The Classification Mechanism of Adaptive Gauss Neural Networks and Application on Target Classification

  • Yanning Zhang
  • Licheng Jiao
Part of the The International Series on Asian Studies in Computer and Information Science book series (ASIS, volume 6)


In this paper, an adaptive Gauss neural network (AGNN) is proposed and used to extract automatically features from noises radiated from three types of ships after Fourier transform preprocessing. The feature extraction mechanism and the classification mechanism of the adaptive Gauss neural network are revealed by mathematical analysis, a local adaptive Gauss neural network is proposed, and an efficient engineering classifier based on the local adaptive Gauss neural network is designed and applied to classify the actual ship noises. The classification experiment results are encouraging, and show that the mathematical analysis of the feature extraction mechanism and the classification mechanism of AGNN is correct, all of which show the method proposed in this paper to be superior.


Feature extraction target classification adaptive Gauss neural network 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Cui, Jintai. Translated by Chen, Zhengxing. An Introduction to Wavelets, Xi’an Jiao Tong University Press, 1995. 1.Google Scholar
  2. Ding, Dong. A Sonar target classification system Based on Neural Network. Acoustic Technology, 1993, 12(2):42–45.Google Scholar
  3. Du, Shuanping. Radar basis function neural network applied in underwater target classification, Proceedings of 1996 national acoustic conference, 1996, 72–77.Google Scholar
  4. Fu, Donglei. The Study of Passive Sonar Signal Classification Based on Neural Network, Northwestern Polytechenical University, Thesis [ Master ], 1993.Google Scholar
  5. Hassob, J. C.; Chen, C. H. On Constructing an Expert System for Contact. Localization and Tracking. Pattern Recognition, 1985, 18(6):465–473.CrossRefGoogle Scholar
  6. Moore, F. Passive Sonar Target Recognition Using A Back-Propagation Neural Network, Naval Postgraduate School, Monterey, Californie, Thesis, June,1991.Google Scholar
  7. Wang, Biquan; Chen, Zuyin.Pattern Recognituon Theory, Method and Application, Earthquake Press, 1989.Google Scholar
  8. Zhang, Wenfeng; He, Mingyi. A Modified method BP Algorithm and Applied to Image Classification in X Machine, Proceedings of 1992 National Neural Network Conference, 1992.Google Scholar
  9. Zhang, Yanning; Jiao, Licheng. A New Method of Classification Based on Neural Network, Proceedings of 1997 Chinese Congress on Neurocomputing Science, 1997 October, 588–591.Google Scholar

Copyright information

© Springer Science+Business Media New York 1999

Authors and Affiliations

  • Yanning Zhang
    • 1
  • Licheng Jiao
    • 1
  1. 1.Key Lab. for radar signal processingXidian UniversityXi’anChina

Personalised recommendations