Local Transforms

Functions
Profile	oevislib_net.Profile.Convolve (Array< float > inKernel, bool inNormalizeKernel, Optional< int > inKernelAnchor, BorderPadding inBorderPadding)
	Applies a convolution to the profile, using the given mask.
Profile	oevislib_net.Profile.Erode (bool inCyclic, int inKernelRadius, Optional< Range > inRange=null)
	Erodes the profile.
Profile	oevislib_net.Profile.Differentiate (bool inCyclic, DifferentiationMethod inDifferentiationMethod)
	Differentiates the profile (computes its derivative).
Profile	oevislib_net.Profile.Differentiate (bool inCyclic, int inStep)
	Differentiate the profile, using a specific difference step.
Profile	oevislib_net.Profile.Dilate (bool inCyclic, int inKernelRadius, Optional< Range > inRange=null)
	Dilates the profile.
Profile	oevislib_net.Profile.GaussianSmooth (float inStdDev, float inKernelRelativeSize, BorderPadding inBorderPadding)
	Smooths a profile by Gaussian-averaging points inside the specified kernel.
Profile	oevislib_net.Profile.Smooth (int inKernelRadius, BorderPadding inBorderPadding)
	Smooths a profile by averaging points inside the specified kernel.

Detailed Description

Function Documentation

Convolve()

Profile oevislib_net.Profile.Convolve ( Array< float > inKernel, bool inNormalizeKernel, Optional< int > inKernelAnchor, BorderPadding inBorderPadding )

inline

Applies a convolution to the profile, using the given mask.

Parameters

inKernel	Convolution kernel.
inNormalizeKernel	If true, the kernel values are normalized in the range [-1, 1].
inKernelAnchor	Index of the kernel element that should be considered as the anchor point. Range [0, n-1] where n is the number of elements of the kernel.
inBorderPadding	Padding type added to both profile borders.

Returns

Output profile.

Differentiate() [1/2]

Profile oevislib_net.Profile.Differentiate ( bool inCyclic, DifferentiationMethod inDifferentiationMethod )

inline

Differentiates the profile (computes its derivative).

Parameters

inCyclic	Whether the profile is cyclic.
inDifferentiationMethod	Differentiation method.

Returns

Derivative.

Differentiate() [2/2]

Profile oevislib_net.Profile.Differentiate ( bool inCyclic, int inStep )

inline

Differentiate the profile, using a specific difference step.

Parameters

inCyclic	Whether the profile is cyclic.
inStep	Difference step. Range: [1, +inf).

Returns

Derivative.

Dilate()

Profile oevislib_net.Profile.Dilate ( bool inCyclic, int inKernelRadius, Optional< Range > inRange = null )

inline

Dilates the profile.

Replaces every value in the range with the maximum value inside the specified kernel.

Parameters

inCyclic	Whether the profile is cyclic.
inKernelRadius	Kernel's radius. Range: [0, +inf).
inRange	Range of indices to consider.

Returns

Dilated profile.

Erode()

Profile oevislib_net.Profile.Erode ( bool inCyclic, int inKernelRadius, Optional< Range > inRange = null )

inline

Erodes the profile.

Replaces every value in the range with the minimum value inside the specified kernel.

Parameters

inCyclic	Whether the profile is cyclic.
inKernelRadius	Kernel's radius. Range: [0, +inf).
inRange	Range of indices to consider.

Returns

Eroded profile.

GaussianSmooth()

Profile oevislib_net.Profile.GaussianSmooth ( float inStdDev, float inKernelRelativeSize, BorderPadding inBorderPadding )

inline

Smooths a profile by Gaussian-averaging points inside the specified kernel.

If either the profile or the kernel have size zero, the profile is returned unmodified.

Parameters

inStdDev	Standard deviation of the Gaussian kernel. Range: [0, +inf).
inKernelRelativeSize	Multiplier of the standard deviation that determines the size of the kernel. Range: [0, +inf).
inBorderPadding	Padding type added to both profile borders.

Returns

Smoothed profile.

Smooth()

Profile oevislib_net.Profile.Smooth ( int inKernelRadius, BorderPadding inBorderPadding )

inline

Smooths a profile by averaging points inside the specified kernel.

If either the profile or the kernel have size zero, the profile is returned unmodified.

Parameters

inKernelRadius	Kernel's radius. Kernel size is computed as (2 * inKernelRadius + 1). Range: [0, +inf).
inBorderPadding	Padding type added to both profile borders.

Returns

Smoothed profile.