Abstract
It has been proved that opinion-unaware (OU) blind image quality assessment (BIQA) methods are of good generalization capability and practical usage. However, so far, no OU-BIQA methods have demonstrated a very high prediction accuracy. In this paper, we proposed a novel OU-BIQA method which incorporated a collection of quality-aware statistical features that are highly related to the distortions of image local structure, the hue, the contrast and the natural scene statistics (NSS). Then, to take the benefit from the psychophysical characteristics of the human visual system (HVS), a new fitting model from the visual saliency modulated multivariate gaussian distribution is proposed to fit these features to predict the perceptual image quality by comparing with a standard model learned from natural image patches. The experimental results show that the proposed method has excellent performance and good generalization capacity.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsReferences
Mittal, A., Soundararajan, R., Bovik, A.C.: Making a completely blind image quality analyzer. IEEE Sig. Process. Lett. 22(3), 209–212 (2013)
Moorthy, A.K., Bovik, A.C.: Blind image quality assessment: from natural scene statistics to perceptual quality. IEEE Trans. Image Process. 20(12), 3350–3364 (2012)
Mittal, A., Moorthy, A.K., Bovik, A.C.: No-reference image quality assessment in the spatial domain. IEEE Trans. Image Process. Publ. IEEE Sig. Process. Soc. 21(12), 4695 (2012)
Zhang, M., Muramatsu, C., Zhou, X., et al.: Blind image quality assessment using the joint statistics of generalized local binary pattern. IEEE Sig. Process. Lett. 22(2), 207–210 (2015)
Saad, M.A., Bovik, A.C., Charrier, C.: Blind image quality assessment: a natural scene statistics approach in the DCT domain. IEEE Trans. Image Process. 21(8) (2012)
Deepti, G., Bovik, A.C.: Perceptual quality prediction on authentically distorted images using a bag of features approach. J. Vis. 17(1), 32 (2017)
Zhang, L., Zhang, L., Bovik, A.C.: A feature-enriched completely blind image quality evaluator. IEEE Trans. Image Process. 24(8), 2579–2591 (2015)
Zhang, M., Li, Y., Chen, Y.: Completely blind image quality assessment using latent quality factor from image local structure representation. In: International Conference on Acoustics, Speech, and Signal Processing (2019)
Farias, M.C.Q., Akamine, W.Y.L.: On performance of image quality metrics enhanced with visual attention computational models. Electron. Lett. 48(11), 631–633 (2012)
Lasmar, N.E., Stitou, Y., Berthoumieu, Y.: Multiscale skewed heavy tailed model for texture analysis. In: IEEE International Conference on Image Processing. IEEE (2010)
Gupta, P., Bampis, C.G., Glover, J.L., Paulter, N.G., Bovik, A.C.: Multivariate statistical approach to image quality tasks. Imaging 4, 117 (2018)
Gómez, E., Gomez-Viilegas, M.A., Marín, J.M.: A multivariate generalization of the power exponential family of distributions. Commun. Stat. 27(3), 12 (1998)
Field, D.J.: Relations between the statistics of natural images and the response properities of cortical cells. J. Opt. Soc. Amer. 4(12), 2379–2394 (1987)
Lyu, S.: Dependency reduction with divisive normalization: justification and effectiveness. Neural Comput. 23(11), 2942–2973 (2011)
Su, C.C., Cormack, L.K., Bovik, A.C.: Closed-form correlation model of oriented bandpass natural images. IEEE Sig. Process. Lett. 22(1), 21–25 (2014)
Pascal, F., Bombrun, L., Tourneret, J.Y., et al.: Parameter estimation for multivariate generalized Gaussian distributions. IEEE Trans. Sig. Process. 61(23), 5960–5971 (2013)
Su, C.C., Cormack, L.K., Bovik, A.C.: oriented correlation models of distorted natural images with application to natural stereopair quality evaluation. IEEE Trans. Image Process. 24(5), 1685–1699 (2015)
Ojala, T., Pietikainen, M., Maenpaa, T.: Multiresolution gray-scale and rotation invariant texture classification with local binary patterns, pp. 971–987. IEEE (2002)
Sinno, Z., Caramanis, C., Bovik, A.C.: Towards a closed form second-order natural scene statistics model. IEEE Trans. Image Process. 27(7), 3194–3209 (2018)
Seo, H.J., Milanfar, P.: Static and space-time visual saliency detection by self-resemblance. J. Vis. 9(12), 15 (2009)
Sheikh, H.R., Sabir, M.F., Bovik, A.C.: A statistical evaluation of recent full reference image quality assessment algorithms. IEEE Trans. Image Process. 15(11), 3440–3451 (2006)
Ghadiyaram, D., Bovik, A.C.: LIVE in the wild image quality challenge database (2015). http://live.ece.utexas.edu/research/ChallengeDB/index.htm
Acknowledgement
This research work is financially supported by National Natural Science Foundation of China (No. 61701404) and partially supported by Major Program of National Natural Science Foundation of China (No. 81727802), Natural Science Foundation of Shaanxi Province of China (No. 2020JM-438), the Key Research and Development Program in Shaanxi Province of China (No. 2017ZDCXL-GY-03-01-01).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2020 Springer Nature Switzerland AG
About this paper
Cite this paper
Zhang, M., Hou, W., Xu, X., Zhang, L., Feng, J. (2020). Completely Blind Image Quality Assessment with Visual Saliency Modulated Multi-feature Collaboration. In: Peng, Y., et al. Pattern Recognition and Computer Vision. PRCV 2020. Lecture Notes in Computer Science(), vol 12305. Springer, Cham. https://doi.org/10.1007/978-3-030-60633-6_56
Download citation
DOI: https://doi.org/10.1007/978-3-030-60633-6_56
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-60632-9
Online ISBN: 978-3-030-60633-6
eBook Packages: Computer ScienceComputer Science (R0)