JPS6174087A - Word reading device - Google Patents

Word reading device

Info

Publication number
JPS6174087A
JPS6174087A JP59195527A JP19552784A JPS6174087A JP S6174087 A JPS6174087 A JP S6174087A JP 59195527 A JP59195527 A JP 59195527A JP 19552784 A JP19552784 A JP 19552784A JP S6174087 A JPS6174087 A JP S6174087A
Authority
JP
Japan
Prior art keywords
word
character
characters
read
recognizing
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
JP59195527A
Other languages
Japanese (ja)
Inventor
Haruo Mizukami
水上 治雄
Yoji Maeda
前田 陽二
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.)
Mitsubishi Electric Corp
Original Assignee
Mitsubishi Electric Corp
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 Mitsubishi Electric Corp filed Critical Mitsubishi Electric Corp
Priority to JP59195527A priority Critical patent/JPS6174087A/en
Publication of JPS6174087A publication Critical patent/JPS6174087A/en
Pending legal-status Critical Current

Links

Landscapes

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

Abstract

PURPOSE:To read words including external character precisely and rapidly by retrieving words at a character position by a recognized character from a word dictionary composed of the words including both the internal character and the external character and outputting as a reading result. CONSTITUTION:An input word, which is written on a slip 1 and should be read, is scanned and photoelectrically converted by a scanning means, and recognized at a character recognizing means 2. Character position information in the word of a mark showing a recognizing result and a recognizing impossibil ity is stored in a recognizing result memory means 3. At a word determining means 4, by using a recognizing result 10 stored in the recognizing result mem ory means 3 and character position information, the word is retrieved from a word dictionary 5 by a word retrieving means 6. By replacing the mark showing recognizing impossibility by the retrieved external character, an input word can be precisely read.

Description

【発明の詳細な説明】 〔産業上の利用分野〕 この発明は、単語を構成する文字を1文字ととに認識し
、その結果を用いて単語を読み取る単語読み取り装置に
関するものである。
DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a word reading device that recognizes characters constituting a word as single characters and uses the results to read the word.

〔従来の技術〕[Conventional technology]

従来の単語読み取り装置では、単語を構成する文字を1
文字ごとに認識し、その結果として得られる認識候補文
字の組み合わせのうち、あらかじめ作成した単語辞書内
にある単語と一致するものを読み取り結果としていた。
In conventional word reading devices, the letters that make up a word are
Each character was recognized, and among the resulting combinations of candidate characters, those that matched words in a word dictionary created in advance were used as reading results.

この様な方式では、認識対象文字(内字)のみで構成さ
れる単語は読み取り可能であるが、認識対象外文字(外
字)を含む単語は外字を認識することができないので、
自動的に読み取ることができなかった。このため、外字
の読み取り結果は必ず誤読又は読み取り不能となり、人
間が判断して読み取り結果を修正する必要があった。ま
た、この発明の出願人がさきに提案した他の方式(特卯
昭        号参照)の単語読み取り装置におい
ては、内字ごとに字形が類似した外字を、あらかじめ使
用頻度順序に記憶しておき、外字を入力する時には、文
字認識手段によって出力される認識候補文字の中から、
入力し様とする外字に字形が類似したものを選択指定す
ることによって、対応する外字を表示させ、その中から
選択入力する様にしていた。
With this method, words that consist only of characters to be recognized (internal characters) can be read, but words that include characters that are not to be recognized (external characters) cannot be recognized.
could not be read automatically. For this reason, the reading results of external characters are always misread or unreadable, and it is necessary for humans to make judgments and correct the reading results. In addition, in a word reading device of another type (see Tokusho No.) previously proposed by the applicant of this invention, external characters with similar shapes for each internal character are stored in advance in order of frequency of use. When inputting external characters, select from among the recognition candidate characters output by the character recognition means.
By selecting and specifying a character whose shape is similar to the custom character to be input, the corresponding custom characters are displayed, and the user can select and input from among them.

〔発明が解決し様とする問題点〕[Problems that the invention is expected to solve]

上記の様な従来の単語読み取り装置では、その一方式に
おいて、人間が判断して読み取り結果を修正する場合に
、最も単純な方法としては、JIS区点番号などの文字
コードで入力することである。しかるに、この方法によ
れば、コードブックなどを用いて入力し様とする文字コ
ードをいちいち調べなければならず、このため、文字を
入力するのに非常に時間がかかるという問題点があった
In conventional word reading devices such as those mentioned above, the simplest method for correcting the reading results based on human judgment is to enter character codes such as JIS Kuten numbers. . However, according to this method, it is necessary to check each character code to be inputted using a code book or the like, and therefore, there is a problem in that it takes a very long time to input characters.

