New Neural Network Pruning and its application to sonar imagery - Université de Bretagne Occidentale Access content directly
Conference Papers Year : 1998

New Neural Network Pruning and its application to sonar imagery

Abstract

In this paper, we propose a neural network approach for automatic classification of underwater objects on sonar images. A major problem with sonar imagery applications is the difficulty to obtain large databases for training. Real sonar devices are costly and staged experiments where objects are well known and manually placed are rare because of cost. We show that the simultaneous use of parameters extraction and neural network pruning can significantly help to obtain good generalization rates (despite the lack of large training databases) and to reduce the complexity of the classifier. Furthermore, a method derived from neural network pruning is proposed to evaluate the significance of each parameter.
Fichier principal
Vignette du fichier
1998_04_cesa_elagage.pdf (452.02 Ko) Télécharger le fichier
Origin Files produced by the author(s)
Licence
Copyright

Dates and versions

hal-03222606 , version 1 (18-03-2023)

Licence

Copyright

Identifiers

  • HAL Id : hal-03222606 , version 1

Cite

Pascal Galerne, Koffi Clément Yao, Gilles Burel. New Neural Network Pruning and its application to sonar imagery. Computational Engineering in Systems Applications (CESA), Apr 1998, Hammamet, Tunisia. ⟨hal-03222606⟩

Collections

UNIV-BREST CNRS
22 View
3 Download

Share

Gmail Mastodon Facebook X LinkedIn More