An eVision.EasyDeepLearning.EDataAugmentation object is responsible for storing the data augmentation parameters and generating new transformed images from existing ones. More...

Public Member Functions
None	__init__ (self)
	Creates an eVision.EasyDeepLearning.EDataAugmentation object.
None	__init__ (self, eVision.EasyDeepLearning.EDataAugmentation other)
	Creates a copy of the eVision.EasyDeepLearning.EDataAugmentation object other.
None	CopyTo (self, eVision.EasyDeepLearning.EDataAugmentation other)
	Copies itself to the eVision.EasyDeepLearning.EDataAugmentation object other.
None	Load (self, str path)
	Loads an eVision.EasyDeepLearning.EDataAugmentation object.
None	Load (self, eVision.ESerializer\|None serializer)
	Loads an eVision.EasyDeepLearning.EDataAugmentation object.
None	Save (self, str path)
	Saves an eVision.EasyDeepLearning.EDataAugmentation object.
None	Save (self, eVision.ESerializer\|None serializer)
	Saves an eVision.EasyDeepLearning.EDataAugmentation object.
bool	HasDataAugmentations (self)
	Whether the eVision.EasyDeepLearning.EDataAugmentation object contains augmentations.
eVision.EBaseROI\|None	Generate (self, eVision.EBaseROI\|None img, eVision.EasyDeepLearning.EDLDataAugmentationType type)
	Generates a new image from a generic eVision.EBaseROI image.
eVision.EBaseROI\|None	Generate (self, eVision.EBaseROI\|None img)
	Generates a new image from a generic eVision.EBaseROI image.
None	Generate (self, eVision.EROIBW8 source, eVision.EROIBW8 dest, eVision.EasyDeepLearning.EDLDataAugmentationType type)
	Generates a new image.
None	Generate (self, eVision.EROIBW8 source, eVision.EROIBW8 dest)
	Generates a new image.
None	Generate (self, eVision.EROIBW16 source, eVision.EROIBW16 dest, eVision.EasyDeepLearning.EDLDataAugmentationType type)
	Generates a new image.
None	Generate (self, eVision.EROIBW16 source, eVision.EROIBW16 dest)
	Generates a new image.
None	Generate (self, eVision.EROIC24 source, eVision.EROIC24 dest, eVision.EasyDeepLearning.EDLDataAugmentationType type)
	Generates a new image.
None	Generate (self, eVision.EROIC24 source, eVision.EROIC24 dest)
	Generates a new image.
int	MaxHorizontalShift (self)
	Return the maximum horizontal shift allowed.
	MaxHorizontalShift (self, int value)
	Return the maximum horizontal shift allowed.
int	MaxVerticalShift (self)
	Return the maximum vertical shift allowed.
	MaxVerticalShift (self, int value)
	Return the maximum vertical shift allowed.
float	MaxRotationAngle (self)
	Returns the maximum rotation angle allowed.
	MaxRotationAngle (self, float value)
	Returns the maximum rotation angle allowed.
bool	EnableVerticalFlip (self)
	Gets whether to use vertically flipped versions of input images.
	EnableVerticalFlip (self, bool value)
	Gets whether to use vertically flipped versions of input images.
bool	EnableHorizontalFlip (self)
	Gets whether to use horizontally flipped versions of input images.
	EnableHorizontalFlip (self, bool value)
	Gets whether to use horizontally flipped versions of input images.
float	MinScale (self)
	Gets the minimum scaling allowed.
	MinScale (self, float value)
	Gets the minimum scaling allowed.
float	MaxScale (self)
	Gets the maximum scaling allowed.
	MaxScale (self, float value)
	Gets the maximum scaling allowed.
float	MaxHorizontalShear (self)
	Gets the maximum absolute horizontal shear, represented as an angle from the vertical direction.
	MaxHorizontalShear (self, float value)
	Gets the maximum absolute horizontal shear, represented as an angle from the vertical direction.
float	MaxVerticalShear (self)
	Gets the maximum absolute vertical shear, represented as an angle from the horizontal direction.
	MaxVerticalShear (self, float value)
	Gets the maximum absolute vertical shear, represented as an angle from the horizontal direction.
float	MaxBrightnessOffset (self)
	Gets the maximum absolute brightness offset.
	MaxBrightnessOffset (self, float value)
	Gets the maximum absolute brightness offset.
float	MaxContrastGain (self)
	Gets the maximum contrast gain.
	MaxContrastGain (self, float value)
	Gets the maximum contrast gain.
float	MinContrastGain (self)
	Gets the minimum contrast gain.
	MinContrastGain (self, float value)
	Gets the minimum contrast gain.
float	MinGamma (self)
	Gets the minimum gamma for gamma correction.
	MinGamma (self, float value)
	Gets the minimum gamma for gamma correction.
float	MaxGamma (self)
	Gets the maximum gamma for gamma correction.
	MaxGamma (self, float value)
	Gets the maximum gamma for gamma correction.
float	MaxHueOffset (self)
	Gets the maximum absolute hue offset.
	MaxHueOffset (self, float value)
	Gets the maximum absolute hue offset.
float	MinSaturationGain (self)
	Gets the minimum saturation gain.
	MinSaturationGain (self, float value)
	Gets the minimum saturation gain.
float	MaxSaturationGain (self)
	Gets the maximum saturation gain.
	MaxSaturationGain (self, float value)
	Gets the maximum saturation gain.
float	GaussianNoiseMaximumStandardDeviation (self)
	The Gaussian noise maximum standard deviation.
	GaussianNoiseMaximumStandardDeviation (self, float value)
	The Gaussian noise maximum standard deviation.
float	GaussianNoiseMinimumStandardDeviation (self)
	The Gaussian noise minimum standard deviation.
	GaussianNoiseMinimumStandardDeviation (self, float value)
	The Gaussian noise minimum standard deviation.
float	SpeckleNoiseMaximumStandardDeviation (self)
	The speckle noise maximum standard deviation.
	SpeckleNoiseMaximumStandardDeviation (self, float value)
	The speckle noise maximum standard deviation.
float	SpeckleNoiseMinimumStandardDeviation (self)
	The speckle noise minimum standard deviation.
	SpeckleNoiseMinimumStandardDeviation (self, float value)
	The speckle noise minimum standard deviation.
float	SaltAndPepperNoiseMaximumDensity (self)
	The maximum density of the salt and pepper noise.
	SaltAndPepperNoiseMaximumDensity (self, float value)
	The maximum density of the salt and pepper noise.
float	SaltAndPepperNoiseMinimumDensity (self)
	The minimum density of the salt and pepper noise.
	SaltAndPepperNoiseMinimumDensity (self, float value)
	The minimum density of the salt and pepper noise.
int	StainDisruptionMaxAnchorPoints (self)
	The maximum number of anchor points to use while disrupting the ellipse.
	StainDisruptionMaxAnchorPoints (self, int value)
	The maximum number of anchor points to use while disrupting the ellipse.
float	StainDisruptionMaxIntensity (self)
	The maximum offset applied to the coordinates (x,y) of a pixel from the ellipse contour.
	StainDisruptionMaxIntensity (self, float value)
	The maximum offset applied to the coordinates (x,y) of a pixel from the ellipse contour.
int	StainEllipseMinRadius (self)
	The minimum radius of the ellipse used to produce the stain.
	StainEllipseMinRadius (self, int value)
	The minimum radius of the ellipse used to produce the stain.
int	StainEllipseMaxRadius (self)
	The maximum radius of the ellipse used to produce the stain.
	StainEllipseMaxRadius (self, int value)
	The maximum radius of the ellipse used to produce the stain.
int	MinStainColor (self)
	The minimum color value from which to draw a color to fill in the stain (we use a gaussian distribution of with a mean between eVision.EasyDeepLearning.EDataAugmentation.MinStainColor and eVision.EasyDeepLearning.EDataAugmentation.MaxStainColor).
	MinStainColor (self, int value)
	The minimum color value from which to draw a color to fill in the stain (we use a gaussian distribution of with a mean between eVision.EasyDeepLearning.EDataAugmentation.MinStainColor and eVision.EasyDeepLearning.EDataAugmentation.MaxStainColor).
int	MaxStainColor (self)
	The maximum color value from which to draw a color to fill in the stain (we use a gaussian distribution of with a mean between eVision.EasyDeepLearning.EDataAugmentation.MinStainColor and eVision.EasyDeepLearning.EDataAugmentation.MaxStainColor).
	MaxStainColor (self, int value)
	The maximum color value from which to draw a color to fill in the stain (we use a gaussian distribution of with a mean between eVision.EasyDeepLearning.EDataAugmentation.MinStainColor and eVision.EasyDeepLearning.EDataAugmentation.MaxStainColor).
int	StainColorVariation (self)
	The variation of the color used to fill in the ellipse (We use an gaussian distribution of standard mean deviation stainColorVariation).
	StainColorVariation (self, int value)
	The variation of the color used to fill in the ellipse (We use an gaussian distribution of standard mean deviation stainColorVariation).
float	StainInterpolationSiteFactor (self)
	The factor to compute the number of interpolation sites to use from the number of anchor points.
	StainInterpolationSiteFactor (self, float value)
	The factor to compute the number of interpolation sites to use from the number of anchor points.
float	StainProbability (self)
	Probability of applying the stain data augmentation (default value: 0, i.e.
	StainProbability (self, float value)
	Probability of applying the stain data augmentation (default value: 0, i.e.
int	StainBlur (self)
	Stain blur.
	StainBlur (self, int value)
	Stain blur.

