JPH065551B2 - Print character recognition method - Google Patents
Print character recognition methodInfo
- Publication number
- JPH065551B2 JPH065551B2 JP61270867A JP27086786A JPH065551B2 JP H065551 B2 JPH065551 B2 JP H065551B2 JP 61270867 A JP61270867 A JP 61270867A JP 27086786 A JP27086786 A JP 27086786A JP H065551 B2 JPH065551 B2 JP H065551B2
- Authority
- JP
- Japan
- Prior art keywords
- character
- spacing
- document
- specific area
- line
- 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 - Lifetime
Links
- 238000000034 method Methods 0.000 title claims description 4
- 238000010586 diagram Methods 0.000 description 4
- 239000000872 buffer Substances 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
Landscapes
- Character Input (AREA)
Description
【発明の詳細な説明】 (イ)産業上の利用分野 本発明は、タイミングマークや文字枠が印刷されていな
い白紙の入力用紙を用い、該入力用紙に認識させようと
する文字を印刷して、この印刷文字を認識する印刷文字
認識方式に関する。DETAILED DESCRIPTION OF THE INVENTION (A) Field of Industrial Application The present invention uses a blank input sheet on which timing marks and character frames are not printed, and prints characters to be recognized on the input sheet. , A print character recognition method for recognizing this print character.
(ロ)従来の技術 特開昭６１−９５４８３号公報には、入力用紙を先ず行
方向に補助走査して２値イメージを得、この２値イメー
ジから行間隔を推定し、次に入力用紙を主走査して２値
イメージを得、この２値イメージから推定した行間隔を
用いて文字行を分離し、分離された１行分の２値イメー
ジから更に文字間隔を推定し、推定した文字間隔を用い
て文字の切出しを行なう技術が開示されている。(B) Prior Art In Japanese Patent Laid-Open No. 61-95483, an input sheet is first auxiliary scanned in the row direction to obtain a binary image, the line spacing is estimated from the binary image, and then the input sheet is A binary image is obtained by main scanning, character lines are separated using the line spacing estimated from this binary image, the character spacing is further estimated from the separated binary image for one line, and the estimated character spacing is calculated. A technique for cutting out a character by using is disclosed.
(ハ)発明が解決しようとする問題点 従来より、走査して得た２値イメージから、行間隔や文
字間隔を推定することは行なわれていたが、この推定の
ための走査の対象は、入力用紙に印刷された認識しよう
とする文字そのものであった。従って、推定の基準とな
る文字列は、複数種類の異なった文字より成る文字列で
あって、このために、正確な行間隔及び文字間隔の推定
ができず、更には、行間隔を推定するために、特別な補
助走査を行なわなくてはならなかった。(C) Problems to be Solved by the Invention Conventionally, line spacing and character spacing have been estimated from a binary image obtained by scanning, but the target of scanning for this estimation is It was the characters that were printed on the input paper and were to be recognized. Therefore, the character string used as the estimation reference is a character string composed of a plurality of different characters. For this reason, it is not possible to accurately estimate the line spacing and the character spacing, and further to estimate the line spacing. Therefore, a special auxiliary scan had to be performed.
(ニ)問題点を解決するための手段 本発明は、認識させようとする文書を入力用紙の特定エ
リア以降に印刷する際、同時に、前記特定エリアに同一
の特定文字を複数行に亙って１行当り複数文字印刷する
ようにし、前記入力用紙に印刷された前記文書を認識す
る際、該文書及び前記特定エリアの内容を２値イメージ
に変換して画像メモリに記憶し、該画像メモリの前記特
定エリアに対応する範囲内の２値イメージに基づいて文
字間隔及び行間隔を推定し、該推定した文字間隔及び行
間隔を用いて前記文書を構成する文字の切出しを行な
い、該切出された文字を認識し、前記入力用紙の特定エ
リアの印刷内容を除く前記文書内容についてのみ、認識
結果を出力するようにして、上記問題点を解決するもの
である。(D) Means for Solving Problems When the document to be recognized is printed after the specific area of the input sheet, at the same time, the same specific character is spread over a plurality of lines in the specific area. When recognizing the document printed on the input sheet by printing a plurality of characters per line, the contents of the document and the specific area are converted into a binary image and stored in an image memory, Character spacing and line spacing are estimated based on a binary image within a range corresponding to the specific area, and the characters constituting the document are cut out using the estimated character spacing and line spacing, and the cut out is performed. The above problem is solved by recognizing the characters and outputting the recognition result only for the document contents except the print contents in the specific area of the input paper.
