Abstract
Reduced-reference (RR) image quality measures aim to predict the visual quality of distorted images with only partial information about the reference images. In this paper, we propose an RR image quality assessment method based on a natural image statistic model in the wavelet transform domain. We use the Kullback-Leibler distance between the marginal probability distributions of wavelet coefficients of the reference and distorted images as a measure of image distortion. A generalized Gaussian model is employed to summarize the marginal distribution of wavelet coefficients of the reference image, so that only a relatively small number of RR features are needed for the evaluation of image quality. The proposed method is easy to implement and computationally efficient. In addition, we find that many well-known types of image distortions lead to significant changes in wavelet coefficient histograms, and thus are readily detectable by our measure. A Matlab implementation of the method has been made available online at http://www.cns.nyu.edu/~lcv/rriqa/.
| Original language | English (US) |
|---|---|
| Title of host publication | Proceedings of SPIE - The International Society for Optical Engineering |
| Editors | B.E. Rogowitz, T.N. Pappas, S.J. Daly |
| Pages | 149-159 |
| Number of pages | 11 |
| Volume | 5666 |
| DOIs | |
| State | Published - 2005 |
| Event | Proceedings of SPIE-IS and T Electronic Imaging - Human Vision and Electronic Imaging X - San Jose, CA, United States Duration: Jan 17 2005 → Jan 20 2005 |
Other
| Other | Proceedings of SPIE-IS and T Electronic Imaging - Human Vision and Electronic Imaging X |
|---|---|
| Country | United States |
| City | San Jose, CA |
| Period | 1/17/05 → 1/20/05 |
Fingerprint
ASJC Scopus subject areas
- Electrical and Electronic Engineering
- Condensed Matter Physics
Cite this
Reduced-reference image quality assessment using a wavelet-domain natural image statistic model. / Wang, Zhou; Simoncelli, Eero.
Proceedings of SPIE - The International Society for Optical Engineering. ed. / B.E. Rogowitz; T.N. Pappas; S.J. Daly. Vol. 5666 2005. p. 149-159 20.Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
}
TY - GEN
T1 - Reduced-reference image quality assessment using a wavelet-domain natural image statistic model
AU - Wang, Zhou
AU - Simoncelli, Eero
PY - 2005
Y1 - 2005
N2 - Reduced-reference (RR) image quality measures aim to predict the visual quality of distorted images with only partial information about the reference images. In this paper, we propose an RR image quality assessment method based on a natural image statistic model in the wavelet transform domain. We use the Kullback-Leibler distance between the marginal probability distributions of wavelet coefficients of the reference and distorted images as a measure of image distortion. A generalized Gaussian model is employed to summarize the marginal distribution of wavelet coefficients of the reference image, so that only a relatively small number of RR features are needed for the evaluation of image quality. The proposed method is easy to implement and computationally efficient. In addition, we find that many well-known types of image distortions lead to significant changes in wavelet coefficient histograms, and thus are readily detectable by our measure. A Matlab implementation of the method has been made available online at http://www.cns.nyu.edu/~lcv/rriqa/.
AB - Reduced-reference (RR) image quality measures aim to predict the visual quality of distorted images with only partial information about the reference images. In this paper, we propose an RR image quality assessment method based on a natural image statistic model in the wavelet transform domain. We use the Kullback-Leibler distance between the marginal probability distributions of wavelet coefficients of the reference and distorted images as a measure of image distortion. A generalized Gaussian model is employed to summarize the marginal distribution of wavelet coefficients of the reference image, so that only a relatively small number of RR features are needed for the evaluation of image quality. The proposed method is easy to implement and computationally efficient. In addition, we find that many well-known types of image distortions lead to significant changes in wavelet coefficient histograms, and thus are readily detectable by our measure. A Matlab implementation of the method has been made available online at http://www.cns.nyu.edu/~lcv/rriqa/.
UR - http://www.scopus.com/inward/record.url?scp=21944449269&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=21944449269&partnerID=8YFLogxK
U2 - 10.1117/12.597306
DO - 10.1117/12.597306
M3 - Conference contribution
AN - SCOPUS:21944449269
VL - 5666
SP - 149
EP - 159
BT - Proceedings of SPIE - The International Society for Optical Engineering
A2 - Rogowitz, B.E.
A2 - Pappas, T.N.
A2 - Daly, S.J.
ER -