Computational Visualization Center University of Texas at Austin   
   
COMPUTATIONAL VISUALIZATION CENTER

  PROJECTS  
Infrastructure | Applications | Remote Visualization
ShastraVisualEyesDiDiAngstromImaging-to-ModellingX-Tierra
 Introduction | Visualization Tool | Filtering | Gradient Vector Diffusion | Segmentation | Skeletonization | References

3. Anisotropic Filtering

Our approach to three dimensional nonlinear noise reduction filters, such as bilateral pre-filtering coupled with an evolution driven anisotropic geometric diffusion PDE (partial differential equation), have shown significant results in enhancing the visualization of macromolecular tomographic imaging. The PDE model is :

The efficacy of our method is based on a careful selection of the anisotropic diffusion tensor based on estimates of the normal and two principal curvatures and curvature directions of a feature isosurface in three dimensions.


Fig. 2 A sub-volume of dataset comprising extracellular links between a stereocilium and the kinociliar bulb in a bullfrog macular sensory epithelia hair bundle.

Project Angstrom     CCV Projects



   Computational Visualization Center University of Texas at Austin