JPS609314B2 - Character recognition method - Google Patents
Character recognition methodInfo
- Publication number
- JPS609314B2 JPS609314B2 JP54113914A JP11391479A JPS609314B2 JP S609314 B2 JPS609314 B2 JP S609314B2 JP 54113914 A JP54113914 A JP 54113914A JP 11391479 A JP11391479 A JP 11391479A JP S609314 B2 JPS609314 B2 JP S609314B2
- Authority
- JP
- Japan
- Prior art keywords
- characters
- character
- written
- writing position
- general
- 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
Links
Landscapes
- Character Discrimination (AREA)
Description
【発明の詳細な説明】
この発明は一般文字と記入位置の異なる特殊文字、小文
字、濁点、半濁点、カンマ、ピリオド、アポストロフィ
などをも読み取る文字認識方式に関するものである。DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a character recognition method that can also read special characters written in different positions from general characters, such as lowercase letters, voiced marks, half-voiced marks, commas, periods, and apostrophes.
第１図は従来のこの種の装置を説明するための図であっ
て、図中１，２，９は読み取り帳票上に記入された文字
パターン列の例である。FIG. 1 is a diagram for explaining a conventional device of this type, and numerals 1, 2, and 9 in the figure are examples of character pattern strings written on a reading form.
従来この種の装置では、識別された文字自身から抽出し
た特徴のみにより識別していたので、上記文字パターン
列中のパターン３のような場合、「アポストロフィ」か
数字「１」であるのか識別するのが困難であり、誤読し
やすく、更にピリオド５と識別するのが非常に困難であ
るという欠点があった。この発明は同一行に記入された
文字の記入位置の差を調べることにより、一般文字と特
殊文字、又は特殊文字どうしを識別することを特徴とし
、その目的は誤読を防ぎ、識別不能を少なくし読み取り
精度を上げることにある。第２図はこの発明の実施例を
示す図で、図中１川ま文字パターンの位置を検出する検
出手段、１１は上記検出手段１０で検出された一般文字
の文字パターンの位置を記憶する記憶手段、１２は上記
検出手段で検出された文字パタ−ンの位置と上記記憶手
段に記憶されている一般文字の文字パターンの位置とを
比較し、この情報を利用し認識する認識手段であり、１
３は上記検出手段から上記記憶手段へのデータ線、１４
は上記検出手段から上記認識手段へのデータ線、１５は
上記記憶手段から上記認識手段へのデータ線、１６は上
記認識手段からの出力である。Conventionally, this type of device has identified only the features extracted from the identified character itself, so in a case like pattern 3 in the character pattern string above, it is necessary to identify whether it is an "apostrophe" or the number "1". This has disadvantages in that it is difficult to read, easy to misread, and very difficult to distinguish from period 5. This invention is characterized by identifying general characters and special characters, or between special characters, by checking the difference in the writing position of characters written on the same line.The purpose of this invention is to prevent misreading and reduce the possibility of indistinguishability. The goal is to improve reading accuracy. FIG. 2 is a diagram showing an embodiment of the present invention. In the figure, 1 is a detection means for detecting the position of a character pattern, and 11 is a memory for storing the position of a character pattern of a general character detected by the detection means 10. Means 12 is a recognition means that compares the position of the character pattern detected by the detection means with the position of the character pattern of the general character stored in the storage means, and uses this information for recognition; 1
3 is a data line from the detection means to the storage means; 14
15 is a data line from the detection means to the recognition means, 15 is a data line from the storage means to the recognition means, and 16 is an output from the recognition means.
次に動作を説明する。検出手段１０で検出された認識す
べき文字の文字パターンの位置はデータ線１３を通して
記憶手段１１へ、データ線１４を通して認識手段１２へ
送られる。同時に既に記憶されている参照すべき一般文
字の文字パターンの位置はデータ線１５を通して認識手
段１２に送られる。認識手段１２においてデータ線１４
，１５を通じて入力されたそれぞれの文字の位置の差を
とることにより、検出された文字パターンが帳票上の記
入枠の中でどの位置に記入されているかの情報を得る。
上記文字パターンを認識する際、複数個の候補文字があ
る場合、上記情報と各候補文字の通常の記入位置を比較
参照することにより容易に認識することができる。Next, the operation will be explained. The position of the character pattern of the character to be recognized detected by the detection means 10 is sent to the storage means 11 through the data line 13 and to the recognition means 12 through the data line 14. At the same time, the position of the already stored character pattern of the general character to be referenced is sent to the recognition means 12 through the data line 15. In the recognition means 12, the data line 14
, 15, information on where the detected character pattern is written in the entry frame on the form is obtained.
