JPH04256193A - Word register processing system for recognition word - Google Patents

Word register processing system for recognition word

Info

Publication number
JPH04256193A
JPH04256193A JP3017313A JP1731391A JPH04256193A JP H04256193 A JPH04256193 A JP H04256193A JP 3017313 A JP3017313 A JP 3017313A JP 1731391 A JP1731391 A JP 1731391A JP H04256193 A JPH04256193 A JP H04256193A
Authority
JP
Japan
Prior art keywords
word
registered
dictionary file
register
character
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.)
Withdrawn
Application number
JP3017313A
Other languages
Japanese (ja)
Inventor
Takeshi Hasegawa
健 長谷川
Yoshihisa Fujishima
敬久 藤島
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.)
Fujitsu Ltd
Original Assignee
Fujitsu 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 Fujitsu Ltd filed Critical Fujitsu Ltd
Priority to JP3017313A priority Critical patent/JPH04256193A/en
Publication of JPH04256193A publication Critical patent/JPH04256193A/en
Withdrawn legal-status Critical Current

Links

Landscapes

  • Character Discrimination (AREA)
  • Document Processing Apparatus (AREA)

Abstract

PURPOSE:To improve the register efficiency of a word dictionary file by registering a minimally necessary word in the word dictionary file since there are problems such as register repetition or the register omission of the word which should be registered in the register processing of the character-recognized word. CONSTITUTION:A word selection processing part 2 displays the word which is character-recognized and decided by a specified method in a display device 4, reads the specified word from the display device 4 by an input command and informs a word register processing part 3 of the word. The word register processing part 3 is constituted to retrieve the word dictionary file 5 with the word as a key and to register the word in the word dictionary file 5 unless the word is registered.

Description

【発明の詳細な説明】[Detailed description of the invention]

【0001】0001

【産業上の利用分野】本発明は認識文字の処理方式に関
する。手書き文字や活字を読み取る文字認識処理におい
て、金融機関等で定常的に入力処理される顧客名や顧客
の住所等を構成する単語は、予め単語辞書ファイルに登
録しておくことにより文字認識処理の品質並びに処理効
率の向上が図られているが、誤認識された単語は該単語
を修正した後に該単語が単語辞書ファイルへ追加登録さ
れるため、単語辞書ファイルに未登録の単語が正常に認
識された場合は該単語の登録の欠落が生じ、また単語フ
ァイルに登録済の単語が誤認識されると該単語の重複登
録が生じるといった問題があった。単語辞書ファイルの
重複登録や登録の欠落を無くすことが要請されている。
BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for processing recognized characters. In character recognition processing for reading handwritten characters and printed characters, words that make up customer names, customer addresses, etc. that are regularly input at financial institutions etc. can be registered in advance in a word dictionary file to facilitate character recognition processing. Although efforts are being made to improve the quality and processing efficiency, words that are incorrectly recognized are added to the word dictionary file after being corrected, so words that are not registered in the word dictionary file may not be correctly recognized. If a word is registered in the word file, the word may be registered incorrectly, and if a word already registered in the word file is incorrectly recognized, the word may be registered redundantly. There is a need to eliminate duplicate registrations and missing registrations in word dictionary files.

【0002】0002

【従来の技術】図5に従来のシステム構成図を、図6に
従来の処理手順を示す。処理装置20は、帳票21に記
載された文字を読み取り、処理ステップ91で文字認識
部15は、帳票記載の文字認識を行い、該文字認識の結
果生成される文字コードを候補文字として作成する。図
4に認識対象の帳票と認識結果の説明図を示す。
2. Description of the Related Art FIG. 5 shows a conventional system configuration diagram, and FIG. 6 shows a conventional processing procedure. The processing device 20 reads the characters written on the form 21, and in processing step 91, the character recognition unit 15 performs character recognition on the form, and creates a character code generated as a result of the character recognition as a candidate character. FIG. 4 shows an explanatory diagram of the form to be recognized and the recognition result.

