OCR

Detects and recognizes characters in a binary image (white characters on black background).
Can use available TwinCAT Job Tasks for executing parallel code regions.

Syntax

Definition:

HRESULT OCR(
    HRESULT          hrPrev,
    ITcVnImage*      ipSrcImage,
    ULONGLONG        eModel,
    ITcVnContainer*& ipCharacters,
    float&           fMinConfidence,
    PCCH             sPattern = "",
    ULONG            eOcrOptions = OO_NONE,
    ITcVnContainer** pipBoundingBoxes = nullptr,
    ITcVnContainer** pipConfidences = nullptr
)

Parameters

Name	Type	Default	Description
hrPrev	HRESULT		HRESULT indicating the result of previous operations (If SUCCEEDED(hrPrev) equals false, no operation is executed.)
ipSrcImage	ITcVnImage*		Source image (USINT elements, 1 or 3 channels). The required number of channels depends on the selected model type.
eModel	ULONGLONG		Selection of OCR model types (ETcVnOcrModelType)
ipCharacters	ITcVnContainer*&		Returns the recognized characters (ContainerType_Vector_String_SINT)
fMinConfidence	float&		Returns the minimum value of the confidences across all recognized characters. Characters rejected by the novelty detector are excluded from this minimum. Always computed, even if ipConfidences is not requested.
sPattern	PCCH	""	String pattern containing the format in which characters are presented. Pattern handling support depends on the selected model type.
eOcrOptions	ULONG	OO_NONE	Specifies which options must be applied to the function (ETcVnOcrOptions)
pipBoundingBoxes	ITcVnContainer**	nullptr	Returns the corresponding bounding boxes of the recognized characters (ContainerType_Vector_TcVnRectangle_DINT, optional, set to 0 if not required). Returns an empty container when bounding boxes are not supported by the selected model type.
pipConfidences	ITcVnContainer**	nullptr	Returns the corresponding classification confidences of the recognized characters (ContainerType_Vector_REAL, optional, set to 0 if not required). When the novelty detector is selected, the confidence value is a combination of the classification confidence and the novelty detection confidence. If a character is identified as unknown, the total confidence equals the novelty detection confidence.

Return value

HRESULT

Required License

TC3 Vision OCR

System Requirements

Development environment	Target platform	PLC libraries to include
TwinCAT V3.1.4024.59 or later	PC or CX (x64) with min. PL50, e.g. Intel 4-core Atom CPU	Tc3_Vision