JPH0512484A - Optical character recognizing device - Google Patents

Optical character recognizing device

Info

Publication number
JPH0512484A
JPH0512484A JP3161216A JP16121691A JPH0512484A JP H0512484 A JPH0512484 A JP H0512484A JP 3161216 A JP3161216 A JP 3161216A JP 16121691 A JP16121691 A JP 16121691A JP H0512484 A JPH0512484 A JP H0512484A
Authority
JP
Japan
Prior art keywords
character code
character
unit
candidate
code
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
JP3161216A
Other languages
Japanese (ja)
Inventor
Yutaka Kajino
豊 梶野
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
NEC Engineering Ltd
Original Assignee
NEC Engineering Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by NEC Engineering Ltd filed Critical NEC Engineering Ltd
Priority to JP3161216A priority Critical patent/JPH0512484A/en
Publication of JPH0512484A publication Critical patent/JPH0512484A/en
Pending legal-status Critical Current

Links

Landscapes

  • Character Discrimination (AREA)

Abstract

PURPOSE:To considerably reduce correction conventionally executed by an operator. CONSTITUTION:A recognition part 2 segments a character from an image signal from a scanner part 1, recognizes it and afterwards stores a character code and a candidate character code in a character code storage memory 6 and a candidate character code storage memory 7, and an arithmetic part 3 collates the validity of the character code. When the code is illegal, a replacement part 5 extracts and outputs the candidate character code as a replaced system. The arithmetic part 3 collates the validity again with the replaced character code, when the proper character code is obtained, the character code in the character code storage memory 6 is replaced with the proper character code, and an output part 4 outputs this character code.

Description

【発明の詳細な説明】Detailed Description of the Invention

【0001】[0001]

【産業上の利用分野】本発明は光学的文字認識装置に関
し、特に演算式あるいは日本語の文脈により文字コード
の妥当性を照合する機能を有する光学的文字認識装置に
関する。
BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an optical character recognition device, and more particularly to an optical character recognition device having a function of verifying the validity of a character code according to an arithmetic expression or the context of Japanese.

【0002】[0002]

【従来の技術】従来の光学的文字認識装置は、認識対象
文字に対して複数の類似した文字コードが候補にあがる
と、読取り不能文字として出力していた。また演算式あ
るいは日本語の文脈処理指定により妥当性を照合し、妥
当性が不正であれば不正コードとして出力していた。読
取り不能文字および不正コードが出力されると、操作員
は帳票あるいは操作卓の画面上に表示されたイメージを
読むことにより修正していた。
2. Description of the Related Art In a conventional optical character recognition device, when a plurality of similar character codes are candidates for a character to be recognized, the character is output as an unreadable character. In addition, the validity was checked by an arithmetic expression or Japanese context processing specification, and if the validity was incorrect, it was output as an invalid code. When unreadable characters and illegal codes are output, the operator corrects them by reading the image displayed on the form or the screen of the console.

【0003】図3は従来の光学的文字認識装置の一例を
示すブロック図である。帳票上の文字はスキャナ部1に
より光電変換され、画像信号として認識部2へ送られ
る。認識部2は認識結果の文字コードを演算部3へ出力
する。演算部3は認識された文字コードを演算式あるい
は日本語の文脈処理指定による文字コードと照合し、正
常な場合は正常データとして、また不正な場合は不正デ
ータとして文字コードを出力部4へ送る。出力部4は文
字コードを媒体または接続機器へ出力する。
FIG. 3 is a block diagram showing an example of a conventional optical character recognition device. The characters on the form are photoelectrically converted by the scanner unit 1 and sent to the recognition unit 2 as an image signal. The recognition unit 2 outputs the character code of the recognition result to the calculation unit 3. The arithmetic unit 3 collates the recognized character code with a character code designated by an arithmetic expression or Japanese context processing specification, and sends the character code to the output unit 4 as normal data when it is normal and as invalid data when it is incorrect. . The output unit 4 outputs the character code to a medium or a connected device.

