EDataAugmentation Class
An 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 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.
Namespace: Euresys::Open_eVision::EasyDeepLearning
Methods
Copies itself to the EDataAugmentation object other .
Creates an EDataAugmentation object.
Generates a new image from a generic EBaseROI image. The caller is responsible for calling freeing the memory of the returned image by calling delete on the image.
Sets whether to use horizontally flipped versions of input images.
Sets whether to use vertically flipped versions of input images.
The Gaussian noise maximum standard deviation.
The Gaussian noise is an additive noise sampled from a Gaussian distribution of deviation between EDataAugmentation and EDataAugmentation.(also called the normal distribution).
Its value must be superior or equal toEDataAugmentation .
The Gaussian noise is an additive noise sampled from a Gaussian distribution of deviation between EDataAugmentation and EDataAugmentation.(also called the normal distribution).
Its value must be superior or equal to
The Gaussian noise minimum standard deviation.
The Gaussian noise is an additive noise sampled from a Gaussian distribution of deviation between EDataAugmentation and EDataAugmentation.(also called the normal distribution).
Its value must be betwteen0 and EDataAugmentation .
The Gaussian noise is an additive noise sampled from a Gaussian distribution of deviation between EDataAugmentation and EDataAugmentation.(also called the normal distribution).
Its value must be betwteen
Sets the maximum absolute brightness offset. It must be between 0 and 1.
Sets the maximum contrast gain. Its value must be strictly positive and over EDataAugmentation::MinContrastGain.
Sets the maximum gamma for gamma correction. Its value must be higher than EDataAugmentation::MinGamma.
Sets the maximum absolute horizontal shear, represented as an angle from the vertical direction. Its value must be between 0 and 90 degrees.
Sets the maximum horizontal shift allowed.
Sets the maximum absolute hue offset. Its value must be between 0 and 180 degrees.
Set the maximum rotation angle allowed. Its value must be between 0 and 180 degrees.
Sets the maximum saturation gain. Its value must be over or equal to EDataAugmentation::MinSaturationGain.
Sets the maximum scaling allowed. Its value must be strictly positive and over EDataAugmentation::MinScale.
The maximum color value from which to draw a color to fill in the stain (we use a gaussian distribution of with a mean betweeen EDataAugmentation::MinStainColor and EDataAugmentation::MaxStainColor)
Its value must be bigger than EDataAugmentation::MinStainColor.
Its value must be bigger than EDataAugmentation::MinStainColor.
Sets the maximum absolute vertical shear, represented as an angle from the horizontal direction. Its value must be between 0 and 90 degrees.
Sets the maximum vertical shift allowed.
Sets the minimum contrast gain. Its value must be strictly positive and below EDataAugmentation::MaxContrastGain.
Sets the minimum gamma for gamma correction. Its value must be strictly positive and below EDataAugmentation::MaxGamma.
Sets the minimum saturation gain. Its value must be strictly positive.
Sets the minimum scaling allowed. Its value must be strictly positive and below EDataAugmentation::MaxScale.
The minimum color value from which to draw a color to fill in the stain (we use a gaussian distribution of with a mean betweeen EDataAugmentation::MinStainColor and EDataAugmentation::MaxStainColor)
Its value must be inferior to255 and EDataAugmentation::MaxStainColor.
Its value must be inferior to
The maximum density of the salt and pepper noise.
The salt and pepper noise sets the value of a number of randomly selected (between EDataAugmentation and EDataAugmentation) pixels to its minimum or maximum value.
Its value must be betwteenEDataAugmentation and 1 .
The salt and pepper noise sets the value of a number of randomly selected (between EDataAugmentation and EDataAugmentation) pixels to its minimum or maximum value.
Its value must be betwteen
The minimum density of the salt and pepper noise.
The salt and pepper noise sets the value of a number of randomly selected (between EDataAugmentation and EDataAugmentation) pixels to its minimum or maximum value.
Its value must be betwteen0 and EDataAugmentation .
The salt and pepper noise sets the value of a number of randomly selected (between EDataAugmentation and EDataAugmentation) pixels to its minimum or maximum value.
Its value must be betwteen
The speckle noise maximum standard deviation.
The speckle noise is a multiplicative noise sampled from a Gamma distribution of deviation between EDataAugmentation and EDataAugmentation.
