Deprecated: The each() function is deprecated. This message will be suppressed on further calls in /home/zhenxiangba/zhenxiangba.com/public_html/phproxy-improved-master/index.php on line 456
JPH0427591B2 - - Google Patents
[go: Go Back, main page]

JPH0427591B2 - - Google Patents

Info

Publication number
JPH0427591B2
JPH0427591B2 JP57025429A JP2542982A JPH0427591B2 JP H0427591 B2 JPH0427591 B2 JP H0427591B2 JP 57025429 A JP57025429 A JP 57025429A JP 2542982 A JP2542982 A JP 2542982A JP H0427591 B2 JPH0427591 B2 JP H0427591B2
Authority
JP
Japan
Prior art keywords
character
category
corrected
reading
misread
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.)
Expired
Application number
JP57025429A
Other languages
Japanese (ja)
Other versions
JPS58142489A (en
Inventor
Yoshitake Tsuji
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 Corp
Original Assignee
Nippon Electric Co 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 Nippon Electric Co Ltd filed Critical Nippon Electric Co Ltd
Priority to JP57025429A priority Critical patent/JPS58142489A/en
Publication of JPS58142489A publication Critical patent/JPS58142489A/en
Publication of JPH0427591B2 publication Critical patent/JPH0427591B2/ja
Granted legal-status Critical Current

Links

Classifications

    • GPHYSICS
    • G06COMPUTING OR CALCULATING; COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00Arrangements for image or video recognition or understanding
    • G06V10/98Detection or correction of errors, e.g. by rescanning the pattern or by human intervention; Evaluation of the quality of the acquired patterns
    • G06V10/987Detection or correction of errors, e.g. by rescanning the pattern or by human intervention; Evaluation of the quality of the acquired patterns with the intervention of an operator

Landscapes

  • Engineering & Computer Science (AREA)
  • Quality & Reliability (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • Theoretical Computer Science (AREA)
  • Character Discrimination (AREA)

Description

【発明の詳細な説明】 本発明は文字読取装置によつて読み取つた文字
カテゴリを操作者の介入により修正する文字読取
修正装置に関するものである。
DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a character reading/correcting device that corrects character categories read by a character reading device through operator intervention.

従来、印刷あるいは手書き文字を光学的に読み
取る装置(以下OCRと称す)は、英数字及びカ
タカナを対象とするものはすでに実用化されてい
る。
Conventionally, devices for optically reading printed or handwritten characters (hereinafter referred to as OCR) have already been put into practical use for alphanumeric characters and katakana.

最近では、手書き漢字を読み取ることも研究レ
ベルにおいて、可能となつてきた。
Recently, it has become possible at the research level to read handwritten kanji.

このように、読み取り対象が手書き漢字まで拡
大されると、読み取り対象文字数の増加及び字形
の複雑さにより、十分な認識精度を得ることが困
難となつている。そのため、OCRによつて読み
取つた結果の修正を行なうことが必要となる。
As described above, when the objects to be read are expanded to include handwritten Chinese characters, it becomes difficult to obtain sufficient recognition accuracy due to the increase in the number of characters to be read and the complexity of the character shapes. Therefore, it is necessary to correct the results read by OCR.

従来の修正方法は、OCRにおいて生じた読取
り不能文字のみを対象としていた。
Traditional correction methods only target unreadable characters that occur in OCR.

また、OCRで生じる誤まつた読取りを減少さ
せるために、OCRの認識精度に応じて、複数個
出力し、出力された複数個の候補文字から、操作
員が正解文字を選択する手法を用いていた。しか
し、いずれの場合でも、OCRによる読取り結果
に正解となる文字カテゴリが含まれないと、文字
修正に時間を要し、全体的にOCRシステムの処
理能力が低下する。
In addition, in order to reduce the number of incorrect readings that occur with OCR, a method is used in which multiple candidate characters are output depending on the recognition accuracy of OCR, and the operator selects the correct character from the multiple output candidate characters. Ta. However, in any case, if the correct character category is not included in the OCR reading result, it takes time to correct the characters, and the overall processing capacity of the OCR system decreases.

