Image binarization techniques for correlation-based pattern recognition

William C. Hasenplaugh, Mark A Neifeld

Research output: Contribution to journalArticle

10 Citations (Scopus)

Abstract

Correlation using binary images is suited to efficient digital realization or convenient optical implementation. Binarization algorithms are required in order to match grayscale imagery to these binary correlation architectures. We present several novel point-wise and block-wise binarization techniques all of which outperform the grayscale matched filter for large values of input signal-to-noise ratio (SNR = 0 dB). We discuss direct binarization methods based on global thresholds, local thresholds, histogram equalization, edge-enhancement, and statistical binarization, as well as indirect methods based on auto- and cross-correlation techniques. These point-wise methods are shown to offer poor noise tolerance and a new block-wise binarization method is introduced to enhance recognition at low values of SNR. This block-wise technique is motivated by vector quantization-based image compression and offers performance superior to the grayscale matched filter for an input SNR as low as -12 dB.

Original languageEnglish (US)
Pages (from-to)1907-1917
Number of pages11
JournalOptical Engineering
Volume38
Issue number11
DOIs
StatePublished - Nov 1999

Fingerprint

matched filters
Matched filters
pattern recognition
Pattern recognition
noise tolerance
vector quantization
thresholds
Binary images
Vector quantization
Image compression
Autocorrelation
histograms
imagery
cross correlation
autocorrelation
Signal to noise ratio
signal to noise ratios
augmentation

ASJC Scopus subject areas

  • Atomic and Molecular Physics, and Optics

Cite this

Image binarization techniques for correlation-based pattern recognition. / Hasenplaugh, William C.; Neifeld, Mark A.

In: Optical Engineering, Vol. 38, No. 11, 11.1999, p. 1907-1917.

Research output: Contribution to journalArticle

Hasenplaugh, William C. ; Neifeld, Mark A. / Image binarization techniques for correlation-based pattern recognition. In: Optical Engineering. 1999 ; Vol. 38, No. 11. pp. 1907-1917.
@article{6cd28be00a0c40b1b527f55d39988791,
title = "Image binarization techniques for correlation-based pattern recognition",
abstract = "Correlation using binary images is suited to efficient digital realization or convenient optical implementation. Binarization algorithms are required in order to match grayscale imagery to these binary correlation architectures. We present several novel point-wise and block-wise binarization techniques all of which outperform the grayscale matched filter for large values of input signal-to-noise ratio (SNR = 0 dB). We discuss direct binarization methods based on global thresholds, local thresholds, histogram equalization, edge-enhancement, and statistical binarization, as well as indirect methods based on auto- and cross-correlation techniques. These point-wise methods are shown to offer poor noise tolerance and a new block-wise binarization method is introduced to enhance recognition at low values of SNR. This block-wise technique is motivated by vector quantization-based image compression and offers performance superior to the grayscale matched filter for an input SNR as low as -12 dB.",
author = "Hasenplaugh, {William C.} and Neifeld, {Mark A}",
year = "1999",
month = "11",
doi = "10.1117/1.602241",
language = "English (US)",
volume = "38",
pages = "1907--1917",
journal = "Optical Engineering",
issn = "0091-3286",
publisher = "SPIE",
number = "11",

}

TY - JOUR

T1 - Image binarization techniques for correlation-based pattern recognition

AU - Hasenplaugh, William C.

AU - Neifeld, Mark A

PY - 1999/11

Y1 - 1999/11

N2 - Correlation using binary images is suited to efficient digital realization or convenient optical implementation. Binarization algorithms are required in order to match grayscale imagery to these binary correlation architectures. We present several novel point-wise and block-wise binarization techniques all of which outperform the grayscale matched filter for large values of input signal-to-noise ratio (SNR = 0 dB). We discuss direct binarization methods based on global thresholds, local thresholds, histogram equalization, edge-enhancement, and statistical binarization, as well as indirect methods based on auto- and cross-correlation techniques. These point-wise methods are shown to offer poor noise tolerance and a new block-wise binarization method is introduced to enhance recognition at low values of SNR. This block-wise technique is motivated by vector quantization-based image compression and offers performance superior to the grayscale matched filter for an input SNR as low as -12 dB.

AB - Correlation using binary images is suited to efficient digital realization or convenient optical implementation. Binarization algorithms are required in order to match grayscale imagery to these binary correlation architectures. We present several novel point-wise and block-wise binarization techniques all of which outperform the grayscale matched filter for large values of input signal-to-noise ratio (SNR = 0 dB). We discuss direct binarization methods based on global thresholds, local thresholds, histogram equalization, edge-enhancement, and statistical binarization, as well as indirect methods based on auto- and cross-correlation techniques. These point-wise methods are shown to offer poor noise tolerance and a new block-wise binarization method is introduced to enhance recognition at low values of SNR. This block-wise technique is motivated by vector quantization-based image compression and offers performance superior to the grayscale matched filter for an input SNR as low as -12 dB.

UR - http://www.scopus.com/inward/record.url?scp=0033225309&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0033225309&partnerID=8YFLogxK

U2 - 10.1117/1.602241

DO - 10.1117/1.602241

M3 - Article

AN - SCOPUS:0033225309

VL - 38

SP - 1907

EP - 1917

JO - Optical Engineering

JF - Optical Engineering

SN - 0091-3286

IS - 11

ER -