Scheduled System Maintenance
On Tuesday, September 26, IEEE Xplore will undergo scheduled maintenance from 1:00-4:00 PM ET.
During this time, there may be intermittent impact on performance. We apologize for any inconvenience.

Neural Networks for Signal Processing VI. Proceedings of the 1996 IEEE Signal Processing Society Workshop

4-6 Sept. 1996

Filter Results

Displaying Results 1 - 25 of 61
  • Neural Networks for Signal Processing VI. Proceedings of the 1996 IEEE Signal Processing Society Workshop

    Publication Year: 1996
    Request permission for commercial reuse | PDF file iconPDF (270 KB)
    Freely Available from IEEE
  • Speech enhancement based on extended Kalman filter and neural predictive hidden Markov model

    Publication Year: 1996, Page(s):302 - 310
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (430 KB)

    To represent the nonlinear and nonstationarity nature of speech, we assume that speech is the output of an NPHMM combining a neural network and hidden Markov model (HMM). The NPHMM is a nonlinear autoregressive process whose time-varying parameters are controlled by a hidden Markov chain. Given some speech data for training, the parameter of NPHMM is estimated by a learning algorithm based on the ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Author index

    Publication Year: 1996
    Request permission for commercial reuse | PDF file iconPDF (60 KB)
    Freely Available from IEEE
  • Neural network approaches for the extraction of the eigenstructure

    Publication Year: 1996, Page(s):23 - 32
    Cited by:  Papers (2)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (372 KB)

    A feedback neural network for eigen-decomposition of a positive semidefinite matrix is presented. In the paper, we have shown the stability and real-time eigen-decomposition computation ability of the proposed neural network. The network can go into the stable state in the magnitude of the circuit time constant. The output voltage of the net, under the smallest energy state, is just the eigenvecto... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Supervised learning for multilayered neural network with non-monotonic activation functions

    Publication Year: 1996, Page(s):13 - 22
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (356 KB)

    We describe the performance of multilayer neural network with hidden units of nonmonotonic activation functions. Our previous work has shown that the network was effective in improving two difficulties: a convergence to local minima and a slow learning speed for the exclusive-OR and the binary addition problems. The purpose of this paper is to evaluate the performance of the proposed network for m... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Recognition of handwritten similar Chinese characters by neural networks

    Publication Year: 1996, Page(s):320 - 329
    Cited by:  Papers (2)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (408 KB)

    This paper presents a multi-stage neural networks for the recognition of similar Chinese characters. In this research, the authors have developed a three stage recognition structure: 1) an overlapped c-means clustering algorithm to implement a coarse classifier; 2) a Bayesian decision based neural network as a fine classifier; and 3) a two-layered feedforward neural network for similar character r... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Design method for a pattern classifier suited to adaptation

    Publication Year: 1996, Page(s):263 - 272
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (396 KB)

    This paper describes a method for designing a pattern classifier that will perform well after it has been adapted to changes in input conditions. Considering the off-line (batch-mode) adaptation methods which are based on the transformation of classifier parameters, we formulate the problem of designing classifiers, and propose a method for training them. In the proposed training method, the class... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Snowfall and rainfall forecasting from the images of weather radar with artificial neural networks

    Publication Year: 1996, Page(s):473 - 481
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (424 KB)

    We discuss problems of the weather forecasting technique with artificial neural networks and describe some solutions. We show that the computational time for learning with an acceleration learning algorithm can be reduced about 10 percent. To overcome the problem of overtraining, the pruning method is introduced and the prediction error is decreased about 20 percent. Using the data obtained over a... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Adaptive orthogonal least squares learning algorithm for the radial basis function network

    Publication Year: 1996, Page(s):3 - 12
    Cited by:  Papers (2)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (400 KB)

    This paper presents an algorithm to select the parameters of a radial basis function network based on the orthogonal least squares (OLS) learning algorithm. To improve the OLS learning process, an additional procedure to modify the selected node's parameter during training is introduced. Using simulation results, we show that significant improvement to the selected model's performance can be achie... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A neural network trained microphone array system for noise reduction

    Publication Year: 1996, Page(s):311 - 319
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (412 KB)

    This paper presents a neural network based microphone array system, which is capable to continuously perform speech enhancement and adaptation to nonuniform quantization, such as A-law and μ-law. Such a quantizer is designed to increase the signal to quantization noise ratio (SQNR) for small amplitudes in telecommunications systems. The proposed method primarily developed for hand-free mobile t... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Adaptive online learning of optimal decision boundary using active sampling

    Publication Year: 1996, Page(s):253 - 262
    Cited by:  Papers (2)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (344 KB)

    An adaptive online learning method is presented to facilitate pattern classification using active sampling to identify optimal decision boundary for a stochastic oracle with a minimum number of training samples. The strategy of sampling at the current estimate of the decision boundary is shown to be optimal in the sense that the probability of convergence toward the true decision boundary at each ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Defect diagnosis of solder joints using fuzzy logic

    Publication Year: 1996, Page(s):502 - 509
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (412 KB)

    This paper describes how the methods of fuzzy logic can be used in the classification of solder joint inspection results. The inspection results of over 900 circuit boards have been analysed for the purpose of distinguishing the essential characters of defective and decent solder joints. Also the effect of the prevailing average solder amount of each component type on the decision making process h... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Earthquake early warning system using real-time signal processing

    Publication Year: 1996, Page(s):463 - 472
    Cited by:  Papers (6)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (616 KB)

