Local adaptive filtering in DCT domain

Leonid Yaroslavsky, Leonid Mitliansky,
Dept. of Interdisciplinary Studies,
Faculty of Engineering,
Tel Aviv University, Tel Aviv, Israel

October 1997

Multi component local adaptive filtering in the domain of Dicrete Cosine Transform Domain (RDCT - Recursive DCT) for image restoration and enhancement, its computer implementation and experimental verification is descriabed. The filters work in a moving window and nonlinearly transform signal DCT spectral coefficients to obtain an estimation of the central pixel of the window. The filters are shown to be capable of efficient signal deblurring and noise suppression with preservation of edges and detail boundaries. The filter computational implementation is based on a recursive computational algorithm which allows to reduce the computational complexity of the filtering to O (size of the window). For the timr being, the filters are implemented for the use on MATLAB platform as MEX-files  in C - language. The filter signal restoration and image enhancement capability has been tested  on  1-D ECG signals, single component (black and white) and three component (RGB) medical images.

References:

1.     L. Yaroslavsky, M. Eden, Fundamentals of Digital Optics, Birkhauser, Boston, 1996

2.     L. P. Yaroslavsky, Linear and Rank Adaptive Filters for Picture Processing, In: Digital Image Processing and Computer Graphics, ed. by E. Wenger and L. Dimitrov., Oestwerreichische Computer Gesellschaft, R. Oldenburg, Wien, Muenchen, 1991, p. 327 - 374.

3.    L. P. Yaroslavsky, Local Adaptive Filrters for Image Restoration and Enhancement, ICAOS’96, 12th International Conference on Analysis and Optimization of Systems, Images, Waveletes and PDEs., Paris, June 26-28, 1996, INRIA, Ceremade, In: Lecture Notes in Control and Information Sciences 219, M-O Berger, R. Deriche, I. Herlin, J. Jaffe and J.-M. Morel, Eds., Springer-Verlag, London, 1996

4.    L.P. Yaroslavsky, Local adaptive image restoration and enhancement with the use of DFT and DCT in a running window (invited paper), in: Proceedings, Wavelet Applications inSignal and Image Processing IV, 6-9 August 1996, Denver, Colorado, SPIE Proc. Series, v. 2825, pp. 1-13.