本発明の目的は上記従来の欠点を筆記者の
OCRにおける読取傾向を記憶することによつて、
解決し、全体的にOCRの処理能力を向上させる
ようにした文字読取修正装置を提供することにあ
る。
The purpose of the present invention is to solve the above-mentioned drawbacks of the conventional scribe.
By memorizing reading trends in OCR,
The object of the present invention is to provide a character reading/correcting device which solves the above problems and improves the overall OCR processing ability.

本発明の他の目的は、従来に用いられて帳票上
の文字の他に文章で書かれた文字を対象とした場
合、筆記者各人によつて、OCRの文字読取りに
一定の傾向を持つことに着目し全体的にOCRの
処理能力を向上させるようにした文字読取修正装
置を提供することにある。
Another object of the present invention is to have a certain tendency in OCR character reading depending on each scribe when targeting characters written in sentences in addition to characters on a form. The object of the present invention is to provide a character reading/correcting device which focuses on this and improves the overall OCR processing ability.

本発明によれば、紙面上の文字を読取る文字読
取装置によつて読み取つた文字カテゴリを操作者
の介入により修正する文字読取修正装置におい
て、文字読取装置によつて誤まつて読取つた文字
カテゴリ(以下誤読文字カテゴリと称す)を操作
者を介して修正し、修正した文字カテゴリ(以下
修正文字カテゴリと称す)と誤読文字カテゴリを
予め設定した第1のキーを打鍵することによつ
て、エラーリスト記憶メモリへ登録する手段と、
予め設定した第2のキーを打鍵することによつ
て、エラーリスト記憶メモリから文字読取装置に
よつて生じた誤読文字カテゴリに対応する修正文
字カテゴリを検出し、検出された修正文字カテゴ
リを順次表示する候補文字検出手段と、候補文字
検出手段によつて検出された修正文字カテゴリを
用いて、読取修正を行なう手段とを有する文字読
取修正装置が得られる。
According to the present invention, in a character reading correction device that corrects a character category read by a character reading device that reads characters on a sheet of paper through operator intervention, the character category read incorrectly by the character reading device ( By correcting the character category (hereinafter referred to as the misread character category) through the operator and pressing the first key in which the corrected character category (hereinafter referred to as the corrected character category) and the misread character category are set in advance, the error list is displayed. a means for registering in a storage memory;
By pressing a second key set in advance, a corrected character category corresponding to the misread character category generated by the character reading device is detected from the error list storage memory, and the detected corrected character categories are sequentially displayed. According to the present invention, there is obtained a character reading correction device having a candidate character detection means for detecting a candidate character, and a means for performing reading correction using a correction character category detected by the candidate character detection means.

以下本発明における具体的一実施例を示す論理
ブロツク図を用いて説明する。図において、1は
OCRである。2は、OCR1において読み取られ
た文字カテゴリを格納する読取文字格納バツフア
である。尚、読取文字格納バツフア2は、複数行
分の入力文字画像に対する読取結果を格納するも
のとする。3は認識制御装置である。認識制御装
置は3は通常のマイクロコンピユータを用いるの
が適当である。
A specific embodiment of the present invention will be explained below using a logic block diagram. In the figure, 1 is
It is OCR. Reference numeral 2 denotes a read character storage buffer that stores character categories read by OCR1. It is assumed that the read character storage buffer 2 stores the read results for input character images of a plurality of lines. 3 is a recognition control device. As the recognition control device 3, it is appropriate to use an ordinary microcomputer.