【0004】[0004]

【発明が解決しようとする課題】この従来の光学的文字
認識装置では、「ソ」と「ン」のように類似した字形を
持つ文字の認識において複数の候補文字コードが発生す
ると、判別不能あるいは誤読となることが多く、操作員
の修正作業が増加するという問題点があった。一方、文
字記入者は、これらの文字の字形の特徴をはっきりと記
入しなければならなかった。
In this conventional optical character recognition device, when a plurality of candidate character codes are generated in recognition of characters having similar character shapes such as "so" and "n", it is impossible to discriminate or There was a problem that it was often misread and the correction work by the operator increased. On the other hand, the character writer had to write down the glyphic features of these characters clearly.

【0005】[0005]

【課題を解決するための手段】本発明の光学的文字認識
装置は、紙面上を走査して文字を画像信号に光電変換す
るスキャナ部と、前記画像信号から1文字分ずつを切り
出して認識した文字コードおよび文字コードの候補を出
力する認識部と、前記文字コードの候補群から演算式や
文脈に適した文字コードを選択して出力する選択出力手
段とを備えている。
The optical character recognition device of the present invention recognizes a character by cutting out a character from the image signal and a scanner section for photoelectrically converting a character into an image signal by scanning on a paper surface. A recognition unit that outputs a character code and a character code candidate, and a selection output unit that selects and outputs a character code suitable for an arithmetic expression or a context from the character code candidate group are provided.

【0006】そして、前記選択出力手段は前記文字コー
ドを予め指定された演算式や文脈処理指定により前記文
字コードの妥当性を照合する演算部と、この演算部によ
って不正とされた文字コードを前記文字コードの候補群
から選択した文字コードと置き換える置換部と、前記妥
当な文字コードおよび置換された文字コードを出力する
出力部とから構成してもよい。
The selection and output means compares the character code with an arithmetic expression specified in advance or a context processing specification to check the validity of the character code, and the character code invalidated by the arithmetic section. A replacement unit that replaces the character code selected from the character code candidate group and an output unit that outputs the valid character code and the replaced character code may be configured.

【0007】[0007]

【実施例】次に、本発明について図面を参照して説明す
る。図1は本発明の光学的文字認識装置の一実施例を示
すブロック図である。
DESCRIPTION OF THE PREFERRED EMBODIMENTS Next, the present invention will be described with reference to the drawings. FIG. 1 is a block diagram showing an embodiment of the optical character recognition device of the present invention.

【0008】帳票上の文字はスキャナ部1により光電変
換され、画像信号として認識部2へ送られる。認識部2
は文字を切り出して認識した後、文字コードを文字コー
ド格納メモリ6に格納し、文字コードの候補および候補
優先度を候補文字コード格納メモリ7へ格納する。認識
部2の処理が終了すると、演算部3の演算器8が演算命
令格納メモリ9の命令に従って文字コード格納メモリ6
内の文字コードの妥当性を照合する。妥当性が不正とな
ったときは置換部5へ不正文字コードおよび文字位置情
報を送る。置換部5では、候補文字位置抽出器10が候
補文字コード格納メモリ7から置換対象となる候補文字
コードを抽出し、候補優先度比較器11が置換対象の候
補文字コードの優先度を比較し、置換コード出力器12
が最優先の候補文字コードを出力する。演算部3は置換
された文字コードで再度妥当性を照合し、不正の場合は
再度置換部5に文字コードの置換を要求する。置換部5
は次に最も優先度の高い候補文字コードを選択して演算
部3へ出力する。このようにして照合と置換をくり返
し、正当な文字コードが得られると文字コード格納メモ
リ6内の文字コードを正当な文字コードに置換する。正
当な文字コードが得られない場合は置換しない。演算が
終了すると、出力部4は文字コード格納メモリ6内の文
字コードを媒体または外部装置へ出力する。
The characters on the form are photoelectrically converted by the scanner unit 1 and sent to the recognition unit 2 as an image signal. Recognition unit 2
After cutting out the character and recognizing it, the character code is stored in the character code storage memory 6, and the character code candidate and the candidate priority are stored in the candidate character code storage memory 7. When the processing of the recognition unit 2 is completed, the arithmetic unit 8 of the arithmetic unit 3 follows the instruction of the arithmetic instruction storage memory 9 and stores the character code storage memory 6
Check the validity of the character code in. When the validity is incorrect, the incorrect character code and the character position information are sent to the replacing unit 5. In the replacement unit 5, the candidate character position extractor 10 extracts the candidate character code to be replaced from the candidate character code storage memory 7, and the candidate priority comparator 11 compares the priorities of the candidate character codes to be replaced, Replacement code output device 12
Outputs the highest priority candidate character code. The calculation unit 3 again checks the validity of the replaced character code, and if the character code is incorrect, requests the replacement unit 5 to replace the character code again. Replacement unit 5
Selects the candidate character code having the next highest priority and outputs it to the arithmetic unit 3. In this way, collation and replacement are repeated, and when a valid character code is obtained, the character code in the character code storage memory 6 is replaced with the valid character code. Do not replace if a valid character code cannot be obtained. When the calculation is completed, the output unit 4 outputs the character code in the character code storage memory 6 to a medium or an external device.