Its value must be strictly higher than EDataAugmentation.
The speckle noise is a multiplicative noise sampled from a Gamma distribution of deviation between EDataAugmentation and EDataAugmentation.
Its value must be strictly higher than EDataAugmentation.
The speckle noise minimum standard deviation.
The speckle noise is a multiplicative noise sampled from a Gamma distribution of deviation between EDataAugmentation and EDataAugmentation.
Its value must be strictly positive and lower than EDataAugmentation.
The speckle noise is a multiplicative noise sampled from a Gamma distribution of deviation between EDataAugmentation and EDataAugmentation.
Its value must be strictly positive and lower than EDataAugmentation.
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.
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.
The variation of the color used to fill in the ellipse (We use an gaussian distribution of standart mean deviation stainColorVariation).
The maximum number of anchor points to use while disrupting the ellipse.
The maximun offset applied to the coordinates (x,y) of a pixel from the ellipse contour.
The maximum radius of the ellipse used to produce the stain.
Its value must be superior toEDataAugmentation::StainEllipseMinRadius .
Its value must be superior to
The minimum radius of the ellipse used to produce the stain.
Its value must be inferior toEDataAugmentation::StainEllipseMaxRadius .
Its value must be inferior to
The factor to compute the number of interpolation sites to use from the number of anchor points.
Probability of applying the stain data augmentation (default value: 0, i.e. no stain).
Whether the EDataAugmentation object contains augmentations.
Loads an EDataAugmentation object. The given ESerializer must have been created for reading.
Copies the EDataAugmentation object other to this object.
Equality operator.
Saves an EDataAugmentation object. The given ESerializer must have been created for writing.
Sets whether to use horizontally flipped versions of input images.
Sets whether to use vertically flipped versions of input images.
The Gaussian noise maximum standard deviation.
The Gaussian noise is an additive noise sampled from a Gaussian distribution of deviation between EDataAugmentation and EDataAugmentation.(also called the normal distribution).
Its value must be superior or equal toEDataAugmentation .
The Gaussian noise is an additive noise sampled from a Gaussian distribution of deviation between EDataAugmentation and EDataAugmentation.(also called the normal distribution).
Its value must be superior or equal to
The Gaussian noise minimum standard deviation.
The Gaussian noise is an additive noise sampled from a Gaussian distribution of deviation between EDataAugmentation and EDataAugmentation.(also called the normal distribution).
Its value must be betwteen0 and EDataAugmentation .
The Gaussian noise is an additive noise sampled from a Gaussian distribution of deviation between EDataAugmentation and EDataAugmentation.(also called the normal distribution).
Its value must be betwteen
Sets the maximum absolute brightness offset. It must be between 0 and 1.
Sets the maximum contrast gain. Its value must be strictly positive and over EDataAugmentation::MinContrastGain.
Sets the maximum gamma for gamma correction. Its value must be higher than EDataAugmentation::MinGamma.
Sets the maximum absolute horizontal shear, represented as an angle from the vertical direction. Its value must be between 0 and 90 degrees.
Sets the maximum horizontal shift allowed.
Sets the maximum absolute hue offset. Its value must be between 0 and 180 degrees.
Set the maximum rotation angle allowed. Its value must be between 0 and 180 degrees.
Sets the maximum saturation gain. Its value must be over or equal to EDataAugmentation::MinSaturationGain.
Sets the maximum scaling allowed. Its value must be strictly positive and over EDataAugmentation::MinScale.
The maximum color value from which to draw a color to fill in the stain (we use a gaussian distribution of with a mean betweeen EDataAugmentation::MinStainColor and EDataAugmentation::MaxStainColor)
Its value must be bigger than EDataAugmentation::MinStainColor.
Its value must be bigger than EDataAugmentation::MinStainColor.
Sets the maximum absolute vertical shear, represented as an angle from the horizontal direction. Its value must be between 0 and 90 degrees.
Sets the maximum vertical shift allowed.
Sets the minimum contrast gain. Its value must be strictly positive and below EDataAugmentation::MaxContrastGain.
Sets the minimum gamma for gamma correction. Its value must be strictly positive and below EDataAugmentation::MaxGamma.
Sets the minimum saturation gain. Its value must be strictly positive.
Sets the minimum scaling allowed. Its value must be strictly positive and below EDataAugmentation::MaxScale.
