A new image quality measure for assessment of histogram equalization-based contrast enhancement techniques

Research output: Contribution to journalArticle

49 Citations (Scopus)

Abstract

Absolute Mean Brightness Error (AMBE) and Entropy are among the two most popular IQMs used to assess Histogram Equalization (HE) based techniques. To the best of author's knowledge, there is no evaluation report on how well the two IQMs correlate to human opinion. This paper reviews and discusses the potential flaws in using AMBE and Entropy to assess HE-based techniques. This paper presents results of a subjective quality assessment in which image quality data obtained from 1935 human observer opinion scores were used to evaluate the IQMs. The statistical evaluation results show that the two IQMs have poor correlation with human mean opinion score (MOS); Pearson Correlation Coefficient (PCC)<0.4, Root Mean Square Error (RMSE)>0.75, Outlier Ratio (OR)>20%. A new IQM which takes into account important properties of human visual perception (HVP) is proposed. It is tested and found to have significantly better correlation (PCC>0.86, RMSE<0.39 and OR=0%). The proposed IQM also outperforms Multi-Scale Structural Similarity (MSSIM) and Information Fidelity Criterion-based (IFC) measure, which are two prominent fidelity-based IQMs.

Original languageEnglish
Pages (from-to)640-647
Number of pages8
JournalDigital Signal Processing: A Review Journal
Volume22
Issue number4
DOIs
Publication statusPublished - 01 Jan 2012

Fingerprint

Image quality
Luminance
Entropy
Defects

All Science Journal Classification (ASJC) codes

  • Signal Processing
  • Electrical and Electronic Engineering

Cite this

@article{5e1971ab29d6462284586d3fc010347f,
title = "A new image quality measure for assessment of histogram equalization-based contrast enhancement techniques",
abstract = "Absolute Mean Brightness Error (AMBE) and Entropy are among the two most popular IQMs used to assess Histogram Equalization (HE) based techniques. To the best of author's knowledge, there is no evaluation report on how well the two IQMs correlate to human opinion. This paper reviews and discusses the potential flaws in using AMBE and Entropy to assess HE-based techniques. This paper presents results of a subjective quality assessment in which image quality data obtained from 1935 human observer opinion scores were used to evaluate the IQMs. The statistical evaluation results show that the two IQMs have poor correlation with human mean opinion score (MOS); Pearson Correlation Coefficient (PCC)<0.4, Root Mean Square Error (RMSE)>0.75, Outlier Ratio (OR)>20{\%}. A new IQM which takes into account important properties of human visual perception (HVP) is proposed. It is tested and found to have significantly better correlation (PCC>0.86, RMSE<0.39 and OR=0{\%}). The proposed IQM also outperforms Multi-Scale Structural Similarity (MSSIM) and Information Fidelity Criterion-based (IFC) measure, which are two prominent fidelity-based IQMs.",
author = "Chen, {Soong Der}",
year = "2012",
month = "1",
day = "1",
doi = "10.1016/j.dsp.2012.04.002",
language = "English",
volume = "22",
pages = "640--647",
journal = "Digital Signal Processing: A Review Journal",
issn = "1051-2004",
publisher = "Elsevier Inc.",
number = "4",

}

TY - JOUR

T1 - A new image quality measure for assessment of histogram equalization-based contrast enhancement techniques

AU - Chen, Soong Der

PY - 2012/1/1

Y1 - 2012/1/1

N2 - Absolute Mean Brightness Error (AMBE) and Entropy are among the two most popular IQMs used to assess Histogram Equalization (HE) based techniques. To the best of author's knowledge, there is no evaluation report on how well the two IQMs correlate to human opinion. This paper reviews and discusses the potential flaws in using AMBE and Entropy to assess HE-based techniques. This paper presents results of a subjective quality assessment in which image quality data obtained from 1935 human observer opinion scores were used to evaluate the IQMs. The statistical evaluation results show that the two IQMs have poor correlation with human mean opinion score (MOS); Pearson Correlation Coefficient (PCC)<0.4, Root Mean Square Error (RMSE)>0.75, Outlier Ratio (OR)>20%. A new IQM which takes into account important properties of human visual perception (HVP) is proposed. It is tested and found to have significantly better correlation (PCC>0.86, RMSE<0.39 and OR=0%). The proposed IQM also outperforms Multi-Scale Structural Similarity (MSSIM) and Information Fidelity Criterion-based (IFC) measure, which are two prominent fidelity-based IQMs.

AB - Absolute Mean Brightness Error (AMBE) and Entropy are among the two most popular IQMs used to assess Histogram Equalization (HE) based techniques. To the best of author's knowledge, there is no evaluation report on how well the two IQMs correlate to human opinion. This paper reviews and discusses the potential flaws in using AMBE and Entropy to assess HE-based techniques. This paper presents results of a subjective quality assessment in which image quality data obtained from 1935 human observer opinion scores were used to evaluate the IQMs. The statistical evaluation results show that the two IQMs have poor correlation with human mean opinion score (MOS); Pearson Correlation Coefficient (PCC)<0.4, Root Mean Square Error (RMSE)>0.75, Outlier Ratio (OR)>20%. A new IQM which takes into account important properties of human visual perception (HVP) is proposed. It is tested and found to have significantly better correlation (PCC>0.86, RMSE<0.39 and OR=0%). The proposed IQM also outperforms Multi-Scale Structural Similarity (MSSIM) and Information Fidelity Criterion-based (IFC) measure, which are two prominent fidelity-based IQMs.

UR - http://www.scopus.com/inward/record.url?scp=84860669159&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84860669159&partnerID=8YFLogxK

U2 - 10.1016/j.dsp.2012.04.002

DO - 10.1016/j.dsp.2012.04.002

M3 - Article

VL - 22

SP - 640

EP - 647

JO - Digital Signal Processing: A Review Journal

JF - Digital Signal Processing: A Review Journal

SN - 1051-2004

IS - 4

ER -