3D-examples
Edge Preserving Smoothing
Nonlinear Gaussian Filter Chains
Prof. Dr. V. Aurich
Institut für Informatik
Heinrich-Heine-Universität
Düsseldorf
Abstract 2D-examples 3D-examples Publications Programs Home Page

Performance Limits
In order to detect the performance limits of the nonlinear Gaussian filter chain synthetical volume data are used. The undisturbed data consist of thin ellipsoidal shells. By adding white Gaussian noise several data set with various signal-to-noise ratios (SNR) have been generated. The images show sections through the centers of the volumes.
undisturbed data SNR=4 SNR=1 SNR=0.5
filter results ->