Operator Reference

HALCON Operator Reference / OCR / Neural Nets

• HDevelop
• .NET
• Python
• C++
• C

ClassesClasses | | Operators

do_ocr_single_class_mlpT_do_ocr_single_class_mlpDoOcrSingleClassMlpDoOcrSingleClassMlpdo_ocr_single_class_mlp (Operator)

do_ocr_single_class_mlpT_do_ocr_single_class_mlpDoOcrSingleClassMlpDoOcrSingleClassMlpdo_ocr_single_class_mlp — Classify a single character with an OCR classifier.

Signature

do_ocr_single_class_mlp(Character, Image : : OCRHandle, Num : Class, Confidence)

Herror T_do_ocr_single_class_mlp(const Hobject Character, const Hobject Image, const Htuple OCRHandle, const Htuple Num, Htuple* Class, Htuple* Confidence)

void DoOcrSingleClassMlp(const HObject& Character, const HObject& Image, const HTuple& OCRHandle, const HTuple& Num, HTuple* Class, HTuple* Confidence)

HTuple HRegion::DoOcrSingleClassMlp(const HImage& Image, const HOCRMlp& OCRHandle, const HTuple& Num, HTuple* Confidence) const

HString HRegion::DoOcrSingleClassMlp(const HImage& Image, const HOCRMlp& OCRHandle, const HTuple& Num, double* Confidence) const

HTuple HOCRMlp::DoOcrSingleClassMlp(const HRegion& Character, const HImage& Image, const HTuple& Num, HTuple* Confidence) const

HString HOCRMlp::DoOcrSingleClassMlp(const HRegion& Character, const HImage& Image, const HTuple& Num, double* Confidence) const

static void HOperatorSet.DoOcrSingleClassMlp(HObject character, HObject image, HTuple OCRHandle, HTuple num, out HTuple classVal, out HTuple confidence)

HTuple HRegion.DoOcrSingleClassMlp(HImage image, HOCRMlp OCRHandle, HTuple num, out HTuple confidence)

string HRegion.DoOcrSingleClassMlp(HImage image, HOCRMlp OCRHandle, HTuple num, out double confidence)

HTuple HOCRMlp.DoOcrSingleClassMlp(HRegion character, HImage image, HTuple num, out HTuple confidence)

string HOCRMlp.DoOcrSingleClassMlp(HRegion character, HImage image, HTuple num, out double confidence)

def do_ocr_single_class_mlp(character: HObject, image: HObject, ocrhandle: HHandle, num: Sequence[int]) -> Tuple[Sequence[str], Sequence[float]]

def do_ocr_single_class_mlp_s(character: HObject, image: HObject, ocrhandle: HHandle, num: Sequence[int]) -> Tuple[str, float]

Description

do_ocr_single_class_mlpdo_ocr_single_class_mlpDoOcrSingleClassMlpDoOcrSingleClassMlpdo_ocr_single_class_mlp computes the best NumNumNumnumnum classes of the character given by the region CharacterCharacterCharactercharactercharacter and the gray values ImageImageImageimageimage with the OCR classifier OCRHandleOCRHandleOCRHandleOCRHandleocrhandle and returns the classes in ClassClassClassclassValclass and the corresponding confidences (probabilities) of the classes in ConfidenceConfidenceConfidenceconfidenceconfidence. Because multiple classes may be returned by do_ocr_single_class_mlpdo_ocr_single_class_mlpDoOcrSingleClassMlpDoOcrSingleClassMlpdo_ocr_single_class_mlp, CharacterCharacterCharactercharactercharacter may only contain a single region (a single character). If multiple characters should be classified in a single call, do_ocr_multi_class_mlpdo_ocr_multi_class_mlpDoOcrMultiClassMlpDoOcrMultiClassMlpdo_ocr_multi_class_mlp must be used. Because do_ocr_multi_class_mlpdo_ocr_multi_class_mlpDoOcrMultiClassMlpDoOcrMultiClassMlpdo_ocr_multi_class_mlp typically is faster than a loop with do_ocr_single_class_mlpdo_ocr_single_class_mlpDoOcrSingleClassMlpDoOcrSingleClassMlpdo_ocr_single_class_mlp and because the confidences can be interpreted as probabilities (see classify_class_mlpclassify_class_mlpClassifyClassMlpClassifyClassMlpclassify_class_mlp and evaluate_class_mlpevaluate_class_mlpEvaluateClassMlpEvaluateClassMlpevaluate_class_mlp), and it is therefore easy to check whether a character has been classified with too much uncertainty, in most cases do_ocr_multi_class_mlpdo_ocr_multi_class_mlpDoOcrMultiClassMlpDoOcrMultiClassMlpdo_ocr_multi_class_mlp should be used, unless the second-best class should be examined explicitly.

