JPH0319589B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Technical Field of the Invention] The present invention relates to an optical character reading device particularly for recognizing handwritten Chinese characters.

[Technical background of the invention and its problems] In recent years, optical character readers (OCR) for handwritten kanji recognition have been equipped with furigana characters (usually katakana characters) that correspond to handwritten kanji in order to improve recognition accuracy. A system has been developed that performs kanji recognition processing using the reading results as a guide (Patent Application Reference 56).
−99573).

In this type of OCR, furigana characters corresponding to kanji are written on the form in advance, and the recognition result is an answer from multiple candidate characters obtained from the kanji recognition results based on the reading results of these furigana characters. is configured to be selected.

However, in the above-mentioned OCR method, the accuracy of kanji recognition depends on the recognition result of furigana characters (katakana characters), and it is a method that becomes effective when furigana character recognition processing is performed with high accuracy. be. For this reason, for example, Tsukimi Sato (Yamanashi),
Because it is difficult to assign the correct furigana in advance to kanji with special pronunciations, such as ``Samidare,'' OCR using the above-mentioned method may result in reading errors. Furthermore, even in the case of normal kanji, it is impossible to completely recognize katakana characters, which are furigana characters, so it is difficult to read kanji with high accuracy.

[Purpose of the invention]

The present invention was made in view of the above circumstances, and
The purpose is to provide an optical character reading device that can recognize kanji with high precision in OCR, which uses the recognition results of furigana characters to perform kanji recognition.

[Summary of the invention]

In the present invention, a character recognition means is provided for recognizing each character in units of characters for a form in which kanji characters and furigana characters corresponding to the kanji characters are recorded. Based on the recognition results of each character by the character recognition means, the word recognition means recognizes words each consisting of kanji and furigana. The word recognition means performs recognition processing for each word by referring to a word recognition table prepared in advance.

The control means collates the recognition results of each word unit from the word recognition means, and when the collation results indicate that the recognition results of each word unit of kanji and furigana characters correspond to each other, converts each recognition result into a final answer. It is configured to output as .

As a result, each character of kanji and furigana can be recognized simultaneously and accurately, and as a result, kanji can be read with high accuracy.

[Embodiments of the invention]

An embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing a partial configuration of an OCR according to an embodiment. In FIG. 1, a scanning unit 10 scans a form (shown in FIG. 2), photoelectrically converts each of the kanji and furigana/katakana characters recorded on the form, and converts each quantization pattern (binary A character pattern consisting of a signal) is output to the character recognition unit 11. This character recognition unit 11 is
Based on the quantization pattern sent from the scanning unit 10, recognition processing of kanji and furigana katakana characters on the form is performed character by character.

The word recognition unit 12 recognizes each word made up of the respective characters based on the recognition results (character by character) of the kanji and katakana characters output from the character recognition unit 11. The word recognition unit 12 stores in advance a word-by-word recognition table (dictionary) for each of kanji and katakana characters. The control unit 13 is
The recognition results for each word of kanji and katakana characters output from the word recognition unit 12 are collated, and if the recognition results for each word correspond to each other in the collation results, each recognition result is output as the final answer. do. Further, the control section 13 controls each operation of the scanning section 10, the character recognition section 11, and the word recognition section 12.

The operation of the OCR configured as described above will be explained. First, a form as shown in FIG. 2 is scanned by the scanning unit 10, and the kanji and furigana/katakana characters recorded on the form are each converted into quantized patterns and sent to the character recognition unit 11. Here, the form includes a character area 21 in which the kanji 20 is written, and the kanji 2
Character area 2 in which furigana katakana characters (hereinafter simply referred to as katakana) 22 corresponding to 0 are written
3 is set in advance. The scanning unit 10 sends each quantization pattern corresponding to the kanji 20 and katakana 22 on the form to the character recognition unit 11 . The character recognition unit 11 performs character-by-character recognition processing on the kanji 20 and katakana 22 based on each quantization pattern from the scanning unit 10.

The recognition results for each character recognized by the character recognition section 11 are sequentially sent to the word recognition section 12. The word recognition unit 12 stores each character recognition result in a buffer memory, word by word, for example. In this case, a format table prepared in advance by the control unit 13 (characters 20 to be written on the form,
Information indicating the location of 22 is stored)
Based on this, the above recognition results are sent to the word recognition section 12 for each word. As a result, the word recognition unit 12 performs recognition processing on a word-by-word basis (that is, for example, a surname and a word for each name).

