access icon free Single-image super-resolution using online kernel adaptive filters

The online kernel adaptive filters are non-linear filters which provide impulse response and are more efficient compared to other kernel algorithms. The performance of kernel adaptive filters depends on dictionary size. Here the single-image super-resolution using online kernel adaptive filters is a learning-based method. The algorithm generates a sparser solution for obtaining high-resolution image from a low-resolution image. It finds out a dictionary with most significant set of basis vectors using the spatial similarity among the dictionaries created from the low-resolution and high-resolution image patches in the training set. The dictionary is utilised to generate the high-resolution image. The algorithm is analysed on three different kernel adaptive filters, extended kernel recursive least squares, kernel recursive least squares tracker and naive online regularised risk minimisation algorithm. The performance of the super-resolution method is evaluated on a large number of images and is compared with the state-of-the art non-linear solutions to the super-resolution. The results show a better progress in peak signal-to-noise ratio up to 1.2 dB.

Inspec keywords: least squares approximations; image classification; Bayes methods; minimisation; learning (artificial intelligence); filtering theory; image resolution; adaptive filters

Other keywords: dictionary; naive online regularised risk minimisation algorithm; kernel algorithms; extended kernel recursive least squares; low-resolution image; nonlinear filters; single-image super-resolution; super-resolution method; kernel recursive least squares tracker; different kernel adaptive filters; noise figure 1.2 dB; online kernel adaptive filters; high-resolution image patches

Subjects: Optimisation techniques; Knowledge engineering techniques; Filtering methods in signal processing; Other topics in statistics; Optical, image and video signal processing; Interpolation and function approximation (numerical analysis); Interpolation and function approximation (numerical analysis); Computer vision and image processing techniques

