A Super-Resolution Reconstruction Algorithm Based on Learning Improvement

  • Han Gao
  • Xinwei Li
  • Aiping JiangEmail author
Conference paper
Part of the Lecture Notes in Electrical Engineering book series (LNEE, volume 517)


In view of the problem of edge blurring and slow reconstruction speed in the existing super-resolution reconstruction of learning-based images, this paper proposes an improvement to the original learning-based reconstruction algorithm and applies Markov random fields to image super-resolution. In the rate reconstruction, during the dictionary training phase, training image blocks are randomly selected, and the texture part of the image is learned and reconstructed. The bicubic interpolation method is used to enlarge the image structure part and color information, the final reconstructed image and the interpolation-amplified image are merged. That is the final result. The experimental results show that the peak signal-to-noise ratio (PSNR) is used to objectively evaluate the image reconstruction effect, and it is concluded that the algorithm of this paper reconstructs the image better, and the reconstruction time also has a certain increase.


Learning-based Super-resolution reconstruction Image decomposition Markov random field 



This work is supported in part by the National Natural Science Foundation China (61601174), in part by the Postdoctoral Research Foundation of Heilongjiang Province (LBH-Q17150), in part by the Science and Technology Innovative Research Team in Higher Educational Institutions of Heilongjiang Province (No. 2012TD007), in part by the Fundamental Research Funds for the Heilongjiang Provincial Universities (KJCXZD201703), and in part by the Science Foundation of Heilongjiang Province of China (F2018026).


  1. 1.
    Yan, Z., Kun, X.U., Yong, L.I.: Remote sensing image super-resolution based on POCS and out-of-core. J. Tsinghua Univ. 50(10), 1743–341 (2010)Google Scholar
  2. 2.
    Kim, K.I., Kwon, Y.: Single-image super-resolution using sparse regression and natural image prior. IEEE Trans. Pattern Anal. Mach. Intell. 32(6), 1127 (2010)CrossRefGoogle Scholar
  3. 3.
    Shou, Z.Y., Liao, M.L., Zhang, T.: Improved image super-resolution via sparse representation based on IBP. Comput. Eng. Des. (2014)Google Scholar
  4. 4.
    Freeman, W.T., Pasztor, E.C.: Learning low-level vision. Int. J. Comput. Vis. 40(1), 25–47 (2000)CrossRefGoogle Scholar
  5. 5.
    Yang, J., Wright, J., Huang, T., et al.: Image super-resolution as a sparse representation of raw image patches. In: IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2008, vol. 2008, pp. 1–8. IEEE (2008)Google Scholar
  6. 6.
    Tang, Y., Yan, P., Li, X.: Single-image super-resolution via local learning. Int. J. Mach. Learn. Cybernet. 2(1), 15–23 (2011)CrossRefGoogle Scholar
  7. 7.
    Freeman, W.T., Jones, T.R., Pasztor, E.C.: Example-Based Super-Resolution. IEEE Computer Society Press (2002)Google Scholar

Copyright information

© Springer Nature Singapore Pte Ltd. 2020

Authors and Affiliations

  1. 1.Heilongjiang UniversityHarbinChina

Personalised recommendations