From two-dimensional nonlinear diffusion to coupled Haar wavelet shrinkage

This paper studies the connections between discrete two-dimensional schemes for shift-invariant Haar wavelet shrinkage on one hand, and nonlinear diffusion on the other. We show that using a single iteration on a single scale, the two methods can be made equivalent by the choice of the nonlinearity which controls each method: the shrinkage function, or the diffusivity function, respectively. In the two-dimensional setting, this diffusion-wavelet connection shows an important novelty compared to the one-dimensional framework or compared to classical 2-D wavelet shrinkage: The structure of two-dimensional diffusion filters suggests to use a coupled, synchronised shrinkage of the individual wavelet coefficient channels. This coupling enables to design Haar wavelet filters with good rotation invariance at a low computational cost. Furthermore, by transferring the channel coupling of vector- and matrix-valued nonlinear diffusion filters to the Haar wavelet setting, we obtain well-synchronised shrinkage methods for colour and tensor images. Our experiments show that these filters perform significantly better than conventional shrinkage methods that process all wavelets independently.

[1]  N. Kingsbury.  Complex Wavelets for Shift Invariant Analysis and Filtering of Signals , 2001 .

[2]  Richard Kronland-Martinet,et al.  A real-time algorithm for signal analysis with the help of the wavelet transform , 1989 .

[3]  Jacques Froment,et al.  Reconstruction of Wavelet Coefficients Using Total Variation Minimization , 2002, SIAM J. Sci. Comput..

[4]  Ph. Tchamitchian,et al.  Wavelets: Time-Frequency Methods and Phase Space , 1992 .

[5]  R. DeVore,et al.  Nonlinear Approximation and the Space BV(R2) , 1999 .

[6]  Emmanuel J. Candès,et al.  New multiscale transforms, minimum total variation synthesis: applications to edge-preserving image reconstruction , 2002, Signal Process..

[7]  Kannan Ramchandran,et al.  Wavelet denoising by recursive cycle spinning , 2002, Proceedings. International Conference on Image Processing.

[8]  P. Basser,et al.  Diffusion tensor MR imaging of the human brain. , 1996, Radiology.

[9]  Fionn Murtagh,et al.  Gray and color image contrast enhancement by the curvelet transform , 2003, IEEE Trans. Image Process..

[10]  Jianhong Shen.  A Note on Wavelets and Diffusions , 2003 .

[11]  Karol Mikula,et al.  An Adaptive Finite Volume Method In Processing Of Color Images , 2000 .

[12]  Thomas Brox,et al.  Diffusion Filters and Wavelets: What Can They Learn from Each Other? , 2006, Handbook of Mathematical Models in Computer Vision.

[13]  Emmanuel J. Candès,et al.  The curvelet transform for image denoising , 2001, Proceedings 2001 International Conference on Image Processing (Cat. No.01CH37205).

[14]  Akram Aldroubi,et al.  Reconstruction of vector and tensor fields from sampled discrete data , 2003 .

[15]  Philip M. Carrion.  On stability of 1D exact inverse methods , 1986 .

[16]  Jitendra Malik,et al.  Scale-Space and Edge Detection Using Anisotropic Diffusion , 1990, IEEE Trans. Pattern Anal. Mach. Intell..

[17]  I. Johnstone,et al.  Ideal spatial adaptation by wavelet shrinkage , 1994 .

[18]  F. Malgouyres,et al.  Mathematical analysis of a model which combines total variation and wavelet for image restoration 1 , 2002 .

[19]  D. Donoho,et al.  Translation-Invariant DeNoising , 1995 .

[20]  Antonin Chambolle,et al.  Nonlinear wavelet image processing: variational problems, compression, and noise removal through wavelet shrinkage , 1998, IEEE Trans. Image Process..

[21]  Truong Q. Nguyen,et al.  Wavelets and filter banks , 1996 .

[22]  François Malgouyres,et al.  Combining total variation and wavelet packet approaches for image deblurring , 2001, Proceedings IEEE Workshop on Variational and Level Set Methods in Computer Vision.

