Multiscale morphology based illumination normalization with enhanced local textures for face recognition

Sanghun Lee, Chul Hee Lee

Research output: Contribution to journalArticle

9 Citations (Scopus)

Abstract

A key challenge of face recognition is to obtain illumination invariant face images while preserving the discriminative features. The locations and shapes of small-scale features (e.g. eyebrows, eyes, nostrils, a mouth, etc.) are usually treated as key features for face recognition. However, it has also been observed that the local texture information of facial regions contains intrinsic facial features and needs to be enhanced to improve performance. To compensate for the illumination effects that appeared while extracting both the small-scale features and the texture information, we used multiscale morphological techniques. We used a generalized dynamic morphological quotient image (GDMQI) method based on Retinex theory and multiscale morphological closing to solve the artifact problem discussed in previous works. The proposed method consisted of two main steps: (i) illumination estimation and (ii) texture enhancement. The proposed method showed improved performance when using the CMU PIE, AR and Extended Yale-B databases.

Original languageEnglish
Pages (from-to)347-357
Number of pages11
JournalExpert Systems with Applications
Volume62
DOIs
Publication statusPublished - 2016 Nov 15

Fingerprint

Face recognition
Textures
Lighting

All Science Journal Classification (ASJC) codes

  • Engineering(all)
  • Computer Science Applications
  • Artificial Intelligence

Cite this

@article{9f9eb5320779426aafb7cdf68e9cacf0,
title = "Multiscale morphology based illumination normalization with enhanced local textures for face recognition",
abstract = "A key challenge of face recognition is to obtain illumination invariant face images while preserving the discriminative features. The locations and shapes of small-scale features (e.g. eyebrows, eyes, nostrils, a mouth, etc.) are usually treated as key features for face recognition. However, it has also been observed that the local texture information of facial regions contains intrinsic facial features and needs to be enhanced to improve performance. To compensate for the illumination effects that appeared while extracting both the small-scale features and the texture information, we used multiscale morphological techniques. We used a generalized dynamic morphological quotient image (GDMQI) method based on Retinex theory and multiscale morphological closing to solve the artifact problem discussed in previous works. The proposed method consisted of two main steps: (i) illumination estimation and (ii) texture enhancement. The proposed method showed improved performance when using the CMU PIE, AR and Extended Yale-B databases.",
author = "Sanghun Lee and Lee, {Chul Hee}",
year = "2016",
month = "11",
day = "15",
doi = "10.1016/j.eswa.2016.06.039",
language = "English",
volume = "62",
pages = "347--357",
journal = "Expert Systems with Applications",
issn = "0957-4174",
publisher = "Elsevier Limited",

}

Multiscale morphology based illumination normalization with enhanced local textures for face recognition. / Lee, Sanghun; Lee, Chul Hee.

In: Expert Systems with Applications, Vol. 62, 15.11.2016, p. 347-357.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Multiscale morphology based illumination normalization with enhanced local textures for face recognition

AU - Lee, Sanghun

AU - Lee, Chul Hee

PY - 2016/11/15

Y1 - 2016/11/15

N2 - A key challenge of face recognition is to obtain illumination invariant face images while preserving the discriminative features. The locations and shapes of small-scale features (e.g. eyebrows, eyes, nostrils, a mouth, etc.) are usually treated as key features for face recognition. However, it has also been observed that the local texture information of facial regions contains intrinsic facial features and needs to be enhanced to improve performance. To compensate for the illumination effects that appeared while extracting both the small-scale features and the texture information, we used multiscale morphological techniques. We used a generalized dynamic morphological quotient image (GDMQI) method based on Retinex theory and multiscale morphological closing to solve the artifact problem discussed in previous works. The proposed method consisted of two main steps: (i) illumination estimation and (ii) texture enhancement. The proposed method showed improved performance when using the CMU PIE, AR and Extended Yale-B databases.

AB - A key challenge of face recognition is to obtain illumination invariant face images while preserving the discriminative features. The locations and shapes of small-scale features (e.g. eyebrows, eyes, nostrils, a mouth, etc.) are usually treated as key features for face recognition. However, it has also been observed that the local texture information of facial regions contains intrinsic facial features and needs to be enhanced to improve performance. To compensate for the illumination effects that appeared while extracting both the small-scale features and the texture information, we used multiscale morphological techniques. We used a generalized dynamic morphological quotient image (GDMQI) method based on Retinex theory and multiscale morphological closing to solve the artifact problem discussed in previous works. The proposed method consisted of two main steps: (i) illumination estimation and (ii) texture enhancement. The proposed method showed improved performance when using the CMU PIE, AR and Extended Yale-B databases.

UR - http://www.scopus.com/inward/record.url?scp=84976320070&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84976320070&partnerID=8YFLogxK

U2 - 10.1016/j.eswa.2016.06.039

DO - 10.1016/j.eswa.2016.06.039

M3 - Article

VL - 62

SP - 347

EP - 357

JO - Expert Systems with Applications

JF - Expert Systems with Applications

SN - 0957-4174

ER -