By Topic

1995 Fourth International Conference on Artificial Neural Networks

26-28 Jun 1995

Filter Results

Displaying Results 1 - 25 of 90
  • Sleep apnoea analysis from neural network post-processing

    Publication Year: 1995, Page(s):427 - 432
    Cited by:  Papers (2)
    IEEE is not the copyright holder of this material | Click to expandAbstract | PDF file iconPDF (488 KB)

    This paper presents methods of analysis of electroencephalogram (EEG) signals using artificial neural networks, and subsequent methods of detection of obstructive sleep apnoea (OSA) from the neural network outputs. EEG signals are measurements of scalp potential differences arising from the brain's electrical activity. Gross changes in the human EEG occur between different types of sleep. Traditio... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Adaptive control of gasoline engine air-fuel ratio using artificial neural networks

    Publication Year: 1995, Page(s):274 - 278
    Cited by:  Papers (2)
    IEEE is not the copyright holder of this material | Click to expandAbstract | PDF file iconPDF (292 KB)

    Adaptive control is seen as playing an important role in meeting the ever tightening legislation on vehicle emissions and the requirement to maintain these low emission levels throughout the lifetime of the vehicle. Due to the highly non-linear air and fuel flow processes involved, adaptive control based on linear techniques is ineffectual. However, artificial neural networks (ANNs) offer the capa... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An improved hardware-realisable learning algorithm for pyramidal feed-forward pRAM based ANNs

    Publication Year: 1995, Page(s):495 - 498
    IEEE is not the copyright holder of this material | Click to expandAbstract | PDF file iconPDF (388 KB)

    Proposes a hardware-realisable training algorithm, modified from that proposed by Guan et al. (1992). Probabilistic random access memory (pRAM) based artificial neural networks (ANNs), trained using the improved algorithm (which lets the network itself decide the output coding it should use for classification), managed to easily overcome the hard learning problem facing architectures that contain ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Applications of linear weight neural networks to fingerprint recognition

    Publication Year: 1995, Page(s):139 - 142
    Cited by:  Papers (1)  |  Patents (1)
    IEEE is not the copyright holder of this material | Click to expandAbstract | PDF file iconPDF (508 KB)

    Fingerprints form an important aspect of evidence in criminal investigations in modern police work. However, the task of searching for a match from a scene-of-crime image (a mark or latent) to the files of prints taken from previous convicts can be labour-intensive. The new approach described in this paper uses a localised ridge direction determination which is generated by applying anisotropic fi... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A hybrid histogram and neural based call admission control for VBR video traffic

    Publication Year: 1995, Page(s):421 - 426
    IEEE is not the copyright holder of this material | Click to expandAbstract | PDF file iconPDF (560 KB)

    We propose a hybrid neural network (NN) approach to estimate cell loss rate of variable bit rate (VBR) video traffic for call admission control (CAC) in an ATM environment. Existing CAC algorithms, which are mostly based on the on-off model, do not appear to apply well to VBR video traffic. In reality, VBR video sources are not two-state on-off sources. Recently, a histogram based model for video ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A neural-network-based flexible assembly controller

    Publication Year: 1995, Page(s):268 - 273
    Cited by:  Papers (2)
    IEEE is not the copyright holder of this material | Click to expandAbstract | PDF file iconPDF (476 KB)

    A neural network-based learning algorithm has been developed for control of robotic peg-in-hole insertions. As simple insertions may comprise up to one-third of all assembly operations, this controller has direct application to automated flexible assembly. A neural network derived from the Cerebellar Model Articulation Controller (CMAC) utilising force feedback from a Cartesian robot is demonstrat... View full abstract»

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

    Publication Year: 1995
    IEEE is not the copyright holder of this material | Click to expandAbstract | PDF file iconPDF (124 KB)

    First Page of the Article
    View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Neural networks implementation with VLSI

    Publication Year: 1995, Page(s):489 - 494
    Cited by:  Papers (1)
    IEEE is not the copyright holder of this material | Click to expandAbstract | PDF file iconPDF (336 KB)

    This paper presents the design of a new and efficient winner-take-all (WTA) cell for the self-organising mapping (SOM) neuron. This cell is implemented in VLSI that provides both faster operation and a reduction in the number of transistors per cell compared to existing designs. The operation of the circuit is described and results of SPICE simulations are presented View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • High performance OCR with syntactic neural networks

    Publication Year: 1995, Page(s):133 - 138
    Cited by:  Papers (3)
    IEEE is not the copyright holder of this material | Click to expandAbstract | PDF file iconPDF (528 KB)