OCR1によつて読み取られた文字カテゴリが
読取文字格納バツフア2へ順次転送されると、認
識制御装置3は、読取文字格納バツフア2から順
次、文字カテゴリを読み出し、表示制御装置4は
文字カテゴリに対応する文字パターンを発生さ
せ、表示装置5に順次転送し、表示する。尚、
OCR1によつて読み取られた文字カテゴリを表
示するにあたり、OCR1に内蔵されたパターン
メモリ(図中省略)から読取結果に対応する入力
文字画像を同時表示することも可能である。
When the character categories read by the OCR 1 are sequentially transferred to the read character storage buffer 2, the recognition control device 3 sequentially reads out the character categories from the read character storage buffer 2, and the display control device 4 corresponds to the character categories. A character pattern is generated, sequentially transferred to the display device 5, and displayed. still,
In displaying the character categories read by the OCR 1, it is also possible to simultaneously display input character images corresponding to the reading results from a pattern memory (not shown) built into the OCR 1.

6は鍵盤である。鍵盤6には予め適当な4個の
キーに特別な符号を割り当てて、誤読文字登録キ
ー61、候補文字検出キー62、修正キー63確
認キー64としたものである。
6 is a keyboard. On the keyboard 6, special codes are assigned in advance to four appropriate keys, which are a misread character registration key 61, a candidate character detection key 62, a correction key 63, and a confirmation key 64.

尚、図において、鍵盤を用いるとしたが、タブ
レツト等の入力装置を用いても良い。
Although a keyboard is used in the figure, an input device such as a tablet may also be used.

7は符号変換回路である。符号変換回路7は鍵
盤6上の誤読文字登録キー61、候補文字検出キ
ー62、修正キー63、確認キー64を打鍵した
か否かを検出する。
7 is a code conversion circuit. The code conversion circuit 7 detects whether or not a misread character registration key 61, a candidate character detection key 62, a correction key 63, and a confirmation key 64 on the keyboard 6 are pressed.

即ち、符号変換回路7は、誤読文字登録キー6
1の打鍵を示す符号が入力されると、登録指令信
号(図中省略)を認識制御装置3へ出力し、候補
文字検出キー13の打鍵を示す符号が入力される
と、候補文字検出指令信号(図中省略)を認識制
御装置3へ出力し、修正キー63の打鍵を示す符
号が入力されると、修正指令信号(図中省略)を
認識制御装置3へ出力し、確認キー64の打鍵を
示す符号が入力されると、確認指令信号(図中省
略)を認識制御装置3へ出力する。
That is, the code conversion circuit 7 converts the misread character registration key 6
When the code indicating the key press 1 is input, a registration command signal (not shown in the figure) is output to the recognition control device 3, and when the code indicating the press of the candidate character detection key 13 is input, the candidate character detection command signal is output. (not shown in the figure) is output to the recognition control device 3, and when a code indicating the keystroke of the correction key 63 is input, a correction command signal (not shown in the figure) is output to the recognition control device 3, and the keystroke of the confirmation key 64 is input. When the code indicating , is input, a confirmation command signal (not shown) is output to the recognition control device 3 .

そこで、操作員が、表示装置5上の読取結果の
確認を行ない、誤読文字の修正を行なうにあた
り、本発明では、筆記者のOCR1による読取傾
向を記憶したエラーリスト記憶メモリ8が用いら
れる。
Therefore, when the operator checks the reading results on the display device 5 and corrects the misread characters, the present invention uses an error list storage memory 8 that stores the scribe's reading tendency based on the OCR 1.

即ちエラーリスト記憶メモリ8にはOCR1に
よつて読み取つた結果、誤読を生じた文字カテゴ
リと修正された文字カテゴリが記憶される。
That is, the error list storage memory 8 stores character categories that have been misread as a result of reading by OCR 1 and corrected character categories.

エラーリスト記憶メモリ8へ誤読文字カテゴリ
と修正文字カテゴリを登録するには、上述したよ
うに、鍵盤6上の誤読文字登録キー61を打鍵す
ることによつて、認識制御装置3へ登録指令信号
を出力し、認識制御装置3によつて読取文字格納
バツフア2に記憶された誤読文字カテゴリと操作
者によつて修正された修正文字カテゴリとをエラ
ー記憶メモリ8へ登録する。尚、エラーリスト記
憶メモリ8への誤読文字カテゴリと修正文字カテ
ゴリの登録は、予め、テストデータを用いて、行
なつても良い。
To register the misread character category and corrected character category in the error list storage memory 8, as described above, by pressing the misread character registration key 61 on the keyboard 6, a registration command signal is sent to the recognition control device 3. The misread character category stored in the read character storage buffer 2 by the recognition control device 3 and the corrected character category corrected by the operator are registered in the error storage memory 8. Incidentally, the misread character category and the corrected character category may be registered in the error list storage memory 8 in advance using test data.

