Chapter 5. Kernel Fisher's Discriminant with Heterogeneous Kernels

  1. Dr Gustavo Camps-Valls B.Sc., Ph.D. professor member3 and
  2. Dr Lorenzo Bruzzone M.S., Ph.D. Postdoctoral Researcher Professor member Chair4
  1. M. Murat Dundar1 and
  2. Glenn Fung2

Published Online: 4 NOV 2009

DOI: 10.1002/9780470748992.ch5

Kernel Methods for Remote Sensing Data Analysis

Kernel Methods for Remote Sensing Data Analysis

How to Cite

Dundar, M. M. and Fung, G. (2009) Kernel Fisher's Discriminant with Heterogeneous Kernels, in Kernel Methods for Remote Sensing Data Analysis (eds G. Camps-Valls and L. Bruzzone), John Wiley & Sons, Ltd, Chichester, UK. doi: 10.1002/9780470748992.ch5

Editor Information

  1. 3

    Image Processing Laboratory (IPL) & Dept. Enginyeria Electrónica, Universitat de Valéncia, Spain

  2. 4

    Dept. Information Engineering and Computer Science, University of Trento, Italy

Author Information

  1. 1

    Indiana University-Purdue University, Indianapolis, USA

  2. 2

    Siemens Medical Solutions Inc., Malvern, USA

Publication History

  1. Published Online: 4 NOV 2009
  2. Published Print: 23 OCT 2009

ISBN Information

Print ISBN: 9780470722114

Online ISBN: 9780470748992



  • Kernel Fisher's discriminant with heterogeneous kernels;
  • framework for obtaining nonlinear version of Fisher's Discriminant (KFD);
  • supervised image classification;
  • Kernel Fisher's discriminant (KFD) using heterogeneous kernel models;
  • supervised classification problems and error probability due to Bayes classifier;
  • Linear Fisher's Discriminant (LFD);
  • automatic kernel selection KFD algorithm;
  • hyperspectral data collected with airborne HYMAP system;
  • gray-level image and ground truth fields for Purdue campus dataset;
  • mathematical programming approach for kernelizing algorithm


This chapter contains sections titled:

  • Introduction

  • Linear Fisher's Discriminant

  • Kernel Fisher Discriminant

  • Kernel Fisher's Discriminant with heterogeneous kernels

  • Automatic kernel selection KFD algorithm

  • Numerical results

  • Conclusion

  • References