【0009】図2は本実施例における文字コードの置換
例を示す説明図で、(a)は数字の演算を指定されてい
るフィールドにおける文字コードの置換例を示し、
(b)は片仮名の読取りにおける日本語の文脈による文
字コードの置換例を示す。
FIG. 2 is an explanatory view showing an example of character code replacement in this embodiment. FIG. 2 (a) shows an example of character code replacement in a field in which a numerical operation is designated.
(B) shows an example of character code substitution in reading Japanese Katakana depending on the Japanese context.

【0010】図2(a)においては、認識部2が文字コ
ードを「01+05=08」と認識し、演算部3は演算
式により「8」が不正コードと判断する。置換部5は候
補文字コード格納メモリ7から候補文字コード「6」を
選択して「8」を「6」に置換し、出力部4は置換後の
文字コード「01+05=06」を出力する。同様に図
2(b)においては、認識部2からの文字コード中の読
取り不能文字「?」を候補文字「ン」,「ソ」のうち文
脈処理指定により妥当性のある「ソ」を選択して置換
し、出力部4は置換後の文字コード「レイゾウコ」を出
力する。
In FIG. 2 (a), the recognition unit 2 recognizes the character code as "01 + 05 = 08", and the calculation unit 3 judges that "8" is an illegal code according to the calculation formula. The replacement unit 5 selects the candidate character code “6” from the candidate character code storage memory 7 and replaces “8” with “6”, and the output unit 4 outputs the replaced character code “01 + 05 = 06”. Similarly, in FIG. 2B, the unreadable character “?” In the character code from the recognition unit 2 is selected from the candidate characters “n” and “so” by selecting the valid “so” by the context processing designation. Then, the output unit 4 outputs the character code “Reizouko” after the replacement.

【0011】[0011]

【発明の効果】以上説明したように本発明は、文字単独
の認識結果があらかじめ指定された演算式や日本語の文
脈の指定と照合して不正である場合には、候補文字群の
中から正当な文字コードを選択して置換することができ
るようにしたので、従来操作員が行っていた修正作業を
大幅に軽減するという効果を有する。
As described above, according to the present invention, when the recognition result of a single character is invalid by collating with a previously designated arithmetic expression or the designation of the Japanese context, it is selected from the candidate character group. Since a legitimate character code can be selected and replaced, the correction work conventionally performed by the operator can be significantly reduced.

【図面の簡単な説明】[Brief description of drawings]

【図1】本発明の光学的文字認識装置の一実施例を示す
ブロック図である。
FIG. 1 is a block diagram showing an embodiment of an optical character recognition device of the present invention.

【図2】本実施例における文字コードの置換例を示す説
明図である。
FIG. 2 is an explanatory diagram showing a replacement example of a character code according to the present embodiment.

【図3】従来の光学的文字認識装置の一例を示すブロッ
ク図である。
FIG. 3 is a block diagram showing an example of a conventional optical character recognition device.

