Skip to main content

Neurocomputing

Objective

The aim of PYGMALION was to create an independent European technological base for the applications, algorithms and software aspects of neurocomputing. The project led to the coordination of research on connectionist computing techniques and provided the means for developing the necessary software tools for productive research.
The aim of PYGMALION was to create an independent European technological base for the applications, algorithms and software aspects of neurocomputing. The project led to the coordination of research on connectionist computing techniques and provided the means for developing the necessary software tools for productive research. The objectives are to promote information exchange within the European research community; to develop a portable European neural network specification language and tools to stand as an interface between applications and emulation architectures; and to demonstrate the potential of a neural approach through industrial applications, mainly of image processing and speech processing.
High and intermediate languages (N and nC) have been specified. They are close to the C++ language. The development of compilers is underway. All the well known neural models are already available in the library.
A first cascadable very large scale integration (VLSI) demonstrator has been produced and tested. This chip offers full connectivity, which allows the programming of the most commonly used neural networks models (BP, Hopfield, MLP).
Results of applications of neural models (BP-constrained) have proved robustness and good restoration for image compression. Applications for real time high definition television (HDTV) compression are under study. Further work on image segmentation and texture is underway, and is building the basis for future image processing using neural networks.
High level imaging, speech recognition and acoustic signal classification show positive results in 3 areas, with some potential short term commercial applications.
The objectives are to:

- promote information exchange within the European research community
- develop a portable European Neural Network Specification Language and tools to stand as an interface between applications and emulation architectures
- demonstrate the potential of a neural approach through industrial applications, mainly of image processing and speech processing.

Coordinator

THOMSON CSF
Address
9150 Rue Des Mathurins
92223 Bagneux
France

Participants (14)

Alcatel SEL AG
Germany
Address
Lorenzstraße 10
70435 Stuttgart
COMPUTER TECHNOLOGY INSTITUTE
Greece
Address
Kolokotroni Street, 326110
26221 Patras
Centre National de la Recherche Scientifique (CNRS)
France
Address
1919 Route De Mende
34405 Montpellier
Centro Studi e Laboratori Telecomunicazioni SpA
Italy
Address
Via G. Reiss Romoli, 274
10148 Torino
Concept Logiciels Expert
France
Address
16-18 Rue Du Dome
92514 Boulogne-billancourt
INSTITUTO DE ENGENHARIA DE SISTEMAS E COMPUTADORES
Portugal
Address
Apartado 10105, Rua Alves Redol, 9
1017 Lisboa
IRIAC
France
Address
10 Rue Andre Vandrezanne
75013 Paris
Institut National Polytechnique de Grenoble
France
Address
46 Avenue Félix Viallet
38031 Grenoble
Laboratoire d'Électronique Philips
France
Address
22 Avenue Descartes
94453 Limeil-brevannes
POLITECNICO DI TORINO
Italy
Address
Corso Duca Degli Abruzzi 24
10129 Torino
SEXTANT AVIONIQUE SA
France
Address
5-7 Rue Jeanne Braconnier
92366 Meudon La Foret
UNIVERSITAT POLITECNICA DE MADRID
Spain
Address
Campus De Montegancedo
28660 Madrid
Birkbeck College, University of London
United Kingdom
Address
Malet Street, Bloomsbury
WC1E 7HX London
Université de Paris V (Université René Descartes)
France
Address
45 Rue Des Saints Peres
75006 Paris