【0003】図4(a)において帳票に記載されている
名前の欄の文字「中」と「川」が文字認識され、同じく
住所の欄の各文字もそれぞれ認識され、それぞれ認識さ
れた文字が標準の文字パターンと比較照合され、標準の
文字との距離(類似度)が所定の値に収まる文字が、文
字認識部15が文字カテゴリーとして保持する文字の中
から抽出されて認識結果の文字(文字コード)として生
成される。即ち図4(b)に示すように類似度の高い順
番に認識結果文字が例えば、第1位候補「中」,「川」
に対し第2位候補「牛」、「ツ」と決まる。住所の欄の
各文字もそれぞれ認識の文字が決まる。
In FIG. 4(a), the characters ``Naka'' and ``Kawa'' in the name field written on the form are recognized, and each character in the address field is also recognized. Characters that are compared with the standard character pattern and whose distance (similarity) to the standard character falls within a predetermined value are extracted from the characters held as character categories by the character recognition unit 15, and the characters ( character code). That is, as shown in FIG. 4(b), the recognition result characters are ranked in order of similarity, for example, the first candidate "Naka" and "Kawa".
However, the second candidates were decided as ``Ushi'' and ``Tsu.'' Recognition characters are determined for each character in the address field.

【0004】処理ステップ92で文字認識部15は、第
1位候補、第2位候補として決まった認識結果を候補単
語として単語選択処理部25へ渡す。処理ステップ93
で単語選択処理部25は、該候補単語をディスプレイ装
置45に表示する。
In processing step 92, the character recognition section 15 passes the recognition results determined as the first and second candidates to the word selection processing section 25 as candidate words. Processing step 93
Then, the word selection processing unit 25 displays the candidate word on the display device 45.

【0005】表示された候補単語は処理ステップ94で
、オペレータが帳票記載の文字と同一か否かを確認し、
修正が必要ならば表示内容を更新して、ディスプレイ装
置45から入力する。処理ステップ95で単語選択処理
部25は、ディスプレイ装置45からオペレータ確認後
の候補単語データを読込み、表示内容に変更があったか
否かをチェックする。表示内容が更新された場合は処理
ステップ96で、更新された単語の登録を単語登録処理
部35へ依頼する。処理ステップ97で単語登録処理部
35は、更新された単語を単語辞書ファイル55へ追加
登録する。表示した候補単語が更新されていなければ単
語の追加登録は行われない。
[0005] In processing step 94, the operator checks whether the displayed candidate word is the same as the character written on the form.
If corrections are necessary, the displayed contents are updated and input from the display device 45. In processing step 95, the word selection processing unit 25 reads candidate word data after operator confirmation from the display device 45, and checks whether or not there has been a change in the displayed content. If the displayed content has been updated, in processing step 96, the word registration processing unit 35 is requested to register the updated word. In processing step 97, the word registration processing section 35 additionally registers the updated word in the word dictionary file 55. If the displayed candidate words have not been updated, no additional words will be registered.

【0006】[0006]

【発明が解決しようとする課題】文字認識処理において
、文字認識の結果生成された単語(候補単語)がオペレ
ータによりチェックされ、修正があった単語のみが単語
登録処理部35によって単語辞書ファイル55に追加登
録されるために、単語辞書ファイル55に未登録で登録
の必要な単語が、正常認識された場合に登録されなかっ
たり、登録済の単語であっても誤認識されてディスプレ
イ装置45より修正後に入力されると、該入力された単
語が重複して単語辞書ファイル55へ登録されるといっ
た欠点があった。
[Problem to be Solved by the Invention] In character recognition processing, words (candidate words) generated as a result of character recognition are checked by an operator, and only the words that have been corrected are stored in the word dictionary file 55 by the word registration processing section 35. In order to be additionally registered, a word that is not registered in the word dictionary file 55 and needs to be registered may not be registered even if it is recognized normally, or even a registered word may be incorrectly recognized and corrected by the display device 45. If the word is input later, there is a drawback that the input word is registered in the word dictionary file 55 in duplicate.

【0007】本発明は必要最小限の単語を単語辞書ファ
イルに登録することを目的とする。
An object of the present invention is to register the minimum necessary number of words in a word dictionary file.

【0008】[0008]

【課題を解決するための手段】図1は本発明の原理ブロ
ック図である。単語選択処理部2と、単語登録処理部3
と、ディスプレイ装置4と、文字認識される単語が登録
されている単語辞書ファイル5とを有し、単語選択処理
部2は、文字認識されて所定の方法で決まる単語をディ
スプレイ装置4に表示し、入力指令で該ディスプレイ装
置4から所定の単語を読込み、該単語を単語登録処理部
3へ通知し、単語登録処理部3は、該単語をキーとして
単語辞書ファイル5を検索し、該単語が登録されていな
ければ該単語を単語辞書ファイル5に登録する。
[Means for Solving the Problems] FIG. 1 is a block diagram of the principle of the present invention. Word selection processing unit 2 and word registration processing unit 3
, a display device 4, and a word dictionary file 5 in which words to be character-recognized are registered, and the word selection processing unit 2 displays words that are determined by character recognition and determined by a predetermined method on the display device 4. , reads a predetermined word from the display device 4 in response to an input command, notifies the word registration processing unit 3 of the word, and the word registration processing unit 3 searches the word dictionary file 5 using the word as a key, and searches the word dictionary file 5 using the word as a key. If the word is not registered, the word is registered in the word dictionary file 5.

