By Topic

Vision, Image and Signal Processing, IEE Proceedings -

Issue 4 • Date Aug 1994

Filter Results

Displaying Results 1 - 9 of 9
  • Contextual image labelling with a neural network

    Page(s): 238 - 244
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (472 KB)  

    A neural network with a multilayer perceptron architecture is shown to be capable of labelling the visible objects in colour images of urban and rural outdoor scenes. The two problems of segmentation and recognition are separated by using `ideal' segmentations, allowing the performance of the recognition method to be studied independently of the effects of using an imperfect real segmentation process. A label clustering transformation is proposed and shown to cause a significant increase in the expected classification accuracy of the network. The deletion of the contextual features from the feature vector is shown to degrade the performance of the network. Measurements of the generalisation performance on unseen test data show that, on average, the system correctly recognises approximately 72% of the area of these images View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Original approach for the localisation of objects in images

    Page(s): 245 - 250
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (432 KB)  

    An original approach is presented for the localisation of objects in an image which approach is neuronal and has two steps. In the first step, a rough localisation is performed by presenting each pixel with its neighbourhood to a neural net which is able to indicate whether this pixel and its neighbourhood are the image of the search object. This first filter does not discriminate for position. From its result, areas which might contain an image of the object can be selected. In the second step, these areas are presented to another neural net which can determine the exact position of the object in each area. This algorithm is applied to the problem of localising faces in images View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Supervised and unsupervised learning in radial basis function classifiers

    Page(s): 210 - 216
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (560 KB)  

    The paper considers a number of strategies for training radial basis function (RBF) classifiers. A benchmark problem is constructed using ten-dimensional input patterns which have to be classified into one of three classes. The RBF networks are trained using a two-phase approach (unsupervised clustering for the first layer followed by supervised learning for the second layer), error backpropagation (supervised learning for both layers) and a hybrid approach. It is shown that RBF classifiers trained with error backpropagation give results almost identical to those obtained with a multilayer perceptron. Although networks trained with the two-phase approach give slightly worse classification results, it is argued that the hidden-layer representation of such networks is much more powerful, especially if it is encoded in the form of a Gaussian mixture model. During training, the number of subclusters present within the training database can be estimated: during testing, the activities in the hidden layer of the classification network can be used to assess the novelty of input patterns and thereby help to validate network outputs View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Robot path planning using VLSI resistive grids

    Page(s): 267 - 272
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (448 KB)  

    The resistive grid algorithm for mobile robot path planning is described. A major advantage of the method is that it is capable of a fine-grained parallel analogue VLSI implementation, which offers a fast, low-power solution to the problem of mobile robot navigation. The results from a small-scale test chip are presented, together with their implications for scaling up to a full-sized path-planning chip View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Some investigations on neural processing of scattered light in water quality assessment

    Page(s): 261 - 266
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (396 KB)  

    Applications of artificial neural networks to in situ assessment of water quality are considered by means of an online optical scatter nephelometer. Light scattered by suspensions of oil in water is investigated for three different oils in the concentration range 0-100 parts per million by volume. An artificial neural network is designed to recognise the oil species and output its concentration to within an accuracy of 5.4%. Applications of the technique to more general classes of suspensions are discussed View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Decision-theoretic approach to visual inspection using neural networks

    Page(s): 223 - 229
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (488 KB)  

    The Bayesian real-time network (BARTIN) is applied to solving a visual-inspection problem requiring translation, rotation and scale (TRS) invariance. The system is capable of classifying n-fold symmetric engineering parts from near-axial views which may contain more than one part. It is evaluated and compared with other approaches using real visual-inspection data. A novel TRS-invariant preprocessor, the polygon transform, which is optimised for near-circular objects, provides information about the line and circle structure in two-dimensional images. An integral part of the polygon transform is a new Hough transform for circle radii used for both scale invariance and image characterisation. The BARTIN formalism is presented from the viewpoint of subjective Bayesian analysis, and this approach demonstrates how the personal probabilities and utilities of BARTIN can be used to optimise an externally provided reward function. A method is given for adjusting the global level of caution. To handle sparse training data, parameter parsimony in the observer was achieved using a structure comprising a stripped-out Parzen-windows classifier followed by a softmax perceptron trim. For real-time operation, the system is initialised by pretraining it using data extracted from design drawings View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Partitioned mixture distribution: an adaptive Bayesian network for low-level image processing

    Page(s): 251 - 260
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (680 KB)  

    Bayesian methods are used to analyse the problem of training a model to make predictions about the probability distribution of data that has yet to be received. Mixture distributions emerge naturally from this framework, but are not ideally matched to the density estimation problems that arise in image processing. An extension, called a partitioned mixture distribution is presented, which is essentially a set of overlapping mixture distributions. An expectation maximisation training algorithm is derived for optimising partitioned mixture distributions according to the maximum likelihood description. Finally, the results of some numerical simulations are presented, which demonstrate that lateral inhibition arises naturally in partitioned mixture distributions, and that the nodes in a partitioned mixture distribution network co-operate in such a way that each mixture distribution in the partitioned mixture distribution receives its necessary complement of computing machinery View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Novelty detection and neural network validation

    Page(s): 217 - 222
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (460 KB)  

    One of the key factors which limits the use of neural networks in many industrial applications has been the difficulty of demonstrating that a trained network will continue to generate reliable outputs once it is in routine use. An important potential source of errors is novel input data; that is, input data which differ significantly from the data used to train the network. The author investigates the relationship between the degree of novelty of input data and the corresponding reliability of the outputs from the network. He describes a quantitative procedure for assessing novelty, and demonstrates its performance by using an application which involves monitoring oil flow in multiphase pipelines View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Two original weight pruning methods based on statistical tests and rounding techniques

    Page(s): 230 - 237
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (556 KB)  

    The authors focus on the use of neural networks to approximate continuous decision functions. In this context, the parameters to be estimated are the synaptic weights of the network. The number of such parameters and the quantity of data (information) available for training greatly influence the quality of the solution obtained. A previous study analysed the influence and interaction of these two features. In order to reach the architecture of the net leading to the best fitting of the training data, two original pruning techniques are proposed. The evolution of the neural network performances, training and test rates, as the number of synaptic weights pruned increases, is shown experimentally. Two kinds of synaptic weights are obvious: irrelevant synaptic weights, which can be suppressed from the model; and relevant synaptic weights, which cannot be removed. In the test problem, it is possible to reduce the size of the network up to 42%. A 4% improvement of the performance in generalisation is observed View full abstract»

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