By the way, for example, as shown in FIG. 3, in the recognition results for katakana in the character recognition unit 11,
Assume that one character 30 of the last name and one character 31 of the first name cannot be specified. Each recognition result 3 at this time
The candidate characters 32 and 33 of 0 and 31 are each two characters. For example, as shown in FIG. 4, it is assumed that one character 40 of the last name and one character 41 of the first name cannot be identified in the recognition results for Chinese characters by the character recognition unit 11. At this time, the candidate characters 42 and 43 of each recognition result 40 and 41 are two characters each. The word recognition unit 12 uses a word recognition table (hereinafter referred to as a word dictionary) shown in FIGS. 5 and 6 for the above-mentioned word units of katakana (FIG. 3) and kanji (FIG. 4). Word-by-word recognition processing is performed based on this. That is,
In the recognition process for Katakana shown in FIG. 3, word recognition is performed based on the word dictionary shown in FIG. The last name "Sakaku" and the first name "Kenio" are removed because they do not exist in the word dictionary. As a result, both the words "Sakai" and "Kunio" shown in FIG. 7 are sent from the word recognition section 12 to the control section 1 as the word recognition results for each katakana of the surname and first name. At this time, multiple kanji 7 corresponding to each furigana of "Sakai" and "Kunio"
0 and 71 are both sent to the control section 13.

In addition, the word recognition unit 1 for kanji shown in FIG.
In the recognition process No. 2, a search is performed based on the word dictionary shown in FIG.
Both the kanji ``Sakai'' and ``Kunio'' corresponding to 3 are selected. As a result, from the word recognition unit 12,
The recognition results (Sakai and Kunio) shown in FIG. 8 are sent to the control unit 13 along with the corresponding katakana 80 and 81.

The control unit 13 collates the recognition results shown in FIGS. 7 and 8 sent from the word recognition unit 12. Then, the recognition results in which both the katakana and kanji characters of each word correspond to each other are output as final answers. That is, in this case, an answer as shown in FIG. 9 is output from the control section 13. As a result of the matching process in the control unit 13, if there are no mutually corresponding recognition results for katakana and kanji, a reject process is performed.

In this way, each word of kanji and katakana written on the form is recognized, and the recognition results are compared. As a result, a combination result in which the kanji and katakana of each word correspond to each other can be output as the final answer. Therefore, even if the character-by-character recognition results for katakana (furigana) are not perfect, the answer is output as a combination of kanji and katakana that correspond to each other on a word-by-word basis, significantly reducing the misreading rate of kanji. can be done. Furthermore, since recognition processing is performed on a word-by-word basis as described above, it is also effective in reading processing for kanji that have special readings (for example, Tsukimi-sato, Satsuki-ame).

〔Effect of the invention〕

As detailed above, according to the present invention, the recognition results for each word of furigana characters and kanji are collated,
By outputting the combination result of furigana characters and kanji that correspond to each other as the final answer, even if the recognition result of each character of furigana characters is incomplete,
The rate of misreading of kanji can be significantly reduced. Therefore, it is possible to recognize kanji with high accuracy as a result.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing the configuration of an optical character reading device according to an embodiment of the present invention, FIG. 2 is a diagram showing an example of a form, and FIGS. 3 and 4 are character recognition of FIG. 1, respectively. 5 and 6 are diagrams each showing an example of a word recognition table prepared in the word recognition section of FIG. 1,
7 and 8 are diagrams each showing an example of the recognition result of the word recognition section of FIG. 1, and FIG. 9 is a diagram showing an example of the final answer of the control section of FIG. 1. 10... Scanning section, 11... Character recognition section, 12...
...Word recognition section, 13...Control section.

Claims (1)

[Claims] 1. A form that has a character area for recording kanji and a character area for recording furigana characters corresponding to the kanji, and a quantization pattern by photoelectrically converting the kanji and furigana characters recorded on this form. a scanning unit that outputs a quantization pattern, a character recognition unit that recognizes each of the above-mentioned kanji and furigana characters character by character based on the quantization pattern output from the scanning unit; Word recognition means that recognizes words made of each of the above-mentioned kanji and furigana characters based on the recognition results by referring to a word recognition table prepared in advance, and recognition of each word unit output from this word recognition means. It is characterized by comprising a control means for collating the results and outputting each recognition result as a final answer when the recognition results for each word of kanji and furigana characters correspond to each other. optical character reading device.