Wafer bin map recognition using a neural network approach

Although the fabrication of modern integrated circuits uses highly automatic and precisely controlled operations, equipment malfunctions or process drifts are still inevitable owing to the high complexity involved in the hundreds of processing steps. To detect the existence of these problems at the earliest stage, some important analytical tools must be applied. Among them is wafer bin map analysis. When the bin map exhibits specific patterns, it is usually a clue that equipment problems or process variations have occurred. The aim was to develop an intelligent system that could automatically recognize wafer bin map patterns and aid in the diagnosis of failure causes. A neural network architecture named Adaptive Resonance Theory Network 1 was adopted for the purpose. Actual data collected from a semiconductor manufacturing company in Taiwan were used for system verification. Experimental results show that with an adequate parameter, the neural network can successfully recognize and distinguish random and systematic wafer bin map patterns.     

Calibration and verification of DEM parameters for dynamic particle flow conditions using a backpropagation neural network

The Discrete Element Method (DEM) requires input parameters to be calibrated and validated in order to accurately model the physical process being simulated. This is typically achieved through experiments that examine the macroscopic behavior of particles, however, it is often difficult to efficiently and accurately obtain a representative parameter set. In this study, a method is presented to identify and select a set of DEM input parameters by applying a backpropagation (BP) neural network to establish the non-linear relationship between dynamic macroscopic particle properties and DEM parameters. Once developed and trained, the BP neural network provides an efficient and accurate method to select the DEM parameter set. The BP neural network can be developed and trained for one or more laboratory calibration experiments, and be applied to a wide range of bulk materials under dynamic flow conditions.     

Incorporating pattern recognition capability in a flexible vibratory bowl feeder using a neural network

Programmable parts feeders that can orientate most of the parts of one or more part families, with short changeover times from one part to the next, are highly sought after in batch production. This study investigates a suitable neural-network-based pattern recognition algorithm for the recognition of parts in a programmable vibratory bowl feeder. Three fibre-optic sensors were mounted on a vibratory bowl feeder to scan the surface of each feeding part. The scanned signatures were used as the input for the different neural network models. The performances of ARTMAP, ART2 and backpropagation neural network models were compared. The results showed that, among the three models, ARTMAP is deemed to be superior, based on the criteria of learning speed, high generalization and flexibility. The better performance obtained with the ARTMAP neural network is mainly the result of its online training and supervised learning capabilities.     

Colored pattern recognition with a neural network model

An algorithm for multicolored pattern recognition is proposed. A scheme for recognizing patterns encoded in three basic colors, i.e., red, green, and blue, is presented. This scheme can be implemented optically with grating structures. Another advantage of this scheme is its capability of pattern recognition with gray levels. This can be accomplished by coding gray levels with different colors.     

An efficient gait recognition based on a selective neural network ensemble

The neural network ensemble is a learning paradigm where a collection of neural networks is trained for the same task. Generally, the ensemble shows better generalization performance than a single neural network. In this article, a selective neural network ensemble is applied to gait recognition. The proposed method selects some neural network based on the minimization of generalization error. Since the selection rule is directly incorporated into the cost function, we can obtain adequate component networks to constitute an ensemble. Experiments are performed with the NLPR database to show the performance of the proposed algorithm. © 2008 Wiley Periodicals, Inc. Int J Imaging Syst Technol, 18, 237–241, 2008; Published online in Wiley InterScience (www.interscience.wiley.com).     

64-channel correlator implementing a Kohonen-like neural network for handwritten-digit recognition

We present an optical implementation of an improved version of the Kohonen map neural network applied to the recognition of handwritten digits taken from a postal code database. Improvements result from the introduction of supervision during the learning stage, a technique that also simplifies the map layer labeling. The experimental implementation is based on a frequency-multiplexed raster computer-generated hologram used to realize the required N(4) interconnection. The setup is shown to be equivalent to a 64-channel correlator. Computer simulations are used to study various detection and classification procedures. The results of the optical experiments, obtained with binary phase computer-generated holograms, are presented and shown to be in excellent agreement with the simulations.     

