Modified Embedded Zerotree Scheme for Efficient Coding of Discrete Wavelet Coded Frames

Abstract:

Zerotree coding is an intelligent approach for quantization of Discrete Wavelet coefficients and thus achieving image compression. Embedded Zerotree Coding by Shapiro is very effective in this regard, having the property that the bits in the bit stream are generated in order of importance, yielding a fully embedded code. In this paper we have proposed a Modified Embedded Zerotree (MEZT) scheme which is computationally more efficient, making it more suitable for real time video compression applications. MEZT is a bit plane wise single pass algorithm, unlike Shapiro?s two pass technique. This leads to reduction in processing time. Depending on the priority of bit budget or compression ratio, the number of bit planes to be encoded can be decided. More over, since most of the necessary tasks are bit operations, it is more suitable for VLSI implementation.

Bibtex:

@INPROCEEDINGS{mba_ezt_iccv_00,
AUTHOR = "N. J. Mitra and P. K. Biswas and T. Acharya",
TITLE = "Modified Embedded Zerotree Scheme for Efficient Coding of Discrete Wavelet Coded Frames",
BOOKTITLE = "Indian Conference on Computer Vision, Graphics and Image Processing",
YEAR = "2000"
}