    This paper describes the application of a special type of syntactic neural network (SNN) to the recognition of hand-written digits. Importantly, it is shown that this class of SNN can be implemented to work at very high classification speeds (similar to that of an N-tuple classifier), but with higher classification accuracy when trained on enough data. Results are reported on the ESSEX and CEDAR d... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Neural processing in an electronic odour sensing system

    Publication Year: 1995, Page(s):415 - 420
    IEEE is not the copyright holder of this material | Click to expandAbstract | PDF file iconPDF (416 KB)

    The past ten years has seen a significant increase in activity in the application of multisensor arrays to odour classification and description. Much of this work has been based around systems consisting of a semiconductor gas sensor array, elementary signal conditioning and microcomputer based pattern recognition (P. Corcoran, 1993). Most interest in the area of pattern recognition has been in th... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Error functions, error signals, and conjugate gradient back propagation

    Publication Year: 1995, Page(s):76 - 81
    IEEE is not the copyright holder of this material | Click to expandAbstract | PDF file iconPDF (372 KB)

    We propose a three class taxonomy of error functions, based on the limit behaviour of the error signal. We classify four established error functions: the quadratic, Fahlman's Quickprop, entropy, and the exception error function. We introduce two new error functions, and benchmark all six on the N-2-N encoder. The two new functions found correct solutions faster and more reliably than the establish... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Robotic task level programming using neural networks

    Publication Year: 1995, Page(s):262 - 267
    IEEE is not the copyright holder of this material | Click to expandAbstract | PDF file iconPDF (500 KB)

    Robot programming is a difficult, complex and time consuming operation. It consists of three main stages, the definition of points/locations, program coding (including error planning) and finally program proving. Due to problems associated with each of these stages, alternative techniques are sought to reduce the programming duration, to simplify the programming complexity and to improve the succe... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Training neural networks to identify coding regions in genomic DNA

    Publication Year: 1995, Page(s):399 - 403
    Cited by:  Papers (1)  |  Patents (1)
    IEEE is not the copyright holder of this material | Click to expandAbstract | PDF file iconPDF (412 KB)

    The four nitrogenous bases of DNA spell out the recipes from which proteins are made. A gene typically contains five thousand or so bases but often only a small percentage of these are protein coding. Computer based prediction systems are increasingly relied upon as submissions to the major genetic databases are growing exponentially. Several systems exist to locate coding regions (exons) and nonc... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • How to modify Kohonen's self-organising feature maps for an efficient digital parallel implementation

    Publication Year: 1995, Page(s):86 - 91
    Cited by:  Papers (1)
    IEEE is not the copyright holder of this material | Click to expandAbstract | PDF file iconPDF (420 KB)

    Two new variants of Kohonen's self-organising feature maps based on batch processing are presented in this work. The motivation is related to the need of exploiting the hardware resources of neurocomputers based on systolic arrays. Ordering and convergence to asymptotic values for 1D maps and 1D continuous input and weight spaces are proved for both variants. Finally, simulations on uniform 2D dat... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Accountability of neural networks trained with “real world” data

    Publication Year: 1995, Page(s):218 - 222
    Cited by:  Papers (3)
    IEEE is not the copyright holder of this material | Click to expandAbstract | PDF file iconPDF (380 KB)

    In the real world neural networks have to be accountable. It is not sufficient to produce output data from a network without any information as to the quality of that data. The quality of the output data from any given neural network for a given generalisation point is dependent on the following: (1) the representative nature of the original training data in relation to the scope of the problem do... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Using self-organising maps to classify radar range profiles

    Publication Year: 1995, Page(s):335 - 340
    Cited by:  Papers (1)
    IEEE is not the copyright holder of this material | Click to expandAbstract | PDF file iconPDF (472 KB)

    A model based approach to radar range profile classification is presented, and it is shown to be equivalent to training a topographic mapping neural network (T. Kohonen, 1984) on each of the range profile categories to be classified. The topographic mapping method is basically a Euclidean distance method of classifying range profiles. However, because it is model based, it offers much more flexibi... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Exploiting multiple degrees of BP parallelism on the highly parallel computer AP1000

    Publication Year: 1995, Page(s):483 - 488
    Cited by:  Papers (4)  |  Patents (1)
    IEEE is not the copyright holder of this material | Click to expandAbstract | PDF file iconPDF (500 KB)

