Sobel3D

MLModule

genre	Kernel
author	MeVis Medical Solutions AG
package	MeVisLab/Standard
dll	MLKernel
definition	MLKernel.def
see also	Convolution, ExtendedConvolution, KernelEditor
keywords	edge, detect, kernel, filter, interval, gradient, ExtendedConvolution, Convolution, Roberts

Purpose

The module Sobel3D computes gradients of a gray-value image by applying a Sobel kernel for each main axis direction.

The module can also compute the gradients strengths and a gradient vector image.

Windows

Default Panel

Input Fields

input0

name: input0, type: Image

Output Fields

output0

name: output0, type: Image

Depending on the mode (Gradient Strength, Edge Detection or Estimation) the output is a grey-scale image, a binary edge-map or a RGB-image, where the rgb-values represent xyz-vector components.

Parameter Fields

Field Index

autoCalcMinMax: Bool	Min. Threshold: Float	useMinMax: Bool
Border Handling: Enum	Non Edge Value: Float
Edge Value: Float	outputMax: Double
Fill Value: Double	outputMin: Double
Filter Mode: Enum	referenceExtentMode: Enum
Max: Double	Scale with voxel size: Bool
Max. Threshold: Float	setAutoMinMax: Trigger
Min: Double	Use: Bool

Visible Fields

Border Handling

name: borderHandling, type: Enum, default: PadSrcClamp, deprecated name: edgeMode

Defines the border handling mode.

See Border Handling in Kernel Operations for details.

Values:

Title	Name
​No Pad	​NoPad
​Pad Src Fill	​PadSrcFill
​Pad Dst Fill	​PadDstFill
​Pad Dst Fill With Orig	​PadDstFillWithOrig
​Pad Src Undefined	​PadSrcUndefined
​Pad Dst Undefined	​PadDstUndefined
​Pad Src Clamp	​PadSrcClamp

Fill Value

name: fillValue, type: Double, default: 0

Sets the fill value for certain Border Handling modes.

Min

name: intervalMinOfFilteredVoxels, type: Double, default: 30, deprecated name: imageIntervalMin

Sets the minimum value threshold for outputting a subset of voxels.

Max

name: intervalMaxOfFilteredVoxels, type: Double, default: 50, deprecated name: imageIntervalMax

Sets the maximum value threshold for outputting a subset of voxels.

Use

name: useIntervalOfFilteredVoxels, type: Bool, default: FALSE

If checked, the module outputs only on a subset of voxels defined by a value range.

Min. Threshold

name: edgeThresholdMinimum, type: Float, default: 31.25, minimum: 0, maximum: 150000, deprecated name: edgeThresholdMinimumFld

Sets the minimum value of one kernel convolution result for thresholding in Edge Detection or in Estimation mode.

Max. Threshold

name: edgeThresholdMaximum, type: Float, default: 46.875, minimum: 0, maximum: 150000, deprecated name: edgeThresholdMaximumFld

Sets the maximum value of one kernel convolution result for thresholding in Edge Detection or in Estimation mode.

Edge Value

name: detectedEdgeValue, type: Float, default: 32, minimum: 0, maximum: 4096, deprecated name: detectedEdgeValueFld

If in Edge Detection mode, this value will be set into the output image if the convolution value lies within the threshold range.

Non Edge Value

name: detectedNonEdgeValue, type: Float, default: 0, minimum: 0, maximum: 4096, deprecated name: detectedNonEdgeValueFld

If in Edge Detection mode, this value will be set into the output image if the convolution value lies outside of the threshold value.

Scale with voxel size

name: useVoxelSize, type: Bool, default: TRUE

If checked and in Gradient Strength mode, each resulting convolution value is divided by the voxel size in the particular (x, y, and z) direction.

Filter Mode

name: filterMode, type: Enum, default: Gradient Strength, deprecated name: FilterMode

Defines the filter mode of this module.

Values:

Title	Name	Description
​Gradient Strength	​Gradient Strength	​The resulting sum of the kernel convolutions is written into the output image, which is a gray-value image representing the strengths of the gradients in each voxel.
​Edge Detection	​Edge Detection	​Thresholds the convolution values in each direction: if any of these values is within the threshold range, an Edge Value is written into the output image; otherwise, a Non Edge Value is written, resulting in a binary edge image.
​Estimation	​Estimation	​The values of the three convolutions are used as vector components, where the x, y, z values are encoded into r, g, b values of the output image, resulting in a RGB-image.
​Gradient Estimation And Strength Packed	​GradientEstimationAndStrengthPacked	​Estimates the gradient and its strength. The gradient and strength are normalized and packed to the range 0-1 using: rgb = (grad / gradStrength + 1.) * 0.5 a = gradStrength / maxGradStrength This is the format that makes it usable for the SoGVRVolumeRenderer and SoPathTracer, since they will convert this input gradient image to RGBA8 and have the same packing convention.

Hidden Fields

referenceExtentMode

name: referenceExtentMode, type: Enum, default: Overlap

Values:

Title	Name
​Overlap	​Overlap
​Input0 Ext Without Fill	​Input0_ExtWithoutFill
​Input0 Ext With Fill	​Input0_ExtWithFill
​Max Extents Without Fill	​MaxExtentsWithoutFill
​Max Extents With Fill	​MaxExtentsWithFill

autoCalcMinMax

name: autoCalcMinMax, type: Bool, default: TRUE

setAutoMinMax

name: setAutoMinMax, type: Trigger

outputMin

name: outputMin, type: Double, default: 1

outputMax

name: outputMax, type: Double, default: 0

useMinMax

name: useMinMax, type: Bool, default: FALSE