また、他の方式では、外字を入力する時、入カレ様とす
る外字の文字コードを調べなくても良いという利点はあ
るが、認識候補文字の中から入カレ様とする外字の字形
に似たものを見付は出すのに、やはりかなりの時間がか
かる。という問題点があった。一方、従来の単語読み取
り装置では、内字の字種を増やすことによって読み取り
対象単語を増加することができるも、この様に内字の字
種を増やすと、認識性能の低下を招くという問題点があ
った。
Other methods have the advantage of not having to check the character code of the external character you want to input when inputting a external character, but they also select recognition candidate characters that resemble the shape of the external character you want to input. After all, it takes a considerable amount of time to generate a headline for something that you want. There was a problem. On the other hand, with conventional word reading devices, it is possible to increase the number of words to be read by increasing the types of inner characters, but the problem is that increasing the number of inner characters in this way leads to a decline in recognition performance. was there.

この発明は、かかる問題点を解決するためになされたも
ので、読み取るべき単語を構成する文字のうちで少なく
とも1つが認識され、かつ少なくとも1つが認識不能と
なった場合に、同字と外字の両方を含む単語で構成した
単語辞書から、認識された文字がその文字位置にある単
語を検索し、認識不能となった文字に対して、その文字
位置に対応する文字を読み取り結果とすることにより、
外字を含む単語を正確だ、かつ迅速に読み取ることがで
きる単語読み取り装置を得ることを目的とするものであ
る。
This invention was made to solve this problem, and when at least one of the characters constituting the word to be read is recognized and at least one is unrecognizable, By searching a word dictionary consisting of words that include both characters for words in which the recognized character is in that character position, and reading the character corresponding to that character position for unrecognized characters. ,
The object of the present invention is to obtain a word reading device that can accurately and quickly read words including external characters.

〔問題点を解決するための手段〕[Means for solving problems]

この発明に係る単語読み取り装置は、文字認識手段で、
読み取るべき単語を構成する文字のうちで少なくとも1
つが認識され、かつ少なくとも1つが認識不能となった
場合に、単語゛検索手段を用いて、単語辞書から認識さ
れた文字のすべてが該当する文字位置にある単語を検索
し、認識不能となった文字位置に存在する検索された単
語内の文字を、認識不能に置き換えて読み取り結果とし
て、出力することにより、単語を読み取る様にするもの
である。
The word reading device according to the present invention includes character recognition means,
At least one of the letters that make up the word to be read
is recognized, and at least one becomes unrecognizable, a word search method is used to search the word dictionary for a word in which all of the recognized characters are in the corresponding character position, and the word becomes unrecognizable. The word is read by replacing the character in the searched word existing at the character position with an unrecognizable character and outputting it as a reading result.

〔作用〕[Effect]

この発明の単語読み取り装置においては、読み取るべき
単語を構成する文字のうちで少なくとも1つ−が認識さ
れ、かつ少なくとも1つが認識不能となった場合に、内
字と外字の両方を含む単語で構成した単語辞書から、認
識された文字がその文字位置にある単語を検索し、認識
不能となった文字に対して、その文字位置に対応する文
字を読み取り結果として出力し、これKより、外字を含
む単語を正確に、かつ迅速に読み取ることを可能にする
In the word reading device of the present invention, when at least one of the characters constituting the word to be read is recognized and at least one is unrecognizable, the word reading device is configured to read a word consisting of both internal and external characters. Search the word dictionary in which the recognized character is at that character position, read the character corresponding to that character position for the unrecognized character, and output it as a result. Enables users to read words accurately and quickly.

〔実施例〕〔Example〕

第1図はこの発明の一実施例である単語読み取り装置の
構成を示すブロック図である。図において、1は帳票、
2は文字を1文字ととに認識する文字認識手段、3は文
字認識手段2による認識結果と単語内の文字位置を格納
する認識結果記憶手段、4は上記認識結果から単語を決
定する単語決定手段、5は内字と外字の両方を含んで構
成される単語を格納した単語辞書、6は特定の内字が特
定の位置に含まれる単語を単語辞書5から検索する単語
検索手段である。
FIG. 1 is a block diagram showing the configuration of a word reading device which is an embodiment of the present invention. In the figure, 1 is a form,
2 is a character recognition means for recognizing characters into individual characters; 3 is a recognition result storage means for storing the recognition result by the character recognition means 2 and the character position within a word; 4 is a word determination means for determining a word from the recognition result. Means 5 is a word dictionary storing words including both internal characters and external characters, and 6 is a word search means for searching the word dictionary 5 for words in which a specific internal character is included in a specific position.

