A NOVEL DIFFUSION FILTER FOR IMAGE RESTORATION AND ENHANCEMENT

We are proposing a new filter for image restoration and enhancement tasks that relies on image orientation analysis techniques for steering and directing smoothing or enhancement processes. The filter is developed under the Partial Differential Equations (PDE) framework using asymmetric orientation estimation operators and it has better junction preservation and noise removal properties than existing classical PDE based methods. We employ an experimental setup involving several computer generated images and a statistical interpretation of the results for proving the efficiency of the method in processing images composed of directional textures and degraded by additive Gaussian noise. The algorithm is compared also with state-of-the-art non-PDE based methods in filtering images containing oriented patterns.

[1]  Joachim Weickert,et al.  Coherence-Enhancing Shock Filters , 2003, DAGM-Symposium.

[2]  P. Lions,et al.  Image selective smoothing and edge detection by nonlinear diffusion. II , 1992 .

[3]  David Tschumperlé,et al.  Fast Anisotropic Smoothing of Multi-Valued Images using Curvature-Preserving PDE's , 2006, International Journal of Computer Vision.

[4]  Jaakko Astola,et al.  From Local Kernel to Nonlocal Multiple-Model Image Denoising , 2009, International Journal of Computer Vision.

[5]  G. G. Stokes.  "J." , 1890, The New Yale Book of Quotations.

[6]  Gjlles Aubert,et al.  Mathematical problems in image processing , 2001 .

[7]  DericheRachid,et al.  Vector-Valued Image Regularization with PDEs , 2005 .

[8]  R. Terebes,et al.  Asymmetric directional diffusion based image filtering and enhancement , 2008, 2008 IEEE International Conference on Automation, Quality and Testing, Robotics.

[9]  Pierre Kornprobst,et al.  Mathematical problems in image processing - partial differential equations and the calculus of variations , 2010, Applied mathematical sciences.

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

[11]  Martin J. Wainwright,et al.  Image denoising using scale mixtures of Gaussians in the wavelet domain , 2003, IEEE Trans. Image Process..

[12]  Jeremy R. Cooperstock,et al.  An Asymmetrical Diffusion Framework for Junction Analysis , 2006, BMVC.

[13]  R. Iman,et al.  Rank Transformations as a Bridge between Parametric and Nonparametric Statistics , 1981 .

[14]  Jerry D. Gibson,et al.  Handbook of Image and Video Processing , 2000 .

[15]  René A. Carmona,et al.  Adaptive smoothing respecting feature directions , 1998, IEEE Trans. Image Process..

[16]  Pierre Baylou,et al.  Flow Coherence Diffusion. Linear and Nonlinear Case , 2005, ACIVS.

[17]  Pierre Baylou,et al.  Estimating local multiple orientations , 2007, Signal Process..

[18]  Yuan Baozong,et al.  A new PDE based approach for image restoration and enhancement using robust diffusion directions and directional derivatives based diffusivities , 2004, Proceedings 7th International Conference on Signal Processing, 2004. Proceedings. ICSP '04. 2004..