CONTENT-ADAPTIVE MULTIRESOLUTION ANALYSES

In this paper we present a technique for building adaptive wavelets by means of an extension of the lifting scheme and analyze the stability of the resulting decompositions. Our scheme comprises an adaptive update lifting and a fixed prediction lifting step. The adaptivity consists hereof that the system can choose between two different update filters, and that this choice is triggered by the local gradient of the original signal. If the gradient is large (in some seminorm sense) it chooses one filter, if it is small the other. We derive necessary and sufficient conditions for the invertibility of such an adaptive system for various scenarios. Furthermore, we present some examples to illustrate our theoretical results. We also discuss the effects of quantization in such an adaptive wavelet decomposition and provide conditions for recovering the original decisions at the synthesis and for relating the reconstruction error to the quantization error. Such an analysis is essential for the application of these adaptive decompositions in image compression.

[1]  H. Heijmans,et al.  Adaptive update lifting with a decision rule based on derivative filters , 2002, IEEE Signal Processing Letters.

[2]  Albert Cohen,et al.  Compact representation of images by edge adapted multiscale transforms , 2001, Proceedings 2001 International Conference on Image Processing (Cat. No.01CH37205).

[3]  David L. Donoho,et al.  Orthonormal Ridgelets and Linear Singularities , 2000, SIAM J. Math. Anal..

[4]  Wim Sweldens,et al.  Lifting scheme: a new philosophy in biorthogonal wavelet constructions , 1995, Optics + Photonics.

[5]  Henk J. A. M. Heijmans,et al.  Adaptive Wavelets for Image Compression Using Update Lifting: Quantization and Error Analysis , 2006, Int. J. Wavelets Multiresolution Inf. Process..

[6]  P. L. Combettes,et al.  Convex multiresolution analysis , 1996, Proceedings of Third International Symposium on Time-Frequency and Time-Scale Analysis (TFTS-96).

[7]  M. Kunt,et al.  High compression image coding using an adaptive morphological subband decomposition , 1995, Proc. IEEE.

[8]  A. Enis Çetin,et al.  Adaptive polyphase subband decomposition structures for image compression , 2000, IEEE Trans. Image Process..

[9]  G. Piella,et al.  Adaptive lifting schemes with perfect reconstruction , 2002, IEEE Trans. Signal Process..

[10]  D. Donoho.  Wedgelets: nearly minimax estimation of edges , 1999 .

[11]  Jean-Christophe Pesquet,et al.  M-band nonlinear subband decompositions with perfect reconstruction , 1998, IEEE Trans. Image Process..

[12]  Beatrice Pesquet-Popescu,et al.  Building nonredundant adaptive wavelets by update lifting , 2002 .

[13]  Richard G. Baraniuk,et al.  Nonlinear wavelet transforms for image coding via lifting , 2003, IEEE Trans. Image Process..

[14]  Emmanuel J. Candès,et al.  The curvelet transform for image denoising , 2002, IEEE Trans. Image Process..

[15]  Henk J. A. M. Heijmans,et al.  Nonlinear multiresolution signal decomposition schemes. II. Morphological wavelets , 2000, IEEE Trans. Image Process..

[16]  Stéphane Mallat,et al.  Image compression with geometrical wavelets , 2000, Proceedings 2000 International Conference on Image Processing (Cat. No.00CH37101).

[17]  Ricardo L. de Queiroz,et al.  Nonexpansive pyramid for image coding using a nonlinear filterbank , 1998, IEEE Trans. Image Process..

[18]  Jean-Christophe Pesquet,et al.  A nonlinear subband decomposition with perfect reconstruction , 1996, 1996 IEEE International Conference on Acoustics, Speech, and Signal Processing Conference Proceedings.

[19]  Robert D. Nowak,et al.  Adaptive wavelet transforms via lifting , 1998, Proceedings of the 1998 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP '98 (Cat. No.98CH36181).

[20]  Luis F. Chaparro,et al.  Adaptive Morphological Representation of Signals: Polynomial and Wavelet Methods , 1997, Multidimens. Syst. Signal Process..