F_VN_MergeRegions

Syntax

Definition:

FUNCTION F_VN_MergeRegions : HRESULT
VAR_INPUT
    ipSrcContainer  : ITcVnContainer;
    ipDestContainer : Reference To  ITcVnContainer;
    nDistance       : UDINT;
    nOptions        : ULINT;
    hrPrev          : HRESULT;
END_VAR

 Inputs

 Return value

HRESULT

Further information

The function F_VN_MergeRegions merges rectangles into a single region if they overlap or if the distance between their edges does not exceed a specified value.

Typically, this function is used for post-processing detection results in which objects have been detected multiple times in adjacent ROIs and the resulting regions overlap or are contiguous. One use case is the post-processing of the code regions detected by F_VN_DetectCodesNeuralNetwork in order to merge codes detected multiple times at the boundaries of adjacent ROIs into a single result region. The grouped regions can then be reused as ROIs and, in this example, passed to a code-reading function.

Parameter

List of rectangles

The list of rectangles to be merged is passed via ipSrcContainer in a container of type ContainerType_Vector_TcVnRectangle_DINT.

List of resulting regions

The aggregated regions are returned in ipDestContainer, which has the same type as ipSrcContainer.

Distance

The parameter nDistance is the maximum pixel distance between two rectangles for them to be merged. Overlapping rectangles are always merged. For non-overlapping rectangles, the following applies: The distance in the x-direction and the distance in the y-direction must each be less than or equal to nDistance. A distance of 0 means that two rectangles are touching each other and will therefore also be merged.

Options

The nOption parameter specifies the method used to merge the individual sections. The methods are defined in ETcVnMergeRegionsOptions:

• TCVN_MRO_UPRIGHT_BOUNDING_RECT: Each aggregated region is represented by the smallest enclosing rectangle parallel to the axes.

Required License

TC3 Vision Base

System Requirements