When recognizing the above-mentioned character pattern, if there are a plurality of candidate characters, they can be easily recognized by comparing and referring to the above-mentioned information and the usual writing position of each candidate character.
例えば文字列９の中の文字パターン５を認識する際、候
補文字としてピリオド「．」とアポストロフィ「′」が
ある場合、文字パターン８と５の記入位置を比較すると
文字パターン５が一般文字のパターン８より下の方に記
入されていることがわかるから通常下の方の位置に記入
されるピリオドと認識することができる。For example, when recognizing character pattern 5 in character string 9, if there are a period "." and an apostrophe "'" as candidate characters, comparing the positions of character patterns 8 and 5, character pattern 5 is a general character pattern. Since it can be seen that it is written below 8, it can be recognized as a period that is normally written in the lower position.
なお、以上は特殊文字の例として、英字、記号の場合に
ついて説明したが、この発明はこれに限らず他の特殊文
字、例えばシャ、シュ、ショなどの小文字を識別する場
合、パピプベポなどの半濁点を識別する場合、バビブベ
ボなどの濁点を識別する場合に使用してもよい。In addition, the above description has been made of alphabetic characters and symbols as examples of special characters, but this invention is not limited to these, but can also be used to identify other special characters, such as lowercase letters such as sha, shu, sho, etc. When identifying a voiced mark, it may be used to identify a voiced mark such as babibubebo.
更に特殊文字が行の先頭にあって識別できない場合は、
その行を２回以上走査することにより、１回目の走査で
検出された一般文字の記入位置を参照して、２回目の走
査で識別すればよい。Furthermore, if a special character is at the beginning of a line and cannot be identified,
By scanning the line two or more times, the characters can be identified in the second scan by referring to the writing position of the general character detected in the first scan.
以上のように、この発明によれば同一行に記入された文
字の記入位置の差を調べることにより、小文字、濁点、
半濁点、カンマ、ピリオド、アポストロフィなど一般文
字と記入位置の異なる特殊文字と一般文字を識別する特
徴情報を得ることができるから、謀認識を防ぎ認識精度
を向上することができる。As described above, according to the present invention, by checking the difference in the writing position of characters written on the same line, lowercase letters, voiced marks,
Since it is possible to obtain characteristic information that distinguishes between special characters and general characters, such as pendakuten, comma, period, and apostrophe, which are written in different positions from general characters, it is possible to prevent false recognition and improve recognition accuracy.
第１図は従来のこの種の装置を説明するための文字列の
例を示す図、第２図はこの発明の実施例を示す図である
。
図中１は「、ＡＢ′と記入された文字列の例、２は「
、ＡＢＩ」と記入さた文字列の例、９は「、ＡＢ．」と
記入された文字列の例、１川ま文字パターンの記入位置
を検出する検出手段、１１は記憶手段、１２は認識手段
、１３，１４，１５はデータ線、１６は出力である。
亀１図
予之図FIG. 1 is a diagram showing an example of a character string for explaining a conventional device of this type, and FIG. 2 is a diagram showing an embodiment of the present invention. In the figure, 1 is an example of a character string written as ", AB', and 2 is an example of a character string written as ", AB'.
