Abstract

A useful discriminant vector for pattern classification is one that maximizes the minimum separation of discriminant function values for two pattern classes. This optimality criterion can prove valuable in many situations because it emphasizes the class elements that are most difficult to classify. A method for computing this discriminant vector by quadratic programming is derived. The resulting calculation scales with training set size rather than number of input variables and hence is well suited to the high dimensionality of image classification tasks. Digitized images are used to demonstrate application of the approach to two class and multiple-class image classification tasks.

© 1991 Optical Society of America

Optical pattern recognition based on convex functions

Uri Mahlab and Joseph Shamir
J. Opt. Soc. Am. A 8(8) 1233-1239 (1991)

Recognition and classification of red blood cells using digital holographic microscopy and data clustering with discriminant analysis

Ran Liu, Dipak K. Dey, Daniel Boss, Pierre Marquet, and Bahram Javidi
J. Opt. Soc. Am. A 28(6) 1204-1210 (2011)

Comparison of statistical pattern-recognition algorithms for hybrid processing. II. Eigenvector-based algorithm

Q. Tian, Y. Fainman, and Sing H. Lee
J. Opt. Soc. Am. A 5(10) 1670-1682 (1988)

Design of an optical linear discriminant filter for classification of subwavelength concave and convex defects on dielectric substrates

Jun-ichiro Sugisaka, Takashi Yasui, and Koichi Hirayama
J. Opt. Soc. Am. A 39(3) 342-351 (2022)

Discriminant analysis for recognition of human face images

Kamran Etemad and Rama Chellappa
J. Opt. Soc. Am. A 14(8) 1724-1733 (1997)