Haar-like compact local binary pattern for illumination-robust feature matching

Bongjoe Kim, Jihoon Choi, Sanghyun Joo, Kwanghoon Sohn

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

A novel method for local image descriptor called a Haarlike compact local binary pattern (HC-LBP) is presented, which is robust to Gaussian noise and illumination changes and reduces the dimension of features by modifying the local binary pattern (LBP) method in two ways. First, the binary Haar-like feature method is applied to calculate region-based rank order. The binary Haar-like feature method maintains only the ordinal relationship and not the difference of intensity. Using a region-based binary operation, the proposed local image descriptor becomes more robust to Gaussian noise and illumination changes. Second, a HC-LBP feature is generated by applying the binary Haar-like feature according to the edge class in order to reduce feature dimensions and increase the discriminating power of each feature. In the feature matching experiment, the proposed method outperforms the popular scale invariant feature transform, center-symmetric local binary pattern and center-symmetric local ternary pattern in the presence of illumination changes, Gaussian noise, image blurring, and viewpoint change. Compared with popular local feature descriptors, the proposed method could get more than 0.1 increases in recall. Also, the HC-LBP descriptor could be used for many computer vision applications such as image retrieval, face recognition, and texture recognition.

Original languageEnglish
Article number43014
JournalJournal of Electronic Imaging
Volume21
Issue number4
DOIs
Publication statusPublished - 2012 Dec 1

Fingerprint

Lighting
illumination
random noise
Image retrieval
Face recognition
Computer vision
Textures
blurring
computer vision
retrieval
textures
Experiments

All Science Journal Classification (ASJC) codes

  • Atomic and Molecular Physics, and Optics
  • Computer Science Applications
  • Electrical and Electronic Engineering

Cite this

@article{939c0095fbcc406eaea67c131c1de2af,
title = "Haar-like compact local binary pattern for illumination-robust feature matching",
abstract = "A novel method for local image descriptor called a Haarlike compact local binary pattern (HC-LBP) is presented, which is robust to Gaussian noise and illumination changes and reduces the dimension of features by modifying the local binary pattern (LBP) method in two ways. First, the binary Haar-like feature method is applied to calculate region-based rank order. The binary Haar-like feature method maintains only the ordinal relationship and not the difference of intensity. Using a region-based binary operation, the proposed local image descriptor becomes more robust to Gaussian noise and illumination changes. Second, a HC-LBP feature is generated by applying the binary Haar-like feature according to the edge class in order to reduce feature dimensions and increase the discriminating power of each feature. In the feature matching experiment, the proposed method outperforms the popular scale invariant feature transform, center-symmetric local binary pattern and center-symmetric local ternary pattern in the presence of illumination changes, Gaussian noise, image blurring, and viewpoint change. Compared with popular local feature descriptors, the proposed method could get more than 0.1 increases in recall. Also, the HC-LBP descriptor could be used for many computer vision applications such as image retrieval, face recognition, and texture recognition.",
author = "Bongjoe Kim and Jihoon Choi and Sanghyun Joo and Kwanghoon Sohn",
year = "2012",
month = "12",
day = "1",
doi = "10.1117/1.JEI.21.4.043014",
language = "English",
volume = "21",
journal = "Journal of Electronic Imaging",
issn = "1017-9909",
publisher = "SPIE",
number = "4",

}

Haar-like compact local binary pattern for illumination-robust feature matching. / Kim, Bongjoe; Choi, Jihoon; Joo, Sanghyun; Sohn, Kwanghoon.

In: Journal of Electronic Imaging, Vol. 21, No. 4, 43014, 01.12.2012.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Haar-like compact local binary pattern for illumination-robust feature matching

AU - Kim, Bongjoe

AU - Choi, Jihoon

AU - Joo, Sanghyun

AU - Sohn, Kwanghoon

PY - 2012/12/1

Y1 - 2012/12/1

N2 - A novel method for local image descriptor called a Haarlike compact local binary pattern (HC-LBP) is presented, which is robust to Gaussian noise and illumination changes and reduces the dimension of features by modifying the local binary pattern (LBP) method in two ways. First, the binary Haar-like feature method is applied to calculate region-based rank order. The binary Haar-like feature method maintains only the ordinal relationship and not the difference of intensity. Using a region-based binary operation, the proposed local image descriptor becomes more robust to Gaussian noise and illumination changes. Second, a HC-LBP feature is generated by applying the binary Haar-like feature according to the edge class in order to reduce feature dimensions and increase the discriminating power of each feature. In the feature matching experiment, the proposed method outperforms the popular scale invariant feature transform, center-symmetric local binary pattern and center-symmetric local ternary pattern in the presence of illumination changes, Gaussian noise, image blurring, and viewpoint change. Compared with popular local feature descriptors, the proposed method could get more than 0.1 increases in recall. Also, the HC-LBP descriptor could be used for many computer vision applications such as image retrieval, face recognition, and texture recognition.

AB - A novel method for local image descriptor called a Haarlike compact local binary pattern (HC-LBP) is presented, which is robust to Gaussian noise and illumination changes and reduces the dimension of features by modifying the local binary pattern (LBP) method in two ways. First, the binary Haar-like feature method is applied to calculate region-based rank order. The binary Haar-like feature method maintains only the ordinal relationship and not the difference of intensity. Using a region-based binary operation, the proposed local image descriptor becomes more robust to Gaussian noise and illumination changes. Second, a HC-LBP feature is generated by applying the binary Haar-like feature according to the edge class in order to reduce feature dimensions and increase the discriminating power of each feature. In the feature matching experiment, the proposed method outperforms the popular scale invariant feature transform, center-symmetric local binary pattern and center-symmetric local ternary pattern in the presence of illumination changes, Gaussian noise, image blurring, and viewpoint change. Compared with popular local feature descriptors, the proposed method could get more than 0.1 increases in recall. Also, the HC-LBP descriptor could be used for many computer vision applications such as image retrieval, face recognition, and texture recognition.

UR - http://www.scopus.com/inward/record.url?scp=84879957349&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84879957349&partnerID=8YFLogxK

U2 - 10.1117/1.JEI.21.4.043014

DO - 10.1117/1.JEI.21.4.043014

M3 - Article

VL - 21

JO - Journal of Electronic Imaging

JF - Journal of Electronic Imaging

SN - 1017-9909

IS - 4

M1 - 43014

ER -