第2図は、第1図の単語読み取り装置において文字認識
動作を説明するための図で為る。図において、7は入力
単語、8,9はそれぞれ入力単語7を構成する文字、1
0は入力単語7の認識結果、11.12はそれぞれ入力
単語7を構成する各文字8,9の認識結果である。
FIG. 2 is a diagram for explaining the character recognition operation in the word reading device of FIG. 1. In the figure, 7 is the input word, 8 and 9 are the characters that make up the input word 7, and 1
0 is the recognition result of the input word 7, and 11.12 is the recognition result of each character 8 and 9 constituting the input word 7, respectively.

第3図は、第1図の単語読み取り装置において、単語辞
書の構成例を示す図である。図において、13は単語辞
書5内の単語、14.15はそれぞれ単語辞書5内の単
語13を構成する内字及び外字である。
FIG. 3 is a diagram showing an example of the configuration of a word dictionary in the word reading device of FIG. 1. In the figure, 13 is a word in the word dictionary 5, and 14 and 15 are internal and external characters constituting the word 13 in the word dictionary 5, respectively.

次に、上記第1図ないし第3図を用いて、この発明に係
る単語読み取り装置の動作について説明する。帳票1上
に記入された読み取るべき入力単語7を、図示されない
走査手段により走査して光電変換し、図示されない前処
−理手段で二値化などの前処理を行う。文字認識手段2
では、入力単語7を構成する各文字8,9を1文字ごと
に、パターンマツチング法等の公知の技術手段を用いて
認識する。今、入力単語7を構成する「長」の文字8を
内字とし、「嶋」の文字9を外字とすると、「長」の文
字8は認識されるが、「嶋」の文字9は認識不能となり
、認識結果10は、「長」の文字8の認識結果11及び
認識不能を表わす記号「◆」12(文字9の認識結果に
相当)で示されているとおりとなる。この認識結果10
と、「長」の文字8の認識結果11及び認識不能を表わ
す記号「◆」12の単語内での文字位置情報は、認識結
果記憶手段3に格納される。
Next, the operation of the word reading device according to the present invention will be explained using FIGS. 1 to 3 above. Input words 7 to be read written on the form 1 are scanned by a scanning means (not shown) and photoelectrically converted, and preprocessing such as binarization is performed by a preprocessing means (not shown). Character recognition means 2
Now, the characters 8 and 9 constituting the input word 7 are recognized character by character using known technical means such as pattern matching. Now, if the character 8 of "cho" that makes up the input word 7 is an internal character and the character 9 of "shima" is an external character, the character 8 of "cho" is recognized, but the character 9 of "shima" is recognized. The recognition result 10 becomes as shown by the recognition result 11 of the character 8 for "long" and the symbol "◆" 12 (corresponding to the recognition result of the character 9) representing unrecognizability. This recognition result 10
The recognition result 11 of the character 8 for "long" and the character position information within the word of the symbol "◆" 12 representing unrecognizability are stored in the recognition result storage means 3.

次いで、単語決定手段4では、以下に述べる様にして単
語を決定する。認識結果記憶手段3に格納された認識結
果10と文字位置情報を用いて、単語検索手段6により
、単語辞書5から第1文字目が「長」の文字8の認識結
果11で、文字数が2の単語を検索する。この場合には
、「長嶋」と云う単語13が検索される。ここで、入力
単語7の第2文字目の「嶋」の文字9の認識結果である
認識不能を表わす記号「◆」12を、検索された「長嶋
」と云う単語13の第2番目の文字である「嶋」の外字
15で置き換えることによって、入力単語7(「長嶋J
)ft正しく読み取ることができる。実際に、手書きで
丁寧に記入された漢字を認識するi合に1内字に対して
は100%近い認識率が得られるし、外字に別してはほ
とんど認識不能となる。
Next, the word determining means 4 determines a word as described below. Using the recognition result 10 and character position information stored in the recognition result storage means 3, the word search means 6 retrieves from the word dictionary 5 the recognition result 11 of the character 8 whose first character is "long" and the number of characters is 2. Search for words. In this case, the word 13 "Nagashima" is searched. Here, the symbol "◆" 12 representing unrecognizability, which is the recognition result of the character 9 of "Shima", which is the second character of the input word 7, is replaced with the second character of the searched word "Nagashima" 13. By replacing it with the external character 15 of "Shima", input word 7 ("Nagashima J
)ft can be read correctly. In fact, when recognizing carefully handwritten kanji, a recognition rate of nearly 100% is obtained for single internal characters, and it is almost impossible to recognize external characters.

