F_VN_SeparableCustomFilter

Syntax

Definition:

FUNCTION F_VN_SeparableCustomFilter : HRESULT
VAR_INPUT
    ipSrcImage  : ITcVnImage;
    ipDestImage : Reference To  ITcVnImage;
    eDestDepth  : ETcVnElementType;
END_VAR
VAR_IN_OUT
    stKernelX   : TcVnMatrix;
    stKernelY   : TcVnMatrix;
END_VAR
VAR_INPUT
    hrPrev      : HRESULT;
END_VAR

Inputs

Name	Type	Description
ipSrcImage	ITcVnImage	Source image
ipDestImage	Reference To ITcVnImage	Destination image (An appropriate destination image will be created if required.)
eDestDepth	ETcVnElementType	Destination image depth
hrPrev	HRESULT	HRESULT indicating the result of previous operations (If SUCCEEDED(hrPrev) equals false, no operation is executed.)

In/Outputs

Name	Type	Description
stKernelX	TcVnMatrix	1D custom row-filter kernel with values of type REAL or LREAL
stKernelY	TcVnMatrix	1D custom column-filter kernel with values of type REAL or LREAL

Return value

HRESULT

Further information

The function F_VN_SeparableCustomFilter applies a separable 2D filter to the input image.

Algorithm

A separable 2D filter can be combined from two 1D filters. It is more efficient to apply a 1D filter to an image twice than to apply a 2D filter once. For example, the well-known Sobel filter can be constructed as follows from two separate one-dimensional filters:

Parameter

Input image

The input image ipSrcImage may have any available format.

Result image

The result image ipDestImage has the same size as the input image ipSrcImage, but the pixel format is determined by the result depth eDestDepth.

X and Y kernel

The two 1-dimensional filter kernels stKernelX and stKernelY are transferred as a matrix. Permitted element types of the matrix are TCVN_ET_REAL and TCVN_ET_LREAL. Corresponding matrices of the type TcVnMatrix can be created with the aid of the function F_VN_InitMatrixStruct.

Expert parameters

The expert version F_VN_SeparableCustomFilterExp contains additional parameters.

Application

The creation and application of a separable 2D filter looks like this, for example:

hr := F_VN_SeparableCustomFilter(
    ipSrcImage  :=  ipImageIn,
    ipDestImage :=  ipImageRes,
    eDestDepth  :=  TCVN_ET_USINT,
    stKernelX   :=  stKernelMatrixX, 
    stKernelY   :=  stKernelMatrixY,
    hrPrev      :=  hr
);