trainableSegmentation.metrics

Class AdjustedRandError

java.lang.Object

trainableSegmentation.metrics.Metrics

trainableSegmentation.metrics.AdjustedRandError

public class AdjustedRandError
extends Metrics

This class implements the adjusted Rand error, defined as the 1 - adjusted Rand index. We follow the Rand index definition described by Lawrence Hubert and Phipps Arabie \cite{Hubert85}. BibTeX:

 @article{Hubert85,
   author    = {Lawrence Hubert and Phipps Arabie},
   title     = {Comparing partitions},
   journal   = {Journal of Classification},
   year      = {1985},
   volume    = {2},
   issue           = {1},
   pages     = {193-218},
   doi       = {10.1007/BF01908075)
 }
 

Field Summary

Fields inherited from class trainableSegmentation.metrics.Metrics

originalLabels, proposedLabels, verbose

Constructor Summary

Constructors

Constructor and Description
AdjustedRandError(ij.ImagePlus originalLabels, ij.ImagePlus proposedLabels) Initialize adjusted Rand error metric.

Method Summary

Modifier and Type	Method and Description
static double	adjustedRandError(ij.process.ImageProcessor label, ij.process.ImageProcessor proposal, double binaryThreshold) Calculate the adjusted Rand error between some 2D original labels and the corresponding proposed labels.
static double	adjustedRandIndex(ij.process.ShortProcessor cluster1, ij.process.ShortProcessor cluster2) Calculate the adjusted Rand index between to clusters, as described by Lawrence Hubert and Phipps Arabie \cite{Rand71}.
Callable<Double>	getAdjustedRandErrorConcurrent(ij.process.ImageProcessor image1, ij.process.ImageProcessor image2, double binaryThreshold) Get adjusted Rand error between two images in a concurrent way (to be submitted to an Executor Service).
double	getMetricValue(double binaryThreshold) Calculate the Rand error in 2D between some original labels and the corresponding proposed labels.

Methods inherited from class trainableSegmentation.metrics.Metrics

getMinimumMetricValue, setVerboseMode

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

AdjustedRandError

public AdjustedRandError(ij.ImagePlus originalLabels,
                         ij.ImagePlus proposedLabels)

Initialize adjusted Rand error metric.

Parameters:

originalLabels - original labels (single 2D image or stack)

proposedLabels - proposed new labels (single 2D image or stack of the same as as the original labels)

Method Detail

getMetricValue

public double getMetricValue(double binaryThreshold)

Calculate the Rand error in 2D between some original labels and the corresponding proposed labels. Both image are binarized. The adjusted Rand error is defined as the 1 - adjusted Rand index, as described by William M. Rand \cite{Rand71}. BibTeX:

 @article{Rand71,
   author    = {William M. Rand},
   title     = {Objective criteria for the evaluation of clustering methods},
   journal   = {Journal of the American Statistical Association},
   year      = {1971},
   volume    = {66},
   number    = {336},
   pages     = {846--850},
   doi       = {10.2307/2284239)
 }
 

Specified by:

getMetricValue in class Metrics

Parameters:

binaryThreshold - threshold value to binarize proposal (larger than 0 and smaller than 1)

Returns:

adjusted Rand error

adjustedRandError

public static double adjustedRandError(ij.process.ImageProcessor label,
                                       ij.process.ImageProcessor proposal,
                                       double binaryThreshold)

Calculate the adjusted Rand error between some 2D original labels and the corresponding proposed labels. Both image are binarized. The adjusted Rand error is defined as the 1 - adjusted Rand index, as described by Lawrence Hubert and Phipps Arabie \cite{Hubert85}. BibTeX:

 @article{Hubert85,
   author    = {Lawrence Hubert and Phipps Arabie},
   title     = {Comparing partitions},
   journal   = {Journal of Classification},
   year      = {1985},
   volume    = {2},
   issue           = {1},
   pages     = {193-218},
   doi       = {10.1007/BF01908075)
 }
 

Parameters:

label - 2D image with the original labels

proposal - 2D image with the proposed labels

binaryThreshold - threshold value to binarize the input images

Returns:

adjusted Rand error

getAdjustedRandErrorConcurrent

public Callable<Double> getAdjustedRandErrorConcurrent(ij.process.ImageProcessor image1,
                                                       ij.process.ImageProcessor image2,
                                                       double binaryThreshold)

Get adjusted Rand error between two images in a concurrent way (to be submitted to an Executor Service). Both images are binarized. The adjusted Rand error is defined as the 1 - adjusted Rand index, as described by Lawrence Hubert and Phipps Arabie \cite{Hubert85}. BibTeX:

 @article{Hubert85,
   author    = {Lawrence Hubert and Phipps Arabie},
   title     = {Comparing partitions},
   journal   = {Journal of Classification},
   year      = {1985},
   volume    = {2},
   issue           = {1},
   pages     = {193-218},
   doi       = {10.1007/BF01908075)
 }
 

Parameters:

image1 - first image

image2 - second image

binaryThreshold - threshold to apply to both images

Returns:

adjusted Rand error

adjustedRandIndex

public static double adjustedRandIndex(ij.process.ShortProcessor cluster1,
                                       ij.process.ShortProcessor cluster2)

Calculate the adjusted Rand index between to clusters, as described by Lawrence Hubert and Phipps Arabie \cite{Rand71}. BibTeX:

 @article{Hubert85,
   author    = {Lawrence Hubert and Phipps Arabie},
   title     = {Comparing partitions},
   journal   = {Journal of Classification},
   year      = {1985},
   volume    = {2},
   issue           = {1},
   pages     = {193-218},
   doi       = {10.1007/BF01908075)
 }
 

Parameters:

cluster1 - 2D segmented image (objects are labeled with different numbers)

cluster2 - 2D segmented image (objects are labeled with different numbers)

Returns:

adjusted Rand index