Detailed Description

An eVision.EasyDeepLearning.EDataAugmentation object is responsible for storing the data augmentation parameters and generating new transformed images from existing ones.

Deep learning algorithms are not invariant to the location, scale, or rotation of the elements of interest in the image. Thus, if the final application requires the deep learning algorithm to be invariant to those characteristics, the dataset must contain images covering the spectrum of those characteristics.

Data augmentation can be used to avoid capturing in the original dataset the whole spectrum of those characteristics by automatically generating new versions of the images in the dataset that are shifted, scaled or rotated to cover this spectrum.

When generating a new image, the eVision.EasyDeepLearning.EDataAugmentation will randomly pick transformation value within the specified limits.

When using data augmentation, one must carefully chose the spectrum of transformations such that the element of interest are still visible in the image.

Remarks:

License(s): EasyClassify or EasySegment or EasyLocate

Constructor & Destructor Documentation

◆ init() [1/2]

None eVision.EasyDeepLearning.EDataAugmentation.__init__ ( self )

Creates an eVision.EasyDeepLearning.EDataAugmentation object.

◆ init() [2/2]

None eVision.EasyDeepLearning.EDataAugmentation.__init__	(		self,
		eVision.EasyDeepLearning.EDataAugmentation	other )

Creates a copy of the eVision.EasyDeepLearning.EDataAugmentation object other.

Parameters

other The other eVision.EasyDeepLearning.EDataAugmentation that we will copy into this one.

