A Minos Training Set from which a classifier can be generated. More...

Inherits INativeHandle.

Classes
class	ImageInfo
	Image that has been added to a training set. More...

class	ImageInfoCollection
	Class that maintains the collection of Training Images inside a Minos Training Set. More...

class	ImageInstanceInfoCollection
	Instance information collection that is attached to a training set image. More...

class	InstanceInfo
	InstanceInformation that has been added to a training set. More...

class	InstanceInfoCollection
	Class that maintains the collection of Training Images inside a Minos Training Set. More...

class	ModelInfo
	InstanceInformation that has been added to a training set. More...

class	ModelInfoCollection
	ModelInformationCollection class that aggregates the collection of Models inside a Minos Training Set. More...

class	ModelInstanceInfoCollection
	Instance information collection that is attached to a training set model. More...

class	TrainingSetObject
	Base class for all objects that can be inside a training set. More...

class	TrainingSetObjectInfoCollection
	Additional inheritance level of the InformationCollection object from the Image Manager wrapper that specifies stronger constraints on the TInfo generic parameter More...

Public Member Functions
	TrainingSet ()
	Create an empty training set. More...

	TrainingSet (string fileName)
	Load a saved training set from a file. More...

	TrainingSet (byte[] buffer)

	TrainingSet (byte[] buffer, int offset, int size)
	Recreate a serialized Minos training set from a byte array. More...

TrainingSet	Transform (Matrix2D transformation)
	Generate a new training set by transforming this training set with a 2x2 transformation matrix. More...

void	Save (string fileName)
	Write the training set to a file. More...

unsafe SearchResult[]	CheckConsistency (double threshold, double density)
	Test all images in the parent training set for instances of this model that might have been forgotten during the training process. More...

void	Dispose ()
	Dispose function to get rid of resources allocated by this object.

override int	GetHashCode ()
	Retrieve a hash code for this object. More...

override bool	Equals (object obj)
	Check two TrainingSet objects for equality. More...

Protected Member Functions
virtual void	OnImageAdded (ImageInfo img)
	Event router for the ImageAdded event. More...

virtual void	OnImageRemoved (ImageRemovedEventArgs args)
	Event router for the ImageRemoved event. More...

virtual void	OnModelAdded (ModelInfo mdl)
	Event router for the ModelAdded event. More...

virtual void	OnModelRemoved (ModelRemovedEventArgs args)
	Event router for the ModelRemoved event. More...

virtual void	OnInstanceAdded (InstanceInfo mdl)
	Event router for the InstanceAdded event. More...

virtual void	OnInstanceRemoved (InstanceRemovedEventArgs args)
	Event router for the InstanceRemoved event. More...

virtual void	Dispose (bool disposing)
	IDispose helper function. More...

Properties
int	ClassCount `[get]`
	The number of identifiable classes inside this training set. More...

InstanceInfoCollection	Instances `[get]`
	The instances contained in this training set. More...

int	InstancesTotal `[get]`
	Total number of instances currently trained in this training set.

ImageInfoCollection	Images `[get]`
	The images contained in this training set. More...

ModelInfoCollection	Models `[get]`
	The models contained in this training set. More...

string	FileName `[get]`
	Name of the file from which this image list was loaded (string.Empty if this image list was neither loaded nor saved since its construction). More...

bool	IsModified `[get, set]`
	Get or sets a flag that informs about unsaved modifications to the training set. Saving the training set will set this flag to false. More...

Rect	Extent `[get]`
	Extent of the classes in the training set relative to the anchor point. More...

Rect	LastFeatureWindow `[get]`
	Feature window of the last model that has been created for the training set. More...

unsafe string	Comment `[get, set]`
	Comment assigned to the training set. More...

double	ExpectationRadius `[get, set]`
	Expectation radius is the radius that is searched for the best occurrence of a sample when adding a new sample to the training set. More...

double	CorrelationThreshold `[get, set]`
	Correlation threshold is the correlation value below which Minos will suggest the generation of a new model rather over the generation of a new instance of an existing model with which the new instance correlates poorly. More...

IntPtr	Handle `[get]`
	Native Minos training set handle. More...

bool	IsDisposed `[get]`
	Tests if the native handle has already been disposed.

Properties inherited from INativeHandle
IntPtr	Handle `[get]`
	The native handle of the object.

bool	IsDisposed `[get]`
	Possibility to check whether the object has already been disposed of.

Events
EventHandler< ConsistencyCheckEventArgs >	ConsistencyCheckProgress
	This event will be used for indicating progress of consistency checks triggered by a call to CheckConsistency.

EventHandler< ImageAddedEventArgs >	ImageAdded
	Event that will be raised when a new image has been added to the training set.

