-
[show abstract]
[hide abstract]
ABSTRACT: Neural networks can be applied for a number of innovative applications in a production environment, ranging from security & safety in the environmental conditions to the product control & diagnosis. Forvisualmonitoring the use of lowresolution images is promising to bridge the time elapse between more elaborous stampings. This facility for biologically motivated decision-making is illustrated by two real-life applications: stain detection and mixture diagnosis.
04/2006: pages 1048-1057;
-
[show abstract]
[hide abstract]
ABSTRACT: Mathematical morphology is a powerful means to specify image manipulations; discrete-time cellular neural networks (DT-CNN) is the fast realization. The attractive combination has been sufficiently shown for simple problems but tends to fail in efficiency for more complex ones. The paper introduces a complement and argument swap (CAS) equivalence that allows to solve an image processing problem through a small library of representative efficient designs.
Cellular Neural Networks and Their Applications, 2002. (CNNA 2002). Proceedings of the 2002 7th IEEE International Workshop on; 08/2002
-
[show abstract]
[hide abstract]
ABSTRACT: This paper describes a system for the automatic identification of vehicles by the contents of its license plate. The system has been developed over the last five years and is one of the four candidate systems for automatic toll collection in the Netherlands. Due to the extreme time and performance requirements placed on the system, advanced techniques like DT–CNNs and classifier combining are used. The first tests on the road show that the system correctly recognizes 85.4% of the passing vehicles, while marking the remaining 14.6% as unrecognizable.
Journal of VLSI Signal Processing 10/1999; 23(2):465-477. · 0.73 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: We report on a structured design methodology for neural hardware
based on a VHDL code which can be implemented on a FPGA or used to
create an ASIC. This code, dubbed virtual neuro-processor (VNP), is
generated automatically from within the neural networks design and
simulation environment and supports several network architectures. Error
backpropagation learning can be supported by the VNP, thus allowing for
the implementation of feedforward networks with on-chip training
capability. The main advantage gained from the VNP-concept is that it
highly automates and structures the design of (application specific)
neural hardware. It thus considerably shortens the development time of
such devices and ensures a high-quality design process
Neural Networks, 1999. IJCNN '99. International Joint Conference on; 02/1999
-
[show abstract]
[hide abstract]
ABSTRACT: We present an automated estimation procedure for linear filter
banks. The motivation for such an automated design procedure comes from
developments in the application of neural networks to time series
prediction and modelling. The estimation procedure is based on neural
implementation of the Karhunen-Loeve transform. FIR filters are obtained
through computation of a principal component transform (PCA) on
tapped-delay lines of a time series. The FIR filter output is used to
predict the time series with a feedforward neural network. The results
of the proposed method are compared to other focused time lagged neural
networks: the tapped-delay line neural network and the Gamma network.
The three techniques are compared by focusing capacities, speed of
convergence, quality of prediction and efficiency i.e. ratio of
prediction quality and memory usage. The temporal PCA method outperforms
the ordinary TDL-network slightly considering prediction error, the
Gamma network however is not beaten, provided that the number of taps is
not over-estimated. Gamma filters appear to be more efficient as they
achieve large time-scopes with a small number of taps. Adaptive Gamma
filters seem to be the more promising technique for automated design of
focusing filters for neural time series prediction
Neural Networks, 1999. IJCNN '99. International Joint Conference on; 02/1999
-
VLSI Signal Processing. 01/1999; 23:465-477.
