Structure Enhancing Bilateral Filtering of Images

We present in this paper a modified bilateral filter that is able to enhance flow-like structure. The shape (contour) of the domain Gaussian filter can change adaptively according to the local orientation obtained from structure tensor. The resulting combined elliptical shaped Gaussian domain filter and range filter is demonstrated to be able to smooth out random noise while preserving and enhancing local structure.

[1]  Joachim Weickert,et al.  Coherence-Enhancing Diffusion Filtering , 1999, International Journal of Computer Vision.

[2]  Jürgen Döllner,et al.  Image and Video Abstraction by Anisotropic Kuwahara Filtering , 2009, Comput. Graph. Forum.

[3]  Ming Zhang,et al.  Bilateral filter in image processing , 2009 .

[4]  John Immerkær,et al.  Fast Noise Variance Estimation , 1996, Comput. Vis. Image Underst..

[5]  Roberto Manduchi,et al.  Bilateral filtering for gray and color images , 1998, Sixth International Conference on Computer Vision (IEEE Cat. No.98CH36271).

[6]  Frédo Durand,et al.  Bilateral Filtering: Theory and Applications , 2009, Found. Trends Comput. Graph. Vis..