Diffusion¶
-
MLModule
¶ genre Diffusion
author MeVis Medical Solutions AG
package MeVisLab/Standard
dll MLDiffusion1
definition MLDiffusion1.def keywords nonlinear
,filter
,perona-malik
,weickert
,aos
Purpose¶
The module Diffusion
implements the isotropic diffusion process in 2D using a semi-implicit AOS scheme.
For a 3D image, the diffusion works slice by slice.
Parameter Fields¶
Field Index¶
Apply : Trigger |
Diffusion Function : Enum |
Edge Parameter : Float |
Number of Iterations : Integer |
Smoothing : Float |
Status : String |
Time Step Size : Float |
Update Mode : Enum |
Visible Fields¶
Update Mode¶
-
name:
updateMode
, type:
Enum
, default:
AutoUpdate
¶ Defines how the module should react on (input) field changes.
Values:
Title | Name | Description |
---|---|---|
Off | Off | The module does nothing on field changes. |
Auto Update | AutoUpdate | The module computes a new output image on any field change. |
Time Step Size¶
-
name:
timeStepSize
, type:
Float
, default:
5
, minimum:
0
, maximum:
200
¶ Sets the time step size for approximation.
The smaller the step, the better the result.
Number of Iterations¶
-
name:
numOfSteps
, type:
Integer
, default:
10
, minimum:
0
, maximum:
200
¶ Sets the number of iteration for the diffusion.
The higher the value, the longer the diffusion time.
Edge Parameter¶
-
name:
edgeParameter
, type:
Float
, default:
3
, minimum:
1e-05
, maximum:
200
¶ Sets a value which is used to detect edges in the diffusivity function.