access icon free Texture and edge preserving multiframe super-resolution

Super-resolution (SR) image reconstruction refers to methods where a higher resolution image is reconstructed using a set of overlapping aliased low-resolution observations of the same scene. Although edge preservation has been a widely explored topic in SR literature, texture-specific regularisation has recently gained interest. In this study, texture-specific regularisation is handled as a post-processing step. A two stage method is proposed, comprising multiple SR reconstructions with different regularisation parameters followed by a restoration step for preserving edges and textures. In the first stage, two maximum-a-posteriori estimators with two different amounts of regularisation are employed. In the second stage, pixel-to-pixel difference between these two estimates is post-processed to restore edges and textures. Frequency selective characteristics of discrete cosine transform and Gabor filters are utilised in the post-processing step. Experiments on synthetically generated images and real experiments demonstrate that the proposed methods give better results compared with the state-of-the-art SR methods especially on textures and edges.

Inspec keywords: image reconstruction; discrete cosine transforms; maximum likelihood estimation; Gabor filters; image texture; image resolution

Other keywords: super resolution image reconstruction; restoration step; edge preserving multiframe super resolution; edge preservation; pixel-to-pixel difference; Gabor fllters; regularisation parameters; post-processing step; discrete cosine transform; texture preserving multiframe super resolution; two stage method; overlapping aliased low-resolution observation; frequency selective characteristics; multiple SR reconstruction; maximum-a-posteriori estimators; texture-speciflc regularisation; SR literature

Subjects: Integral transforms; Optical, image and video signal processing; Graphics techniques; Filtering methods in signal processing; Integral transforms; Computer vision and image processing techniques; Other topics in statistics; Other topics in statistics

References

    1. 1)
      • 25. Tipping, M.E., Bishop, C.M.: ‘Bayesian image super-resolution’. In Becker, S., Thrun, T., Oberymayer, K. (Ed.): ‘Advances in neural information processing systems’ (MIT Press, 2003), pp. 13031310.
    2. 2)
      • 14. Turgay, E., Akar, G.B.: ‘Directionally adaptive super-resolution’. Proc. IEEE Int. Conf. Image Processing, November 2009, pp. 120104.
    3. 3)
    4. 4)
    5. 5)
    6. 6)
    7. 7)
      • 17. Turgay, E., Akar, G.B.: ‘Context based super resolution image reconstruction’. Proc. IEEE Local and Non-Local Approximation in Image Processing, August 2009, pp. 5461.
    8. 8)
    9. 9)
      • 3. Tekalp, A.M., Ozkan, M.K., Sezan, M.I.: ‘High-resolution image reconstruction from lower-resolution image sequences and space varying image restoration’. Proc. IEEE Int. Conf. on Acoustics, Speech and Signal Processing (ICASSP), IEEE San Francisco, CA, March 1992, vol. 3, pp. 169172.
    10. 10)
    11. 11)
      • 12. Pickup, L.C., Capel, D.P., Roberts, S.J., Zisserman, A.: ‘Bayesian image super-resolution, continued’, in Schölkopf, B., Roberts, S.J., Zisserman, A. (Ed.): ‘Advances in neural information processing systems’ (MIT Press, 2006), pp. 10891096.
    12. 12)
    13. 13)
    14. 14)
    15. 15)
    16. 16)
    17. 17)
      • 19. Pickup, L.C., Roberts, S.J., Zisserman, A.: ‘A sampled texture prior for image super-resolution’. Proc. Advances in Neural Information Processing Systems 16 (NIPS ‘03), December 2004, pp. 15871594.
    18. 18)
      • 32. Trandon, T.: ‘Brodatz textures’, http://www.ux.uis.no/~tranden/brodatz.html.
    19. 19)
    20. 20)
    21. 21)
    22. 22)
    23. 23)
    24. 24)
    25. 25)
    26. 26)
      • 1. Tsai, R.Y., Huang, T.S.: ‘Multiframe image restoration and registration’. Advances in Computer Vision and Image Processing, 1984, vol. 1, no 2, pp. 317339.
    27. 27)
      • 9. Bose, N.K., Kim, S.P., Valenzuela, H.M.: ‘Recursive implementation of total least squares algorithm for image reconstruction from noisy, undersampled multiframes’. Proc. IEEE Int. Conf. Acoustics, Speech and Signal Processing, 1993, vol. 5, pp. 269272.
    28. 28)
    29. 29)
      • 39. Keren, D., Peleg, S., Brada, R.: ‘Image sequence enhancement using sub-pixel displacement’. IEEE Conf. Computer Vision and Pattern Recognition, June 1988, pp. 742746.
    30. 30)
    31. 31)
    32. 32)
      • 36. Sorwar, G., Abraham, A.: ‘DCT based texture classification using a soft computing approach’, Malays. J. Comput. Sci., 2004, 17, (1), pp. 1223.
    33. 33)
      • 37. Shewchuk, J.R.: ‘An introduction to the conjugate gradient method without the agonizing pain’. Technical Report, Pittsburgh, PA, USA, 1994.
    34. 34)
    35. 35)
    36. 36)
      • 2. Milanfar, P.: ‘Super-resolution imaging’ (CRC Press, 2010), vol. 1, pp. 520.
    37. 37)
      • 15. Su, H., Tang, L., Tretter, D., Zhou, J.: ‘A practical and adaptive framework for super-resolution’. Proc. IEEE Int. Conf. Image Processing, November 2008, pp. 123639.
    38. 38)
      • 26. Kumar, L., Kumar, C.N.R.: ‘Novel super resolution reconstruction of low resolution images progressively using DCT and zonal filter based denoising’, Int. J. Comput. Sci. Inf. Technol., 2011, 3, (1), pp. 5472.
    39. 39)
http://iet.metastore.ingenta.com/content/journals/10.1049/iet-ipr.2013.0342
Loading

Related content

content/journals/10.1049/iet-ipr.2013.0342
pub_keyword,iet_inspecKeyword,pub_concept
6
6
Loading