EventHandler< ImageRemovedEventArgs >	ImageRemoved
	Event that will be raised when an image has been removed from the training set. Note that removing an image usually also means that a number of instances (the instances that were marked on the removed image) have been removed as well, but the if the reason for the removal of these instances was the removal of an image, there will be no notification event for the implicitly removed instances. Likewise, the removal of an image may also trigger the removal of a model. Again, there will be no separate notification about this if the reason was the removal of an image. This means that to be on the safe side, it usually makes sense to update the model and instance views as well, if an image has been removed.

EventHandler< ModelAddedEventArgs >	ModelAdded
	Event that will be raised when a new model has been added to the training set.

EventHandler< ModelRemovedEventArgs >	ModelRemoved
	Event that will be raised when an image has been removed from the training set. Note that removing an image usually also means that a number of instances (the instances that were marked on the removed image) have been removed as well, but the if the reason for the removal of these instances was the removal of an image, there will be no notification event for the implicitly removed instances. Likewise, the removal of an image may also trigger the removal of a model. Again, there will be no separate notification about this if the reason was the removal of an image. This means that to be on the safe side, it usually makes sense to update the model and instance views as well, if an image has been removed.

EventHandler< InstanceAddedEventArgs >	InstanceAdded
	Event that will be raised when a new model has been added to the training set.

EventHandler< InstanceRemovedEventArgs >	InstanceRemoved
	Event that will be raised when an image has been removed from the training set. Note that removing an image usually also means that a number of instances (the instances that were marked on the removed image) have been removed as well, but the if the reason for the removal of these instances was the removal of an image, there will be no notification event for the implicitly removed instances. Likewise, the removal of an image may also trigger the removal of a model. Again, there will be no separate notification about this if the reason was the removal of an image. This means that to be on the safe side, it usually makes sense to update the model and instance views as well, if an image has been removed.

NativeHandleEventDelegate	ObjectDisposing
	Raised when this object is about to be disposed via the IDisposable.Dispose method. More...

Events inherited from INativeHandle
NativeHandleEventDelegate	ObjectDisposing
	Raised when this object is about to be disposed via the IDisposable.Dispose method. More...

Detailed Description

A Minos Training Set from which a classifier can be generated.

Constructor & Destructor Documentation

◆ TrainingSet() [1/4]

TrainingSet ( )

Create an empty training set.

Exceptions

OutOfMemoryException If a new training set could not be allocated

◆ TrainingSet() [2/4]

TrainingSet ( string fileName )

Load a saved training set from a file.

Parameters

fileName name of the file to be loaded

Exceptions

FileNotFoundException	If the training set file does not exist
FileLoadException	If loading an existing training set file failed
ArgumentNullException	If fileName is null

◆ TrainingSet() [3/4]

TrainingSet ( byte[] buffer )

Recreate a serialized Minos Training Set from a byte array.

Parameters

buffer byte array from which to create the training set.

◆ TrainingSet() [4/4]

TrainingSet	(	byte[]	buffer,
		int	offset,
		int	size
	)

Recreate a serialized Minos training set from a byte array.

Parameters

buffer	Byte array from which to create the training set
offset	offset at which to start
size	number of bytes to use

Exceptions

ArgumentNullException	If buffer is null
ArgumentException	If offset is less than zero or size is less than zero
ArgumentOutOfRangeException	If offset and size have values that together exceed the length of buffer
InvalidDataException	If buffer does not contain a deserializable Minos training set at the given offset.

Member Function Documentation

◆ CheckConsistency()

unsafe SearchResult[] CheckConsistency	(	double	threshold,
		double	density
	)

Test all images in the parent training set for instances of this model that might have been forgotten during the training process.

Parameters

threshold	correlation threshold to be used for the consistency check
density	density to be used for the consistency check

Returns: potentially forgotten positive samples that may be added to the training set

Exceptions

ObjectDisposedException	If the training set has already been disposed
ArgumentOutOfRangeException	If invalid values for threshold or density have been passed (both parameters must be within the range [0...1.0]
CvbException	If the execution of the consistency check failed

◆ Dispose()

virtual void Dispose ( bool disposing )

protectedvirtual

IDispose helper function.

Parameters

disposing true when called via IDisposable.Dispose, false when called by the finalizer.

◆ Equals()

override bool Equals ( object obj )

Check two TrainingSet objects for equality.

Parameters

obj	other object to check

Returns: true if the objects are equal, false otherwise

Exceptions

ObjectDisposedException If this training set has already been disposed

◆ GetHashCode()

override int GetHashCode ( )

Retrieve a hash code for this object.

Returns: Hash code for this object

◆ OnImageAdded()

virtual void OnImageAdded ( ImageInfo img )

protectedvirtual

Event router for the ImageAdded event.

Parameters

img

◆ OnImageRemoved()

virtual void OnImageRemoved ( ImageRemovedEventArgs args )

protectedvirtual

Event router for the ImageRemoved event.

Parameters

args

◆ OnInstanceAdded()

virtual void OnInstanceAdded ( InstanceInfo mdl )

protectedvirtual

Event router for the InstanceAdded event.

