do_ocr_multi_class_cnnT_do_ocr_multi_class_cnnDoOcrMultiClassCnnDoOcrMultiClassCnndo_ocr_multi_class_cnn (算子)

名称

do_ocr_multi_class_cnnT_do_ocr_multi_class_cnnDoOcrMultiClassCnnDoOcrMultiClassCnndo_ocr_multi_class_cnn — 使用基于 CNN 的 OCR 分类器对多个字符进行分类。

签名

do_ocr_multi_class_cnn(Character, Image : : OCRHandle : Class, Confidence)

描述

do_ocr_multi_class_cnndo_ocr_multi_class_cnnDoOcrMultiClassCnnDoOcrMultiClassCnnDoOcrMultiClassCnndo_ocr_multi_class_cnn computes the best class for each of the characters given by the regions CharacterCharacterCharacterCharactercharactercharacter and the gray values ImageImageImageImageimageimage with the OCR classifier OCRHandleOCRHandleOCRHandleOCRHandleOCRHandleocrhandle and returns the classes in ClassClassClassClassclassValclass and the corresponding confidences (probabilities) of the classes in ConfidenceConfidenceConfidenceConfidenceconfidenceconfidence。

In contrast to do_ocr_single_class_cnndo_ocr_single_class_cnnDoOcrSingleClassCnnDoOcrSingleClassCnnDoOcrSingleClassCnndo_ocr_single_class_cnn, do_ocr_multi_class_cnndo_ocr_multi_class_cnnDoOcrMultiClassCnnDoOcrMultiClassCnnDoOcrMultiClassCnndo_ocr_multi_class_cnn can classify multiple characters in one call, and therefore typically is faster than a loop that uses do_ocr_single_class_cnndo_ocr_single_class_cnnDoOcrSingleClassCnnDoOcrSingleClassCnnDoOcrSingleClassCnndo_ocr_single_class_cnn to classify single characters. However, do_ocr_multi_class_cnndo_ocr_multi_class_cnnDoOcrMultiClassCnnDoOcrMultiClassCnnDoOcrMultiClassCnndo_ocr_multi_class_cnn can only return the best class of each character. Because the confidences can be interpreted as probabilities, it is easier to check whether a character has been classified with too much uncertainty. This is usually not a disadvantage, except in cases where the classes overlap so much that in many cases the second best class must be examined to be able to decide the class of the character. In these cases, do_ocr_single_class_cnndo_ocr_single_class_cnnDoOcrSingleClassCnnDoOcrSingleClassCnnDoOcrSingleClassCnndo_ocr_single_class_cnn should be used.

A string of the number '\032'"\032""\032""\032""\032""\032" (alternatively displayed as '\0x1A'"\0x1A""\0x1A""\0x1A""\0x1A""\0x1A") in ClassClassClassClassclassValclass signifies that the region has been classified as rejection class.

执行信息

多线程类型：可重入（与非独占算子并行运行）。
多线程作用域：全局（可从任何线程调用）。
在内部数据级别上自动并行化。

参数

CharacterCharacterCharacterCharactercharactercharacter (输入对象) region(-array) → object

Characters to be recognized.

ImageImageImageImageimageimage (输入对象) singlechannelimage → object (byte / uint2)

字符的灰度值。

OCRHandleOCRHandleOCRHandleOCRHandleOCRHandleocrhandle (输入控制) ocr_cnn → (handle)

Handle of the OCR classifier.

ClassClassClassClassclassValclass (输出控制) string(-array) → (string)

Result of classifying the characters with the CNN.

元素数量： Class == Character

ConfidenceConfidenceConfidenceConfidenceconfidenceconfidence (输出控制) real(-array) → (real)

Confidence of the class of the characters.

元素数量： Confidence == Character

结果

如果参数有效，算子 do_ocr_multi_class_cnndo_ocr_multi_class_cnnDoOcrMultiClassCnnDoOcrMultiClassCnnDoOcrMultiClassCnndo_ocr_multi_class_cnn 返回值 2 ( H_MSG_TRUE )。如有必要，则抛出异常。

可能的前趋

read_ocr_class_cnnread_ocr_class_cnnReadOcrClassCnnReadOcrClassCnnReadOcrClassCnnread_ocr_class_cnn

替代

do_ocr_word_cnndo_ocr_word_cnnDoOcrWordCnnDoOcrWordCnnDoOcrWordCnndo_ocr_word_cnn, do_ocr_single_class_cnndo_ocr_single_class_cnnDoOcrSingleClassCnnDoOcrSingleClassCnnDoOcrSingleClassCnndo_ocr_single_class_cnn

模块

光学字符识别/光学字符验证

算子