-
[show abstract]
[hide abstract]
ABSTRACT: Morphology provides the algebraic means to specify operations on
images. Discrete-time cellular neural networks (DT-CNNs) mechanize the
execution of operations on images. The paper first shows the equivalence
between morphological functions and DT-CNNs. Then, the argument is
extended to the synthesis of optimal DT-CNN structures from complex
morphological expressions. It is shown that morphological specifications
may be freely derived, to be subsequently transformed and adopted to the
needs of a specific target terminology. This process of technology
mapping can be automated along the well-trodden path in CAD for
microelectronics
IEEE Transactions on Circuits and Systems I Fundamental Theory and Applications 10/1998;
-
[show abstract]
[hide abstract]
ABSTRACT: The architecture and design of a CMOS core module for neural signal processing is presented. A logarithmic number representation is exploited to allow for multiplication with adaptive accuracy. This creates the facility to exchange accuracy for speed, both for the single instruction execution as during the parallelisation of a number of instructions. The core module is generated from a process independent silicon assembler. Details are given on a specific pipelined sign-magnitude implementation
Neural Networks Proceedings, 1998. IEEE World Congress on Computational Intelligence. The 1998 IEEE International Joint Conference on; 06/1998
-
[show abstract]
[hide abstract]
ABSTRACT: One of the main issues in the analysis of a time series is its
forecasting. Many questions arise in the design of a neural network that
aims to capture the dynamics of a temporal sequence in order to predict
it. In a reproducible way we want to find decision strategies for the
preprocessing and the architecture of the network. In this paper we
introduce a novel technique to extract important data features, called
the data conflict plot. The conflict plot is used to design a modified
architecture for the prediction of signals with distinct periodic
components. Instead of a single delay line, this architecture is
preceded by several incompletely connected delay lines
Neural Networks Proceedings, 1998. IEEE World Congress on Computational Intelligence. The 1998 IEEE International Joint Conference on; 06/1998
-
[show abstract]
[hide abstract]
ABSTRACT: Automatic license plate recognition requires a series of complex
image processing steps. For practical use, the amount of data to be
processed must be minimized at early stage. This paper shows that the
computationally most intensive steps can be realized by discrete time
cellular neural networks (DTCNNs). Moreover, high-level operations like
`finding the license plate in the image' and `finding the characters on
the plate' need only a small number of DTCNNs. Real-life tests show that
the DTCNNs are capable of correctly identifying more than 85% out of all
license plates while leaving only 0.5% of the original information to be
inspected for actual recognition
Cellular Neural Networks and Their Applications Proceedings, 1998 Fifth IEEE International Workshop on; 05/1998
-
[show abstract]
[hide abstract]
ABSTRACT: Most image processing tasks, like pattern matching, are defined in
terms of large-neighborhood discrete time cellular neural network
(DTCNN) templates, while most hardware implementations support only
direct-neighborhood ones (3×3). Literature on DTCNN template
decomposition shows that such large-neighborhood functions can be
implemented as a sequence of successive direct-neighborhood templates.
However, for this procedure the number of templates in the decomposition
is exponential in the size of the original template. This paper shows
how template decomposition is induced by the decomposition of
structuring elements in the morphological design process. It is proved
that an upper bound for the number of templates found in this way is
quadratic in the size of the original template. For many cases more
efficient and even optimal decompositions can be obtained
Cellular Neural Networks and Their Applications Proceedings, 1998 Fifth IEEE International Workshop on; 05/1998
-
[show abstract]
[hide abstract]
ABSTRACT: One of the main issues in the research on time series is its prediction. Using a time-delayed neural network we determine the optimal network size from the signal correlation time. Then the biofeedbackdriven neurocontrol for artificial human walking is approached with much detail on the signal preprocessing. Finally we indicate the need for a more heterogeneous set-up to eliminate oscillations inherent to handling the human motor control problem. 1 Introduction The most important aspect in time series prediction is the modelling of the series. Before we are able to predict the future values based on its history, we have to derive a model for the underlying behaviour. Statistics gives useful tools for this, but a major objection to most of these techniques (for example the well-known ARMA method [1]) is that they assume a priori that the time series was generated by a linear process. Most series one encounters in practice, however, are of a nonlinear nature; therefore much effort has b...
04/1998;
-
[show abstract]
[hide abstract]
ABSTRACT: Real production data are vague and irreproducible. This suggests
fuzzy knowledge acquisition for an online learned combined fuzzy/neural
network. The paper advocates a modular neural only network based on the
injection of knowledge from a multiple rule block fuzzy specification. A
typical controller takes 10-25 blocks with 10 rules on average, which
after injection in a neural net can be personalized and adapted by small
(typical 100) measurement sets
EUROMICRO 97. 'New Frontiers of Information Technology'., Proceedings of the 23rd EUROMICRO Conference; 10/1997
-
[show abstract]
[hide abstract]
ABSTRACT: The popular multilayer perceptron (MLP) topology with an
error-backpropagation learning rule doesn't allow the developer to use
the (explicit) engineering knowledge as available in real-life problems.