Parameters

mdl

◆ OnInstanceRemoved()

virtual void OnInstanceRemoved ( InstanceRemovedEventArgs args )

protectedvirtual

Event router for the InstanceRemoved event.

Parameters

args

◆ OnModelAdded()

virtual void OnModelAdded ( ModelInfo mdl )

protectedvirtual

Event router for the ModelAdded event.

Parameters

mdl

◆ OnModelRemoved()

virtual void OnModelRemoved ( ModelRemovedEventArgs args )

protectedvirtual

Event router for the ModelRemoved event.

Parameters

args

◆ Save()

void Save ( string fileName )

Write the training set to a file.

As a side effect this function sets the IsModified flag to false if the training set was saved successfully.

Parameters

fileName

Exceptions

ObjectDisposedException	If the training set has already been disposed
ArgumentNullException	when fileName is null

◆ Transform()

TrainingSet Transform ( Matrix2D transformation )

Generate a new training set by transforming this training set with a 2x2 transformation matrix.

Parameters

transformation matrix with which to transform the geometry of this training set

Returns: transformed training set

Exceptions

ObjectDisposedException	If the training set has already been disposed
DivideByZeroException	when providing a transformation matrix that is singular.

Property Documentation

◆ ClassCount

int ClassCount

get

The number of identifiable classes inside this training set.

The number of classes cannot be higher than the number of trained models (see property this.Models.Count). If two models have the same name, the classifier can no longer discriminate these two models, reducing the ClassCount property.

Exceptions

ObjectDisposedException If the training set has already been disposed

◆ Comment

unsafe string Comment

getset

Comment assigned to the training set.

Exceptions

ObjectDisposedException If the training set has already been disposed

◆ CorrelationThreshold

double CorrelationThreshold

getset

Correlation threshold is the correlation value below which Minos will suggest the generation of a new model rather over the generation of a new instance of an existing model with which the new instance correlates poorly.

Exceptions

ObjectDisposedException	If the training set has already been disposed
ArgumentOutOfRangeException	when trying to set a correlation threshold less than 0 or greater than 1

◆ ExpectationRadius

double ExpectationRadius

getset

Expectation radius is the radius that is searched for the best occurrence of a sample when adding a new sample to the training set.

Exceptions

ObjectDisposedException	If the training set has already been disposed
ArgumentOutOfRangeException	when trying to set an expectation radius less than 0

◆ Extent

Rect Extent

get

Extent of the classes in the training set relative to the anchor point.

Exceptions

ObjectDisposedException If the training set has already been disposed

◆ FileName

string FileName

get

Name of the file from which this image list was loaded (string.Empty if this image list was neither loaded nor saved since its construction).

Exceptions

ObjectDisposedException If the training set has already been disposed

◆ Handle

IntPtr Handle

get

Native Minos training set handle.

It is normally not necessary to work with this handle in CVB.Net. Note that using this handle in connection with the legacy CVB managed wrappers may actually be disruptive to your application.

Implements INativeHandle.

◆ Images

ImageInfoCollection Images

get

The images contained in this training set.

Exceptions

ObjectDisposedException If the training set has already been disposed

◆ Instances

InstanceInfoCollection Instances

get

The instances contained in this training set.

Exceptions

ObjectDisposedException If the training set has already been disposed

◆ IsModified

bool IsModified

getset

Get or sets a flag that informs about unsaved modifications to the training set. Saving the training set will set this flag to false.

Exceptions

ObjectDisposedException If the training set has already been disposed

◆ LastFeatureWindow

Rect LastFeatureWindow

get

Feature window of the last model that has been created for the training set.

Exceptions

ObjectDisposedException If the training set has already been disposed

◆ Models

ModelInfoCollection Models

get

The models contained in this training set.

Exceptions

ObjectDisposedException If the training set has already been disposed

Event Documentation

◆ ObjectDisposing

NativeHandleEventDelegate ObjectDisposing

Raised when this object is about to be disposed via the IDisposable.Dispose method.

This event is raised right before this object is disposed. The dispose itself cannot be canceled.

Classes

Public Member Functions

Protected Member Functions

Properties

Events

Detailed Description

Constructor & Destructor Documentation

◆ TrainingSet() [1/4]

◆ TrainingSet() [2/4]

◆ TrainingSet() [3/4]

◆ TrainingSet() [4/4]

Member Function Documentation

◆ CheckConsistency()

◆ Dispose()

◆ Equals()

◆ GetHashCode()

◆ OnImageAdded()

◆ OnImageRemoved()

◆ OnInstanceAdded()

◆ OnInstanceRemoved()

◆ OnModelAdded()

◆ OnModelRemoved()

◆ Save()

◆ Transform()

Property Documentation

◆ ClassCount

◆ Comment

◆ CorrelationThreshold

◆ ExpectationRadius

◆ Extent

◆ FileName

◆ Handle

◆ Images

◆ Instances

◆ IsModified

◆ LastFeatureWindow

◆ Models

Event Documentation

◆ ObjectDisposing