Abstract
Sparse representation has shown to be a successful technique in the fields of image restoration and other image processing tasks over the past few years. The key of the approach with application to image recovery is how to sparsely represent the true-image data, which seriously affects its recovery quality. One such representative method is the widely used wavelet tight frame. However, the filters of the classical wavelet frame are fixed and unchanged for different input images. To overcome the limitation, a data-driven tight frame was proposed to learn a good sparse representation from the image itself very recently. However, the filters of the tight frame are only learned from the selected local image patches, and it doesn’t utilize the non-local similarity of image patches. In this paper, we further propose a 3D data-driven tight frame learning scheme which considers both the local and non-local sparse representation of image patches, and establish a variational model for image recovery based on the new sparse representation method. Numerical experiments demonstrate that our approach is competitive with the recently proposed state-of-the-art methods.
Similar content being viewed by others
References
Steidl, G.: A note on the dual treatment of higher-order regularization functionals. Computing 76(1–2), 135–148 (2006)
Bredies, K., Kunisch, K., Pock, T.: Total generalized variation. SIAM J. Imaging Sci. 3(3), 492–526 (2010)
Chen, D.Q., Cheng, L.Z., Su, F.: A new TV-Stokes model with augmented Lagrangian method for image denoising and deconvolution. J. Sci. Comput. 51(3), 505–526 (2012)
Lefkimmiatis, S., Ward, J.P., Unser, M.: Hessian Schatten-norm regularization for linear inverse problems. IEEE Trans. Image Process. 22(5), 1873 (2013)
Candès, E.J., Donoho, D.L.: New tight frames of curvelets and optimal representations of objects with piecewise C2 singularities. Commun. Pure Appl. Math. 57(2), 219–266 (2004)
Le Pennec, E., Mallat, S.: Sparse geometric image representations with bandelets. IEEE Trans. Image Process. 14(4), 423–438 (2005)
Easley, G., Labate, D., Lim, W.Q.: Sparse directional image representations using the discrete shearlet transform. Appl. Comput. Harmon. Anal. 25(1), 25–46 (2008)
Le Pennec, E., Mallat, S.: Bandelet image approximation and compression. Multiscale Model. Simul. 4(3), 992–1039 (2005)
Aharon, M., Elad, M., Bruckstein, A.: K-SVD: an algorithm for designing overcomplete dictionaries for sparse representation. IEEE Trans. Signal Process. 54(11), 4311–4322 (2006)
Elad, M., Aharon, M.: Image denoising via sparse and redundant representations over learned dictionaries. IEEE Trans. Image Process. 15(12), 3736–3745 (2006)
Mairal, J., Sapiro, G., Elad, M.: Learning multiscale sparse representations for image and video restoration. Multiscale Model. Simul. 7(1), 214–241 (2008)
Cai, J.F., Ji, H., Shen, Z., et al.: Data-driven tight frame construction and image denoising. Appl. Comput. Harmon. Anal. 37(1), 89–105 (2014)
Chen, D.Q.: Image denoising based on improved data-driven sparse representation. arXiv preprint arXiv:1502.03273 (2015)
Buades, A., Coll, B., Morel, J.M.: A review of image denoising algorithms, with a new one. Multiscale Model. Simul. 4(2), 490–530 (2005)
Kindermann, S., Osher, S., Jones, P.W.: Deblurring and denoising of images by nonlocal functionals. Multiscale Model. Simul. 4(4), 1091–1115 (2005)
Zhang, X., Burger, M., Bresson, X., et al.: Bregmanized nonlocal regularization for deconvolution and sparse reconstruction. SIAM J. Imaging Sci. 3(3), 253–276 (2010)
Deledalle, C.A., Denis, L., Tupin, F.: How to compare noisy patches? Patch similarity beyond Gaussian noise. Int. J. Comput. Vis. 99(1), 86–102 (2012)