    During the few last years, several neurocomputers have been developed, but still general-purpose computers are an alternative to these special-purpose computers. This paper describes a mapping of the backpropagation (BP) learning algorithm onto a large 2D torus architecture. The parallel algorithm was implemented on a 512-processor AP1000 and evaluated using NETtalk and other applications. To obta... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Homomorphic graph matching using self-organising Hopfield network

    Publication Year: 1995, Page(s):59 - 64
    IEEE is not the copyright holder of this material | Click to expandAbstract | PDF file iconPDF (548 KB)

    In the past, the Hopfield network has been employed to solve pattern recognition problems by subgraph isomorphism which naturally constrains the scene to have at most one occurrence of any object model. Recently, the author proposed a novel programming procedure to generate a homomorphic mapping which enables simultaneous recognition of multiple instances of any particular object model in the scen... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Neural networks in dynamic process state estimation and non-linear predictive control

    Publication Year: 1995, Page(s):284 - 289
    Cited by:  Papers (1)
    IEEE is not the copyright holder of this material | Click to expandAbstract | PDF file iconPDF (540 KB)

    The much wider availability and power of computing systems, together with new theoretical research studies, is resulting in expanding areas of neural network application. It is particularly significant in these circumstances that the extremely important aspects involved in developing complex industrial process applications is emphasised, especially where safety critical perspectives are prominent.... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Probabilistic Fuzzy ARTMAP: an autonomous neural network architecture for Bayesian probability estimation

    Publication Year: 1995, Page(s):148 - 153
    Cited by:  Papers (5)
    IEEE is not the copyright holder of this material | Click to expandAbstract | PDF file iconPDF (520 KB)

    A hybrid utilisation of the Fuzzy ARTMAP (FAM) neural network and the Probabilistic Neural Network (PNN) is proposed for online learning and prediction tasks. FAM is used as an underlying clustering algorithm to classify the input patterns into different recognition categories during the learning phase. Subsequently, a non parametric probability estimation procedure in accordance with the PNN para... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Modelling the visual cortex using artificial neural networks for visual image reconstruction

    Publication Year: 1995, Page(s):127 - 132
    Cited by:  Papers (1)
    IEEE is not the copyright holder of this material | Click to expandAbstract | PDF file iconPDF (644 KB)

    Presents an artificial neural network model with some special neurons that are designed to function as the feature detectors found in the visual cortex, and applies it to the reconstruction of visual images. The model is a multilayer feedforward neural network. The neurons in the first hidden layer of the network are feature detectors of various scales and orientations. The connection strengths be... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Comparison between different neural network architectures for odour discrimination

    Publication Year: 1995, Page(s):410 - 414
    IEEE is not the copyright holder of this material | Click to expandAbstract | PDF file iconPDF (400 KB)

    With the attempt to develop an artificial olfactory system able to mimic the discrimination ability of the natural system, several artificial neural network architectures were considered and evaluated on the basis of their performances and similarities with the neurophysiological models of the biological system. The neural network architectures were analysed and tested with experimental data from ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Self-consistent training of a neural network with a step edge model for probabilistic edge labelling

    Publication Year: 1995, Page(s):98 - 103
    IEEE is not the copyright holder of this material | Click to expandAbstract | PDF file iconPDF (676 KB)

    Presents a robust neural network edge labelling strategy in which a network is trained with data from an imaging model of an ideal step edge. In addition to the Sobel operator, we employ preprocessing steps on image data to exploit the known invariances due to lighting variation and rotation, and so reduce the complexity of the mapping which the network has to learn. The composition of the trainin... View full abstract»

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

    Publication Year: 1995, Page(s):228 - 233
    IEEE is not the copyright holder of this material | Click to expandAbstract | PDF file iconPDF (432 KB)

    We present a method for validating implementations of connectionist algorithms by comparison with known versions. The method is dependent on a statistical model derived from work in multiversion programming techniques. We present example results which reveal that two variations of the backpropagation algorithm which are widely accepted as mere implementation variants do actually produce networks w... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • On-line learning using hierarchical mixtures of experts

    Publication Year: 1995, Page(s):347 - 351
    Cited by:  Papers (2)
    IEEE is not the copyright holder of this material | Click to expandAbstract | PDF file iconPDF (428 KB)

    In the hierarchical mixtures of experts (HME) framework, outputs from several function approximators specializing in different parts of the input space are combined. Fast learning algorithms derived from the expectation-maximization algorithm have previously been proposed, but they are predominantly for batch learning. In this paper, several online learning algorithms are developed for the HME. Th... View full abstract»

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