Package org.anchoranalysis.image.inference.bean.segment.instance

Class SegmentStackIntoObjectsScaleDecode<T,S extends ImageInferenceModel<T>>

Object

AnchorBean<SegmentStackIntoObjectsPooled<S>>

InitializableBean<SegmentStackIntoObjectsPooled<S>,ImageInitialization>

ImageBean<SegmentStackIntoObjectsPooled<S>>

SegmentationBean<SegmentStackIntoObjectsPooled<S>>

SegmentStackIntoObjectsPooled<S>

SegmentStackIntoObjectsScaleDecode<T,S>

Type Parameters:

T - tensor-type inputted and outputted to model

S - model-type

Direct Known Subclasses:

SegmentObjectsFromONNXModel, SegmentObjectsFromTensorFlowModel

public abstract class SegmentStackIntoObjectsScaleDecode<T,S extends ImageInferenceModel<T>>
extends SegmentStackIntoObjectsPooled<S>

A SegmentStackIntoObjectsScaleDecode that scales the input image, before performing inference, and then decodes the output.

Author:

Owen Feehan

Constructor Summary

Constructors

Constructor	Description
SegmentStackIntoObjectsScaleDecode()

Method Summary

Modifier and Type	Method	Description
protected abstract T	deriveInput(Stack stack, Optional<double[]> subtractMeans)	Derives the input tensor from an image.
String	getClassLabelsPath()	Relative-path to the class-labels file, a text file where each line specifies a class label in order, relative to the models/ directory in the Anchor distribution.
DecodeInstanceSegmentation<T>	getDecode()	Decodes inference output into segmented objects.
StackDisplayer	getDisplayer()	How to convert an image to be displayed to the user.
Interpolator	getInterpolator()	The interpolator to use for scaling images.
ScaleCalculator	getScaleInput()	Any scaling to be applied to the input-image before being input to the model for inference.
DoubleList	getSubtractMean()	A constant intensity for each respective channel to be subtracted before performing inference.
protected abstract Optional<String>	inputName()	The name of the tensor in the model which the input-image is mapped to.
SegmentedObjects	segment(Stack stack, ConcurrentModelPool<S> modelPool, ExecutionTimeRecorder executionTimeRecorder)	Segments a stack to produce an object-collection.
void	setClassLabelsPath(String classLabelsPath)	Relative-path to the class-labels file, a text file where each line specifies a class label in order, relative to the models/ directory in the Anchor distribution.
void	setDecode(DecodeInstanceSegmentation<T> decode)	Decodes inference output into segmented objects.
void	setDisplayer(StackDisplayer displayer)	How to convert an image to be displayed to the user.
void	setInterpolator(Interpolator interpolator)	The interpolator to use for scaling images.
void	setScaleInput(ScaleCalculator scaleInput)	Any scaling to be applied to the input-image before being input to the model for inference.
void	setSubtractMean(DoubleList subtractMean)	A constant intensity for each respective channel to be subtracted before performing inference.

Methods inherited from class org.anchoranalysis.image.inference.bean.segment.instance.SegmentStackIntoObjectsPooled

createModelPool, resolve, segment

Methods inherited from class org.anchoranalysis.bean.initializable.InitializableBean

getInitialization, getLogger, getPropertyInitializer, initialize, initializeRecursive, initRecursiveWithInitializer, isInitialized, onInitialization

Methods inherited from class org.anchoranalysis.bean.AnchorBean

checkMisconfigured, describeBean, describeChildren, duplicateBean, fields, findFieldsOfClass, getBeanName, getLocalPath, localise, toString

Methods inherited from class java.lang.Object

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

Constructor Details

SegmentStackIntoObjectsScaleDecode

public SegmentStackIntoObjectsScaleDecode()

Method Details

segment

public SegmentedObjects segment(Stack stack,
 ConcurrentModelPool<S> modelPool,
 ExecutionTimeRecorder executionTimeRecorder)
                         throws SegmentationFailedException

Description copied from class: SegmentStackIntoObjectsPooled

Segments a stack to produce an object-collection.

Any created objects will always exist inside the stack's Extent.

Specified by:

segment in class SegmentStackIntoObjectsPooled<S extends ImageInferenceModel<T>>

Parameters:

stack - the stack to segment.

modelPool - the pool of model instances which can each be used for inference (in parallel).

executionTimeRecorder - measures execution-times of particular operations.

Returns:

a collection of objects with corresponding confidence scores.

Throws:

SegmentationFailedException - if anything goes wrong during the segmentation.

deriveInput

protected abstract T deriveInput(Stack stack,
 Optional<double[]> subtractMeans)
                          throws OperationFailedException

Derives the input tensor from an image.

Parameters:

stack - the image which is mapped into an input tensor.

subtractMeans - respective intensity values that are subtracted from the voxels before being added to the tensor (respectively for each channel).

Returns:

the tensor, representing the input image.

Throws:

OperationFailedException - if an input tensor cannot be created.

inputName

protected abstract Optional<String> inputName()

The name of the tensor in the model which the input-image is mapped to.

Returns:

the name.

getScaleInput

public ScaleCalculator getScaleInput()

Any scaling to be applied to the input-image before being input to the model for inference.

setScaleInput

public void setScaleInput(ScaleCalculator scaleInput)

Any scaling to be applied to the input-image before being input to the model for inference.

getDecode

public DecodeInstanceSegmentation<T> getDecode()

Decodes inference output into segmented objects.

setDecode

public void setDecode(DecodeInstanceSegmentation<T> decode)

Decodes inference output into segmented objects.

getSubtractMean

public DoubleList getSubtractMean()

A constant intensity for each respective channel to be subtracted before performing inference.

If set, this should create an list, with as many elements as channels inputted to the inference model.

setSubtractMean

public void setSubtractMean(DoubleList subtractMean)

A constant intensity for each respective channel to be subtracted before performing inference.

If set, this should create an list, with as many elements as channels inputted to the inference model.

getClassLabelsPath

public String getClassLabelsPath()

Relative-path to the class-labels file, a text file where each line specifies a class label in order, relative to the models/ directory in the Anchor distribution.

If empty, then no such file is specified.

setClassLabelsPath

public void setClassLabelsPath(String classLabelsPath)

Relative-path to the class-labels file, a text file where each line specifies a class label in order, relative to the models/ directory in the Anchor distribution.

If empty, then no such file is specified.

getInterpolator

public Interpolator getInterpolator()

The interpolator to use for scaling images.

setInterpolator

public void setInterpolator(Interpolator interpolator)

The interpolator to use for scaling images.

getDisplayer

public StackDisplayer getDisplayer()

How to convert an image to be displayed to the user.

setDisplayer

public void setDisplayer(StackDisplayer displayer)

How to convert an image to be displayed to the user.