NonlinEnhance

MLModule

genre	Filters
status	stable
author	Regina Ochotzki
package	MeVisLab/Standard
dll	MLWaveBox
definition	MLWaveBox.def
keywords	wavebox, fourier, transform, mallat, enhance

Purpose

The module NonlinEnhance enhances the contrast of images with wavelet transformations.

Windows

Default Panel

Input Fields

input0

name: input0, type: Image

Output Fields

output0

name: output0, type: Image

Parameter Fields

Field Index

Adjust min and max values: Bool	Min. Value: Double
Apply: Trigger	Q:: Float
Auto Apply: Bool	T:: Float
deepestLevel: Integer	TN: Float
Function Select: Enum	Transform Level: Integer
Get/Set Current Values: Trigger	Wavelet Select: Enum
K:: Float
Max Value: Double

Visible Fields

Adjust min and max values

name: minMaxAdjust, type: Bool, default: TRUE

If checked, the min and max value can be adujsted.

Min. Value

name: minValue, type: Double, default: 0

Shows the min value.

Max Value

name: maxValue, type: Double, default: 0

Shows the max value.

Get/Set Current Values

name: setCurrentValues, type: Trigger

If pressed, the min and the max value are set anew.

Wavelet Select

name: waveletSelect, type: Enum, default: Mallat_p0n1_Wavelet

Defines the wavelet function.

Values:

Title	Name
p0n1 Wavelet	Mallat_p0n1_Wavelet
p1n0 Wavelet	Mallat_p1n0_Wavelet

Function Select

name: functionSelect, type: Enum, default: E(x)

Defines the other function.

Values:

Title	Name	Description
E(x)	E(x)	In the transformed image, the low contrast values between -T and T are spread by factor K while the high contrast values are shifted by (+/-) K*T to guarantee a continuous mapping.
EN(x)	EN(x)	To enhance robustness, in the transformed image, the lowest values between -TN and TN are set to zero and the values between TN and T are spread by factor K (negative values respectively). The high contrast values are shifted accordingly to guarantee for a continuous mapping
E2(x)	E2(x)	Nonlinear amplification of low contrast values. In the transformed image, the low contrast values between -T and T are amplified by factor (1+K*(1-(x/T))^Q), giving a more smooth transition to the high contrast values.
sigma(x)	sigma(x)	In the transformed image, the low contrast values between -T and T are spread by factor K while the high contrast values are just set to (+/-) T respectively.

Apply

name: apply, type: Trigger

If pressed, the module computes anew.

Auto Apply

name: autoApply, type: Bool, default: TRUE

If checked, the module comutes anew on each field or input change.

Transform Level

name: level, type: Integer, default: 1, minimum: 1, maximum: :field:`deepestLevel`

Sets the depth of the transform.

T:

name: T, type: Float, default: 0.5, minimum: 0

Sets the value of the function parameter T.

TN

name: TN, type: Float, default: 0.5, minimum: 0

Sets the value of the function parameter TN.

K:

name: K, type: Float, default: 20, minimum: 0

Sets the value of the function parameter K.

Q:

name: Q, type: Float, default: 2, minimum: 0

Sets the value of the function parameter Q.

Hidden Fields

deepestLevel

name: deepestLevel, type: Integer, default: 1