References

    1. 1)
      • 12. Timofte, R., De Smet, V., Van Gool, L.: ‘A + : adjusted anchored neighborhood regression for fast super-resolution’. Computer Vision, Heidelberg, Germany, 2014, pp. 111126.
    2. 2)
      • 1. Rhee, S., Kang, M.G.: ‘Discrete cosine transform based regularized high-resolution image reconstruction algorithm’, Opt. Eng., 1999, 38, pp. 13481356.
    3. 3)
      • 10. Chang, H., Yeung, D.Y., Xiong, Y.: ‘Super-resolution through neighbor embedding’. Proc. Int. IEEE Conf. CVPR, Washington D.C, USA, 2004, vol. 1, pp. 275282.
    4. 4)
      • 16. Ming, C., Cheng, W., Jonathan, L.: ‘Single-image super-resolution in RGB space via group sparse representation’, IET Image Process., 2015, 9, pp. 461467.
    5. 5)
      • 21. Fisher, I.: ‘Pattern recognition algorithms for symbol strings’, PhD thesis, Department of Computer Architecture, University of Tubingen, 2003.
    6. 6)
      • 25. Kivinen, J., Smola, A.J., Williamson, R.C.: ‘Online learning with kernels’, IEEE Trans. Signal Process., 2004, 52, (8), pp. 21652176.
    7. 7)
      • 8. Freeman, W.T., Pasztor, E., Carmichael, O.: ‘Learning low-level vision’, Int. J. Comput. Vis., 2000, 40, (1), pp. 2547.
    8. 8)
      • 23. Liu, W., Park, I.M., Wang, Y., et al: ‘Extended kernel recursive least squares algorithm’, IEEE Trans. Signal Process., 2009, 57, pp. 38013814.
    9. 9)
      • 6. Schultz, R.R., Stevenson, R.L.: ‘A Bayesian approach to image expansion for improved definition’, IEEE Trans. Image Process., 1994, 3, (3), pp. 233242.
    10. 10)
      • 13. Timofte, R., De Smet, V., Van Gool, L.: ‘Anchored neighborhood regression for fast example-based super-resolution’. Proc. IEEE Int. Conf. Comput. Vis., Sydney, Australia, December 2013, pp. 19201927.
    11. 11)
      • 26. Van, V., Santamar Ignacio, S.: ‘A comparative study of kernel adaptive filtering algorithms’, IEEE Digital Signal Processing (DSP) Workshop and IEEE Signal Processing Education (SPE), Napa, California, USA, 2013, Software: http://sourceforge.net/projects/kafbox/.
    12. 12)
      • 18. Kim, J., Lee, J. K., Lee, K. M.: ‘Accurate image super-resolution using very deep convolutional networks’. Proc. IEEE Conf. Comput. Vis. Pattern Recognit. (CVPR), Las Vegas, USA, June 2016, pp. 16461654.
    13. 13)
      • 11. Sun, J., Zheng, N.N., Tao, H., et al: ‘Image hallucination with primal sketch priors’. Proc. Int. IEEE Conf. Computer Vision and Pattern Recognition, Madison, USA, 2003, Vol. 2, pp. 729736.
    14. 14)
      • 9. Fattal, R.: ‘Image upsampling via imposed edge statistics’, ACM Trans. Graph., 2007, 26, (3), pp. 95:195:8.
    15. 15)
      • 14. Cruz, C., Mehta, R., Katkovnik, V., et al: ‘Single image super-resolution based on Wiener filter in similarity domain’, IEEE Trans. Image Process., 2018, 27, (3), pp. 13761389.
    16. 16)
      • 4. Dai, S., Han, M., Xu, W., et al: ‘Softcuts: a soft edge smoothness prior for color image super-resolution’. IEEE Trans. Image Process., 2009, 18, (5), pp. 969981.
    17. 17)
      • 24. Van, V., Lazaro-Gredilla, S., Santamaria, M.I.: ‘Kernel recursive least-squares tracker for time-varying regression’, IEEE Trans. Neural Netw. Learn. Syst., 2012, 23, pp. 13131326.
    18. 18)
      • 22. Liu, W., Prıncipe, J., Haykin, S.: ‘Kernel adaptive filtering: a comprehensive introduction’ (Wiley, New York, 2010).
    19. 19)
      • 19. Ni, K.S., Nguyen, T.Q.: ‘Image super resolution using support vector regression’. IEEE Trans. Image Process., 2007, 16, (6), pp. 15961610.
    20. 20)
      • 15. Kim, J., Lee, J.K., Lee, K.M.: ‘Deeply-recursive convolutional network for image super-resolution’. Proc. IEEE Conf. Comput. Vis. Pattern Recognit. (CVPR), Las Vegas, USA, June 2016, pp. 16461654.
    21. 21)
      • 2. Nguyen, N., Milanfar, P.: ‘A wavelet-based interpolation-restoration method for superresolution (wavelet superresolution)’, Circuits Syst. Signal Process., 2000, 19, (4), pp. 321338.
    22. 22)
      • 3. Ben-Ezra, M., Zomet, A., Nayar, S.K.: ‘Video super-resolution using controlled subpixel detector shifts’, IEEE Trans. Pattern Anal. Mach. Intell., 2005, 27, (6), pp. 977987.
    23. 23)
      • 17. Liu, D., Wang, Z., Wen, B., et al: ‘Robust single image super-resolution via deep networks with sparse prior’, IEEE Trans. Image Process., 2016, 25, (7), pp. 31943207.
    24. 24)
      • 5. Patti, A.J., Altunbasak, Y.: ‘Artifact reduction for set theoretic super resolution image reconstruction with edge adaptive constraints and higher-order interpolants’, IEEE Trans. Image Process., 2001, 10, (1), pp. 179186.
    25. 25)
      • 27. Li, X., Orchard, M.T.: ‘New edge-directed interpolation’, IEEE Trans. Image Process., 2001, 10, (10), pp. 15211527.
    26. 26)
      • 7. Hardie, R.C., Barnard, K.J., Armstrong, E.E.: ‘Join MAP registration and high resolution image estimation using a sequence of under sampled images’, IEEE Trans. Image Process., 1997, 6, (12), pp. 16211633.
    27. 27)
      • 20. Jesna, A., Abdulla, P.: ‘Single-image super-resolution using kernel recursive least squares’, Signal Image Video Process., 2016, 10, (8), pp. 15511558.
http://iet.metastore.ingenta.com/content/journals/10.1049/iet-ipr.2018.5319
Loading

Related content

content/journals/10.1049/iet-ipr.2018.5319
pub_keyword,iet_inspecKeyword,pub_concept
6
6
Loading