net.imglib2.algorithm.metrics.segmentation

Class LazyMultiMetrics

java.lang.Object

net.imglib2.algorithm.metrics.segmentation.LazyMultiMetrics

public class LazyMultiMetrics
extends Object

The LazyMultiMetrics computes a running MultiMetrics over all images added by calling addTimePoint. The images are expected to be of dimension XYZ, where Z can be of depth 1.

The scores are calculated by accumulating the number of TP, FP, FN and the sum of the IoU for each image. Each contribution is computed by adding images using addTimePoint. The final score can be queried by computeScore().

Each image's contributions are calculated independently. Therefore, the same LazyMultiMetrics object can be called from multiple threads in order to speed up computation. For instance, if the total stack does not fit in memory, lazy loading and multithreading can be used to compute the metrics scores by splitting the XYZ images between threads and adding them one by one. The final scores can be calculated once all threads have finished.

The MultiMetrics scores are calculated at a certain threshold. This threshold is the minimum IoU between a ground-truth and a prediction label at which two labels are considered a potential match. The threshold can only be set at instantiation.

Author:

Joran Deschamps

See Also:

MultiMetrics

Constructor Summary

Constructors

Constructor and Description
LazyMultiMetrics() Constructor with a default threshold of 0.5.
LazyMultiMetrics(double threshold) Constructor that sets the threshold value.

Method Summary

Modifier and Type	Method and Description
protected void	addPoint(MultiMetrics.MetricsSummary newResult) Update the atomic aggregates with the values held by the MultiMetrics.MetricsSummary.
<T,I extends IntegerType<I>,U,J extends IntegerType<J>> void	addTimePoint(ImgLabeling<T,I> groundTruth, ImgLabeling<U,J> prediction) Add a new image pair and compute its contribution to the metrics scores.
<I extends IntegerType<I>,J extends IntegerType<J>> void	addTimePoint(RandomAccessibleInterval<I> groundTruth, RandomAccessibleInterval<J> prediction) Add a new image pair and compute its contribution to the metrics scores.
HashMap<MultiMetrics.Metrics,Double>	computeScore() Compute the total MultiMetrics scores.

Methods inherited from class java.lang.Object

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

Constructor Detail

LazyMultiMetrics

public LazyMultiMetrics()

Constructor with a default threshold of 0.5.

LazyMultiMetrics

public LazyMultiMetrics(double threshold)

Constructor that sets the threshold value.

Parameters:

threshold - Threshold

Method Detail

addTimePoint

public <T,I extends IntegerType<I>,U,J extends IntegerType<J>> void addTimePoint(ImgLabeling<T,I> groundTruth,
                                                                                 ImgLabeling<U,J> prediction)

Add a new image pair and compute its contribution to the metrics scores. The current scores can be computed by calling computeScore(). This method is not compatible with ImgLabeling with intersecting labels.

Type Parameters:

T - Label type associated to the ground-truth

I - Ground-truth pixel type

U - Label type associated to the prediction

J - Prediction pixel type

Parameters:

groundTruth - Ground-truth image

prediction - Predicted image

addTimePoint

public <I extends IntegerType<I>,J extends IntegerType<J>> void addTimePoint(RandomAccessibleInterval<I> groundTruth,
                                                                             RandomAccessibleInterval<J> prediction)

Add a new image pair and compute its contribution to the metrics scores. The current scores can be computed by calling computeScore().

Type Parameters:

I - Ground-truth pixel type

J - Prediction pixel type

Parameters:

groundTruth - Ground-truth image

prediction - Predicted image

computeScore

public HashMap<MultiMetrics.Metrics,Double> computeScore()

Compute the total MultiMetrics scores. If no image was added, or all images were empty, then the metrics scores are TP=FP=FN=0 and NaN for the others.

Returns:

Metrics scores

addPoint

protected void addPoint(MultiMetrics.MetricsSummary newResult)

Update the atomic aggregates with the values held by the MultiMetrics.MetricsSummary.

Parameters:

newResult - Result to add to the aggregates