Cell formation using a Fuzzy Min-Max neural network

This paper proposes the application of a Fuzzy Min-Max neural network for part family formation in a cellular manufacturing environment. Once part families have been formed, a minimum cost flow model is used to form the corresponding machine cells. For simplicity, the input data are in the form of a binary part- machine incidence matrix, although the algorithm can work with an incidence matrix with continuous values. The application of Fuzzy Min-Max is interpreted in physical terms and compared with a related neural network applied previously for cell formation, the Fuzzy ART network. Both neural networks have similarities and differences that are outlined. The algorithms have been programmed and applied to a large set of problems from the literature. Fuzzy Min-Max generally outperforms Fuzzy ART, and the computational times are small and similar in both algorithms.     

Adaptive split spectrum processing using a neural network

The polarity thresholding algorithm for split spectrum processing (SSP) is known to work well once properly tuned. However, there are several problems related to the finding of the right split parameters such as the number of filters and the information carrying spectral range. Here we show that the polarity thresholding method can be formulated as a multilayer perceptron (MLP) neural network with binary neurons and binary input signals operating in feedforward mode. Then the method is generalized to process nonbinary data using an adaptive MLP with graded neurons. Experiments with real ultrasonic NDE signals are presented using the conventional backpropagation optimization algorithm (BP) and a second order optimization method (BFGS) with exact line search. Finally, alternative adaptive algorithms based on a decomposition of the network into single neurons or linear discriminants are briefly discussed.     

Boundary defect recognition using neural networks

This research presents schemes for automated visual inspection for boundary defects and classification using neural networks. An efficient method for representing circular boundaries is proposed utilizing a curvature and circular fitting algorithm. For classification, two types of neural network modelling schemes are established. First, a multi-layer perceptron is discussed for defect classification problems. Second, a Hopfield network is modelled to be used for continuous-type variables by a minimizing energy function. Extensive tests are conducted on the casting parts, then the results of neural networks are compared with those of traditional pattern classifiers.     

Improving EEG signal peak detection using feature weight learning of a neural network with random weights for eye event-related applications

The optimization of peak detection algorithms for electroencephalogram (EEG) signal analysis is an ongoing project; previously existing algorithms have been used with different models to detect EEG peaks in various applications. However, none of the existing techniques perform adequately in eye event-related applications. Therefore, we aimed to develop a general procedure for eye event-related applications based on feature weight learning (FWL), through the use of a neural network with random weights (NNRW) as the classifier. The FWL is performed using a particle swarm optimization algorithm, applied to the well-studied Dumpala, Acir, Liu and Dingle peak detection models, where the associated features are considered as inputs to the NNRW with and without FWL. The combination of all the associated features from the four models is also considered, as a comprehensive model for validation purposes. Real EEG data recorded from two channels of 20 healthy volunteers were used to perform the model simulations. The data set consisted of 40 peaks arising in the frontal eye field in association with a change of horizontal eye gaze direction. It was found that the NNRW in conjunction with FWL has better performance than NNRW alone for all four peak detection models, of which the Dingle model gave the highest performance, with 74% accuracy.     

磁流变阻尼器的性能试验与神经网络建模

磁流变阻尼器是一种新型的智能振动控制装置.通过磁流变阻尼器的性能试验,研究了在不同电流输入下阻尼力-位移、阻尼力-速度之间的关系,分析了摩擦型磁流变阻尼器的主要特点.采用BP神经网络,建立了磁流变阻尼器的正向模型和逆向模型.仿真结果显示,神经网络模型能准确地预测磁流变阻尼器的阻尼力和控制电流,证明该方法的有效性.与已有的模型相比,具有精度高,计算简便等特点.     

Research on citation mention times and contributions using a neural network

Scientometrics - Changes in patterns of collaboration between Russian universities after the commencement of the Russian university excellence initiative (Project 5-100) are studied in this paper....     

基于神经网络的在线计算机图像识别系统

以二值型自适应共振理论（ART－1）神经网络为识别核心设计了一个应用于生产流水线的计算机识别系统,它可以对生产线上的零件和产品的文字和符号进行实时识别,作自动记录。该系统具有学习和识别速度快、识别率高（＞96％）,可以灵活改变识别对象,应用范围广等特点。     

Modelling of plasma etching using a generalized regression neural network

Plasma etching was modelled by using a generalized regression neural network (GRNN). The etching process was characterized with a statistical experimental design. Three etch responses were modelled, which include two etch rates of aluminium and silica and etching profile. GRNN prediction ability was optimized as a function of training factor. Three types of models were constructed depending on the type of prepared data. Type I model corresponds to the model constructed with the original, non-classified data. Type II and III models were built for the classified data without and with the control of data interface, respectively. Compared to type I models, type II models for two etch rates demonstrated more than 25% improvement. By the control of data interface, type III models exhibited more than 15% improvement over type II models. Classification-based models in conjunction with data control thus illustrated much improved prediction of GRNN over those for non-classified models.     