(ホ)作用 本発明では、認識させようとする文書の前の特定エリア
に、特定文字を複数行に亙って１行当り複数文字印刷す
るようにし、この特定エリアに対応する範囲内の２値イ
メージに基づいて文字間隔及び行間隔を推定するように
したので、認識させようとする文書内容に関係なく、常
に同一の文字列に基づいて文字間隔及び行間隔を推定で
き、従って、安定して正確な推定が実行できる。又、特
定エリアへの特定文字の印刷は、予め行なっておくので
はなく、文書を印刷する際同時に行なうようにしたの
で、文書と同一の文字間隔及び行間隔で、確実に特定文
字が印刷され、推定がより正確となる。(E) Operation In the present invention, a specific character is printed over a plurality of lines in a specific area in front of a document to be recognized, and a plurality of characters are printed per line. Since the character spacing and line spacing are estimated based on the value image, the character spacing and line spacing can always be estimated based on the same character string regardless of the document content to be recognized, and therefore stable. Accurate estimation can be performed. Further, the printing of the specific characters in the specific area is not performed in advance but is performed at the same time when the document is printed. Therefore, the specific characters are surely printed at the same character spacing and line spacing as the document. , The estimation will be more accurate.
(ヘ)実施例 第１図は本発明における文字認識装置の構成を示すブロ
ック図であり、(1)は入力用紙(2)を走査して２値イメー
ジを得る文字観測部、(3)は得られた２値イメージを記
憶する画像メモリ、(4)及び(5)は画像メモリ(3)の２値
イメージから行間隔Ｙ0及び文字間隔Ｘ0を各々推定し、
バッファ(6)及び(7)に推定結果を記憶する行間隔推定部
及び文字間隔推定部、(8)は推定された行間隔Ｙ0及び文
字間隔Ｘ0を用いて、画像メモリ(3)の２値イメージから
文字を切出す文字切出し部、(9)は切出された文字の認
識を行なう認識部、(10)は認識結果をホストコンピュー
タ(11)に出力する認識結果出力部であって、ホストコン
ピュータ(11)は制御部(12)及び表示装置(13)を備えてい
る。(F) Embodiment FIG. 1 is a block diagram showing a configuration of a character recognition device according to the present invention. (1) is a character observing section for scanning a input paper (2) to obtain a binary image, and (3) is An image memory for storing the obtained binary image, (4) and (5) respectively estimate a line interval Y 0 and a character interval X 0 from the binary image of the image memory (3),
The line spacing estimation unit and the character spacing estimation unit that store the estimation results in the buffers (6) and (7), and (8) uses the estimated line spacing Y 0 and the character spacing X 0 to store in the image memory (3). A character cutout unit that cuts out characters from a binary image, (9) is a recognition unit that recognizes the cut out characters, and (10) is a recognition result output unit that outputs the recognition result to the host computer (11). The host computer (11) includes a control unit (12) and a display device (13).
ところで、本発明では、入力用紙(2)は白紙であるが、
第２図の破線で示すような特定エリア(14)を仮想的に定
めており、認識させようとする文書は、この特定エリア
以降に印刷するように規定している。そして、プリンタ
等の印刷装置で、行間隔及び文字間隔を指定して、この
被認識文書を印刷する際、同時に、特定エリア(14)に、
特定文字、例えば「＊」を、第２図に示すように、複数
行に亙って１行当り複数文字印刷するようにしている。By the way, in the present invention, the input paper (2) is a blank paper,
The specific area (14) as shown by the broken line in FIG. 2 is virtually defined, and the document to be recognized is defined to be printed after this specific area. Then, when printing the recognized document by specifying the line spacing and the character spacing with a printing device such as a printer, at the same time, in the specific area (14),
As shown in FIG. 2, a specific character, for example, "*", is printed over a plurality of lines by a plurality of characters per line.