Ram, I., Elad, M., Cohen, I.: Image processing using smooth ordering of its patches. IEEE Trans. Image Process. 22(7), 2764–2774 (2013)
Dabov, K., Foi, A., Katkovnik, V., Egiazarian, K.: Image denoising by sparse 3D transform-domain collaborative filtering. IEEE Trans. Image Process. 16(8), 2080–2095 (2007)
Dabov, K., Foi, A., Katkovnik, V., Egiazarian, K.: Image restoration by sparse 3D transform-domain collaborative filtering. In: Proceedings of SPIE, the International Society for Optical Engineering, pp. 681207-1–681207-12. Bellingham, Wash, Society of Photo-Optical Instrumentation Engineers (2008)
Chatterjee, P., Milanfar, P.: Clustering-based denoising with locally learned dictionaries. IEEE Trans. Image Process. 18(7), 1438–1451 (2009)
Danielyan, A., Katkovnik, V., Egiazarian, K.: BM3D frames and variational image deblurring. IEEE Trans. Image Process. 21(4), 1715–1728 (2012)
Dong, W., Li, X., Zhang, D., et al.: Sparsity-based image denoising via dictionary learning and structural clustering. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2011, pp. 457–464
Mairal, J., Bach, F., Ponce, J., et al.: Non-local sparse models for image restoration. In: IEEE International Conference on Computer Vision (ICCV), 2009, pp. 2272–2279
Dong, W., Shi, G., Li, X.: Nonlocal image restoration with bilateral variance estimation: a low-rank approach. IEEE Trans. Image Process. 22(2), 700–711 (2013)
Quan, Y., Ji, H., Shen, Z.: Data-driven multi-scale non-local wavelet frame construction and image recovery. J. Sci. Comput. 63(2), 307–329 (2014)
Chen, D.Q., Zhou, Y.: Wavelet frame based image restoration via combined sparsity and nonlocal prior of coefficients. J. Sci. Comput. 66(1), 196–224 (2016)
Ram, I., Elad, M., Cohen, I.: Generalized tree-based wavelet transform. IEEE Trans. Signal Process. 59(9), 4199–4209 (2011)
Daubechies, I., Han, B., Ron, A., et al.: Framelets: MRA-based constructions of wavelet frames. Appl. Comput. Harmon. Anal. 14(1), 1–46 (2003)
Dong, B., Shen, Z.: MRA based wavelet frames and applications. In: IAS Lecture Notes Series, Summer Program on The Mathematics of Image Processing, Park City Mathematics Institute, 2010
Wipf, D., Nagarajan, S.: Iterative reweighted l1 and l2 methods for finding sparse solutions. IEEE J. Sel. Top. Signal Process. 4(2), 317–329 (2010)
Zhang, Y., Dong, B., Lu, Z.: \(l_0\) Minimization for wavelet frame based image restoration. Math. Comput. 82(282), 995–1015 (2013)
Dong, B., Zhang, Y.: An efficient algorithm for \(l_0\) minimization in wavelet frame based image restoration. J. Sci. Comput. 54(2–3), 350–368 (2013)
Ram, I., Elad, M., Cohen, I.: Redundant wavelets on graphs and high dimensional data clouds. IEEE Signal Process. Lett. 19(5), 291–294 (2012)
Wang, Z., Bovik, A.C., Sheikh, H.R., et al.: Image quality assessment: from error visibility to structural similarity. IEEE Trans. Image Process. 13(4), 600–612 (2004)
Tikhonov, A.N., Arsenin, V.I.A.: Solutions of Ill-Posed Problems. Wiley, New York (1977)
Eckstein, J., Bertsekas, D.P.: On the Douglas Rachford splitting method and the proximal point algorithm for maximal monotone operators. Math. Program. 55(1–3), 293–318 (1992)
Acknowledgments
The work was supported in part by the National Natural Science Foundation of China under Grant 61401473, and the Social Livelihood Science and Technology Innovation Special Project of CSTC (cstc2015shmszx120002). We appreciate the constructive comments of the anonymous reviewers, which led to great improvements in this manuscript.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Chen, DQ. Data-Driven Tight Frame Learning Scheme Based on Local and Non-local Sparsity with Application to Image Recovery. J Sci Comput 69, 461–486 (2016). https://doi.org/10.1007/s10915-016-0205-x
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10915-016-0205-x