次に、操作者がエラーリスト記憶メモリ8を用
いて、誤読文字カテゴリの修正を行なう方法を説
明する。操作者が表示画面上の読み取り結果の確
認を行ない、誤読文字カテゴリを発見すると、鍵
盤6を用い、カーソルを誤読文字カテゴリ位置ま
で移動させる。次に操作員が鍵盤6上の候補文字
検出キー62を打鍵すると、符号変換回路6は、
上述したように、候補文字検出指令信号を認識制
御装置3に転送する。認識制御装置3は候補文字
検出指令信号が入力されると、表示画面上のカー
ソル位置で示された誤読文字カテゴリに対応する
修正文字カテゴリをエラー記憶メモリ8より検索
し、得られた修正文字カテゴリを順次表示制御装
置4を介して、表示装置5上の表示画面上に表示
する。操作者は、表示画面上に表示された修正文
字カテゴリに正解があれば、鍵盤6上の修正キー
63を打鍵することによつて、修正指令信号が認
識制御装置3へ転送され、認識制御装置3によつ
て表示画面上の誤読文字カテゴリと修正文字カテ
ゴリとの置換が行なわれる。
Next, a method for the operator to use the error list storage memory 8 to correct misread character categories will be explained. When the operator checks the reading results on the display screen and finds a misread character category, he uses the keyboard 6 to move the cursor to the misread character category position. Next, when the operator presses the candidate character detection key 62 on the keyboard 6, the code conversion circuit 6
As described above, the candidate character detection command signal is transferred to the recognition control device 3. When the candidate character detection command signal is input, the recognition control device 3 searches the error storage memory 8 for a corrected character category corresponding to the misread character category indicated by the cursor position on the display screen, and selects the corrected character category obtained from the error storage memory 8. are sequentially displayed on the display screen of the display device 5 via the display control device 4. If the operator has a correct answer in the corrected character category displayed on the display screen, by pressing the correction key 63 on the keyboard 6, a correction command signal is transferred to the recognition control device 3, and the recognition control device 3, the misread character category on the display screen is replaced with the corrected character category.

表示画面上の読取り結果の修正が終了し、操作
者が鍵盤6上の確認キー64を打鍵することによ
つて、表示画面上の読取り結果が外部記憶装置9
に転送され格納される。
When the correction of the reading result on the display screen is completed and the operator presses the confirmation key 64 on the keyboard 6, the reading result on the display screen is saved in the external storage device 9.
is transferred and stored.

尚、図における符号変換回路7の機能は認識制
御装置3と別なものとして説明したが、認識制御
装置3へ組み込むことも可能である。
Although the function of the code conversion circuit 7 in the figure has been explained as being separate from the recognition control device 3, it can also be incorporated into the recognition control device 3.

以上述べたように、本発明によつて、筆記者の
OCRにおける読取傾向を利用することによつて、
OCRの認識精度を向上させることが可能となり、
全体的にOCRの処理能力が向上する。
As described above, the present invention allows scribes to
By using reading trends in OCR,
It is now possible to improve OCR recognition accuracy,
Overall OCR processing power is improved.

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

図は本発明における一実施例を示す論理ブロツ
ク図である。 図において、1はOCR、2は読取文字格納バ
ツフア、3は認識制御装置、4は表示制御装置、
5は表示装置、6は鍵盤、7は符号変換回路、8
はエラーリスト記憶メモリ、9は外部記憶装置で
ある。
The figure is a logic block diagram showing one embodiment of the present invention. In the figure, 1 is an OCR, 2 is a reading character storage buffer, 3 is a recognition control device, 4 is a display control device,
5 is a display device, 6 is a keyboard, 7 is a code conversion circuit, 8
9 is an error list storage memory, and 9 is an external storage device.

