Adaptive Spatial and Temporal Prefiltering for Video Compression

  • Astrid Lundmark
  • Leif Haglund
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 2749)

Abstract

When compressing video sequences, noise is fatal to compression performance. Imaging is inherently a noisy process since the number of photons hitting a detector is a statistical process. When scene illumination is high, the statistical nature of the number of photons hitting each detector is of no importance, since other effects such as quantization dominate. For other applications, military for example, where the imaging has to be done with whatever illumination is naturally available, the noise can be both annoying for the observer and make transmission of the imagery consume excessive amounts of bandwidth. We show results of using content based spatial prefiltering combined with motion vector certainty controlled temporal prefiltering to reduce the noise and thus improve both visual impression and transmission properties.

Keywords

Test Sequence Video Code Quantization Parameter Video Compression Digital Imagery 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

References

  1. 1.
    van Roosmalen, P. M. B., Kokaram, A. C., Biemond, J.: Noise Reduction of Image Sequences as Preprocessing for MPEG2 Encoding. European Conference on Signal Processing (EUSIPCO) (1998) 2253–2256Google Scholar
  2. 2.
    Tsuji, H. Sakatani, T., Yashima, Y., Kobayashi, N.: A Nonlinear Spatio-Temporal Diffusion and its Application to Prefiltering in MPEG-4 Video Coding. International Conference on Image Processing (2002) 85–88Google Scholar
  3. 3.
    Karunaratne, P. V., Segall, C. A., Katsaggelos, A. K.: A Rate-Distortion Optimal Video Pre-Processing Algorithm. International Conference on Image Processing (2001) 481–484Google Scholar
  4. 4.
    Vasconcelos, N., Dufaux, F.: Pre and Post-Filtering for Low Bit-Rate Video Coding. International Conference on Image Processing (1997) 291–294Google Scholar
  5. 5.
    Gottardo F., Mir, J., Parladori, G., Siracuranza, G. L.: Nonlinear Preprocessing for MPEG-2 Video Sequences. Applied Signal Processing (6):61–70, 1999.CrossRefGoogle Scholar
  6. 6.
    Donoho, D. L.: De-Noising by Soft-Thresholding. IEEE Transactions on Information Theory, 41(3):613–627, May 1995.MATHCrossRefMathSciNetGoogle Scholar
  7. 7.
    Graña, M., Echave, I. Ruiz-Cabello, J.: VQ Based Bayesian Image Filtering. International Conference on Image Processing (2000)Google Scholar
  8. 8.
    Panchapakesan, K., Bilgin, A., Sheppard, D. G., Marcellin, M. W., Hunt, B. R.: Simultaneous Compression and Denoising of Imagery using Non-Linear Interpolative Vector Quantization. IEEE Digital Signal Processing Workshop (1998)Google Scholar
  9. 9.
    Faber, J., Ihlenburg, L., Mejers, T., Ruschin, D.: Irrelevancy due to Visual Tracking Errors. Picture Coding Symposium (1997)Google Scholar
  10. 10.
    Biloslavo, M., Ramponi, G., Oliveiri, S., Albani, L.: Joint Kalman-Based Noise Filtering and Motion Compensated Video Coding for Low Bit Rate Videoconferencing. International Conference on Image Processing (2000)Google Scholar
  11. 11.
    Bigün, J.: Local Symmetry Features in Image Processing. PhD Thesis, Linköping University, Sweden, 1988. Dissertation No 179, ISBN 91-7870-334-4.Google Scholar
  12. 12.
    Knutsson, H., Wilson, R. and Granlund, G. H.: Anisotropic Non-Stationary Image Estimaiton and its Applications — part I: Restoration of Noisy Images. IEEE Transactions on Communications, COM-31(3):388–397, March 1983.CrossRefGoogle Scholar
  13. 13.
    Haglund, L.: Adaptive Multidimensional Filtering. PhD Thesis, Linköping University, Sweden, 1992. Dissertation No 284, ISBN 91-7870-988-1.Google Scholar
  14. 14.
    Lundmark, A.: Hierarchical Structures and Extended Motion Information for Video Coding. PhD Thesis, Linköping University, Sweden, 2001. Dissertation No. 683, ISBN 91-7219-997-0.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2003

Authors and Affiliations

  • Astrid Lundmark
    • 1
  • Leif Haglund
    • 1
  1. 1.Saab Bofors Dynamics ABSweden

Personalised recommendations