The minimum color value from which to draw a color to fill in the stain (we use a gaussian distribution of with a mean betweeen EDataAugmentation::MinStainColor and EDataAugmentation::MaxStainColor)
Its value must be inferior to255 and EDataAugmentation::MaxStainColor.
Its value must be inferior to
The maximum density of the salt and pepper noise.
The salt and pepper noise sets the value of a number of randomly selected (between EDataAugmentation and EDataAugmentation) pixels to its minimum or maximum value.
Its value must be betwteenEDataAugmentation and 1 .
The salt and pepper noise sets the value of a number of randomly selected (between EDataAugmentation and EDataAugmentation) pixels to its minimum or maximum value.
Its value must be betwteen
The minimum density of the salt and pepper noise.
The salt and pepper noise sets the value of a number of randomly selected (between EDataAugmentation and EDataAugmentation) pixels to its minimum or maximum value.
Its value must be betwteen0 and EDataAugmentation .
The salt and pepper noise sets the value of a number of randomly selected (between EDataAugmentation and EDataAugmentation) pixels to its minimum or maximum value.
Its value must be betwteen
The speckle noise maximum standard deviation.
The speckle noise is a multiplicative noise sampled from a Gamma distribution of deviation between EDataAugmentation and EDataAugmentation.
Its value must be strictly higher than EDataAugmentation.
The speckle noise is a multiplicative noise sampled from a Gamma distribution of deviation between EDataAugmentation and EDataAugmentation.
Its value must be strictly higher than EDataAugmentation.
The speckle noise minimum standard deviation.
The speckle noise is a multiplicative noise sampled from a Gamma distribution of deviation between EDataAugmentation and EDataAugmentation.
Its value must be strictly positive and lower than EDataAugmentation.
The speckle noise is a multiplicative noise sampled from a Gamma distribution of deviation between EDataAugmentation and EDataAugmentation.
Its value must be strictly positive and lower than EDataAugmentation.
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.
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.
The variation of the color used to fill in the ellipse (We use an gaussian distribution of standart mean deviation stainColorVariation).
The maximum number of anchor points to use while disrupting the ellipse.
The maximun offset applied to the coordinates (x,y) of a pixel from the ellipse contour.
The maximum radius of the ellipse used to produce the stain.
Its value must be superior toEDataAugmentation::StainEllipseMinRadius .
Its value must be superior to
The minimum radius of the ellipse used to produce the stain.
Its value must be inferior toEDataAugmentation::StainEllipseMaxRadius .
Its value must be inferior to
The factor to compute the number of interpolation sites to use from the number of anchor points.
Probability of applying the stain data augmentation (default value: 0, i.e. no stain).
EDataAugmentation Class
An 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 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.
Namespace: Euresys.Open_eVision.EasyDeepLearning
Properties
Sets whether to use horizontally flipped versions of input images.
Sets whether to use vertically flipped versions of input images.
The Gaussian noise maximum standard deviation.
The Gaussian noise is an additive noise sampled from a Gaussian distribution of deviation between EDataAugmentation and EDataAugmentation.(also called the normal distribution).
Its value must be superior or equal toEDataAugmentation .
The Gaussian noise is an additive noise sampled from a Gaussian distribution of deviation between EDataAugmentation and EDataAugmentation.(also called the normal distribution).
Its value must be superior or equal to
The Gaussian noise minimum standard deviation.
The Gaussian noise is an additive noise sampled from a Gaussian distribution of deviation between EDataAugmentation and EDataAugmentation.(also called the normal distribution).
Its value must be betwteen0 and EDataAugmentation .
The Gaussian noise is an additive noise sampled from a Gaussian distribution of deviation between EDataAugmentation and EDataAugmentation.(also called the normal distribution).
Its value must be betwteen
Sets the maximum absolute brightness offset. It must be between 0 and 1.
Sets the maximum contrast gain. Its value must be strictly positive and over EDataAugmentation::MinContrastGain.
Sets the maximum gamma for gamma correction. Its value must be higher than EDataAugmentation::MinGamma.
Sets the maximum absolute horizontal shear, represented as an angle from the vertical direction. Its value must be between 0 and 90 degrees.
Sets the maximum horizontal shift allowed.
Sets the maximum absolute hue offset. Its value must be between 0 and 180 degrees.