Member Function Documentation

◆ CopyTo()

None eVision.EasyDeepLearning.EDataAugmentation.CopyTo	(		self,
		eVision.EasyDeepLearning.EDataAugmentation	other )

Copies itself to the eVision.EasyDeepLearning.EDataAugmentation object other.

Parameters

other The other eVision.EasyDeepLearning.EDataAugmentation that we will copy into this one.

◆ EnableHorizontalFlip() [1/2]

bool eVision.EasyDeepLearning.EDataAugmentation.EnableHorizontalFlip ( self )

Gets whether to use horizontally flipped versions of input images.

◆ EnableHorizontalFlip() [2/2]

eVision.EasyDeepLearning.EDataAugmentation.EnableHorizontalFlip	(		self,
		bool	value )

Gets whether to use horizontally flipped versions of input images.

◆ EnableVerticalFlip() [1/2]

bool eVision.EasyDeepLearning.EDataAugmentation.EnableVerticalFlip ( self )

Gets whether to use vertically flipped versions of input images.

◆ EnableVerticalFlip() [2/2]

eVision.EasyDeepLearning.EDataAugmentation.EnableVerticalFlip	(		self,
		bool	value )

Gets whether to use vertically flipped versions of input images.

◆ GaussianNoiseMaximumStandardDeviation() [1/2]

float eVision.EasyDeepLearning.EDataAugmentation.GaussianNoiseMaximumStandardDeviation ( self )

The Gaussian noise maximum standard deviation.

The Gaussian noise is an additive noise sampled from a Gaussian distribution of deviation between eVision.EasyDeepLearning.EDataAugmentation.GaussianNoiseMinimumStandardDeviation and eVision.EasyDeepLearning.EDataAugmentation.GaussianNoiseMaximumStandardDeviation.(also called the normal distribution).

Its value must be superior or equal to eVision.EasyDeepLearning.EDataAugmentation.GaussianNoiseMinimumStandardDeviation.

Remarks:

This noise is computed before the salt and paper noise.

◆ GaussianNoiseMaximumStandardDeviation() [2/2]

eVision.EasyDeepLearning.EDataAugmentation.GaussianNoiseMaximumStandardDeviation	(		self,
		float	value )

The Gaussian noise maximum standard deviation.

The Gaussian noise is an additive noise sampled from a Gaussian distribution of deviation between eVision.EasyDeepLearning.EDataAugmentation.GaussianNoiseMinimumStandardDeviation and eVision.EasyDeepLearning.EDataAugmentation.GaussianNoiseMaximumStandardDeviation.(also called the normal distribution).

Its value must be superior or equal to eVision.EasyDeepLearning.EDataAugmentation.GaussianNoiseMinimumStandardDeviation.

Remarks:

This noise is computed before the salt and paper noise.

◆ GaussianNoiseMinimumStandardDeviation() [1/2]

float eVision.EasyDeepLearning.EDataAugmentation.GaussianNoiseMinimumStandardDeviation ( self )

The Gaussian noise minimum standard deviation.

The Gaussian noise is an additive noise sampled from a Gaussian distribution of deviation between eVision.EasyDeepLearning.EDataAugmentation.GaussianNoiseMinimumStandardDeviation and eVision.EasyDeepLearning.EDataAugmentation.GaussianNoiseMaximumStandardDeviation.(also called the normal distribution).

Its value must be between 0 and eVision.EasyDeepLearning.EDataAugmentation.GaussianNoiseMaximumStandardDeviation.

Remarks:

This noise is computed before the salt and paper noise.

◆ GaussianNoiseMinimumStandardDeviation() [2/2]

eVision.EasyDeepLearning.EDataAugmentation.GaussianNoiseMinimumStandardDeviation	(		self,
		float	value )

The Gaussian noise minimum standard deviation.

The Gaussian noise is an additive noise sampled from a Gaussian distribution of deviation between eVision.EasyDeepLearning.EDataAugmentation.GaussianNoiseMinimumStandardDeviation and eVision.EasyDeepLearning.EDataAugmentation.GaussianNoiseMaximumStandardDeviation.(also called the normal distribution).

Its value must be between 0 and eVision.EasyDeepLearning.EDataAugmentation.GaussianNoiseMaximumStandardDeviation.

Remarks:

This noise is computed before the salt and paper noise.

◆ Generate() [1/8]

eVision.EBaseROI \| None eVision.EasyDeepLearning.EDataAugmentation.Generate	(		self,
		eVision.EBaseROI \| None	img )

Generates a new image from a generic eVision.EBaseROI image.

The caller is responsible for calling freeing the memory of the returned image by calling delete on the image.

Remarks:

If the data augmentation fails, the method will throw an exception.

Parameters

img	The image to transform.

◆ Generate() [2/8]

eVision.EBaseROI \| None eVision.EasyDeepLearning.EDataAugmentation.Generate	(		self,
		eVision.EBaseROI \| None	img,
		eVision.EasyDeepLearning.EDLDataAugmentationType	type )

Generates a new image from a generic eVision.EBaseROI image.

The caller is responsible for calling freeing the memory of the returned image by calling delete on the image.

Remarks:

If the data augmentation fails, the method will throw an exception.

Parameters

img	The image to transform.
type	The type of transformation to generate.

◆ Generate() [3/8]

None eVision.EasyDeepLearning.EDataAugmentation.Generate	(		self,
		eVision.EROIBW16	source,
		eVision.EROIBW16	dest )

Generates a new image.

Remarks:

If the data augmentation fails, the method will throw an exception.

