Pattern recognition and image reconstruction using improved digital zernike moments

Huibao Lin, Jennie Si, Glen P. Abousleman

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)

Abstract

Zernike moments are one of the most effective orthogonal, rotation-invariant moments in continuous space. Unfortunately, the digitization process necessary for use with digital imagery results in compromised orthogonality. In this work, we introduce improved digital Zernike moments that exhibit much better orthogonality, while preserving their inherent invariance to rotation. We then propose a novel pattern recognition algorithm that is based on the improved digital Zernike moments. With the improved orthogonality, targets can be represented by fewer moments, thus minimizing computational complexity. Additionally, the rotation invariance enables our algorithm to recognize targets with arbitrary orientation. Because our algorithm eliminates the segmentation step that is typically applied in other techniques, it is better suited to low-quality imagery. Simulations on real images demonstrate these aspects of the proposed algorithm.

Original languageEnglish (US)
Title of host publicationProceedings of SPIE - The International Society for Optical Engineering
EditorsD.P. Casasent, T.-H. Chao
Pages211-220
Number of pages10
Volume5816
DOIs
StatePublished - 2005
EventOptical Pattern Recognition XVI - Orlando, FL, United States
Duration: Mar 31 2005Apr 1 2005

Other

OtherOptical Pattern Recognition XVI
CountryUnited States
CityOrlando, FL
Period3/31/054/1/05

Fingerprint

image reconstruction
Image reconstruction
pattern recognition
Pattern recognition
orthogonality
moments
Invariance
imagery
invariance
Analog to digital conversion
Computational complexity
preserving
simulation

Keywords

  • Feature
  • Image reconstruction
  • Moments
  • Pattern recognition
  • Zernike moments

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Condensed Matter Physics

Cite this

Lin, H., Si, J., & Abousleman, G. P. (2005). Pattern recognition and image reconstruction using improved digital zernike moments. In D. P. Casasent, & T-H. Chao (Eds.), Proceedings of SPIE - The International Society for Optical Engineering (Vol. 5816, pp. 211-220). [26] https://doi.org/10.1117/12.604076

Pattern recognition and image reconstruction using improved digital zernike moments. / Lin, Huibao; Si, Jennie; Abousleman, Glen P.

Proceedings of SPIE - The International Society for Optical Engineering. ed. / D.P. Casasent; T.-H. Chao. Vol. 5816 2005. p. 211-220 26.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Lin, H, Si, J & Abousleman, GP 2005, Pattern recognition and image reconstruction using improved digital zernike moments. in DP Casasent & T-H Chao (eds), Proceedings of SPIE - The International Society for Optical Engineering. vol. 5816, 26, pp. 211-220, Optical Pattern Recognition XVI, Orlando, FL, United States, 3/31/05. https://doi.org/10.1117/12.604076
Lin H, Si J, Abousleman GP. Pattern recognition and image reconstruction using improved digital zernike moments. In Casasent DP, Chao T-H, editors, Proceedings of SPIE - The International Society for Optical Engineering. Vol. 5816. 2005. p. 211-220. 26 https://doi.org/10.1117/12.604076
Lin, Huibao ; Si, Jennie ; Abousleman, Glen P. / Pattern recognition and image reconstruction using improved digital zernike moments. Proceedings of SPIE - The International Society for Optical Engineering. editor / D.P. Casasent ; T.-H. Chao. Vol. 5816 2005. pp. 211-220
@inproceedings{52c67f951e6f416ab1b0988fac0a5fa8,
title = "Pattern recognition and image reconstruction using improved digital zernike moments",
abstract = "Zernike moments are one of the most effective orthogonal, rotation-invariant moments in continuous space. Unfortunately, the digitization process necessary for use with digital imagery results in compromised orthogonality. In this work, we introduce improved digital Zernike moments that exhibit much better orthogonality, while preserving their inherent invariance to rotation. We then propose a novel pattern recognition algorithm that is based on the improved digital Zernike moments. With the improved orthogonality, targets can be represented by fewer moments, thus minimizing computational complexity. Additionally, the rotation invariance enables our algorithm to recognize targets with arbitrary orientation. Because our algorithm eliminates the segmentation step that is typically applied in other techniques, it is better suited to low-quality imagery. Simulations on real images demonstrate these aspects of the proposed algorithm.",
keywords = "Feature, Image reconstruction, Moments, Pattern recognition, Zernike moments",
author = "Huibao Lin and Jennie Si and Abousleman, {Glen P.}",
year = "2005",
doi = "10.1117/12.604076",
language = "English (US)",
volume = "5816",
pages = "211--220",
editor = "D.P. Casasent and T.-H. Chao",
booktitle = "Proceedings of SPIE - The International Society for Optical Engineering",

}

TY - GEN

T1 - Pattern recognition and image reconstruction using improved digital zernike moments

AU - Lin, Huibao

AU - Si, Jennie

AU - Abousleman, Glen P.

PY - 2005

Y1 - 2005

N2 - Zernike moments are one of the most effective orthogonal, rotation-invariant moments in continuous space. Unfortunately, the digitization process necessary for use with digital imagery results in compromised orthogonality. In this work, we introduce improved digital Zernike moments that exhibit much better orthogonality, while preserving their inherent invariance to rotation. We then propose a novel pattern recognition algorithm that is based on the improved digital Zernike moments. With the improved orthogonality, targets can be represented by fewer moments, thus minimizing computational complexity. Additionally, the rotation invariance enables our algorithm to recognize targets with arbitrary orientation. Because our algorithm eliminates the segmentation step that is typically applied in other techniques, it is better suited to low-quality imagery. Simulations on real images demonstrate these aspects of the proposed algorithm.

AB - Zernike moments are one of the most effective orthogonal, rotation-invariant moments in continuous space. Unfortunately, the digitization process necessary for use with digital imagery results in compromised orthogonality. In this work, we introduce improved digital Zernike moments that exhibit much better orthogonality, while preserving their inherent invariance to rotation. We then propose a novel pattern recognition algorithm that is based on the improved digital Zernike moments. With the improved orthogonality, targets can be represented by fewer moments, thus minimizing computational complexity. Additionally, the rotation invariance enables our algorithm to recognize targets with arbitrary orientation. Because our algorithm eliminates the segmentation step that is typically applied in other techniques, it is better suited to low-quality imagery. Simulations on real images demonstrate these aspects of the proposed algorithm.

KW - Feature

KW - Image reconstruction

KW - Moments

KW - Pattern recognition

KW - Zernike moments

UR - http://www.scopus.com/inward/record.url?scp=27644571897&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=27644571897&partnerID=8YFLogxK

U2 - 10.1117/12.604076

DO - 10.1117/12.604076

M3 - Conference contribution

AN - SCOPUS:27644571897

VL - 5816

SP - 211

EP - 220

BT - Proceedings of SPIE - The International Society for Optical Engineering

A2 - Casasent, D.P.

A2 - Chao, T.-H.

ER -