Perceptual quality assessment of color images using adaptive signal representation

Umesh Rajashekar, Zhou Wang, Eero Simoncelli

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Perceptual image distortion measures can play a fundamental role in evaluating and optimizing imaging systems and image processing algorithms. Many existing measures are formulated to represent "just noticeable differences" (JNDs), as measured in psychophysical experiments on human subjects. But some image distortions, such as those arising from small changes in the intensity of the ambient illumination, are far more tolerable to human observers than those that disrupt the spatial structure of intensities and colors. Here, we introduce a framework in which we quantify these perceptual distortions in terms of "just intolerable differences" (JIDs). As in (Wang & Simoncelli, Proc. ICIP 2005), we first construct a set of spatio-chromatic basis functions to approximate (as a first-order Taylor series) a set of "non-structural" distortions that result from changes in lighting/imaging/viewing conditions. These basis functions are defined on local image patches, and are adaptive, in that they are computed as functions of the undistorted reference image. This set is then augmented with a complete basis arising from a linear approximation of the CIELAB color space. Each basis function is weighted by a scale factor to convert it into units corresponding to JIDs. Each patch of the error image is represented using this weighted overcomplete basis, and the overall distortion metric is computed by summing the squared coefficients over all such (overlapping) patches. We implement an example of this metric, incorporating invariance to small changes in the viewing and lighting conditions, and demonstrate that the resulting distortion values are more consistent with human perception than those produced by CIELAB or S-CIELAB.

Original languageEnglish (US)
Title of host publicationProceedings of SPIE-IS and T Electronic Imaging - Human Vision and Electronic Imaging XV
Volume7527
DOIs
StatePublished - 2010
EventHuman Vision and Electronic Imaging XV - San Jose, CA, United States
Duration: Jan 18 2010Jan 21 2010

Other

OtherHuman Vision and Electronic Imaging XV
CountryUnited States
CitySan Jose, CA
Period1/18/101/21/10

Fingerprint

Quality Assessment
Color Image
Color
color
Patch
Basis Functions
Lighting
illuminating
Metric
Human Perception
Taylor series
Scale factor
Color Space
Spatial Structure
Linear Approximation
Invariance
Imaging System
Imaging systems
Convert
image processing

Keywords

  • Adaptive signal decomposition
  • CIELAB
  • Color image quality assessment
  • S-CIELAB

ASJC Scopus subject areas

  • Applied Mathematics
  • Computer Science Applications
  • Electrical and Electronic Engineering
  • Electronic, Optical and Magnetic Materials
  • Condensed Matter Physics

Cite this

Rajashekar, U., Wang, Z., & Simoncelli, E. (2010). Perceptual quality assessment of color images using adaptive signal representation. In Proceedings of SPIE-IS and T Electronic Imaging - Human Vision and Electronic Imaging XV (Vol. 7527). [75271L] https://doi.org/10.1117/12.845312

Perceptual quality assessment of color images using adaptive signal representation. / Rajashekar, Umesh; Wang, Zhou; Simoncelli, Eero.

Proceedings of SPIE-IS and T Electronic Imaging - Human Vision and Electronic Imaging XV. Vol. 7527 2010. 75271L.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Rajashekar, U, Wang, Z & Simoncelli, E 2010, Perceptual quality assessment of color images using adaptive signal representation. in Proceedings of SPIE-IS and T Electronic Imaging - Human Vision and Electronic Imaging XV. vol. 7527, 75271L, Human Vision and Electronic Imaging XV, San Jose, CA, United States, 1/18/10. https://doi.org/10.1117/12.845312
Rajashekar U, Wang Z, Simoncelli E. Perceptual quality assessment of color images using adaptive signal representation. In Proceedings of SPIE-IS and T Electronic Imaging - Human Vision and Electronic Imaging XV. Vol. 7527. 2010. 75271L https://doi.org/10.1117/12.845312
Rajashekar, Umesh ; Wang, Zhou ; Simoncelli, Eero. / Perceptual quality assessment of color images using adaptive signal representation. Proceedings of SPIE-IS and T Electronic Imaging - Human Vision and Electronic Imaging XV. Vol. 7527 2010.
@inproceedings{db59d4741ca345d2ba75513cbd96edd3,
title = "Perceptual quality assessment of color images using adaptive signal representation",
abstract = "Perceptual image distortion measures can play a fundamental role in evaluating and optimizing imaging systems and image processing algorithms. Many existing measures are formulated to represent {"}just noticeable differences{"} (JNDs), as measured in psychophysical experiments on human subjects. But some image distortions, such as those arising from small changes in the intensity of the ambient illumination, are far more tolerable to human observers than those that disrupt the spatial structure of intensities and colors. Here, we introduce a framework in which we quantify these perceptual distortions in terms of {"}just intolerable differences{"} (JIDs). As in (Wang & Simoncelli, Proc. ICIP 2005), we first construct a set of spatio-chromatic basis functions to approximate (as a first-order Taylor series) a set of {"}non-structural{"} distortions that result from changes in lighting/imaging/viewing conditions. These basis functions are defined on local image patches, and are adaptive, in that they are computed as functions of the undistorted reference image. This set is then augmented with a complete basis arising from a linear approximation of the CIELAB color space. Each basis function is weighted by a scale factor to convert it into units corresponding to JIDs. Each patch of the error image is represented using this weighted overcomplete basis, and the overall distortion metric is computed by summing the squared coefficients over all such (overlapping) patches. We implement an example of this metric, incorporating invariance to small changes in the viewing and lighting conditions, and demonstrate that the resulting distortion values are more consistent with human perception than those produced by CIELAB or S-CIELAB.",
keywords = "Adaptive signal decomposition, CIELAB, Color image quality assessment, S-CIELAB",
author = "Umesh Rajashekar and Zhou Wang and Eero Simoncelli",
year = "2010",
doi = "10.1117/12.845312",
language = "English (US)",
isbn = "9780819479204",
volume = "7527",
booktitle = "Proceedings of SPIE-IS and T Electronic Imaging - Human Vision and Electronic Imaging XV",

