Woolz Image Processing
Version 1.8.3
|
WlzContourGeomFilter [-h] [-o<output file>] [-l#] [-m#] [-n#] [-p#] [-s#] [-P#] [-S#] [<input file>]
-N | Allow non manifild vertices to be filtered. |
-l | Filter parameter \(\lambda\). |
-m | Filter parameter \(\mu\). |
-n | The number of itterations. If the number of itterations is not given, then the pass and stop band ripple determine the number of itterations. |
-p | The band pass frequency parameter (kPB) |
-s | The band stop frequency parameter (kSB). |
-P | The maximum pass band ripple (dPB) |
-S | The maximum stop band ripple (dSB) |
-o | Output object file name. |
-h | Help - print help message |
For simple use to smooth a geometric model from a 2D or 3D contour don't use lambda and mu. Instead use: Values of kPB in the range [0.0100-0.1000], values of kSB in the range [1.000-1.9999]. The default ripple values are probably acceptable for most simple filters. Simple example, which filters a contour smoothing out voxel edges: prompt% WlzContourGeomFilter -o out.wlz -p 0.1 -s 1.5 -n 100 in.wlz This example passes the geometric model from the contour object in in.wlz through a low pass filter with a transition band starting at 0.1 and ending with the stop band at 1.5. A maximum of 100 itterations are performed. If the output from the filter was not sufficiently smooth then the ripple values could be decreased. If you want to design a more complex filter or one that is not a low pass (smoothing) filter then see either: Gabriel Taubin Curve and Surface Smoothing without Shrinkage International Conference on Computer Vision (ICCV'95) Conference Procedings 827-857 1995 or Gabriel Taubin A Signal Processing Approach to Fair Surface Design Computer Graphics, 29, 351-358 1995. Use the -h flag to see the default parameter values.
WlzContourGeomFilter -o smooth.wlz in.wlzSmooths the input contour model in.wlz and writes the smoothed result to smooth.wlz.