Parameters

source	The image to transform.
dest	The transformed image.

◆ Generate() [4/8]

None eVision.EasyDeepLearning.EDataAugmentation.Generate	(		self,
		eVision.EROIBW16	source,
		eVision.EROIBW16	dest,
		eVision.EasyDeepLearning.EDLDataAugmentationType	type )

Generates a new image.

Remarks:

If the data augmentation fails, the method will throw an exception.

Parameters

source	The image to transform.
dest	The transformed image.
type	The type of transformation to generate.

◆ Generate() [5/8]

None eVision.EasyDeepLearning.EDataAugmentation.Generate	(		self,
		eVision.EROIBW8	source,
		eVision.EROIBW8	dest )

Generates a new image.

Remarks:

If the data augmentation fails, the method will throw an exception.

Parameters

source	The image to transform.
dest	The transformed image.

◆ Generate() [6/8]

None eVision.EasyDeepLearning.EDataAugmentation.Generate	(		self,
		eVision.EROIBW8	source,
		eVision.EROIBW8	dest,
		eVision.EasyDeepLearning.EDLDataAugmentationType	type )

Generates a new image.

Remarks:

If the data augmentation fails, the method will throw an exception.

Parameters

source	The image to transform.
dest	The transformed image.
type	The type of transformation to generate.

◆ Generate() [7/8]

None eVision.EasyDeepLearning.EDataAugmentation.Generate	(		self,
		eVision.EROIC24	source,
		eVision.EROIC24	dest )

Generates a new image.

Remarks:

If the data augmentation fails, the method will throw an exception.

Parameters

source	The image to transform.
dest	The transformed image.

◆ Generate() [8/8]

None eVision.EasyDeepLearning.EDataAugmentation.Generate	(		self,
		eVision.EROIC24	source,
		eVision.EROIC24	dest,
		eVision.EasyDeepLearning.EDLDataAugmentationType	type )

Generates a new image.

Remarks:

If the data augmentation fails, the method will throw an exception.

Parameters

source	The image to transform.
dest	The transformed image.
type	The type of transformation to generate.

◆ HasDataAugmentations()

bool eVision.EasyDeepLearning.EDataAugmentation.HasDataAugmentations ( self )

Whether the eVision.EasyDeepLearning.EDataAugmentation object contains augmentations.

◆ Load() [1/2]

None eVision.EasyDeepLearning.EDataAugmentation.Load	(		self,
		eVision.ESerializer \| None	serializer )

Loads an eVision.EasyDeepLearning.EDataAugmentation object.

The given ESerializer must have been created for reading.

Parameters

serializer The serializer.

◆ Load() [2/2]

None eVision.EasyDeepLearning.EDataAugmentation.Load	(		self,
		str	path )

Loads an eVision.EasyDeepLearning.EDataAugmentation object.

The given ESerializer must have been created for reading.

Parameters

path	The file path.

◆ MaxBrightnessOffset() [1/2]

float eVision.EasyDeepLearning.EDataAugmentation.MaxBrightnessOffset ( self )

Gets the maximum absolute brightness offset.

◆ MaxBrightnessOffset() [2/2]

eVision.EasyDeepLearning.EDataAugmentation.MaxBrightnessOffset	(		self,
		float	value )

Gets the maximum absolute brightness offset.

◆ MaxContrastGain() [1/2]

float eVision.EasyDeepLearning.EDataAugmentation.MaxContrastGain ( self )

Gets the maximum contrast gain.

Remarks:

Contrast transformation is performed by multiplying each mean-centered pixel value by a gain value taken between eVision.EasyDeepLearning.EDataAugmentation.MinContrastGain and eVision.EasyDeepLearning.EDataAugmentation.MaxContrastGain. A contrast transformation does not change the overall brightness of an image.

◆ MaxContrastGain() [2/2]

eVision.EasyDeepLearning.EDataAugmentation.MaxContrastGain	(		self,
		float	value )

Gets the maximum contrast gain.

Remarks:

Contrast transformation is performed by multiplying each mean-centered pixel value by a gain value taken between eVision.EasyDeepLearning.EDataAugmentation.MinContrastGain and eVision.EasyDeepLearning.EDataAugmentation.MaxContrastGain. A contrast transformation does not change the overall brightness of an image.

◆ MaxGamma() [1/2]

float eVision.EasyDeepLearning.EDataAugmentation.MaxGamma ( self )

Gets the maximum gamma for gamma correction.

Remarks:

Gamma correction transformation is performed by raising each normalized pixel value to the power of gamma with gamma taken between eVision.EasyDeepLearning.EDataAugmentation.MinGamma and eVision.EasyDeepLearning.EDataAugmentation.MaxGamma.

◆ MaxGamma() [2/2]

eVision.EasyDeepLearning.EDataAugmentation.MaxGamma	(		self,
		float	value )

Gets the maximum gamma for gamma correction.

Remarks:

Gamma correction transformation is performed by raising each normalized pixel value to the power of gamma with gamma taken between eVision.EasyDeepLearning.EDataAugmentation.MinGamma and eVision.EasyDeepLearning.EDataAugmentation.MaxGamma.

◆ MaxHorizontalShear() [1/2]

float eVision.EasyDeepLearning.EDataAugmentation.MaxHorizontalShear ( self )

Gets the maximum absolute horizontal shear, represented as an angle from the vertical direction.

◆ MaxHorizontalShear() [2/2]

eVision.EasyDeepLearning.EDataAugmentation.MaxHorizontalShear	(		self,
		float	value )

Gets the maximum absolute horizontal shear, represented as an angle from the vertical direction.