Set the maximum rotation angle allowed. Its value must be between 0 and 180 degrees.
Sets the maximum saturation gain. Its value must be over or equal to EDataAugmentation::MinSaturationGain.
Sets the maximum scaling allowed. Its value must be strictly positive and over EDataAugmentation::MinScale.
The maximum color value from which to draw a color to fill in the stain (we use a gaussian distribution of with a mean betweeen EDataAugmentation::MinStainColor and EDataAugmentation::MaxStainColor)
Its value must be bigger than EDataAugmentation::MinStainColor.
Its value must be bigger than EDataAugmentation::MinStainColor.
Sets the maximum absolute vertical shear, represented as an angle from the horizontal direction. Its value must be between 0 and 90 degrees.
Sets the maximum vertical shift allowed.
Sets the minimum contrast gain. Its value must be strictly positive and below EDataAugmentation::MaxContrastGain.
Sets the minimum gamma for gamma correction. Its value must be strictly positive and below EDataAugmentation::MaxGamma.
Sets the minimum saturation gain. Its value must be strictly positive.
Sets the minimum scaling allowed. Its value must be strictly positive and below EDataAugmentation::MaxScale.
The minimum color value from which to draw a color to fill in the stain (we use a gaussian distribution of with a mean betweeen EDataAugmentation::MinStainColor and EDataAugmentation::MaxStainColor)
Its value must be inferior to255 and EDataAugmentation::MaxStainColor.
Its value must be inferior to
The maximum density of the salt and pepper noise.
The salt and pepper noise sets the value of a number of randomly selected (between EDataAugmentation and EDataAugmentation) pixels to its minimum or maximum value.
Its value must be betwteenEDataAugmentation and 1 .
The salt and pepper noise sets the value of a number of randomly selected (between EDataAugmentation and EDataAugmentation) pixels to its minimum or maximum value.
Its value must be betwteen
The minimum density of the salt and pepper noise.
The salt and pepper noise sets the value of a number of randomly selected (between EDataAugmentation and EDataAugmentation) pixels to its minimum or maximum value.
Its value must be betwteen0 and EDataAugmentation .
The salt and pepper noise sets the value of a number of randomly selected (between EDataAugmentation and EDataAugmentation) pixels to its minimum or maximum value.
Its value must be betwteen
The speckle noise maximum standard deviation.
The speckle noise is a multiplicative noise sampled from a Gamma distribution of deviation between EDataAugmentation and EDataAugmentation.
Its value must be strictly higher than EDataAugmentation.
The speckle noise is a multiplicative noise sampled from a Gamma distribution of deviation between EDataAugmentation and EDataAugmentation.
Its value must be strictly higher than EDataAugmentation.
The speckle noise minimum standard deviation.
The speckle noise is a multiplicative noise sampled from a Gamma distribution of deviation between EDataAugmentation and EDataAugmentation.
Its value must be strictly positive and lower than EDataAugmentation.
The speckle noise is a multiplicative noise sampled from a Gamma distribution of deviation between EDataAugmentation and EDataAugmentation.
Its value must be strictly positive and lower than EDataAugmentation.
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.
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.
The variation of the color used to fill in the ellipse (We use an gaussian distribution of standart mean deviation stainColorVariation).
The maximum number of anchor points to use while disrupting the ellipse.
The maximun offset applied to the coordinates (x,y) of a pixel from the ellipse contour.
The maximum radius of the ellipse used to produce the stain.
Its value must be superior toEDataAugmentation::StainEllipseMinRadius .
Its value must be superior to
The minimum radius of the ellipse used to produce the stain.
Its value must be inferior toEDataAugmentation::StainEllipseMaxRadius .
Its value must be inferior to
The factor to compute the number of interpolation sites to use from the number of anchor points.
Probability of applying the stain data augmentation (default value: 0, i.e. no stain).
Methods
Copies itself to the EDataAugmentation object other .
Creates an EDataAugmentation object.
Generates a new image from a generic EBaseROI image. The caller is responsible for calling freeing the memory of the returned image by calling delete on the image.
Whether the EDataAugmentation object contains augmentations.
Loads an EDataAugmentation object. The given ESerializer must have been created for reading.
Copies the EDataAugmentation object other to this object.
Equality operator.
Saves an EDataAugmentation object. The given ESerializer must have been created for writing.