F_VN_CheckColorRange

Syntax

Definition:

FUNCTION F_VN_CheckColorRange : HRESULT
VAR_INPUT
    ipSrcImage   : ITcVnImage;
    ipDestImage  : Reference To  ITcVnImage;
END_VAR
VAR_IN_OUT
    aLowerBounds : TcVnVector4_LREAL;
    aUpperBounds : TcVnVector4_LREAL;
END_VAR
VAR_INPUT
    hrPrev       : HRESULT;
END_VAR

 Inputs

 In/Outputs

 Return value

HRESULT

Further information

The F_VN_CheckColorRange function segments the input image with regard to a specific value range. This means that it does not matter which color space is used, as the segmentation is purely numerical.

• Image pixels that lie within the range are assigned the maximum value 255 (white).
• Image pixels that lie outside of the color spectrum are given the value 0 (black).

Parameter

Input image

The input image ipSrcImage can have any format.

Result image

The result image ipDestImage returns the segmented 1-channel 8-bit binary image.

Range of Values

The value range to be segmented is defined by the lower limit aLowerBounds and upper limit aUpperBounds.

For cyclic value ranges such as the Hue channel in HSV space, the lower limit value can exceed the upper limit value in order to segment color ranges across the zero crossing. A range from the lower limit [250, ...] to the upper limit [5, ...] corresponds to the combination of a function call with [250, ...] - [MAX, ...] and [0, ...] - [5, ...]. This makes it possible to directly select color tones that lie in the transition of the value range, such as the red range (+-20° around 0°) in the Hue channel, with a function call.

Initialize boundary vectors Always be sure to initialize the boundary vectors with at least as many elements as the input image has channels. For example, if you segment a 4-channel image (e.g. with alpha channel) and initialize the color vectors with only three elements, the fourth elements of the vectors are 0 and the result image turns black.

Application

For reliable color segmentations, it is generally advisable to work in a color space other than the additive RGB space. For example, in HSV space, hue (H) is separated from lightness (V) and saturation (S), which means that color differentiation is typically less sensitive to fluctuations in lighting.

The segmentation of an HSV image (HSV_FULL) for a red color spectrum looks like this, for example:

VAR
    aLower      :   TcVnVector4_LREAL := [250, 150, 100];
    aUpper      :   TcVnVector4_LREAL := [5, 255, 255];
END_VAR

hr := F_VN_CheckColorRange(
    ipSrcImage      := ipImageIn,
    ipDestImage     := ipImageRes,
    aLowerBounds    := aLower,
    aUpperBounds    := aUpper,
    hrPrev          := hr);

Examples

• Check Color Range

Required License

TC3 Vision Base

System Requirements