Title :
Recursive optimization of an extended Fisher discriminant criterion
Author :
Aladjem, Mayer E.
Author_Institution :
Dept. of Electr. & Comput. Eng., Ben-Gurion Univ. of the Negev, Beer-Sheva, Israel
Abstract :
A method for recursive optimization of an extended Fisher (ExF) discriminant criterion is proposed. The method consists of obtaining a discriminant direction which optimizes the ExF criterion, transforming the data along it into data with greater class overlap, and iterating to obtain the next discriminant direction. An application to a medical dataset indicates the potential of the proposed method for finding a sequence of oblique directions with significant class separation
Keywords :
covariance matrices; eigenvalues and eigenfunctions; image classification; image sampling; image sequences; iterative methods; matrix decomposition; medical image processing; optimisation; recursive estimation; class overlap; class separation; data transformation; discriminant direction; eigenvalue-eigenvector decomposition; extended Fisher discriminant criterion; iteration; linear mapping; medical dataset; pooled sample covariance matrix; projection pursuit procedure; recursive optimization; sequence of oblique directions; sphered design samples; Covariance matrix; Matrix decomposition; Neodymium; Optimization methods; Research and development; Scattering parameters; Tellurium; Vectors;
Conference_Titel :
Electronics, Circuits, and Systems, 1996. ICECS '96., Proceedings of the Third IEEE International Conference on
Conference_Location :
Rodos
Print_ISBN :
0-7803-3650-X
DOI :
10.1109/ICECS.1996.584460