Remote sensing image segmentation using a Kalman filter-trained neural network

This article describes the application of a neural network to the segmentation of remote sensing images of multispectral SPOT and fully polarimetric SAR data. The structure of the network is a modified multilayer perceptron and is trained by the Kalman filter theory. The internal activity of the network is a nonlinear function, while the function at output layer is linearized through the use of a polynomial basis function, thus allowing us employ the theory of Kalman filtering as the learning rule. The network is therefore called the dynamic learning (DL) neural network. It is found that, when applied to SPOT and SAR data, the DL neural network gives a good segmentation results, while the learning rate is very promising compared to the standard backpropagation network and other fast-learning networks. In particular, for polarimetric SAR data, optimum polarizations for discriminating between different terrains are automatically built in through the use of the Kalman filter technique. The suitability and effectiveness of the proposed DL neural network to the segmentation of remote sensing images is demonstrated. © 1996 John Wiley & Sons, Inc.     

基于平移不变Zernike矩和模块化神经网络的掌纹识别方法

在对掌纹原始图像进行去噪、分割等预处理之后,利用平移不变的Zernike矩特征矢量（TIZMs）作为掌纹特征建立特征库,根据已知分类信息建立样本集。并将问题分解为多个小规模的两类问题,然后采用模块化神经网络（MNN）作为分类器进行掌纹识别。对香港理工大学的Polyu PalmprintDB数据库中的3200个掌纹进行实验,在响应时间和识别精度等方面获得了很好的结果。     

基于粗糙集和神经网络的联合模式识别系统

提出了一种划分属性离散区间的新方法．针对这种划分，提出一种约简和去噪的方法．随后，建立了粗糙集和LVQ神经网络的联合模式识别系统．最后，比较了用该系统和仅用神经网络进行识别的效果，证明了该方法的有效性．     

Efficient document-image super-resolution using convolutional neural network

Experiments performed by us using optical character recognizers (OCRs) show that the character level accuracy of the OCR reduces significantly with decrease in the spatial resolution of document images. There are real life scenarios, where high-resolution (HR) images are not available, where it is desirable to enhance the resolution of the low-resolution (LR) document image. In this paper, our objective is to construct a HR image, given a single LR binary image. The works reported in the literature mostly deal with super-resolution of natural images, whereas we try to overcome the spatial resolution problem in document images. We have trained and obtained a novel convolutional model based on neural networks, which achieves significant improvement in terms of the peak-signal-to-noise ratio (PSNR) of the reconstructed HR images. Using parametric rectified linear units, mean PSNR improvements of 2.32, 4.38, 6.43 and 8.92 dB have been achieved over those of LR input images of 50, 75, 100 and 150 dots per inch (dpi) resolution and average word level accuracy of almost 43%, 45% and 57% on 75 dpi Tamil, English and Kannada images, respectively.     

Recognition of control chart concurrent patterns using a neural network approach

Pattern recognition is an important issue in Statistical Process Control, as unnatural patterns exhibited by control charts can be associated with specific assignable causes adversely affecting the process. Neural network approaches to recognition of control chart patterns have been developed by several researchers in recent years, but to date these have been focused on recognition and analysis of single patterns such as sudden shifts, linear trends or cyclic patterns. This paper investigates the detection of concurrent patterns where more than one pattern exists simultaneously. The topology and training of a Back-Propagation Network (BPN) system is described. Extensive performance evaluation has been carried out using simulated data to develop a range of average run length-related performance indices, including new performance indices that are proposed to describe concurrent patterns recognition performance. Two evaluation scenarios were evaluated: in the first, unnatural patterns are already present; while in the second, patterns may appear progressively at any time. Numerical results are provided that indicate that the pattern recognizer can perform very well in the first scenario, while it performs effectively but not without deficiencies for some specific pattern combinations in the second evaluation approach. Limitations and potential improvements in the concurrent pattern recognition scheme are also discussed.