【0009】[0009]

【作用】文字認識された文字から単語選択処理部2は、
第1位候補の候補単語をディスプレイ装置4に表示して
文字認識が正常か否かの確認をとり、オペレータの入力
指示を受けて読込んだ文字が単語辞書ファイル5に未登
録の文字のみを単語辞書ファイル5へ登録することによ
り、登録効率の最適化が可能となる。
[Operation] The word selection processing unit 2 from the recognized characters,
The candidate word of the first candidate is displayed on the display device 4 to confirm whether character recognition is normal or not, and only the characters read in response to input instructions from the operator are not registered in the word dictionary file 5. By registering in the word dictionary file 5, it is possible to optimize the registration efficiency.

【0010】0010

【実施例】図2に実施例のシステム構成図を示し、図3
に実施例の処理手順を示す。処理装置10は、帳票11
に記載された文字を読み取り、処理ステップ81で文字
認識部1は、帳票記載の文字認識を行い、該文字認識の
結果生成される文字コードを候補文字として作成する。 即ち図4(a)の帳票の文字認識が行われた結果、図4
(b)の如く第1位候補と第2位候補の候補単語が決ま
る。
[Example] Figure 2 shows a system configuration diagram of the example, and Figure 3
shows the processing procedure of the example. The processing device 10 processes the form 11
In processing step 81, the character recognition unit 1 performs character recognition on the form, and creates a character code generated as a result of the character recognition as a candidate character. That is, as a result of character recognition of the form shown in Figure 4(a),
As shown in (b), the first and second candidate words are determined.

【0011】処理ステップ82で文字認識部1は、候補
単語を単語選択処理部2へ渡す。処理ステップ83で単
語選択処理部2は、該候補単語をディスプレイ装置4に
表示する。表示された候補単語は処理ステップ84で、
オペレータが帳票記載の文字と同一か否かを確認し、修
正が必要ならば表示内容を更新して、ディスプレイ装置
4から入力する。
[0011] In processing step 82, the character recognition section 1 passes the candidate word to the word selection processing section 2. In processing step 83, the word selection processing unit 2 displays the candidate word on the display device 4. The displayed candidate words are processed in processing step 84.
The operator checks whether the characters are the same as those written on the form, and if corrections are necessary, the displayed contents are updated and inputted from the display device 4.

【0012】処理ステップ85で単語選択処理部2は、
ディスプレイ装置4からオペレータ確認後の候補単語デ
ータを読込み単語登録処理部3へ渡す。処理ステップ8
6で単語登録処理部3は、候補単語が既に単語辞書ファ
イル5に登録されているか否かを単語毎にチェックし、
未登録の単語があれば単語辞書ファイル5に登録する。
[0012] In processing step 85, the word selection processing section 2
Candidate word data after operator confirmation is read from the display device 4 and passed to the word registration processing section 3. Processing step 8
In step 6, the word registration processing unit 3 checks for each word whether the candidate word has already been registered in the word dictionary file 5,
If there are unregistered words, they are registered in the word dictionary file 5.

【0013】[0013]

【発明の効果】文字が文字認識された後の単語辞書ファ
イル5への単語登録において、認識された単語が既に単
語辞書ファイル5に登録済か否かをチェックすることに
より、単語登録の重複や登録の欠落を防ぐことが出来る
Effects of the Invention: When registering a word in the word dictionary file 5 after a character is recognized, by checking whether the recognized word has already been registered in the word dictionary file 5, duplicate word registration can be avoided. This can prevent registration failures.

