AN IMPROVED PREDICTION OF DCT-BASED IMAGE FILTERS EFFICIENCY USING REGRESSION ANALYSIS

Authors

DOI:

https://doi.org/10.20535/2411-2976.12014.30-41

Abstract

Efficiency of DCT-based filters for a wide-class of images is investigated. The study is carried out for additive white Gaussian noise (AWGN) case with several intensity levels. Local DCT-based filter is used as basic denoising technique. Nonlocal BM3D filter known as the state-of-the-art technique for AWGN removal is also exploited. A precise prediction method of denoising efficiency for several quality metrics is proposed. It is shown that statistics of DCT coefficients provides useful information. Regression models for analyzed filters and metrics are presented. The obtained dependence approximations of quality metrics on DCT statistics have high goodness of fit. One-parameter and multi-parameter fitting cases are considered. The most valuable DCT statistics are found.

Author Biographies

Oleksii S. Rubel, National Aerospace University "KhAI"

Postgraduate

Volodymyr V. Lukin, National Aerospace University "KhAI"

Dr.Sc, professor

References

Pratt, W. K. Digital Image Processing. Fourth Edition / W. K. Pratt. – N. Y.: Wiley-Interscience. – USA. – 2007. – 1429 p.

Donoho, D. Nonlinear wavelet methods for recovery of signals, densities, and spectra from indirect and noisy data / D. Donoho // Proceedings Symposium Appl. Math.. – P. 173–205. – 1994.

Lukin, V. Image filtering based on discrete cosine transform / V. Lukin, R. Oktem, N. Ponomarenko, K. Egiaza-rian // Telecommunications and Radio Engineering. – Vol. 66, No. 18. – P. 1685-1701. – 2007.

Dabov, K. Image denoising by sparse 3D transform-domain collaborative filtering / K. Dabov, A. Foi, V. Kat-kovnik, K Egiazarian // IEEE Transactions on Image Processing.- Vol. 16, No. 8. – August 2007. – P. 2080-2095.

Image Filtering: Potential efficiency and current problems / V. Lukin, S. Abramov, N. Ponomarenko, K. Egiaza-rian, J. Astola // Proceedings of ICASSP. – May 2011. – P. 1433-1436.

Zhu, X. Automatic parameter selection for denoising algorithms using a no-reference measure of image content / X. Zhu, P. Milanfar // IEEE Transactions on image processing. – Vol. 19, No. 12. – December 2010. – P. 3116-3132.

Chatterjee, P. Practical Bounds on Image Denoising: From estimation to information / P. Chatterjee, P. Milanfar // IEEE Transactions on Image Processing. – May 2011. – vol. 20, no. 5. – P. 1221-1233.

Fevralev, D. Efficiency analysis of DCT-based filters for color image database / D. Fevralev, V. Lukin, N. Ponomarenko, S. Abramov, K. Egiazarian, J. Astola // Proceedings of SPIE Conference Image Processing: Algorithms and Systems VII, San Francisco, USA. – 2011. – Vol. 7870. – 12 p.

Lam, E. A Mathematical analysis of the DCT coefficient distributions for images / E. Y. Lam, J. W. Goodman // IEEE Transactions on Image Processing. – 2000. – vol. 9, no. 10. – P. 1661-1666.

Zoran, D. Scale invariance and noise in natural images / D. Zoran, Y. Weiss // IEEE 12th International Conference on Computer Vision. – September 2009. – P. 2209-2216.

Pogrebnyak, O. Wiener DCT Based Image Filtering / O. Pogrebnyak, V. Lukin // Journal of Electronic Imaging. – 2012. – No.4. – 14 p.

Prediction of Filtering efficiency for DCT-based Image Denoising / S. Abramov, S. Krivenko, A. Roenko, V. Lukin, I. Djurovic, M. Chobanu // 2-nd Mediterrian Conference on Embedded Computing MECO. – June 2013. – P. 97-100.

An R-squared measure of goodness of fit for some common nonlinear regression models / C. Cameron, A. Windmeijer, A. G. Frank, H. Gramajo, D. E. Cane, C. Khosla // Journal of Econometrics. – 1997. – vol. 77, no. 2. – 16 p.

Rubel, A.S. Prediction of filtering efficiency for discrete cosine transform based removal of additive noise on images / A.S. Rubel, V.V. Lukin // Radio-electronic and computer systems. – 2013. - №4 (63). – P. 35-45. [in Russian].

New full-reference quality metrics based on HVS / K. Egiazarian, J. Astola, N. Ponomarenko, V. Lukin, F. Battisti, M. Carli // Proceedings of the Second International Workshop on Video Processing and Quality Metrics, Scottsdale, USA. – 2006. – 4 p.

Color Image Database TID2013: Peculiarities and Preliminary Results / N. Ponomarenko, O. Ieremeiev, V. Lukin, K. Egiazarian, L. Jin, J. Astola, B. Vozel, K. Chehdi, M. Car-li, F. Battisti, C.-C. Jay Kuo // 4th European Workshop on Visual Information Processing EUVIP2013, Paris, France. – 2013. – 6 p.

Downloads

Issue

Section

Статті