Label Map Smoothing


This filter smoothes a binary label map. With a label map as input, this filter runs an anti-alising algorithm followed by a Gaussian smoothing algorithm. The output is a smoothed label map.

Panels and their use

Label Selection Parameters: Parameters for selecting the label to smooth

  • Label to smooth (labelToSmooth): The label to smooth. All others will be ignored. If no label is selected by the user, the maximum label in the image is chosen by default.

AntiAliasing Parameters: Parameters for the AntiAliasing algorithm

  • Number of Iterations (numberOfIterations): The number of iterations of the level set AntiAliasing algorithm

  • Maximum RMS Error (maxRMSError): The maximum RMS error.

Gaussian Smoothing Parameters: Parameters for Gaussian Smoothing

  • Sigma (gaussianSigma): The standard deviation of the Gaussian kernel

IO: Input/output parameters

  • Input Volume (inputVolume): Input label map to smooth

  • Output Volume (outputVolume): Smoothed label map


Dirk Padfield (GE), Josh Cates (Utah), Ross Whitaker (Utah)


This work is part of the National Alliance for Medical Image Computing (NAMIC), funded by the National Institutes of Health through the NIH Roadmap for Medical Research, Grant U54 EB005149. This filter is based on work developed at the University of Utah, and implemented at GE Research.