◆ MaxHorizontalShift() [1/2]

int eVision.EasyDeepLearning.EDataAugmentation.MaxHorizontalShift ( self )

Return the maximum horizontal shift allowed.

◆ MaxHorizontalShift() [2/2]

eVision.EasyDeepLearning.EDataAugmentation.MaxHorizontalShift	(		self,
		int	value )

Return the maximum horizontal shift allowed.

◆ MaxHueOffset() [1/2]

float eVision.EasyDeepLearning.EDataAugmentation.MaxHueOffset ( self )

Gets the maximum absolute hue offset.

◆ MaxHueOffset() [2/2]

eVision.EasyDeepLearning.EDataAugmentation.MaxHueOffset	(		self,
		float	value )

Gets the maximum absolute hue offset.

◆ MaxRotationAngle() [1/2]

float eVision.EasyDeepLearning.EDataAugmentation.MaxRotationAngle ( self )

Returns the maximum rotation angle allowed.

The unit of the returned angle is in the current angle unit (see eVision.Easy.SetAngleUnit).

◆ MaxRotationAngle() [2/2]

eVision.EasyDeepLearning.EDataAugmentation.MaxRotationAngle	(		self,
		float	value )

Returns the maximum rotation angle allowed.

The unit of the returned angle is in the current angle unit (see eVision.Easy.SetAngleUnit).

◆ MaxSaturationGain() [1/2]

float eVision.EasyDeepLearning.EDataAugmentation.MaxSaturationGain ( self )

Gets the maximum saturation gain.

◆ MaxSaturationGain() [2/2]

eVision.EasyDeepLearning.EDataAugmentation.MaxSaturationGain	(		self,
		float	value )

Gets the maximum saturation gain.

◆ MaxScale() [1/2]

float eVision.EasyDeepLearning.EDataAugmentation.MaxScale ( self )

Gets the maximum scaling allowed.

◆ MaxScale() [2/2]

eVision.EasyDeepLearning.EDataAugmentation.MaxScale	(		self,
		float	value )

Gets the maximum scaling allowed.

◆ MaxStainColor() [1/2]

int eVision.EasyDeepLearning.EDataAugmentation.MaxStainColor ( self )

The maximum color value from which to draw a color to fill in the stain (we use a gaussian distribution of with a mean between eVision.EasyDeepLearning.EDataAugmentation.MinStainColor and eVision.EasyDeepLearning.EDataAugmentation.MaxStainColor).

Its value must be bigger than eVision.EasyDeepLearning.EDataAugmentation.MinStainColor.

◆ MaxStainColor() [2/2]

eVision.EasyDeepLearning.EDataAugmentation.MaxStainColor	(		self,
		int	value )

The maximum color value from which to draw a color to fill in the stain (we use a gaussian distribution of with a mean between eVision.EasyDeepLearning.EDataAugmentation.MinStainColor and eVision.EasyDeepLearning.EDataAugmentation.MaxStainColor).

Its value must be bigger than eVision.EasyDeepLearning.EDataAugmentation.MinStainColor.

◆ MaxVerticalShear() [1/2]

float eVision.EasyDeepLearning.EDataAugmentation.MaxVerticalShear ( self )

Gets the maximum absolute vertical shear, represented as an angle from the horizontal direction.

◆ MaxVerticalShear() [2/2]

eVision.EasyDeepLearning.EDataAugmentation.MaxVerticalShear	(		self,
		float	value )

Gets the maximum absolute vertical shear, represented as an angle from the horizontal direction.

◆ MaxVerticalShift() [1/2]

int eVision.EasyDeepLearning.EDataAugmentation.MaxVerticalShift ( self )

Return the maximum vertical shift allowed.

◆ MaxVerticalShift() [2/2]

eVision.EasyDeepLearning.EDataAugmentation.MaxVerticalShift	(		self,
		int	value )

Return the maximum vertical shift allowed.

◆ MinContrastGain() [1/2]

float eVision.EasyDeepLearning.EDataAugmentation.MinContrastGain ( self )

Gets the minimum contrast gain.

Remarks:

Contrast transformation is performed by multiplying each mean-centered pixel value by a gain value taken between eVision.EasyDeepLearning.EDataAugmentation.MinContrastGain and eVision.EasyDeepLearning.EDataAugmentation.MaxContrastGain. A contrast transformation does not change the overall brightness of an image.

◆ MinContrastGain() [2/2]

eVision.EasyDeepLearning.EDataAugmentation.MinContrastGain	(		self,
		float	value )

Gets the minimum contrast gain.

Remarks:

Contrast transformation is performed by multiplying each mean-centered pixel value by a gain value taken between eVision.EasyDeepLearning.EDataAugmentation.MinContrastGain and eVision.EasyDeepLearning.EDataAugmentation.MaxContrastGain. A contrast transformation does not change the overall brightness of an image.

◆ MinGamma() [1/2]

float eVision.EasyDeepLearning.EDataAugmentation.MinGamma ( self )

Gets the minimum gamma for gamma correction.

Remarks:

Gamma correction transformation is performed by raising each normalized pixel value to the power of gamma with gamma taken between eVision.EasyDeepLearning.EDataAugmentation.MinGamma and eVision.EasyDeepLearning.EDataAugmentation.MaxGamma.

◆ MinGamma() [2/2]

eVision.EasyDeepLearning.EDataAugmentation.MinGamma	(		self,
		float	value )

Gets the minimum gamma for gamma correction.