    An earthquake warning system has been developed to provide a time series profile from which vital parameters such as time until strong shaking begins, intensity of shaking, and duration of shaking, can be derived. Interaction of different types of ground motion and changes in the elastic properties of geological media throughout the propagation path result in a highly nonlinear function. We use ne... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A stochastic natural gradient descent algorithm for blind signal separation

    Publication Year: 1996, Page(s):433 - 442
    Cited by:  Papers (2)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (340 KB)

    A new blind separation algorithm is derived based on minimizing the mutual information of the output of the de-mixing system using natural gradient descent method. The algorithm can be easily implemented on a neural network with data dependent activation functions. A new performance function which depends only on the output and the de-mixing matrix is introduced. The new performance function is ev... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An incremental learning method with relearning of recalled interfered patterns

    Publication Year: 1996, Page(s):243 - 252
    Cited by:  Papers (2)  |  Patents (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (496 KB)

    This paper presents a new incremental learning method for neural networks. If a neural network is trained to memorize novel patterns only by their presentation, the network will forget some patterns that have been already learnt. This problem is caused by the fact that the learning of novel patterns usually interfere in the internal representation corresponding to the old training patterns. In the... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An Arabic character recognition system using neural network

    Publication Year: 1996, Page(s):340 - 348
    Cited by:  Papers (2)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (288 KB)

    An optical character recognition system, which uses a multilayer perceptron classifier, is described. A new approach for the classification of Arabic characters is presented. The technique used is invariant to translation, scale and rotation. Present day artificial neural network (ANN) architecture for invariant character recognition is too complex for our present technology. An alternative proced... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An adaptive Kalman filter model for stable perception of visual space unaffected by eye movement

    Publication Year: 1996, Page(s):492 - 501
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (556 KB)

    A number of visual psychological-physical (PP) models, i.e., the saccadic suppression model, exist for explaining the phenomenon of visually stable perception in visual space regardless of eye movement (EM). However, most models ignore the visual system noise that is added between the retina and the high visual information processing system. This paper introduces a new visual model based on an ada... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A neural network controlled ATM switch

    Publication Year: 1996, Page(s):518 - 526
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (328 KB)

    The Banyan network is a kind of multistage interconnection networks (MINs), and is very popular to be used as a basic building block in ATM switch designs. The neural network scheduling of multistage interconnection networks (MINs) have been proposed in some papers, but its hardware implementation is complicated. In this paper we propose a high throughput ATM switch which consists of neural networ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Classification of seismic waveforms by integrating ensembles of neural networks

    Publication Year: 1996, Page(s):453 - 462
    Cited by:  Patents (2)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (592 KB)

    The problem considered is the discrimination between natural and artificial seismic events, based on their waveform recording. We build a classification environment consists of several ensembles of neural networks trained on boot-strap sample sets, using various data representations and architectures. The integration of the different ensembles is made in a nonconstant signal adaptive manner, using... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • 3-D heart contour delineation and motion tracking of ultrasound images using continuous distance transform neural networks

    Publication Year: 1996, Page(s):361 - 370
    Cited by:  Papers (2)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (548 KB)

    We apply the previously proposed continuous distance transform neural network (CDTNN) to effectively represent the 3-D endocardial (inner) and epicardial (outer) contours and track the motion of the left ventricle (principal pumping chamber) of the heart from ultrasound images. This CDTNN has many good properties as the conventional distance transforms which are suitable for 3-D object representat... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Unsupervised segmentation of multispectral images using hierarchical MRF model

    Publication Year: 1996, Page(s):381 - 390
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (480 KB)

    This paper proposes an Markov random field (MRF) model-based method for unsupervised segmentation of multispectral images, in which the intra-class correlation of multispectral data as well as the class correlation are taken into account. In this method a set of multispectral images is modeled by a hierarchical MRF model. The proposed segmentation method is an iterative method composed of paramete... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A method for estimating coding gain of subband filter considering higher order statistic

    Publication Year: 1996, Page(s):401 - 410
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (436 KB)

    In transform, or subband coding, a coding gain has been widely used for predicting coding efficiency and designing a subband filter. It is a useful criterion which represents performance of the transform, or the subband filter, for object coding. First of all, this paper points out that the estimated coding gain includes error when the coding object has non-normality, such as a signal operated by ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Blind separation of convolved sources based on information maximization

    Publication Year: 1996, Page(s):423 - 432
    Cited by:  Papers (78)  |  Patents (9)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (476 KB)

    Blind separation of independent sources from their convolutive mixtures is a problem in many real world multi-sensor applications. In this paper we present a solution to this problem based on the information maximization principle, which was proposed by Bell and Sejnowski (1995) for the case of blind separation of instantaneous mixtures. We present a feedback network architecture capable of coping... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Genetic programming techniques that evolve recurrent neural network architectures for signal processing

    Publication Year: 1996, Page(s):139 - 148
    Cited by:  Papers (2)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (416 KB)

    We propose a novel design paradigm for recurrent neural networks. This employs a two-stage genetic programming/simulated annealing hybrid algorithm to produce a neural network which satisfies a set of design constraints. The genetic programming part of the algorithm is used to evolve the general topology of the network, along with specifications for the neuronal transfer functions, while the simul... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An RNN-based noise estimation and likelihood compensation for noisy speech recognition

    Publication Year: 1996, Page(s):293 - 301
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (376 KB)

    In this paper, a novel integration of RNN and PMC (parallel model combination) is presented for noisy speech recognition. It first employs an RNN to make the noise/speech discrimination. Then, by viewing the RNN outputs as the membership functions of noise and speech, an online noise tracking is performed for noise estimation. Also, a confidence measure is defined to represent the degree of the re... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.