【図面の簡単な説明】[Brief explanation of the drawing]

【図1】  本発明の原理ブロック図[Figure 1] Block diagram of the principle of the present invention

【図2】  実施例のシステム構成図[Figure 2] System configuration diagram of the embodiment

【図3】  実施例の処理手順[Figure 3] Processing procedure of the example

【図4】  認識対象の帳票と認識結果の説明図[Figure 4] Explanatory diagram of forms to be recognized and recognition results

【図5
】  従来のシステム構成図
[Figure 5
] Conventional system configuration diagram

【図6】  従来の処理手順[Figure 6] Conventional processing procedure

【符号の説明】[Explanation of symbols]

1、15は文字認識部、10、20は処理装置、11、
21は帳票、2、25は単語選択処理部、3、35は単
語登録処理部、4、45はディスプレイ装置、5、55
は単語辞書ファイル、81〜86、91〜97は処理ス
テップを表す。
1 and 15 are character recognition units; 10 and 20 are processing units; 11;
21 is a form; 2 and 25 are word selection processing units; 3 and 35 are word registration processing units; 4 and 45 are display devices; 5 and 55
represents a word dictionary file, and 81 to 86 and 91 to 97 represent processing steps.

Claims (1)

【特許請求の範囲】[Claims] 【請求項1】  単語選択処理部(2)と、単語登録処
理部(3)と、ディスプレイ装置(4)と、文字認識さ
れる単語が登録されている単語辞書ファイル(5)とを
有し、単語選択処理部(2)は、文字認識されて所定の
方法で決まる単語をディスプレイ装置(4)に表示し、
入力指令で該ディスプレイ装置(4)から所定の単語を
読込み、該単語を単語登録処理部(3)へ通知し、単語
登録処理部(3)は、該単語をキーとして単語辞書ファ
イル(5)を検索し、該単語が登録されていなければ該
単語を単語辞書ファイル(5)に登録することを特徴と
する認識文字の単語登録処理方式。
1. A word selection processing unit (2), a word registration processing unit (3), a display device (4), and a word dictionary file (5) in which words to be character recognized are registered. , a word selection processing unit (2) displays words determined by character recognition in a predetermined manner on a display device (4);
A predetermined word is read from the display device (4) in response to an input command, the word is notified to the word registration processing section (3), and the word registration processing section (3) uses the word as a key to create a word dictionary file (5). A word registration processing method for recognized characters, characterized in that the word is searched for, and if the word is not registered, the word is registered in a word dictionary file (5).
JP3017313A 1991-02-08 1991-02-08 Word register processing system for recognition word Withdrawn JPH04256193A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP3017313A JPH04256193A (en) 1991-02-08 1991-02-08 Word register processing system for recognition word

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP3017313A JPH04256193A (en) 1991-02-08 1991-02-08 Word register processing system for recognition word

Publications (1)

Publication Number Publication Date
JPH04256193A true JPH04256193A (en) 1992-09-10

Family

ID=11940525

Family Applications (1)

Application Number Title Priority Date Filing Date
JP3017313A Withdrawn JPH04256193A (en) 1991-02-08 1991-02-08 Word register processing system for recognition word

Country Status (1)

Country Link
JP (1) JPH04256193A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH06251204A (en) * 1993-02-26 1994-09-09 Nec Corp Character recognition device

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH06251204A (en) * 1993-02-26 1994-09-09 Nec Corp Character recognition device

Similar Documents

Publication Publication Date Title
US5752020A (en) Structured document retrieval system
US5161245A (en) Pattern recognition system having inter-pattern spacing correction
JPH05258099A (en) Character recognition processor
JP5669041B2 (en) Document processing apparatus and document processing method
JPH04256193A (en) Word register processing system for recognition word
JPH064717A (en) Kanji address correction processing method
JPH10105654A (en) Character recognition device for form
JP2829002B2 (en) Character recognition device
JP2002032704A (en) System and method for document processing, storage medium, and document
JP2932667B2 (en) Information retrieval method and information storage device
JP3455643B2 (en) Method of updating learning dictionary in character recognition device and character recognition device
JPH10302025A (en) Handwritten character recognizing device and its program recording medium
JP3335494B2 (en) Character recognition processor
JP2000132635A (en) Recognizing data confirming method
JP3045886B2 (en) Character processing device with handwriting input function
JPH05135212A (en) Address and word collation method
JP2839515B2 (en) Character reading system
JP3001618B2 (en) How to copy characters on paper and how to recognize symbols
JPH04283853A (en) Japanese syllabary-to-chinese character converting method
JP3058706B2 (en) How to convert address kana to kanji
JPH02103690A (en) Character reading device
JPH1196224A (en) Method for inspecting application contents and system for issuing certificate
JPH07210623A (en) Document picture processor
JPH09204473A (en) Slip managing device
JP2865443B2 (en) Kanji conversion device for Kana name or Kana corporation name

Legal Events

Date Code Title Description
A300 Application deemed to be withdrawn because no request for examination was validly filed

Free format text: JAPANESE INTERMEDIATE CODE: A300

Effective date: 19980514