[23]  Joachim Weickert,et al.  Coherence-enhancing diffusion of colour images , 1999, Image Vis. Comput..

[24]  Antonin Chambolle,et al.  Interpreting translation-invariant wavelet shrinkage as a new image smoothing scale space , 2001, IEEE Trans. Image Process..

[25]  A. Bruce,et al.  WAVESHRINK WITH FIRM SHRINKAGE , 1997 .

[26]  Hong-Ye Gao,et al.  Wavelet Shrinkage Denoising Using the Non-Negative Garrote , 1998 .

[27]  Nikos Paragios,et al.  Handbook of Mathematical Models in Computer Vision , 2005 .

[28]  Hans Knutsson,et al.  Signal processing for computer vision , 1994 .

[29]  P. Morettin.  Wavelets in Statistics , 1997 .

[30]  Guido Gerig,et al.  Vector-Valued Diffusion , 1994, Geometry-Driven Diffusion in Computer Vision.

[31]  Achilleas S. Frangakis,et al.  Wavelet transform filtering and nonlinear anisotropic diffusion assessed for signal reconstruction performance on multidimensional biomedical data , 2001, IEEE Transactions on Biomedical Engineering.

[32]  I. Daubechies,et al.  Harmonic analysis of the space BV. , 2003 .

[33]  Yves Meyer,et al.  Oscillating Patterns in Image Processing and Nonlinear Evolution Equations: The Fifteenth Dean Jacqueline B. Lewis Memorial Lectures , 2001 .

[34]  Gilbert Strang,et al.  On Wavelet Fundamental Solutions to the Heat Equation—Heatlets , 2000 .

[35]  Rachid Deriche,et al.  Diffusion tensor regularization with constraints preservation , 2001, Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001.

[36]  Joachim Weickert,et al.  On Iterations and Scales of Nonlinear Filters , 2003 .

[37]  Ronald R. Coifman,et al.  New Methods of Controlled Total Variation Reduction for Digital Functions , 2001, SIAM J. Numer. Anal..

[38]  Joachim Weickert,et al.  Diffusion-Inspired Shrinkage Functions and Stability Results for Wavelet Denoising , 2005, International Journal of Computer Vision.

[39]  Ronald R. Coifman,et al.  Combining the Calculus of Variations and Wavelets for Image Enhancement , 2000 .

[40]  Joachim Weickert,et al.  Rotationally Invariant Wavelet Shrinkage , 2003, DAGM-Symposium.

[41]  R. Stollberger,et al.  Nonlinear anisotropic diffusion filtering for multiscale edge enhancement , 2002 .

[42]  D. L. Donoho,et al.  Ideal spacial adaptation via wavelet shrinkage , 1994 .

[43]  Joachim Weickert,et al.  Correspondences between Wavelet Shrinkage and Nonlinear Diffusion , 2003, Scale-Space.

[44]  S. Mallat.  A wavelet tour of signal processing , 1998 .

[45]  David L. Donoho,et al.  De-noising by soft-thresholding , 1995, IEEE Trans. Inf. Theory.

[46]  David L. Donoho,et al.  Translation- and direction-invariant denoising of 2D and 3D images: experience and algorithms , 1996, Optics & Photonics.

[47]  Hamid Krim,et al.  Towards Bridging Scale-Space and Multiscale Frame Analyses , 2001 .

[48]  Tony F. Chan,et al.  Total variation improved wavelet thresholding in image compression , 2000, Proceedings 2000 International Conference on Image Processing (Cat. No.00CH37101).

[49]  Thomas Brox,et al.  On the Equivalence of Soft Wavelet Shrinkage, Total Variation Diffusion, Total Variation Regularization, and SIDEs , 2004, SIAM J. Numer. Anal..

[50]  Joachim Weickert,et al.  From Tensor-Driven Diffusion to Anisotropic Wavelet Shrinkage , 2006, ECCV.

[51]  Stephen L. Keeling,et al.  Nonlinear anisotropic diffusion filters for wide range edge sharpening , 2000, Medical Imaging: Image Processing.