, Negrinha de Freixo, Santulhana and Verdeal Transmontana) were collected in different groves and during four crop years. Five quantitative morphological parameters were evaluated for each fruit and endocarp, respectively. In total, ten biometrical parameters were used together with a multilayer perceptron artificial neural network allowing the implementation of a classification model. Its performance was compared with that obtained using linear discriminant analysis. The best results were obtained using artificial neural networks. In fact, the external validation procedure for linear discriminant analysis, using olive data from olive trees not included in the model development, showed an overall sensibility and specificity in the order of 70% and varying between 45 and 97% for the individual cultivars. On the other hand, the artificial neural network model was able to correctly classify the same unknown olives with a global sensibility and specificity around 75%, varying from 58 and 95% for each cultivar. The predictive results of the artificial neural network model selected was further confirmed since, in general, it correctly or incorrectly classified the unknown olive fruits in each one of the six cultivars studied with, respectively, higher and lower probabilities than those that could be expected by chance. The satisfactory results achieved, even when compared with previous published works, regarding olive cultivar's classification, show that the neural networks could be used by olive oil producers as a preventive and effective tool for avoiding adulterations of Protected Designation of Origin or monovarietal olive oils with olives of non-allowed cultivars.