Pattern Recognition using Neural and Functional Networks

  • Vasantha Kalyani David
  • Sundaramoorthy Rajasekaran

Part of the Studies in Computational Intelligence book series (SCI, volume 160)

Table of contents

  1. Front Matter
  2. Vasantha Kalyani David, Sundaramoorthy Rajasekaran
    Pages 1-7
  3. Vasantha Kalyani David, Sundaramoorthy Rajasekaran
    Pages 9-13
  4. Vasantha Kalyani David, Sundaramoorthy Rajasekaran
    Pages 27-49
  5. Vasantha Kalyani David, Sundaramoorthy Rajasekaran
    Pages 51-71
  6. Vasantha Kalyani David, Sundaramoorthy Rajasekaran
    Pages 73-91
  7. Vasantha Kalyani David, Sundaramoorthy Rajasekaran
    Pages 93-113
  8. Vasantha Kalyani David, Sundaramoorthy Rajasekaran
    Pages 115-122
  9. Vasantha Kalyani David, Sundaramoorthy Rajasekaran
    Pages 123-134
  10. Vasantha Kalyani David, Sundaramoorthy Rajasekaran
    Pages 135-136
  11. Vasantha Kalyani David, Sundaramoorthy Rajasekaran
    Pages 137-137
  12. Back Matter

About this book

Introduction

The concept of pattern is universal in intelligence and discovery. The patterns in biological data contain knowledge. Discrimination of signal pattern allows personal identification by voice, hand writing, finger prints, facial images, recognition of speech, written characters and also scenes in images like identification of military targets based on radar, infrared, and video images. Possibilities are enormous in geologic, climatic, meteorologic, personality, cultural, historical, spectral, electromagnetic as well as from microscopic images of cells to macroscopic images of regions of the earth obtained from satellite scans and radio telescope images of galaxies. It is up to the researcher in some area to glean the essentials and begin to explore the classification and recognition of patterns in data that will lead to discoveries of associations and cause – effect relationships.

Two outlines are suggested as the possible tracks for pattern recognition. They are neural networks and functional networks. A new approach to pattern recognition using microARTMAP and wavelet transforms in the context of hand written characters, gestures and signatures have been dealt. The Kohonen Network, Back Propagation Networks and Competitive Hopfield Neural Network have been considered for various applications. Functional networks, being a generalized form of Neural Networks where functions are learned rather than weights is compared with Multiple Regression Analysis for some applications and the results are seen to be coincident.

Keywords

Processing Wavelet algorithms architecture biologically inspired cognition computational intelligence intelligence learning linear optimization neural network neural networks pattern pattern recognition programming

Authors and affiliations

  • Vasantha Kalyani David
    • 1
  • Sundaramoorthy Rajasekaran
    • 2
  1. 1.Avinashilingam Deemed University India
  2. 2.PSG College of TechnologyIndia

Bibliographic information

  • DOI https://doi.org/10.1007/978-3-540-85130-1
  • Copyright Information Springer-Verlag Berlin Heidelberg 2009
  • Publisher Name Springer, Berlin, Heidelberg
  • eBook Packages Engineering
  • Print ISBN 978-3-540-85129-5
  • Online ISBN 978-3-540-85130-1
  • Series Print ISSN 1860-949X
  • Series Online ISSN 1860-9503
  • About this book
Industry Sectors
Automotive
Chemical Manufacturing
Biotechnology
Electronics
Telecommunications
Consumer Packaged Goods
Energy, Utilities & Environment
Aerospace
Oil, Gas & Geosciences