そこで、第２図に示すように、特定エリア(14)に特定文
字が印刷され、特定エリア(14)以降に被認識文書が印刷
された入力用紙(2)が、文字観測部(1)に入力されると、
文字観測部(1)は入力用紙(2)を走査して、特定エリア(1
4)の内容及び文書を２値イメージに変換し、画像メモリ
(3)に記憶する。そして、行間隔推定部(4)及び文字間隔
推定部(5)は、画像メモリ(3)の特定エリア(14)に対応す
る範囲内の２値イメージに基づいて、行間隔Ｙ0及び文
字間隔Ｘ0を推定し、推定結果を各バッファ(6)及び(7)
に記憶する。例えば、行間隔推定部(4)は、複数行に亙
る特定文字「＊」の各上端から次の上端までの長さＹ1
〜Ｙ2、あるいは行方向の各文字中心から次の文字中心
までの長さを求め、それらの長さを平均して行間隔Ｙ0
を算出し、文字間隔推定部(5)は、特定文字「＊」の各
左端から次の左端までの長さＸ1〜Ｘ4、あるいは、列方
向の各文字中心から次の文字中心までの長さを求め、そ
れらの長さを平均して文字間隔Ｘ0を算出する。Therefore, as shown in FIG. 2, an input sheet (2) on which a specific character is printed in the specific area (14) and a recognized document is printed in the specific area (14) and thereafter is displayed in the character observing section (1). Once entered,
The character observation part (1) scans the input form (2) and
Convert the contents of 4) and the document into a binary image, and use the image memory.
Store in (3). Then, the line spacing estimation unit (4) and the character spacing estimation unit (5) use the line spacing Y 0 and the character spacing based on the binary image within the range corresponding to the specific area (14) of the image memory (3). Estimate X 0 , and use the estimation results for each buffer (6) and (7)
Remember. For example, the line spacing estimation unit (4) uses the length Y 1 from each upper end of the specific character “*” over a plurality of lines to the next upper end.
To Y 2 , or the length from each character center in the line direction to the next character center, and averaging the lengths, the line spacing Y 0
The character spacing estimation unit (5) calculates the length X 1 to X 4 from each left end of the specific character “*” to the next left end, or from each character center in the column direction to the next character center. The lengths are obtained and the lengths are averaged to calculate the character spacing X 0 .
このようにして、行間隔Ｙ0及び文字間隔Ｘ0を推定した
ら、文字切出し部(8)は、画像メモリ(3)の被認識文書に
対応する２値イメージから、推定した行間隔Ｙ0及び文
字間隔Ｘ0を用いて、文字の切出しを行なう。具体的に
は、第３図の説明図を参照しながら説明すると、特定エ
リア(14)以降の２値イメージに対して、先ず、２値イメ
ージの垂直射影データを求め、この垂直射影データと閾
値ＬY及び「０」レベルとの比較から、行開始位置ＹS及
び行終了位置ＹEを求める。次に、このＹSとＹEで囲ま
れる１行分の２値イメージについて、水平射影データを
求め、同様に、この水平射影データと閾値ＬX及び
「０」レベルとの比較から、文字左端位置ＸS及び文字
右端位置ＸEを求め、ＹS，ＹE，ＸS，ＸEで囲まれた２
値イメージを、１文字として切出す。この際、文字右端
位置ＸEから右方へ文字間隔Ｘ0だけ探索しても、水平射
影データが閾値ＬXを越えなければ、その位置は空白と
判断され、文字切出し部(8)から空白コードが認識結果
出力部(10)に送出される。同様に、文字行終了位置ＹE
から下方へ行間隔Ｙ0だけ探索しても、垂直射影データ
が閾値ＬYを越えなければ、その位置は空白行として判
断され、この場合は、文字切出し部(8)から改行コード
が認識結果出力部(10)に送出される。After the line spacing Y 0 and the character spacing X 0 are estimated in this way, the character cutout unit (8) estimates the line spacing Y 0 and the line spacing Y 0 from the binary image corresponding to the recognized document in the image memory (3). Characters are cut out using the character spacing X 0 . Specifically, referring to the explanatory diagram of FIG. 3, first, for the binary image after the specific area (14), the vertical projection data of the binary image is first obtained, and the vertical projection data and the threshold value are calculated. The row start position Y S and the row end position Y E are obtained from the comparison with L Y and the “0” level. Next, the horizontal projection data is obtained for the binary image for one line surrounded by Y S and Y E , and similarly, the horizontal projection data is compared with the threshold L X and the “0” level to determine the left end of the character. The position X S and the right end position X E of the character are calculated, and 2 surrounded by Y S , Y E , X S , and X E
