Abstract
The heavy-tailed hyper-Laplacian prior has been successfully applied in image restoration tasks, in which the unified distribution is adopted for the whole image. However, the gradient distribution of natural image is reasonably assumed to be spatially variant, e.g., gradient distribution of the region with less texture is more heavy-tailed. In this paper, we propose to model the gradient distribution of natural images as spatially variant hyper-Laplacian. The proposed model adapts the hyper-Laplacian parameters to each pixel. Within the maximum a posterior model, we address the problem using an alternating optimization method. Also the proposed model is free from tedious tuning trade-off parameters. Compared with the deconvolution algorithm using uniform hyper-Laplacian prior, the experimental results validate that the proposed model can achieve better restoration results in terms of visual quality and quantitative indicators.
Similar content being viewed by others
References
Chan, T.F., Shen, J.H., Zhou, H.M.: Total variation wavelet inpainting. J. Math. Imaging Vis. 25(1), 107–125 (2006)
Li, M., Nguyen, T.Q.: Markov random field model-based edge-directed image interpolation. IEEE Trans. Image Process. 17(7), 1121–1128 (2008)
Joshi, N., et al.: Image deblurring and denoising using color priors. IEEE Conf. Comput. Vis. Pattern Recognit. 2009, 1550–1557 (2009)
Arietta, S., Lawrence, J.: Building and using a database of one trillion natural-image patches. IEEE Comput. Graph. Appl. 31(1), 9–19 (2010)
Levin, A., Nadler, B.: Natural image denoising: Optimality and inherent bounds. IEEE Conf. Comput. Vis. Pattern Recognit. 2011, 2833–2840 (2011)
Zoran, D., Weiss, Y.: From learning models of natural image patches to whole image restoration. IEEE Int. Conf. Comput. Vis 2011, 479–486 (2011)
Tang, S., et al.: Non-blind image deblurring method by local and nonlocal total variation models. Sig. Process. 94(1), 339–349 (2014)
Jidesh, P., Bini, A.A.: Image despeckling and deblurring via regularized complex diffusion. SIViP 11(6), 977–984 (2017)
Hong, H.Y., Hua, X., Zhang, X.H., Shi, Y.: Multi-frame real image restoration based on double loops with alternative maximum likelihood estimation. SIViP 10(8), 1489–1495 (2016)
Fatih, K., Cabir, V.: Blind restoration and resolution enhancement of images based on complex filtering. SIViP 10(6), 1159–1167 (2016)
Lai, Z., et al.: Sparse tensor discriminant analysis. IEEE Trans. Image Process. 22(10), 3904–3915 (2013)
Gu, S., et al.: Weighted nuclear norm minimization with application to image denoising. IEEE Conf. Comput. Vis. Pattern Recognit. 2014, 2862–2869 (2014)
Rudin, L.I., Osher, S., Fatemi, E.: Nonlinear total variation based noise removal algorithms. Phys. D 60(1), 259–268 (1992)
Cho, T.S., et al.: Image restoration by matching gradient distributions. IEEE Trans. Pattern Anal. Mach. Intell. 34(4), 683–694 (2012)
Krishnan, D., Fergus, R.: Fast image deconvolution using hyper-Laplacian priors. In: NIPS’09 Proceeding of the 22nd International Conference on Neural Information Processing Systems, pp. 1033–1041 (2009)
Cho, T.S., et al.: A content-aware image prior. IEEE Conf. Comput. Vis. Pattern Recognit. 2010, 169–176 (2010)
Levin, A., et al.: Image and depth from a conventional camera with a coded aperture. ACM Trans. Graph. 26(3), 70 (2007)
Couzinie-Devy, F., et al.: Learning to estimate and remove non-uniform image blu. IEEE Conf. Comput. Vis. Pattern Recognit. 2013, 1075–1082 (2013)
Cheng, Q., et al.: Inpainting for remotely sensed images with a multichannel nonlocal total variation model. IEEE Trans. Geosci. Remote Sens. 52(1), 175–187 (2014)
Liu, H., et al.: Non-local extension of total variation regularization for image restoration. IEEE Int. Symp. Circuits Syst. (ISCAS) 2014, 1102–1105 (2014)
Liu, R.W., et al.: Generalized total variation-based MRI Rician denoising model with spatially adaptive regularization parameters. Magn. Reson. Imaging 32(6), 702–720 (2014)
Wu, H., Wu, Y., Wen, Z.: Texture smoothing based on adaptive total variation. Found. Intell. Syst. 277, 43–54 (2014)
Bishop, T.E., Molina, R., Hopgood, J.R.: Nonstationary blind image restoration using variational methods. ICIP 1, 125–128 (2007)
Pogue, B.W., et al.: Spatially variant regularization improves diffuse optical tomography. Appl. Opt. 38(13), 2950–2961 (1999)
Xu, H., et al.: Iterative nonlocal total variation regularization method for image restoration. PLoS ONE 8(6), e65865 (2013)
Zuo, W., et al.: A generalized iterated shrinkage algorithm for non-convex sparse coding. IEEE Int. Conf. Comput. Vis. 2013, 217–224 (2013)
Levin, A., et al.: Understanding and evaluating blind deconvolution algorithms. IEEE Conf. Comput. Vis. Pattern Recognit. 2009, 1964–1971 (2009)
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Cheng, J., Gao, Y., Guo, B. et al. Image restoration using spatially variant hyper-Laplacian prior. SIViP 13, 155–162 (2019). https://doi.org/10.1007/s11760-018-1340-7
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11760-018-1340-7