A string of the number (alternatively displayed as ) in ClassClassClassclassValclass signifies that the region has been classified as rejection class.

Before calling do_ocr_single_class_mlpdo_ocr_single_class_mlpDoOcrSingleClassMlpDoOcrSingleClassMlpdo_ocr_single_class_mlp, the classifier must be trained with trainf_ocr_class_mlptrainf_ocr_class_mlpTrainfOcrClassMlpTrainfOcrClassMlptrainf_ocr_class_mlp.

Execution Information

• Multithreading type: reentrant (runs in parallel with non-exclusive operators).
• Multithreading scope: global (may be called from any thread).
• Processed without parallelization.

Parameters

CharacterCharacterCharactercharactercharacter (input_object)  region → objectHRegionHObjectHObjectHobject

Character to be recognized.

ImageImageImageimageimage (input_object)  singlechannelimage → objectHImageHObjectHObjectHobject (byte / uint2)

Gray values of the character.

OCRHandleOCRHandleOCRHandleOCRHandleocrhandle (input_control)  ocr_mlp → HOCRMlp, HTupleHHandleHTupleHtuple (handle) (IntPtr) (HHandle) (handle)

Handle of the OCR classifier.

NumNumNumnumnum (input_control)  integer-array → HTupleSequence[int]HTupleHtuple (integer) (int / long) (Hlong) (Hlong)

Number of best classes to determine.

Default: 1

Suggested values: 1, 2, 3, 4, 5

ClassClassClassclassValclass (output_control)  string(-array) → HTupleSequence[str]HTupleHtuple (string) (string) (HString) (char*)

Result of classifying the character with the MLP.

ConfidenceConfidenceConfidenceconfidenceconfidence (output_control)  real(-array) → HTupleSequence[float]HTupleHtuple (real) (double) (double) (double)

Confidence(s) of the class(es) of the character.

Result

If the parameters are valid, the operator do_ocr_single_class_mlpdo_ocr_single_class_mlpDoOcrSingleClassMlpDoOcrSingleClassMlpdo_ocr_single_class_mlp returns the value 2 ( H_MSG_TRUE) . If necessary, an exception is raised.

Possible Predecessors

trainf_ocr_class_mlptrainf_ocr_class_mlpTrainfOcrClassMlpTrainfOcrClassMlptrainf_ocr_class_mlp, read_ocr_class_mlpread_ocr_class_mlpReadOcrClassMlpReadOcrClassMlpread_ocr_class_mlp

Alternatives

do_ocr_multi_class_mlpdo_ocr_multi_class_mlpDoOcrMultiClassMlpDoOcrMultiClassMlpdo_ocr_multi_class_mlp

See also

create_ocr_class_mlpcreate_ocr_class_mlpCreateOcrClassMlpCreateOcrClassMlpcreate_ocr_class_mlp, classify_class_mlpclassify_class_mlpClassifyClassMlpClassifyClassMlpclassify_class_mlp

Module

OCR/OCV

---

ClassesClasses | | Operators