Cut out the value image as one character. At this time, if the horizontal projection data does not exceed the threshold L X even if the character spacing X 0 is searched from the character right end position X E to the right, the position is determined to be blank, and the character cutout portion (8) is blank. The code is sent to the recognition result output unit (10). Similarly, character line end position Y E
If the vertical projection data does not exceed the threshold value L Y even if the line spacing Y 0 is searched downward, the position is determined as a blank line, and in this case, the line feed code is recognized from the character cutout unit (8) as a recognition result. It is sent to the output unit (10).
一方、文字切出し部(8)で切出された文字は、従来通り
認識部(9)で認識され、認識結果として文字コードが出
力される。認識結果出力部(10)へ入力された文字コード
及び空白コードは、順次ホストコンピュータ(11)に送ら
れ、認識結果が表示装置(13)の画面に表示される。On the other hand, the character cut out by the character cutout unit (8) is recognized by the recognition unit (9) as usual, and the character code is output as the recognition result. The character code and the blank code input to the recognition result output unit (10) are sequentially sent to the host computer (11), and the recognition result is displayed on the screen of the display device (13).
従って、表示装置(13)には、入力用紙(2)の特定エリア
(14)に印刷された文字を除く文書内容についてのみ、認
識結果が表示されることとなる。Therefore, the display device (13) has a specific area of the input paper (2).
The recognition result will be displayed only for the document contents excluding the characters printed in (14).
(ト)発明の効果 本発明に依れば、行間隔と文字間隔の推定精度が向上
し、このため、正確な文字切出しができるようになり、
認識率が向上する。又、推定のための特別な補助走査も
不要となる。(G) Effect of the Invention According to the present invention, the accuracy of estimating the line spacing and the character spacing is improved, which enables accurate character cutout,
The recognition rate is improved. Further, no special auxiliary scanning for estimation is required.
【図面の簡単な説明】 第１図は本発明における文字認識装置の構成を示すブロ
ック図、第２図は本発明における入力用紙の印刷例を示
す図、第３図は本発明における文字切出しを説明するた
めの説明図である。 (1)…文字観測部、(2)…入力用紙、(3)…画像メモリ、
(4)…行間隔推定部、(5)…文字間隔推定部、(8)…文字
切出し部、(9)…認識部。BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a block diagram showing a configuration of a character recognition device according to the present invention, FIG. 2 is a diagram showing an example of printing on an input sheet according to the present invention, and FIG. 3 is a character cutout according to the present invention. It is an explanatory view for explaining. (1) ... Character observation section, (2) ... Input paper, (3) ... Image memory,
(4) ... Line spacing estimation unit, (5) ... Character spacing estimation unit, (8) ... Character cutout unit, (9) ... Recognition unit.