Remarks:

Gamma correction transformation is performed by raising each normalized pixel value to the power of gamma with gamma taken between eVision.EasyDeepLearning.EDataAugmentation.MinGamma and eVision.EasyDeepLearning.EDataAugmentation.MaxGamma.

◆ MinSaturationGain() [1/2]

float eVision.EasyDeepLearning.EDataAugmentation.MinSaturationGain ( self )

Gets the minimum saturation gain.

◆ MinSaturationGain() [2/2]

eVision.EasyDeepLearning.EDataAugmentation.MinSaturationGain	(		self,
		float	value )

Gets the minimum saturation gain.

◆ MinScale() [1/2]

float eVision.EasyDeepLearning.EDataAugmentation.MinScale ( self )

Gets the minimum scaling allowed.

◆ MinScale() [2/2]

eVision.EasyDeepLearning.EDataAugmentation.MinScale	(		self,
		float	value )

Gets the minimum scaling allowed.

◆ MinStainColor() [1/2]

int eVision.EasyDeepLearning.EDataAugmentation.MinStainColor ( self )

The minimum color value from which to draw a color to fill in the stain (we use a gaussian distribution of with a mean between eVision.EasyDeepLearning.EDataAugmentation.MinStainColor and eVision.EasyDeepLearning.EDataAugmentation.MaxStainColor).

Its value must be inferior to 255 and eVision.EasyDeepLearning.EDataAugmentation.MaxStainColor.

◆ MinStainColor() [2/2]

eVision.EasyDeepLearning.EDataAugmentation.MinStainColor	(		self,
		int	value )

The minimum color value from which to draw a color to fill in the stain (we use a gaussian distribution of with a mean between eVision.EasyDeepLearning.EDataAugmentation.MinStainColor and eVision.EasyDeepLearning.EDataAugmentation.MaxStainColor).

Its value must be inferior to 255 and eVision.EasyDeepLearning.EDataAugmentation.MaxStainColor.

◆ SaltAndPepperNoiseMaximumDensity() [1/2]

float eVision.EasyDeepLearning.EDataAugmentation.SaltAndPepperNoiseMaximumDensity ( self )

The maximum density of the salt and pepper noise.

The salt and pepper noise sets the value of a number of randomly selected (between eVision.EasyDeepLearning.EDataAugmentation.SaltAndPepperNoiseMinimumDensity and eVision.EasyDeepLearning.EDataAugmentation.SaltAndPepperNoiseMaximumDensity) pixels to its minimum or maximum value.

Its value must be between eVision.EasyDeepLearning.EDataAugmentation.SaltAndPepperNoiseMinimumDensity and 1.

Remarks:

This noise is computed after all the other noises.

◆ SaltAndPepperNoiseMaximumDensity() [2/2]

eVision.EasyDeepLearning.EDataAugmentation.SaltAndPepperNoiseMaximumDensity	(		self,
		float	value )

The maximum density of the salt and pepper noise.

The salt and pepper noise sets the value of a number of randomly selected (between eVision.EasyDeepLearning.EDataAugmentation.SaltAndPepperNoiseMinimumDensity and eVision.EasyDeepLearning.EDataAugmentation.SaltAndPepperNoiseMaximumDensity) pixels to its minimum or maximum value.

Its value must be between eVision.EasyDeepLearning.EDataAugmentation.SaltAndPepperNoiseMinimumDensity and 1.

Remarks:

This noise is computed after all the other noises.

◆ SaltAndPepperNoiseMinimumDensity() [1/2]

float eVision.EasyDeepLearning.EDataAugmentation.SaltAndPepperNoiseMinimumDensity ( self )

The minimum density of the salt and pepper noise.

The salt and pepper noise sets the value of a number of randomly selected (between eVision.EasyDeepLearning.EDataAugmentation.SaltAndPepperNoiseMinimumDensity and eVision.EasyDeepLearning.EDataAugmentation.SaltAndPepperNoiseMaximumDensity) pixels to its minimum or maximum value.

Its value must be between 0 and eVision.EasyDeepLearning.EDataAugmentation.SaltAndPepperNoiseMinimumDensity.

Remarks:

This noise is computed after all the other noises.

◆ SaltAndPepperNoiseMinimumDensity() [2/2]

eVision.EasyDeepLearning.EDataAugmentation.SaltAndPepperNoiseMinimumDensity	(		self,
		float	value )

The minimum density of the salt and pepper noise.

The salt and pepper noise sets the value of a number of randomly selected (between eVision.EasyDeepLearning.EDataAugmentation.SaltAndPepperNoiseMinimumDensity and eVision.EasyDeepLearning.EDataAugmentation.SaltAndPepperNoiseMaximumDensity) pixels to its minimum or maximum value.

Its value must be between 0 and eVision.EasyDeepLearning.EDataAugmentation.SaltAndPepperNoiseMinimumDensity.

Remarks:

This noise is computed after all the other noises.

◆ Save() [1/2]

None eVision.EasyDeepLearning.EDataAugmentation.Save	(		self,
		eVision.ESerializer \| None	serializer )

Saves an eVision.EasyDeepLearning.EDataAugmentation object.

The given ESerializer must have been created for writing.

Parameters

serializer The eVision.ESerializer object that is written to.

◆ Save() [2/2]

None eVision.EasyDeepLearning.EDataAugmentation.Save	(		self,
		str	path )

Saves an eVision.EasyDeepLearning.EDataAugmentation object.

The given ESerializer must have been created for writing.

Parameters

path	The file path.

