From KitwarePublic
Jump to navigationJump to search
  Description: Given multiple, registered images and foreground and
  background masks, computes multivariate PDFs for inside and outside
  classes, and then performs competitive region growing.


  ./SegmentConnectedComponentsUsingParzenPDFs  [--returnparameterfile
                                       <std::string>] [--xml] [--echo]
                                       [--saveClassPDFBase <std::string>]
                                       [--loadClassPDFBase <std::string>]
                                       <double>] [--objectPDFWeight
                                       [--holeFillIterations <int>]
                                       [--erodeRadius <int>] [--voidId
                                       <int>] [--objectId
                                       <std::vector<int>>] [--inputVolume4
                                       <std::string>] [--inputVolume3
                                       <std::string>] [--inputVolume2
                                       <std::string>] [--] [--version]
                                       [-h] <std::string> <std::string>


  --returnparameterfile <std::string>
    Filename in which to write simple return parameters (int, float,
    int-vector, etc.) as opposed to bulk return parameters (image,
    geometry, transform, measurement, table).
  --processinformationaddress <std::string>
    Address of a structure to store process information (progress, abort,
    etc.). (default: 0)
    Produce xml description of command line arguments (default: 0)
    Echo the command line arguments (default: 0)
  --saveClassPDFBase <std::string>
    Save images that represent probability density functions.
  --loadClassPDFBase <std::string>
    Load images that represent probability density functions.
  --saveClassProbabilityVolumeBase <std::string>
    Save images where each represents the probability of being a
    particular object at each voxel. Image files create =
    Force classification using simple maximum likelihood. (default: 0)
    Perform classification on all non-void voxels. (default: 0)
    Perform classification on voxels within the object mask. (default: 0)
    Generate draft results. (default: 0)
  --histogramSmoothingStdDev <double>
    Standard deviation of blur applied to convert the histogram to a
    probability density function estimate (default: 5)
  --probImageSmoothingStdDev <double>
    Standard deviation of blur applied to probability images prior to
    computing maximum likelihood of each class at each pixel (default: 1)
  --objectPDFWeight <std::vector<double>>
    Relative weight (multiplier) of each PDF. (default: 1)
    Performs dilation then erosion (versus opposite order) to help fill-in
    sparse models. (default: 0)
  --holeFillIterations <int>
    Number of iterations for hole filling. (default: 1)
  --erodeRadius <int>
    Radius of noise to clip from edges. (default: 1)
  --voidId <int>
    Value that represents 'nothing' in the label map. (default: 0)
  --objectId <std::vector<int>>
    List of values that represent the objects in the label map. (default:
  --inputVolume4 <std::string>
    Input volume 4.
  --inputVolume3 <std::string>
    Input volume 3.
  --inputVolume2 <std::string>
    Input volume 2.
  --,  --ignore_rest
    Ignores the rest of the labeled arguments following this flag.
    Displays version information and exits.
  -h,  --help
    Displays usage information and exits.
    (required)  Input volume 1.
    (required)  Label map that designates the object of interest and
    (required)  Segmentation results.
  Author(s): Stephen R. Aylward (Kitware)
  Acknowledgements: This work is part of the TubeTK project at Kitware.