Claims (1)
エリア以降に印刷する際、同時に、前記特定エリアに同
一の特定文字を複数行に亙って１行当り複数文字印刷す
るようにし、前記入力用紙に印刷された前記文書を認識
する際、該文書及び前記特定エリアの内容を２値イメー
ジに変換して画像メモリに記憶し、該画像メモリの前記
特定エリアに対応する範囲内の２値イメージに基づいて
文字間隔及び行間隔を推定し、該推定した文字間隔及び
行間隔を用いて前記文書を構成する文字の切出しを行な
い、該切出された文字を認識し、前記入力用紙の特定エ
リアの印刷内容を除く前記文書内容についてのみ、認識
結果を出力するようにしたことを特徴とする印刷文字認
識方式。1. When printing a document to be recognized after a specific area on an input sheet, at the same time, the same specific character is printed in a plurality of lines in a plurality of characters per line in the specific area. When recognizing the document printed on the input sheet, the contents of the document and the specific area are converted into a binary image and stored in an image memory, and the range of 2 within a range corresponding to the specific area of the image memory is stored. Character spacing and line spacing are estimated based on the value image, the characters constituting the document are cut out using the estimated character spacing and line spacing, the cut out characters are recognized, and the input paper A print character recognition method, wherein a recognition result is output only for the document contents excluding the print contents of a specific area.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP61270867A JPH065551B2 (en) | 1986-11-12 | 1986-11-12 | Print character recognition method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP61270867A JPH065551B2 (en) | 1986-11-12 | 1986-11-12 | Print character recognition method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| JPS63123181A JPS63123181A (en) | 1988-05-26 |
| JPH065551B2 true JPH065551B2 (en) | 1994-01-19 |
Family
ID=17492073
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| JP61270867A Expired - Lifetime JPH065551B2 (en) | 1986-11-12 | 1986-11-12 | Print character recognition method |
Country Status (1)
| Country | Link |
|---|---|
| JP (1) | JPH065551B2 (en) |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPS59206987A (en) * | 1983-05-10 | 1984-11-22 | Toshiba Corp | Letter recognizing device |
| JPS59206989A (en) * | 1983-05-11 | 1984-11-22 | Nec Corp | Letter segmenting device |
| JPS61100877A (en) * | 1984-10-23 | 1986-05-19 | Omron Tateisi Electronics Co | Method of detecting deficiency of graphic in graphic recognizing device |
-
1986
- 1986-11-12 JP JP61270867A patent/JPH065551B2/en not_active Expired - Lifetime
Also Published As
| Publication number | Publication date |
|---|---|
| JPS63123181A (en) | 1988-05-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107867090B (en) | Printing apparatus, control method of printing apparatus, and recording medium | |
| US5659638A (en) | Method and system for converting bitmap data into page definition language commands | |
| JPH02233275A (en) | Bidirectional graphic print method | |
| US5050221A (en) | Image generating apparatus | |
| JP4071310B2 (en) | Printing control method and printing apparatus in printing apparatus | |
| JP2021128444A (en) | Information processing equipment and programs | |
| US4858171A (en) | Word processor with selective placement of printhead for printing of newly input print data after interruption of printing | |
| US20060285161A1 (en) | Overlay printing device | |
| CN100530219C (en) | Image processing apparatus | |
| JP2001052110A (en) | Document processing method, recording medium recording document processing program, and document processing apparatus | |
| US5319746A (en) | Automatic hyphenation apparatus displaying grammatically correct suggestions for hyphenation of a word isolated on a single display line | |
| JPS63123181A (en) | Printed character recognition system | |
| JPH0916582A (en) | Document creation device and recognition result output method used in the device | |
| JP2863671B2 (en) | Print format creation device | |
| JPH0596806A (en) | Printing method and apparatus thereof | |
| US5113520A (en) | Data processor enabling prioritizing of concurrent tasks | |
| JPS63208990A (en) | Character pattern segmenting device | |
| JPH0581266A (en) | Information processing equipment | |
| JP2636866B2 (en) | Information processing method | |
| JP2005035024A (en) | Inkjet recording apparatus, recording time calculation method, and computer-readable recording medium storing a program for executing the method | |
| JP2691637B2 (en) | Bar code printer | |
| JPH0793479A (en) | Optical character reader | |
| JP2572048B2 (en) | Word processing method | |
| JPH06126991A (en) | Recording method and apparatus | |
| JPH08265551A (en) | Facsimile information extraction processing method and facsimile apparatus |