Research on the single image super-resolution method based on sparse Bayesian estimation
- 57 Downloads
Aiming at the problem that the super-resolution effect on different low-resolution images of existing super-resolution algorithms have large difference, a novel single image super-resolution method based on sparse Bayesian estimation is proposed. In this method, the single image super-resolution problem is regarded as a regression problem. The Kronecker pulse functions are adopted as the regression basis functions, and the optimal sparse solution of the specific prediction is obtained by combining the local information and global information of the image. The Bayesian method is adopted to estimate the weights, so as to reconstructe the super-resolution image. The experimental results show that our proposed method can obtain high average peak signal to noise ratio, small variance and less time-consuming, when carried out on 14 testing images for single image super-resolution. It is proved that our proposed method has good super-resolution effect, strong adaptability, and high efficiency.
KeywordsSingle image super-resolution Super-resolution Bayesian estimations Regression Sparse representation
- 7.Dong, C., Loy, C.C., He, K., et al.: Learning a deep convolutional network for image super-resolution [CComputer Vision–ECCV, pp. 184–199. Springer International Publishing, New York (2014)Google Scholar
- 9.Timofte, R., De, V., Gool, L.: Anchored neighborhood regression for fast example-based super-resolution[C]//Computer Vision (ICCV). In: Proceedings of the IEEE International Conference on. IEEE, pp. 1920–1927 (2013)Google Scholar
- 10.Schulter S, Leistner. C., Bischof, H.: Fast and accurate image upscaling with super-resolution forests. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. pp. 3791–3799 (2015)Google Scholar
- 13.Grave, E., Obozinski, G.R., Bach, F.R.: Trace lasso: a trace norm regularization for correlated designs. Adv. Neur. Inf. Process. Syst. 8, 2187–2195 (2011)Google Scholar