Abstract

Five pattern recognition methods are compared for their ability to classify binary infrared spectra. Included is a discussion of the time vs success balance for each of the techniques. Predictive ability decreases in the order maximum likelihood > distance > Tanimoto similarity ~ Hamming distance > dot product. The time required for each prediction after the classifier has been developed increases in order maximum likelihood ~ distance ~ dot product < Tanimoto similarity ~ Hamming distance.

Spatio-spectral binary patterns based on multispectral filter arrays for texture classification

Sofiane Mihoubi, Olivier Losson, Benjamin Mathon, and Ludovic Macaire
J. Opt. Soc. Am. A 35(9) 1532-1542 (2018)

Spectral pattern recognition in IR remote sensing

Alexander S. Zachor
Appl. Opt. 22(17) 2699-2703 (1983)

Robust validation of pattern classification methods for laser-induced breakdown spectroscopy

Jeremiah Remus and Kehinde S. Dunsin
Appl. Opt. 51(7) B49-B56 (2012)

Hybrid optical processor for pattern recognition and classification using a generalized set of pattern functions

James R. Leger and Sing H. Lee
Appl. Opt. 21(2) 274-287 (1982)

Simulated annealing algorithm for binary phase only filters in pattern classification

Myung Soo Kim and Clark C. Guest
Appl. Opt. 29(8) 1203-1208 (1990)