◆ SpeckleNoiseMaximumStandardDeviation() [1/2]

float eVision.EasyDeepLearning.EDataAugmentation.SpeckleNoiseMaximumStandardDeviation ( self )

The speckle noise maximum standard deviation.

The speckle noise is a multiplicative noise sampled from a Gamma distribution of deviation between eVision.EasyDeepLearning.EDataAugmentation.SpeckleNoiseMinimumStandardDeviation and eVision.EasyDeepLearning.EDataAugmentation.SpeckleNoiseMaximumStandardDeviation.

Its value must be strictly higher than eVision.EasyDeepLearning.EDataAugmentation.SpeckleNoiseMinimumStandardDeviation.

Remarks:

This noise is computed before the salt and paper noise.

◆ SpeckleNoiseMaximumStandardDeviation() [2/2]

eVision.EasyDeepLearning.EDataAugmentation.SpeckleNoiseMaximumStandardDeviation	(		self,
		float	value )

The speckle noise maximum standard deviation.

The speckle noise is a multiplicative noise sampled from a Gamma distribution of deviation between eVision.EasyDeepLearning.EDataAugmentation.SpeckleNoiseMinimumStandardDeviation and eVision.EasyDeepLearning.EDataAugmentation.SpeckleNoiseMaximumStandardDeviation.

Its value must be strictly higher than eVision.EasyDeepLearning.EDataAugmentation.SpeckleNoiseMinimumStandardDeviation.

Remarks:

This noise is computed before the salt and paper noise.

◆ SpeckleNoiseMinimumStandardDeviation() [1/2]

float eVision.EasyDeepLearning.EDataAugmentation.SpeckleNoiseMinimumStandardDeviation ( self )

The speckle noise minimum standard deviation.

The speckle noise is a multiplicative noise sampled from a Gamma distribution of deviation between eVision.EasyDeepLearning.EDataAugmentation.SpeckleNoiseMinimumStandardDeviation and eVision.EasyDeepLearning.EDataAugmentation.SpeckleNoiseMaximumStandardDeviation.

Its value must be strictly positive and lower than eVision.EasyDeepLearning.EDataAugmentation.SpeckleNoiseMaximumStandardDeviation.

Remarks:

This noise is computed before the salt and paper noise.

◆ SpeckleNoiseMinimumStandardDeviation() [2/2]

eVision.EasyDeepLearning.EDataAugmentation.SpeckleNoiseMinimumStandardDeviation	(		self,
		float	value )

The speckle noise minimum standard deviation.

The speckle noise is a multiplicative noise sampled from a Gamma distribution of deviation between eVision.EasyDeepLearning.EDataAugmentation.SpeckleNoiseMinimumStandardDeviation and eVision.EasyDeepLearning.EDataAugmentation.SpeckleNoiseMaximumStandardDeviation.

Its value must be strictly positive and lower than eVision.EasyDeepLearning.EDataAugmentation.SpeckleNoiseMaximumStandardDeviation.

Remarks:

This noise is computed before the salt and paper noise.

◆ StainBlur() [1/2]

int eVision.EasyDeepLearning.EDataAugmentation.StainBlur ( self )

Stain blur.

The stain blur is the half kernel size of a Gaussian filter that is applied on the stain to smooth its edges and therefore the transition between the original image and the stain.

◆ StainBlur() [2/2]

eVision.EasyDeepLearning.EDataAugmentation.StainBlur	(		self,
		int	value )

Stain blur.

The stain blur is the half kernel size of a Gaussian filter that is applied on the stain to smooth its edges and therefore the transition between the original image and the stain.

◆ StainColorVariation() [1/2]

int eVision.EasyDeepLearning.EDataAugmentation.StainColorVariation ( self )

The variation of the color used to fill in the ellipse (We use an gaussian distribution of standard mean deviation stainColorVariation).

◆ StainColorVariation() [2/2]

eVision.EasyDeepLearning.EDataAugmentation.StainColorVariation	(		self,
		int	value )

The variation of the color used to fill in the ellipse (We use an gaussian distribution of standard mean deviation stainColorVariation).

◆ StainDisruptionMaxAnchorPoints() [1/2]

int eVision.EasyDeepLearning.EDataAugmentation.StainDisruptionMaxAnchorPoints ( self )

The maximum number of anchor points to use while disrupting the ellipse.

◆ StainDisruptionMaxAnchorPoints() [2/2]

eVision.EasyDeepLearning.EDataAugmentation.StainDisruptionMaxAnchorPoints	(		self,
		int	value )

The maximum number of anchor points to use while disrupting the ellipse.

◆ StainDisruptionMaxIntensity() [1/2]

float eVision.EasyDeepLearning.EDataAugmentation.StainDisruptionMaxIntensity ( self )

The maximum offset applied to the coordinates (x,y) of a pixel from the ellipse contour.

◆ StainDisruptionMaxIntensity() [2/2]

eVision.EasyDeepLearning.EDataAugmentation.StainDisruptionMaxIntensity	(		self,
		float	value )

The maximum offset applied to the coordinates (x,y) of a pixel from the ellipse contour.

◆ StainEllipseMaxRadius() [1/2]

int eVision.EasyDeepLearning.EDataAugmentation.StainEllipseMaxRadius ( self )

The maximum radius of the ellipse used to produce the stain.

Its value must be superior to eVision.EasyDeepLearning.EDataAugmentation.StainEllipseMinRadius.

◆ StainEllipseMaxRadius() [2/2]

eVision.EasyDeepLearning.EDataAugmentation.StainEllipseMaxRadius	(		self,
		int	value )

The maximum radius of the ellipse used to produce the stain.

