A no-reference texture regularity metric based on visual saliency

Srenivas Varadarajan, Lina Karam

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

This paper presents a no-reference perceptual metric that quantifies the degree of perceived regularity in textures. The metric is based on the similarity of visual attention (VA) of the textural primitives and the periodic spatial distribution of foveated fixation regions throughout the image. A ground-truth eye-tracking database for textures is also generated as part of this paper and is used to evaluate the performance of the most popular VA models. Using the saliency map generated by the best VA model, the proposed texture regularity metric is computed. It is shown through subjective testing that the proposed metric has a strong correlation with the mean opinion score for the perceived regularity of textures. The proposed texture regularity metric can be used to improve the quality and performance of many image processing applications like texture synthesis, texture compression, and content-based image retrieval.

Original languageEnglish (US)
Article number7069250
Pages (from-to)2784-2796
Number of pages13
JournalIEEE Transactions on Image Processing
Volume24
Issue number9
DOIs
StatePublished - Sep 1 2015

Fingerprint

Textures
Subjective testing
Image retrieval
Spatial distribution
Image processing

Keywords

  • Image Quality Assessment
  • Randomness
  • Texture Analysis
  • Visual Attention

ASJC Scopus subject areas

  • Computer Graphics and Computer-Aided Design
  • Software

Cite this

A no-reference texture regularity metric based on visual saliency. / Varadarajan, Srenivas; Karam, Lina.

In: IEEE Transactions on Image Processing, Vol. 24, No. 9, 7069250, 01.09.2015, p. 2784-2796.

Research output: Contribution to journalArticle

@article{de9bbcf42d6348f0a6c4f981b934ef01,
title = "A no-reference texture regularity metric based on visual saliency",
abstract = "This paper presents a no-reference perceptual metric that quantifies the degree of perceived regularity in textures. The metric is based on the similarity of visual attention (VA) of the textural primitives and the periodic spatial distribution of foveated fixation regions throughout the image. A ground-truth eye-tracking database for textures is also generated as part of this paper and is used to evaluate the performance of the most popular VA models. Using the saliency map generated by the best VA model, the proposed texture regularity metric is computed. It is shown through subjective testing that the proposed metric has a strong correlation with the mean opinion score for the perceived regularity of textures. The proposed texture regularity metric can be used to improve the quality and performance of many image processing applications like texture synthesis, texture compression, and content-based image retrieval.",
keywords = "Image Quality Assessment, Randomness, Texture Analysis, Visual Attention",
author = "Srenivas Varadarajan and Lina Karam",
year = "2015",
month = "9",
day = "1",
doi = "10.1109/TIP.2015.2416632",
language = "English (US)",
volume = "24",
pages = "2784--2796",
journal = "IEEE Transactions on Image Processing",
issn = "1057-7149",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "9",

}

TY - JOUR

T1 - A no-reference texture regularity metric based on visual saliency

AU - Varadarajan, Srenivas

AU - Karam, Lina

PY - 2015/9/1

Y1 - 2015/9/1

N2 - This paper presents a no-reference perceptual metric that quantifies the degree of perceived regularity in textures. The metric is based on the similarity of visual attention (VA) of the textural primitives and the periodic spatial distribution of foveated fixation regions throughout the image. A ground-truth eye-tracking database for textures is also generated as part of this paper and is used to evaluate the performance of the most popular VA models. Using the saliency map generated by the best VA model, the proposed texture regularity metric is computed. It is shown through subjective testing that the proposed metric has a strong correlation with the mean opinion score for the perceived regularity of textures. The proposed texture regularity metric can be used to improve the quality and performance of many image processing applications like texture synthesis, texture compression, and content-based image retrieval.

AB - This paper presents a no-reference perceptual metric that quantifies the degree of perceived regularity in textures. The metric is based on the similarity of visual attention (VA) of the textural primitives and the periodic spatial distribution of foveated fixation regions throughout the image. A ground-truth eye-tracking database for textures is also generated as part of this paper and is used to evaluate the performance of the most popular VA models. Using the saliency map generated by the best VA model, the proposed texture regularity metric is computed. It is shown through subjective testing that the proposed metric has a strong correlation with the mean opinion score for the perceived regularity of textures. The proposed texture regularity metric can be used to improve the quality and performance of many image processing applications like texture synthesis, texture compression, and content-based image retrieval.

KW - Image Quality Assessment

KW - Randomness

KW - Texture Analysis

KW - Visual Attention

UR - http://www.scopus.com/inward/record.url?scp=84930939850&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84930939850&partnerID=8YFLogxK

U2 - 10.1109/TIP.2015.2416632

DO - 10.1109/TIP.2015.2416632

M3 - Article

VL - 24

SP - 2784

EP - 2796

JO - IEEE Transactions on Image Processing

JF - IEEE Transactions on Image Processing

SN - 1057-7149

IS - 9

M1 - 7069250

ER -