Abstract

A Fourier transform gives a first dimensionally reduced description of optical data. But it is not sensitive to statistical variations characterizing class properties and allowing clustering and statistical recognition. A Karhunen-Loève transform of Fourier spectra leads to a more classifying space: it is shown, through examples of writings, that clustering of optical data (especially recognition of scriptors) is achieved in a 2-D Karhunen-Loève space. Inner evolution of data belonging to a given class is described in a 3-D KL space, allowing the dating of texts.

© 1976 Optical Society of America

Handwriting synthesis and classification by means of space-variant transform and Karhunen–Loève analysis

J. Duvernoy
J. Opt. Soc. Am. 65(11) 1331-1336 (1975)

Analysis of optical near-field images by Karhunen–Loève transformation

Daniel Charraut, Daniel Courjon, Claudine Bainier, and Laurent Moulinier
Appl. Opt. 35(20) 3853-3861 (1996)

Recognition of rotated images by invariant Karhunen–Loeve expansion

Ran Hilai and Jacob Rubinstein
J. Opt. Soc. Am. A 11(5) 1610-1618 (1994)

Hilbert-space Karhunen–Loève transform with application to image analysis

A. Levy and J. Rubinstein
J. Opt. Soc. Am. A 16(1) 28-35 (1999)

Surface classification by an optoelectronic implementation of the Karhunen–Loève expansion

C. Gorecki
Appl. Opt. 30(31) 4548-4553 (1991)