Its value must be superior to eVision.EasyDeepLearning.EDataAugmentation.StainEllipseMinRadius.

◆ StainEllipseMinRadius() [1/2]

int eVision.EasyDeepLearning.EDataAugmentation.StainEllipseMinRadius ( self )

The minimum radius of the ellipse used to produce the stain.

Its value must be inferior to eVision.EasyDeepLearning.EDataAugmentation.StainEllipseMaxRadius.

◆ StainEllipseMinRadius() [2/2]

eVision.EasyDeepLearning.EDataAugmentation.StainEllipseMinRadius	(		self,
		int	value )

The minimum radius of the ellipse used to produce the stain.

Its value must be inferior to eVision.EasyDeepLearning.EDataAugmentation.StainEllipseMaxRadius.

◆ StainInterpolationSiteFactor() [1/2]

float eVision.EasyDeepLearning.EDataAugmentation.StainInterpolationSiteFactor ( self )

The factor to compute the number of interpolation sites to use from the number of anchor points.

◆ StainInterpolationSiteFactor() [2/2]

eVision.EasyDeepLearning.EDataAugmentation.StainInterpolationSiteFactor	(		self,
		float	value )

The factor to compute the number of interpolation sites to use from the number of anchor points.

◆ StainProbability() [1/2]

float eVision.EasyDeepLearning.EDataAugmentation.StainProbability ( self )

Probability of applying the stain data augmentation (default value: 0, i.e.

no stain).

◆ StainProbability() [2/2]

eVision.EasyDeepLearning.EDataAugmentation.StainProbability	(		self,
		float	value )

Probability of applying the stain data augmentation (default value: 0, i.e.

no stain).

Public Member Functions

Detailed Description

Constructor & Destructor Documentation

◆ __init__() [1/2]

◆ __init__() [2/2]

Member Function Documentation

◆ CopyTo()

◆ EnableHorizontalFlip() [1/2]

◆ EnableHorizontalFlip() [2/2]

◆ EnableVerticalFlip() [1/2]

◆ EnableVerticalFlip() [2/2]

◆ GaussianNoiseMaximumStandardDeviation() [1/2]

◆ GaussianNoiseMaximumStandardDeviation() [2/2]

◆ GaussianNoiseMinimumStandardDeviation() [1/2]

◆ GaussianNoiseMinimumStandardDeviation() [2/2]

◆ Generate() [1/8]

◆ Generate() [2/8]

◆ Generate() [3/8]

◆ Generate() [4/8]

◆ Generate() [5/8]

◆ Generate() [6/8]

◆ Generate() [7/8]

◆ Generate() [8/8]

◆ HasDataAugmentations()

◆ Load() [1/2]

◆ Load() [2/2]

◆ MaxBrightnessOffset() [1/2]

◆ MaxBrightnessOffset() [2/2]

◆ MaxContrastGain() [1/2]

◆ MaxContrastGain() [2/2]

◆ MaxGamma() [1/2]

◆ MaxGamma() [2/2]

◆ MaxHorizontalShear() [1/2]

◆ MaxHorizontalShear() [2/2]

◆ MaxHorizontalShift() [1/2]

◆ MaxHorizontalShift() [2/2]

◆ MaxHueOffset() [1/2]

◆ MaxHueOffset() [2/2]

◆ MaxRotationAngle() [1/2]

◆ MaxRotationAngle() [2/2]

◆ MaxSaturationGain() [1/2]

◆ MaxSaturationGain() [2/2]

◆ MaxScale() [1/2]

◆ MaxScale() [2/2]

◆ MaxStainColor() [1/2]

◆ MaxStainColor() [2/2]

◆ MaxVerticalShear() [1/2]

◆ MaxVerticalShear() [2/2]

◆ MaxVerticalShift() [1/2]

◆ MaxVerticalShift() [2/2]

◆ MinContrastGain() [1/2]

◆ MinContrastGain() [2/2]

◆ MinGamma() [1/2]

◆ MinGamma() [2/2]

◆ MinSaturationGain() [1/2]

◆ MinSaturationGain() [2/2]

◆ MinScale() [1/2]

◆ MinScale() [2/2]

◆ MinStainColor() [1/2]

◆ MinStainColor() [2/2]

◆ SaltAndPepperNoiseMaximumDensity() [1/2]

◆ SaltAndPepperNoiseMaximumDensity() [2/2]

◆ SaltAndPepperNoiseMinimumDensity() [1/2]

◆ SaltAndPepperNoiseMinimumDensity() [2/2]

◆ Save() [1/2]

◆ Save() [2/2]

◆ SpeckleNoiseMaximumStandardDeviation() [1/2]

◆ SpeckleNoiseMaximumStandardDeviation() [2/2]

◆ SpeckleNoiseMinimumStandardDeviation() [1/2]

◆ SpeckleNoiseMinimumStandardDeviation() [2/2]

◆ StainBlur() [1/2]

◆ StainBlur() [2/2]

◆ StainColorVariation() [1/2]

◆ StainColorVariation() [2/2]

◆ StainDisruptionMaxAnchorPoints() [1/2]

◆ StainDisruptionMaxAnchorPoints() [2/2]

◆ StainDisruptionMaxIntensity() [1/2]

◆ StainDisruptionMaxIntensity() [2/2]

◆ StainEllipseMaxRadius() [1/2]

◆ StainEllipseMaxRadius() [2/2]

◆ init() [1/2]

◆ init() [2/2]