Design procedures described in literature start either with a random
initialization or with a `smart' initialization of the weight values
based on statistical properties of the training data. This article
presents a design methodology that enables the insertion of pre-trained
parts in a MLP network topology and illustrates the advantages of such a
modular approach. Furthermore we will discuss the differences between
the modular approach and a hybrid approach, where explicit knowledge is
captured by mathematical models. In a hybrid design a mathematical model
is embedded in the modular neural network as an optimization of one of
the pre-trained subnetworks or because the designer wants to obtain a
certain degree of transparency of captured knowledge in the modular
design
Neural Networks,1997., International Conference on; 07/1997
-
[show abstract]
[hide abstract]
ABSTRACT: The morphological design of discrete-time cellular neural networks
(DTCNNs) has been presented in a companion paper (1996). DTCNN templates
have been given for the elemental morphological operators. One way to
obtain realizations for more complex operators is cascading the DTCNN
equivalences of the constituent elemental operators. Here it is shown
that this straightforward mapping mostly yields a non-optimal solution
with respect to the required amount of hardware. A hardware reduction
scheme of morphologically designed DTCNNs is proposed which includes the
introduction of time variant templates and the identification of
non-elementary expressions for which a single layer DTCNN exists
Cellular Neural Networks and their Applications, 1996. CNNA-96. Proceedings., 1996 Fourth IEEE International Workshop on; 07/1996
-
[show abstract]
[hide abstract]
ABSTRACT: Morphology is a mathematical discipline that provides algebraic
manipulation for images. Such a formal framework is very well suited to
the design of discrete-time cellular neural networks (DTCNNs). It is
shown how morphological expressions can be transformed by algebraic
rules to provide an efficient DTCNN implementation for complex
operations on images. In fact, most specialized DTCNNs published in
literature prove to be straight morphological derivatives, as
exemplified in the design of a public transport ticketing system
Neural Networks, 1996., IEEE International Conference on; 07/1996
-
[show abstract]
[hide abstract]
ABSTRACT: A method is presented to optimize a trained neural network for
physical realization styles. Target architectures are embedded
microcontrollers or standard cell based ASIC designs. The approach
exploits the redundancy in the network, required for successful
training, to replace the synaptic weighting and the neuron transfer
functions by ones that can be implemented with smaller cost. Redundancy
indices are used to identify the network elements that are candidates
for optimization to be performed by the judicious application of local,
behaviour-invariant transformations. The usefulness of the presented
approach is illustrated by a image processing application realized in
our lab
Neural Networks, 1996., IEEE International Conference on; 07/1996
-
[show abstract]
[hide abstract]
ABSTRACT: Time-delay neural networks are well-suited for prediction purposes. A particular implementation is the Finite Impulse Response neural net. The GREMLIN architecture is introduced to accommodate such networks. It can be micropipelined to achieve a 85 MCPS performance on a conventional connection-serial structure and allows from its Logic-Enhance Memory nature an easily parametrized design. A typical design for biomedical applications can be trained in a Cascade fashion and subsequently mapped
Microelectronics for Neural Networks, 1996., Proceedings of Fifth International Conference on; 03/1996
-
[show abstract]
[hide abstract]
ABSTRACT: Time-delay neural networks are well-suited for prediction purposes. A particular implementation is the finite impulse response (FIR) neural net. A major design problem exists in establishing the optimal order of such filters while minimizing the number of weights. Here, a constructive solution inspired by cascade learning is outlined and illustrated by some typical case-studies
Neural Networks, 1995. Proceedings., IEEE International Conference on; 12/1995
-
[show abstract]
[hide abstract]
ABSTRACT: A car license plate recognition system (CLPR-system) has been
developed to identify vehicles by the contents of their license plate
for speed-limit enforcement. This type of application puts high demands
on the reliability of the CLPR-system. A combination of neural and fuzzy
techniques is used to guarantee a very low error rate at an acceptable
recognition rate. First experiments along highways in the Netherlands
show that the system has an error rate, of 0.02% at a recognition rate
of 98.51%. These results are also compared with other published
CLPR-systems
Neural Networks, 1995. Proceedings., IEEE International Conference on; 12/1995