}

TY - GEN

T1 - Perceptual quality assessment of color images using adaptive signal representation

AU - Rajashekar, Umesh

AU - Wang, Zhou

AU - Simoncelli, Eero

PY - 2010

Y1 - 2010

N2 - Perceptual image distortion measures can play a fundamental role in evaluating and optimizing imaging systems and image processing algorithms. Many existing measures are formulated to represent "just noticeable differences" (JNDs), as measured in psychophysical experiments on human subjects. But some image distortions, such as those arising from small changes in the intensity of the ambient illumination, are far more tolerable to human observers than those that disrupt the spatial structure of intensities and colors. Here, we introduce a framework in which we quantify these perceptual distortions in terms of "just intolerable differences" (JIDs). As in (Wang & Simoncelli, Proc. ICIP 2005), we first construct a set of spatio-chromatic basis functions to approximate (as a first-order Taylor series) a set of "non-structural" distortions that result from changes in lighting/imaging/viewing conditions. These basis functions are defined on local image patches, and are adaptive, in that they are computed as functions of the undistorted reference image. This set is then augmented with a complete basis arising from a linear approximation of the CIELAB color space. Each basis function is weighted by a scale factor to convert it into units corresponding to JIDs. Each patch of the error image is represented using this weighted overcomplete basis, and the overall distortion metric is computed by summing the squared coefficients over all such (overlapping) patches. We implement an example of this metric, incorporating invariance to small changes in the viewing and lighting conditions, and demonstrate that the resulting distortion values are more consistent with human perception than those produced by CIELAB or S-CIELAB.

AB - Perceptual image distortion measures can play a fundamental role in evaluating and optimizing imaging systems and image processing algorithms. Many existing measures are formulated to represent "just noticeable differences" (JNDs), as measured in psychophysical experiments on human subjects. But some image distortions, such as those arising from small changes in the intensity of the ambient illumination, are far more tolerable to human observers than those that disrupt the spatial structure of intensities and colors. Here, we introduce a framework in which we quantify these perceptual distortions in terms of "just intolerable differences" (JIDs). As in (Wang & Simoncelli, Proc. ICIP 2005), we first construct a set of spatio-chromatic basis functions to approximate (as a first-order Taylor series) a set of "non-structural" distortions that result from changes in lighting/imaging/viewing conditions. These basis functions are defined on local image patches, and are adaptive, in that they are computed as functions of the undistorted reference image. This set is then augmented with a complete basis arising from a linear approximation of the CIELAB color space. Each basis function is weighted by a scale factor to convert it into units corresponding to JIDs. Each patch of the error image is represented using this weighted overcomplete basis, and the overall distortion metric is computed by summing the squared coefficients over all such (overlapping) patches. We implement an example of this metric, incorporating invariance to small changes in the viewing and lighting conditions, and demonstrate that the resulting distortion values are more consistent with human perception than those produced by CIELAB or S-CIELAB.

KW - Adaptive signal decomposition

KW - CIELAB

KW - Color image quality assessment

KW - S-CIELAB

UR - http://www.scopus.com/inward/record.url?scp=77949462456&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77949462456&partnerID=8YFLogxK

U2 - 10.1117/12.845312

DO - 10.1117/12.845312

M3 - Conference contribution

AN - SCOPUS:77949462456

SN - 9780819479204

VL - 7527

BT - Proceedings of SPIE-IS and T Electronic Imaging - Human Vision and Electronic Imaging XV

ER -