F_VN_CalibrateCamera

Syntax

Definition:

FUNCTION F_VN_CalibrateCamera : HRESULT
VAR_INPUT
    ipSrcImage              : ITcVnImage;
    ipReferencePoints       : ITcVnContainer;
END_VAR
VAR_IN_OUT
    aCameraMatrix           : TcVnMatrix3x3_LREAL;
    aDistortionCoefficients : TcVnArray8_LREAL;
    aRotationMatrix         : TcVnMatrix3x3_LREAL;
    aTranslationVector      : TcVnVector3_LREAL;
    fReprojError            : LREAL;
    stBlobDetectionParams   : TcVnParamsBlobDetection;
    stCalibrationOptions    : TcVnCameraCalibrationOptions;
END_VAR
VAR_INPUT
    hrPrev                  : HRESULT;
END_VAR

 Inputs

 In/Outputs

 Return value

HRESULT

Further information

The function F_VN_CalibrateCamera calculates the intrinsic and extrinsic camera parameters on the basis of an input image containing a calibration pattern. The elements of the calibration pattern must be detectable as a blob.

Input image

The input image ipSrcImage must be a 1-channel grayscale image with a bit depth of 8 bits. This image must contain a calibration pattern with circles that matches the reference points ipReferencePoints. The function expects black filled circles on a white background. To invert the image, the ETcVnThresholdType = TCVN_TT_BINARY_INV of the Blob function can be used.

The expert version F_VN_CalibrateCameraExp is required for a calibration with several input images.

Reference points

The container ipReferencePoints must contain the reference points that are illustrated on the input image ipSrcImage as the calibration pattern. The points must be specified as 3-dimensional points of the type REAL so that the container is of the type ContainerType_Vector_TcVnPoint3_REAL.

The reference points can be generated as follows:

• Manual creation of an XML file and reading with the function block FB_VN_ReadCalibrationPattern.
• Configuration in the calibration assistants, saving in the TcCOM object by clicking Write Results and reading with the method GetCalibPatternRef of the function block FB_VN_GevCameraControl. The set center point is taken into account.
• Inserting reference points individually into a container.

Matrices of the camera parameters

These matrices return the calculated parameters of the camera model. They can then be used for the compensation of the objective distortions and for the conversion between pixel coordinates and world coordinates.

• Camera matrix
• Distortion coefficients
• Rotation matrix
• Translation vector

Rounding errors On account of rounding errors, the calculated parameter values can deviate slightly from the values calculated in the calibration assistants.

Reprojection errors

Parameters for blob detection

No chessboard pattern As the reference points are detected as a blob structure, the use of chessboard patterns is not possible.

Options for calibration

Default settings To get the default settings from the calibration wizard, the bits bFixAspectRatio, bFixPrincipalPoint, bFixK5, bFixK6 and bRationalModel must be set to TRUE. bFixPrincipalPoint = „Fix Center Point“ The bit bFixPrincipalPoint corresponds to the option "Fix Center Point" in the camera calibration assistant.

Samples

• Calibration Assistant

Required License

TC3 Vision Metrology 2D

System Requirements