【符号の説明】[Explanation of symbols]

1 スキャナ部 2 認識部 3 演算部 4 出力部 5 置換部 6 文字コード格納メモリ 7 候補文字コード格納メモリ 8 演算器 9 演算命令格納メモリ 10 候補文字位置抽出器 11 候補優先度比較器 12 置換コード出力器 1 Scanner section 2 recognition section 3 operation part 4 Output section 5 Replacement part 6 Character code storage memory 7 Candidate character code storage memory 8 arithmetic unit 9 Operation instruction storage memory 10 Candidate character position extractor 11 Candidate priority comparator 12 Replacement code output device

Claims (2)

【特許請求の範囲】[Claims] 【請求項1】 紙面上を走査して文字を画像信号に光電
変換するスキャナ部と、前記画像信号から1文字分ずつ
を切り出して認識した文字コードおよび文字コードの候
補を出力する認識部と、前記文字コードの候補群から演
算式や文脈に適した文字コードを選択して出力する選択
出力手段とを備えることを特徴とする光学的文字認識装
置。
1. A scanner unit for photoelectrically converting a character into an image signal by scanning on a paper surface, and a recognition unit for outputting a character code and a character code candidate recognized by cutting out one character from the image signal. An optical character recognition device, comprising: a selection output unit that selects and outputs a character code suitable for an arithmetic expression or a context from the character code candidate group.
【請求項2】 前記選択出力手段は前記文字コードを予
め指定された演算式や文脈処理指定により前記文字コー
ドの妥当性を照合する演算部と、この演算部によって不
正とされた文字コードを前記文字コードの候補群から選
択した文字コードと置き換える置換部と、前記妥当な文
字コードおよび置換された文字コードを出力する出力部
とから成ることを特徴とする光学的文字認識装置。
2. The selecting and outputting means compares the character code with an arithmetic unit that preliminarily specifies an arithmetic expression or a context process specification to check the validity of the character code, and the character code that is illegal by the arithmetic unit. An optical character recognition device comprising: a replacement unit that replaces a character code selected from a group of character code candidates, and an output unit that outputs the valid character code and the replaced character code.
JP3161216A 1991-07-02 1991-07-02 Optical character recognizing device Pending JPH0512484A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP3161216A JPH0512484A (en) 1991-07-02 1991-07-02 Optical character recognizing device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP3161216A JPH0512484A (en) 1991-07-02 1991-07-02 Optical character recognizing device

Publications (1)

Publication Number Publication Date
JPH0512484A true JPH0512484A (en) 1993-01-22

Family

ID=15730826

Family Applications (1)

Application Number Title Priority Date Filing Date
JP3161216A Pending JPH0512484A (en) 1991-07-02 1991-07-02 Optical character recognizing device

Country Status (1)

Country Link
JP (1) JPH0512484A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6651120B2 (en) 2000-09-14 2003-11-18 Fujitsu Limited Image data converting system and a storage medium thereof

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6651120B2 (en) 2000-09-14 2003-11-18 Fujitsu Limited Image data converting system and a storage medium thereof

Similar Documents

Publication Publication Date Title
JPH0512484A (en) Optical character recognizing device
JP3727422B2 (en) Character recognition apparatus and method
JP3725953B2 (en) Character recognition device
JPH07152856A (en) Optical character reader
JPH05189599A (en) Optical character reader
JP2570571B2 (en) Optical character reader
JP2848560B2 (en) Image recognition method and apparatus
JPH06333083A (en) Optical character reader
JP2924356B2 (en) Optical character reader
JP2925270B2 (en) Character reader
JPH03212779A (en) Device for recognizing character
JPH06251192A (en) Optical character reader
JPH03268089A (en) Optical character reader
JP2674475B2 (en) Character reader
JPH04242874A (en) Optical character reader
JPH09305712A (en) Method, device for recognizing character and storage medium storing program for character recognition
JPH06251187A (en) Method and device for correcting character recognition error
JPH07141462A (en) Document system
JP2002133358A (en) Document reading method
JPH08202811A (en) Character reader
JPH04255088A (en) Optical character reader
JPH08194772A (en) Optical character reader
JPH01217686A (en) Character reader
JPH0758497B2 (en) Character recognition device
JPH0830717A (en) Character recognition method and device therefor