, ABI", 9 is an example of a character string written as ",AB.", 1 is a detection means for detecting the position where a character pattern is written, 11 is a storage means, and 12 is a recognition means. 13, 14, 15 are data lines, and 16 is an output. Turtle 1 illustration
Claims (1)
上の文字記入枠の中ほどに印字、又は記入される一般文
字と異なり、カナ小文字、濁点、半濁点、カンマ、ピリ
オド、アポストロフイなどのように一般文字に比較して
、帳票上の文字記入枠の上下に片寄って印字又は記入さ
れる特殊文字をも読み取る文字読み取り装置において、
読み取り帳票上に印字又は記入された文字の記入位置を
検出する検出手段と、上記検出手段で検出された文字の
記入位置を記憶する記憶手段と、上記記憶手段に記憶さ
れている文字記入位置を文字パターンを認識する際の情
報の１つとして加え認識する認識手段を備え、ある文字
パターンを認識する際に文字記入位置が異なる複数個の
候補文字がある場合、上記記憶手段に記憶されている同
一行の一般文字の記入位置と認識すべき文字パターンの
記入位置を比較し、この情報を利用して一般文字と特殊
文字、又はカンマとアポストロフイなどのような特殊文
字どうしを識別、判定することを特徴とする文字認識方
式。1 Unlike general characters that are printed or written in the middle of the character entry frame on the form, such as 1 123, ABC, +-#¥, IU, etc., lowercase kana characters, voiced marks, half-voiced marks, commas, periods, apostrophes, etc. In a character reading device that also reads special characters that are printed or written on the top and bottom of the character entry frame on a form compared to general characters, such as
A detection means for detecting the writing position of the characters printed or written on the reading form, a storage means for storing the writing position of the characters detected by the detection means, and a character writing position stored in the storage means. It is equipped with a recognition means for adding and recognizing as one of the information when recognizing a character pattern, and when there are multiple candidate characters with different character entry positions when recognizing a certain character pattern, the candidate characters are stored in the storage means. Compare the writing position of general characters on the same line with the writing position of the character pattern to be recognized, and use this information to identify and judge general characters and special characters, or special characters such as commas and apostrophes. A character recognition method characterized by
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP54113914A JPS609314B2 (en) | 1979-09-05 | 1979-09-05 | Character recognition method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP54113914A JPS609314B2 (en) | 1979-09-05 | 1979-09-05 | Character recognition method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| JPS5638684A JPS5638684A (en) | 1981-04-13 |
| JPS609314B2 true JPS609314B2 (en) | 1985-03-09 |
Family
ID=14624344
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| JP54113914A Expired JPS609314B2 (en) | 1979-09-05 | 1979-09-05 | Character recognition method |
Country Status (1)
| Country | Link |
|---|---|
| JP (1) | JPS609314B2 (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPS6155791A (en) * | 1984-08-28 | 1986-03-20 | Nec Corp | Character recognition device |
| JP2521915B2 (en) * | 1986-07-04 | 1996-08-07 | 日本電気株式会社 | Type character recognition device |
-
1979
- 1979-09-05 JP JP54113914A patent/JPS609314B2/en not_active Expired
Also Published As
| Publication number | Publication date |
|---|---|
| JPS5638684A (en) | 1981-04-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US4408342A (en) | Method for recognizing a machine encoded character | |
| JPS609314B2 (en) | Character recognition method | |
| JPH0430070B2 (en) | ||
| US20200242390A1 (en) | Information processing apparatus and non-transitory computer readable medium storing program | |
| JPS5919388B2 (en) | character reader | |
| JPS5932829B2 (en) | How to read characters | |
| JPS5853390B2 (en) | pattern recognition device | |
| JPS6054082A (en) | Optical character reader | |
| JP2877380B2 (en) | Optical character reader | |
| JPS6160184A (en) | Optical character reader | |
| JPS5972577A (en) | Drawing reader | |
| JPS6095689A (en) | Optical character reader | |
| JP2784004B2 (en) | Character recognition device | |
| JPS623472B2 (en) | ||
| JP3310063B2 (en) | Document processing device | |
| JP2993533B2 (en) | Information processing device and character recognition device | |
| JPS62134767A (en) | Automatic extracting device for symbol name and segment name | |
| JPH0475557B2 (en) | ||
| JPH0122958B2 (en) | ||
| JPH04318687A (en) | Character recognition unit | |
| JPS63188284A (en) | Character reader | |
| KR900005141B1 (en) | Character Recognition Device | |
| JP2624124B2 (en) | Character recognition device and character recognition method | |
| JPS5854431B2 (en) | Misreading detection method for character recognition device | |
| JPS6020786B2 (en) | character reading device |