Abstract
We propose a super-resolution reconstruction model based on the fusion of convolutional neural networks and regularization constraints. Our model not only takes advantage of the convolutional neural network’s prominent capability for nonlinear mapping between low-resolution and high-resolution images, but also takes the image inherent tendency to have bountiful repeated structural information into accounts. We derive our total variation regularization constraints based on the image local similarity and non-local similarity. Through coalescence of convolutional nerual network and delicately devised adaptive regularization constraints, our model yields a state-of-the-art restoration quality from a single image. Besides, our system can be expanded to tackle more low-level vision problems as well.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Hou, H.S., Andrews, H.: Cubic splines for image interpolation and digital filtering. IEEE Trans. Acoust. Speech Signal Process. 26(6), 508–517 (1978)
Irani, M., Peleg, S.: Super resolution from image sequences. In: Pattern Recognition, pp. 115–120 (1990)
Tom, B.C., Katsaggelos, A.K., Galatsanos, N.P.: Reconstruction of a high resolution image from registration and restoration of low resolution images. In: ICIP (1994)
Kaltenbacher, E., Hardie, R.C.: High resolution infrared image reconstruction using multiple, low resolution, aliased frames. In: NAECON (1996)
Hardie, R.C., Barnard, K.J., Armstrong, E.E.: Joint MAP registration and high-resolution image estimation using a sequence of undersampled images. ICIP 6(16), 1621–1633 (1997)
Schultz, R.R., Stevenson, R.L.: Extraction of high-resolution frames from video sequences. ICIP 5(6), 996–1011 (1996)
Chung, J., Haber, E., Nagy, J.: Numerical methods for coupled superresolution. Inverse Prob. 22(4), 1261–1272 (2006)
Tom, B.C., Katsaggelos, A.K.: Reconstruction of a high-resolution image by simultaneous registration, restoration, and interpolation of low-resolution images. In: ICIP (1995)
Michael, E.T., Christopher M.B.: Bayesian image superresolution. In: Proceedings of Advances in Neural Information Proceeding Systems, pp. 1279–1286 (2003)
Freeman, W.T., Jones, T.R., Pasztor, E.C.: Example-based super-resolution. Comput. Graphics Appl. 22(2), 56–65 (2002)
Sun, J., Zheng, N.N., Tao, H., Shum, H.Y.: Image hallucination with primal sketch priors. In: CVPR (2003)
Wang, Q., Tang, X., Shum, H.: Patch based blind image super resolution. In: ICCV (2005)
Chang, H., Yeung, D.Y., Xiong, Y.: Super-resolution through neighbor embedding. In: CVPR (2004)
Yang, J., Wright, J., Huang, T., Ma, Y.: Image super-resolution as sparse representation of raw image patches. In: CVPR (2008)
Nadir, V., Hinton G.E.: Rectified linear units improve restricted Boltzmann machines. In: ICML, pp. 807–814 (2010)
Lan, G.: An optimal method for stochastic convex optimization. Technical report, Georgia Institute of Technology (2009)
Lecun, Y., Bottou, L., Bengio Y., Haffner: Gradient-based learning applied to document recognition. In: Proceedings of the IEEE, vol. 86, no.11, pp. 2278-2324 (1998)
Takeda, H., Farsiu, S., Milanfar, P.: kernel regression for image processing and reconstruction. Image Process. 16(2), 349–366 (2007)
Goldstein, T., Osher, S.: The split Bregman method for L1-regularized problems. J. SIAM J. Imaging Sci. 2(2), 323–343 (2009)
Timofte, R., De V., Van Gool, L.: Anchored Neighborhood Regression for Fast Example-Based Super-Resolution. In: ICCV, pp. 1920–1927 (2013)
Dong, C., Loy, C., He K., Tang X.: Image super-resolution using deep convolutional networks. In: Pattern Analysis and Machine Intelligence (2014)
Vedaldi, A., Lenc, K.: MatConvNet - convolutional neural networks for MATLAB (2014). arXiv:1412.4564
Zhang, K., Gao, X., Tao, D., Li, X.: Single image super-resolution with non-local means and steering kernel regression. Image Process. 21(11), 4544–4556 (2012)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2016 Springer International Publishing Switzerland
About this paper
Cite this paper
Qu, Y., Shi, C., Liu, J., Peng, L., Du, X. (2016). Single Image Super-Resolution via Convolutional Neural Network and Total Variation Regularization. In: Tian, Q., Sebe, N., Qi, GJ., Huet, B., Hong, R., Liu, X. (eds) MultiMedia Modeling. MMM 2016. Lecture Notes in Computer Science(), vol 9517. Springer, Cham. https://doi.org/10.1007/978-3-319-27674-8_3
Download citation
DOI: https://doi.org/10.1007/978-3-319-27674-8_3
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-27673-1
Online ISBN: 978-3-319-27674-8
eBook Packages: Computer ScienceComputer Science (R0)