Claims (1)

【特許請求の範囲】[Claims] 1 紙面上の文字を読取る文字読取装置によつて
読み取つた文字カテゴリを、操作者の介入により
修正する文字読取修正装置において、前記文字読
取装置によつて誤まつて読み取つた文字カテゴリ
(以下誤読文字カテゴリと称す)を操作者を介し
て修正し、修正した文字カテゴリ(以下修正文字
カテゴリと称す)と前記誤読文字カテゴリとを予
め設定した第1のキーを打鍵することによつて、
エラーリスト記憶メモリへ登録する手段と、予め
設定した第2のキーを打鍵することによつて、前
記エラーリスト記憶メモリより前記文字読取装置
によつて生じた前記誤読文字カテゴリに対応する
前記修正文字カテゴリを検出し、検出された前記
修正文字カテゴリを順次表示する候補文字検出手
段と、前記候補文字検出手段によつて検出された
前記修正文字カテゴリを用いて読取修正を行なう
手段とを有することを特徴とする文字読取修正装
置。
1. In a character reading correction device that corrects character categories read by a character reading device that reads characters on paper through the intervention of an operator, character categories that are incorrectly read by the character reading device (hereinafter referred to as misread characters) (hereinafter referred to as "corrected character category") is corrected by the operator, and by pressing a first key in which the corrected character category (hereinafter referred to as "corrected character category") and the misread character category are set in advance,
means for registering in an error list storage memory, and by pressing a preset second key, the corrected character corresponding to the misread character category generated by the character reading device from the error list storage memory; Candidate character detection means for detecting a category and sequentially displaying the detected corrected character categories, and means for performing reading correction using the corrected character categories detected by the candidate character detection means. Characteristic character reading correction device.
JP57025429A 1982-02-18 1982-02-18 Character reading and correcting device Granted JPS58142489A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP57025429A JPS58142489A (en) 1982-02-18 1982-02-18 Character reading and correcting device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP57025429A JPS58142489A (en) 1982-02-18 1982-02-18 Character reading and correcting device

Publications (2)

Publication Number Publication Date
JPS58142489A JPS58142489A (en) 1983-08-24
JPH0427591B2 true JPH0427591B2 (en) 1992-05-12

Family

ID=12165721

Family Applications (1)

Application Number Title Priority Date Filing Date
JP57025429A Granted JPS58142489A (en) 1982-02-18 1982-02-18 Character reading and correcting device

Country Status (1)

Country Link
JP (1) JPS58142489A (en)

Also Published As

Publication number Publication date
JPS58142489A (en) 1983-08-24

Similar Documents

Publication Publication Date Title
US4566039A (en) Facsimile system
US4132976A (en) Operator readable and machine readable character recognition systems
JP2740335B2 (en) Table reader with automatic cell attribute determination function
JPH08329187A (en) Document reader
JPH0427591B2 (en)
JPS5841542B2 (en) optical character reader
US5984189A (en) Sheet for data codes and method of recognizing these codes
JP2676993B2 (en) Character reading correction device
JPS592191A (en) Recognizing and processing system of handwritten japanese sentence
JPH01222393A (en) Information code sheet and its code recognizing method
JPS61272882A (en) Information recognizing device
JPS61150081A (en) Character recognizing device
JP2578747B2 (en) Handwritten information processing method
JPS5875278A (en) Character and symbol recognizing device
JP2578748B2 (en) Handwritten information processing method
JPS58125182A (en) Character reader
JPH0661112B2 (en) Pattern recognition method
JPS5668869A (en) Optical character reader
JPH09138835A (en) Character recognition device
JPS62123579A (en) Character recognizing device
JPS62295192A (en) Optical character image reader
JP3310063B2 (en) Document processing device
JPH0362280A (en) Character reading device
JPS6095689A (en) Optical character reader
JP2706086B2 (en) Character recognition device