Moment-based local binary patterns

A novel descriptor for invariant pattern recognition applications

G. A. Papakostas, D. E. Koulouriotis, E. G. Karakasis, V. D. Tourassis

Research output: Contribution to journalArticle

38 Citations (Scopus)

Abstract

A novel descriptor able to improve the classification capabilities of a typical pattern recognition system is proposed in this paper. The introduced descriptor is derived by incorporating two efficient region descriptors, namely image moments and local binary patterns (LBP), commonly used in pattern recognition applications, in the last decades. The main idea behind this novel feature extraction methodology is the need of improved recognition capabilities, a goal achieved by the combinative use of these descriptors. This collaboration aims to make use of the major advantages each one presents, by simultaneously complementing each other, in order to elevate their weak points. In this way, the useful properties of the moments and moment invariants regarding their robustness to the noise presence, their global information coding mechanism and their invariant behaviour under scaling, translation and rotation conditions, along with the local nature of the LBP, are combined in a single concrete methodology. As a result a novel descriptor invariant to common geometric transformations of the described object, capable to encode its local characteristics, is formed and its classification capabilities are investigated through massive experimental scenarios. The experiments have shown the superiority of the introduced descriptor over the moment invariants, the LBP operator and other well-known from the literature descriptors such as HOG, HOG-LBP and LBP-HF.

Original languageEnglish
Pages (from-to)358-371
Number of pages14
JournalNeurocomputing
Volume99
DOIs
Publication statusPublished - Jan 1 2013
Externally publishedYes

Fingerprint

Pattern recognition
Automated Pattern Recognition
Pattern recognition systems
Noise
Feature extraction
Concretes
Experiments

Keywords

  • Computer-robotic vision
  • Feature extraction
  • Image moments
  • Local binary patterns
  • Moment invariants
  • Momentgram
  • Pattern recognition

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Science Applications
  • Cognitive Neuroscience

Cite this

Moment-based local binary patterns : A novel descriptor for invariant pattern recognition applications. / Papakostas, G. A.; Koulouriotis, D. E.; Karakasis, E. G.; Tourassis, V. D.

In: Neurocomputing, Vol. 99, 01.01.2013, p. 358-371.

Research output: Contribution to journalArticle

Papakostas, G. A. ; Koulouriotis, D. E. ; Karakasis, E. G. ; Tourassis, V. D. / Moment-based local binary patterns : A novel descriptor for invariant pattern recognition applications. In: Neurocomputing. 2013 ; Vol. 99. pp. 358-371.
@article{52cc14bf59974c50bc0a5e3c7a06d44e,
title = "Moment-based local binary patterns: A novel descriptor for invariant pattern recognition applications",
abstract = "A novel descriptor able to improve the classification capabilities of a typical pattern recognition system is proposed in this paper. The introduced descriptor is derived by incorporating two efficient region descriptors, namely image moments and local binary patterns (LBP), commonly used in pattern recognition applications, in the last decades. The main idea behind this novel feature extraction methodology is the need of improved recognition capabilities, a goal achieved by the combinative use of these descriptors. This collaboration aims to make use of the major advantages each one presents, by simultaneously complementing each other, in order to elevate their weak points. In this way, the useful properties of the moments and moment invariants regarding their robustness to the noise presence, their global information coding mechanism and their invariant behaviour under scaling, translation and rotation conditions, along with the local nature of the LBP, are combined in a single concrete methodology. As a result a novel descriptor invariant to common geometric transformations of the described object, capable to encode its local characteristics, is formed and its classification capabilities are investigated through massive experimental scenarios. The experiments have shown the superiority of the introduced descriptor over the moment invariants, the LBP operator and other well-known from the literature descriptors such as HOG, HOG-LBP and LBP-HF.",
keywords = "Computer-robotic vision, Feature extraction, Image moments, Local binary patterns, Moment invariants, Momentgram, Pattern recognition",
author = "Papakostas, {G. A.} and Koulouriotis, {D. E.} and Karakasis, {E. G.} and Tourassis, {V. D.}",
year = "2013",
month = "1",
day = "1",
doi = "10.1016/j.neucom.2012.06.031",
language = "English",
volume = "99",
pages = "358--371",
journal = "Neurocomputing",
issn = "0925-2312",
publisher = "Elsevier",

}

TY - JOUR

T1 - Moment-based local binary patterns

T2 - A novel descriptor for invariant pattern recognition applications

AU - Papakostas, G. A.

AU - Koulouriotis, D. E.

AU - Karakasis, E. G.

AU - Tourassis, V. D.

PY - 2013/1/1

Y1 - 2013/1/1

N2 - A novel descriptor able to improve the classification capabilities of a typical pattern recognition system is proposed in this paper. The introduced descriptor is derived by incorporating two efficient region descriptors, namely image moments and local binary patterns (LBP), commonly used in pattern recognition applications, in the last decades. The main idea behind this novel feature extraction methodology is the need of improved recognition capabilities, a goal achieved by the combinative use of these descriptors. This collaboration aims to make use of the major advantages each one presents, by simultaneously complementing each other, in order to elevate their weak points. In this way, the useful properties of the moments and moment invariants regarding their robustness to the noise presence, their global information coding mechanism and their invariant behaviour under scaling, translation and rotation conditions, along with the local nature of the LBP, are combined in a single concrete methodology. As a result a novel descriptor invariant to common geometric transformations of the described object, capable to encode its local characteristics, is formed and its classification capabilities are investigated through massive experimental scenarios. The experiments have shown the superiority of the introduced descriptor over the moment invariants, the LBP operator and other well-known from the literature descriptors such as HOG, HOG-LBP and LBP-HF.

AB - A novel descriptor able to improve the classification capabilities of a typical pattern recognition system is proposed in this paper. The introduced descriptor is derived by incorporating two efficient region descriptors, namely image moments and local binary patterns (LBP), commonly used in pattern recognition applications, in the last decades. The main idea behind this novel feature extraction methodology is the need of improved recognition capabilities, a goal achieved by the combinative use of these descriptors. This collaboration aims to make use of the major advantages each one presents, by simultaneously complementing each other, in order to elevate their weak points. In this way, the useful properties of the moments and moment invariants regarding their robustness to the noise presence, their global information coding mechanism and their invariant behaviour under scaling, translation and rotation conditions, along with the local nature of the LBP, are combined in a single concrete methodology. As a result a novel descriptor invariant to common geometric transformations of the described object, capable to encode its local characteristics, is formed and its classification capabilities are investigated through massive experimental scenarios. The experiments have shown the superiority of the introduced descriptor over the moment invariants, the LBP operator and other well-known from the literature descriptors such as HOG, HOG-LBP and LBP-HF.

KW - Computer-robotic vision

KW - Feature extraction

KW - Image moments

KW - Local binary patterns

KW - Moment invariants

KW - Momentgram

KW - Pattern recognition

UR - http://www.scopus.com/inward/record.url?scp=84867887875&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84867887875&partnerID=8YFLogxK

U2 - 10.1016/j.neucom.2012.06.031

DO - 10.1016/j.neucom.2012.06.031

M3 - Article

VL - 99

SP - 358

EP - 371

JO - Neurocomputing

JF - Neurocomputing

SN - 0925-2312

ER -