Title :
Complexity-constrained best-basis wavelet packet algorithm for image compression
Author :
Marpe, D. ; Cycon, H.L. ; Li, W.
Author_Institution :
Heinrich-Hertz-Inst. fur Nachrichtentech. Berlin GmbH, Germany
fDate :
12/1/1998 12:00:00 AM
Abstract :
The concept of adapted waveform analysis using a best-basis selection out of a predefined library of wavelet packet (WP) bases allows an efficient image representation for the purpose of compression. Image coding methods based on the best-basis WP representation have shown significant coding gains for some image classes compared with methods using a fixed dyadic structured wavelet basis, at the expense however, of considerably higher computational complexity. A modification of the best-basis method, the so-called complexity constrained best-basis algorithm (CCBB), is proposed which parameterises the complexity gap between the fast (standard) wavelet transform and the best wavelet packet basis of a maximal WP library. This new approach allows a `suboptimal´ best basis to be found with respect to a given budget of computational complexity or, in other words, it offers an instrument to control the trade-off between compression speed and, coding efficiency. Experimental results are presented for image coding applications showing a highly nonlinear relationship between the rate-distortion performance and the computational complexity in such a way that a relatively small increase in complexity with respect to the standard wavelet basis results in a relatively high rate distortion gain
Keywords :
computational complexity; data compression; discrete wavelet transforms; image coding; image representation; rate distortion theory; transform coding; adapted waveform analysis; best-basis wavelet packet algorithm; coding efficiency; complexity constrained best-basis algorithm; compression speed; computational complexity; image coding; image compression; image representation; rate-distortion performance; suboptimal best basis; wavelet packet basis; wavelet transform;
Journal_Title :
Vision, Image and Signal Processing, IEE Proceedings -
DOI :
10.1049/ip-vis:19982457