A Novel Illumination Invariant Feature Extraction Method Based on Improved Local-Gravity-Face

Xin Chen, Xiwei Peng, Jin Tang, Bai Luo

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

In recent years, face recognition technology has made great progress and achieved high accuracy. In the case of extreme illumination changes, the accuracy of face recognition drops sharply. Based on Local-Gravity-Face(LG face), an excellent illumination invariant feature extraction method, this paper proposes an improved LG face and a novel distance measurement method. Combined with Gamma Correction and Difference of Gaussian (DOG) filtering, it achieves a high correct recognition rate on Yale B+, and reduces the amount of calculation, thus can achieve real-time detection.

Original languageEnglish
Title of host publicationProceedings of the 40th Chinese Control Conference, CCC 2021
EditorsChen Peng, Jian Sun
PublisherIEEE Computer Society
Pages7143-7147
Number of pages5
ISBN (Electronic)9789881563804
DOIs
Publication statusPublished - 26 Jul 2021
Event40th Chinese Control Conference, CCC 2021 - Shanghai, China
Duration: 26 Jul 202128 Jul 2021

Publication series

NameChinese Control Conference, CCC
Volume2021-July
ISSN (Print)1934-1768
ISSN (Electronic)2161-2927

Conference

Conference40th Chinese Control Conference, CCC 2021
Country/TerritoryChina
CityShanghai
Period26/07/2128/07/21

Keywords

  • Complex Illumination
  • Face Recognition
  • Facial Feature
  • Illumination Invariants
  • Image Processing

Fingerprint

Dive into the research topics of 'A Novel Illumination Invariant Feature Extraction Method Based on Improved Local-Gravity-Face'. Together they form a unique fingerprint.

Cite this