An evidential reasoning based classification algorithm and its application for face recognition with class noise

Xiaodong Wang, F. Liu, L. C. Jiao, Zhiguo Zhou, Jingjing Yu, Bing Li, Jianrui Chen, Jiao Wu, Fanhua Shang

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

For classification problems, in practice, real-world data may suffer from two types of noise, attribute noise and class noise. It is the key for improving recognition performance to remove as much of their adverse effects as possible. In this paper, a formalism algorithm is proposed for classification problems with class noise, which is more challenging than those with attribute noise. The proposed formalism algorithm is based on evidential reasoning theory which is a powerful tool to deal with uncertain information in multiple attribute decision analysis and many other areas. Thus, it may be more effective alternative to handle noisy label information. And then a specific algorithm - Evidential Reasoning based Classification algorithm (ERC) is derived to recognize human faces under class noise conditions. The proposed ERC algorithm is extensively evaluated on five publicly available face databases with class noise and yields good performance.

Original languageEnglish (US)
Pages (from-to)4117-4128
Number of pages12
JournalPattern Recognition
Volume45
Issue number12
DOIs
StatePublished - Dec 1 2012

Fingerprint

Face recognition
Decision theory
Labels

Keywords

  • Class noise
  • Evidential reasoning
  • Face recognition
  • Linear regression classification (LRC)
  • Sparse representation-based classification (SRC)

ASJC Scopus subject areas

  • Software
  • Signal Processing
  • Computer Vision and Pattern Recognition
  • Artificial Intelligence

Cite this

An evidential reasoning based classification algorithm and its application for face recognition with class noise. / Wang, Xiaodong; Liu, F.; Jiao, L. C.; Zhou, Zhiguo; Yu, Jingjing; Li, Bing; Chen, Jianrui; Wu, Jiao; Shang, Fanhua.

In: Pattern Recognition, Vol. 45, No. 12, 01.12.2012, p. 4117-4128.

Research output: Contribution to journalArticle

Wang, X, Liu, F, Jiao, LC, Zhou, Z, Yu, J, Li, B, Chen, J, Wu, J & Shang, F 2012, 'An evidential reasoning based classification algorithm and its application for face recognition with class noise', Pattern Recognition, vol. 45, no. 12, pp. 4117-4128. https://doi.org/10.1016/j.patcog.2012.06.005
Wang, Xiaodong ; Liu, F. ; Jiao, L. C. ; Zhou, Zhiguo ; Yu, Jingjing ; Li, Bing ; Chen, Jianrui ; Wu, Jiao ; Shang, Fanhua. / An evidential reasoning based classification algorithm and its application for face recognition with class noise. In: Pattern Recognition. 2012 ; Vol. 45, No. 12. pp. 4117-4128.
@article{037845ea72a8424ea84a0154bb948949,
title = "An evidential reasoning based classification algorithm and its application for face recognition with class noise",
abstract = "For classification problems, in practice, real-world data may suffer from two types of noise, attribute noise and class noise. It is the key for improving recognition performance to remove as much of their adverse effects as possible. In this paper, a formalism algorithm is proposed for classification problems with class noise, which is more challenging than those with attribute noise. The proposed formalism algorithm is based on evidential reasoning theory which is a powerful tool to deal with uncertain information in multiple attribute decision analysis and many other areas. Thus, it may be more effective alternative to handle noisy label information. And then a specific algorithm - Evidential Reasoning based Classification algorithm (ERC) is derived to recognize human faces under class noise conditions. The proposed ERC algorithm is extensively evaluated on five publicly available face databases with class noise and yields good performance.",
keywords = "Class noise, Evidential reasoning, Face recognition, Linear regression classification (LRC), Sparse representation-based classification (SRC)",
author = "Xiaodong Wang and F. Liu and Jiao, {L. C.} and Zhiguo Zhou and Jingjing Yu and Bing Li and Jianrui Chen and Jiao Wu and Fanhua Shang",
year = "2012",
month = "12",
day = "1",
doi = "10.1016/j.patcog.2012.06.005",
language = "English (US)",
volume = "45",
pages = "4117--4128",
journal = "Pattern Recognition",
issn = "0031-3203",
publisher = "Elsevier Limited",
number = "12",

}

TY - JOUR

T1 - An evidential reasoning based classification algorithm and its application for face recognition with class noise

AU - Wang, Xiaodong

AU - Liu, F.

AU - Jiao, L. C.

AU - Zhou, Zhiguo

AU - Yu, Jingjing

AU - Li, Bing

AU - Chen, Jianrui

AU - Wu, Jiao

AU - Shang, Fanhua

PY - 2012/12/1

Y1 - 2012/12/1

N2 - For classification problems, in practice, real-world data may suffer from two types of noise, attribute noise and class noise. It is the key for improving recognition performance to remove as much of their adverse effects as possible. In this paper, a formalism algorithm is proposed for classification problems with class noise, which is more challenging than those with attribute noise. The proposed formalism algorithm is based on evidential reasoning theory which is a powerful tool to deal with uncertain information in multiple attribute decision analysis and many other areas. Thus, it may be more effective alternative to handle noisy label information. And then a specific algorithm - Evidential Reasoning based Classification algorithm (ERC) is derived to recognize human faces under class noise conditions. The proposed ERC algorithm is extensively evaluated on five publicly available face databases with class noise and yields good performance.

AB - For classification problems, in practice, real-world data may suffer from two types of noise, attribute noise and class noise. It is the key for improving recognition performance to remove as much of their adverse effects as possible. In this paper, a formalism algorithm is proposed for classification problems with class noise, which is more challenging than those with attribute noise. The proposed formalism algorithm is based on evidential reasoning theory which is a powerful tool to deal with uncertain information in multiple attribute decision analysis and many other areas. Thus, it may be more effective alternative to handle noisy label information. And then a specific algorithm - Evidential Reasoning based Classification algorithm (ERC) is derived to recognize human faces under class noise conditions. The proposed ERC algorithm is extensively evaluated on five publicly available face databases with class noise and yields good performance.

KW - Class noise

KW - Evidential reasoning

KW - Face recognition

KW - Linear regression classification (LRC)

KW - Sparse representation-based classification (SRC)

UR - http://www.scopus.com/inward/record.url?scp=84864287114&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84864287114&partnerID=8YFLogxK

U2 - 10.1016/j.patcog.2012.06.005

DO - 10.1016/j.patcog.2012.06.005

M3 - Article

AN - SCOPUS:84864287114

VL - 45

SP - 4117

EP - 4128

JO - Pattern Recognition

JF - Pattern Recognition

SN - 0031-3203

IS - 12

ER -