SeparableKernelExample¶

MLModule¶

genre	`Kernel`
author	`MeVis Medical Solutions AG`
package	`MeVisLab/Examples`
dll	`MLKernelExamples`
definition	MLKernelExamples.def
see also	`Convolution`, `ExtendedConvolution`, `Rank`, `Sigma`, `BitMorphology`, `Morphology`, `LoG`, `KernelEditor`, `StandardDeviation`
keywords	`enhance`, `kernel`, `filter`

Purpose¶

This module is an example module to show how to implement a separable convolution filter.

It contains some options and parameters which are available from the KernelBaseOp and KernelOp classes.

Windows¶

Default Panel¶

../../../Modules/ML/MLKernelExamples/mhelp/Images/Screenshots/SeparableKernelExample._default.png

Input Fields¶

input0¶

name: input0, type: Image¶

Output Fields¶

output0¶

name: output0, type: Image¶

Parameter Fields¶

Field Index¶

`Border Handling`: `Enum`	`X`: `Integer`
`C`: `Integer`	`Y`: `Integer`
`Fill Value`: `Double`	`Z`: `Integer`
`Info`: `String`
`Kernel Output`: `String`
`referenceExtentMode`: `Enum`
`T`: `Integer`
`U`: `Integer`

Visible Fields¶

Border Handling¶

name: borderHandling, type: Enum, default: PadSrcClamp¶: Defines the border handling of the kernel operation.

Values:

Title	Name	Description
No Pad	NoPad	Only those voxels are passed to the output which can be filtered correctly by the entire kernel. So the output image usually shrinks by the extents of the kernel - 1.
Pad Src Fill	PadSrcFill	The input image is virtually expanded by so many voxels that all input image voxels can be filtered correctly with the kernel. The area added around the input image is filled with the value specified in the `Fill Value` field.
Pad Dst Fill	PadDstFill	All pixels of the output image which can be filtered correctly with the kernel without accessing voxels outside the output image are filtered with the kernel. All other voxels in the image are filled with the value specified in the `Fill Value` field.
Pad Dst Fill With Orig	PadDstFillWithOrig	All pixels of the output image which can be filtered correctly with the kernel without accessing voxels outside the output image are filtered with the kernel. All other voxels in the image are filled with their corresponding values from the input image.
Pad Src Undefined	PadSrcUndefined	The input image is virtually expanded by so many voxels that all input image voxels can be filtered correctly with the kernel. The area added around the input image is not filled with a certain value and remains undefined. Note that this mode saves time for border filling but causes unpredictable values in the border regions of the output image.
Pad Dst Undefined	PadDstUndefined	All pixels of the output image which can be filtered correctly with the kernel without accessing voxels outside the output image are filtered with the kernel. All other voxels in the image are left undefined. Note that this mode saves time for border filling but causes unpredictable values in the border regions of the output image.
Pad Src Clamp	PadSrcClamp	The input image is virtually expanded by so many voxels that all input image voxels can be filtered correctly with the kernel. The contents of the area added around the input image is filled with those voxels of the input image which are nearest to the filled ones.

Fill Value¶

name: fillValue, type: Double, default: 0¶: Sets a fill value that is used for filling the border in certain modes of Border Handling.

Info¶

name: kernelInfo, type: String, persistent: no¶: Shows information about the module.

Kernel Output¶

name: kernelOutput, type: String, persistent: no¶: Shows the used kernel as a string.

X¶

name: kernelX, type: Integer, default: 3, minimum: 1¶: Sets the size of the kernel in x-direction.

Y¶

name: kernelY, type: Integer, default: 3, minimum: 1¶: Sets the size of the kernel in y-direction.

Z¶

name: kernelZ, type: Integer, default: 1, minimum: 1¶: Sets the size of the kernel in z-direction.

C¶

name: kernelC, type: Integer, default: 1, minimum: 1¶: Sets the size of the kernel in c-direction.

T¶

name: kernelT, type: Integer, default: 1, minimum: 1¶: Sets the size of the kernel in t-direction.

U¶

name: kernelU, type: Integer, default: 1, minimum: 1¶: Sets the size of the kernel in u-direction.

Hidden Fields¶

referenceExtentMode¶

name: referenceExtentMode, type: Enum, default: Overlap¶