この様にして、この発明の単語読み取り装置によれば、
内字の字種を増やすことなく、内字と外字の両方を含ん
で構成される単語を、自動的に、かつ迅速に読み取るこ
とができる。
In this way, according to the word reading device of this invention,
To automatically and quickly read words including both internal characters and external characters without increasing the number of internal characters.

なお、上記実施例では、文字認識手段2による認識結果
を用いて単語辞書5を検索した結果、該当する単語がた
だ1つ存在する場合について説明したが、この発明はこ
れに限定されることなく、該当する単語が単語辞書5内
に複数個存在する場合には、出現頻度の最も高いものを
出力し、それが所望のものでない時に、単語辞書5内の
すべての該当する単語を順次に出力して選択できる様に
しても良い。
In the above embodiment, a case has been described in which only one matching word exists as a result of searching the word dictionary 5 using the recognition result by the character recognition means 2, but the present invention is not limited to this. , if there are multiple corresponding words in the word dictionary 5, output the one with the highest frequency of appearance, and if it is not the desired one, sequentially output all the corresponding words in the word dictionary 5. It may also be possible to make the selection possible.

〔発明の効果〕〔Effect of the invention〕

この発明は以上説明したとおり、単語読み取り装置にお
いて、読み取るべき単語を構成する文字のうちで少なく
とも1つが認識され、かつ少なくとも1つが認識不能と
なった場合に、内字と外字の両方を含む単語で構成した
単語辞書から、認識された文字がその文字位置にある単
語を検索し、認識不能となった文字に対して、その文字
位置に対応する文字を読み取り結果とするので、内字の
字種を増やすことなく、内字と外字の両方を含んで構成
される単語を正確に、かつ迅速に読み取ることができる
という優れた効果を奏するもので、ある。
As explained above, in a word reading device, when at least one of the characters constituting a word to be read is recognized and at least one is unrecognizable, a word containing both internal characters and external characters can be read. From the word dictionary constructed with It has the excellent effect of being able to accurately and quickly read words that include both internal and external characters without increasing the number of characters.

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

第1図はこの発明の一実施例である単語読み取り装置の
構成を示すブロック図、第2図は第1図の単語読み取り
装置において、文字認識動作を説明するための図、第3
図は、第1図の単語読み取り装置において、単語辞書の
律成例を示す図である0 図において、1・・・帳票、2は文字認識手段、3・・
・認識結果記憶手段、4・・・単語決定手段、5・・・
単語辞書、6・・・単語検索手段、7・・・入力単語、
10・・・認識結果、13・・・単語辞書5内の単語で
ある。 なお、各図中、同一符号は同一、又は相当部分を示す。
FIG. 1 is a block diagram showing the configuration of a word reading device which is an embodiment of the present invention, FIG. 2 is a diagram for explaining character recognition operation in the word reading device of FIG. 1, and FIG.
The figure is a diagram showing an example of the rules for a word dictionary in the word reading device of FIG.
- Recognition result storage means, 4... word determination means, 5...
word dictionary, 6... word search means, 7... input word,
10... Recognition results, 13... Words in the word dictionary 5. In each figure, the same reference numerals indicate the same or equivalent parts.

Claims (1)

【特許請求の範囲】[Claims] 複数の文字から成る単語を認識して読み取る単語読み取
り装置において、単語を構成する文字を1文字ごとに認
識する文字認識手段と、この文字認識手段で認識された
結果と前記単語内での文字位置を格納する認識結果記憶
手段と、認識対象文字(内字)と認識対象外文字(外字
)の両方を含んで構成される単語を格納した単語辞書と
、この単語辞書の中から特定の内字を特定の位置に含む
単語を検索する単語検索手段を備え、前記文字認識手段
で、読み取るべき単語を構成する文字のうちで少なくと
も1つが認識され、かつ少なくとも1つが認識不能とな
つた場合に、前記単語検索手段を用いて、前記単語辞書
から前記認識された文字のすべてが該当する文字位置に
ある単語を検索し、前記認識不能となつた文字位置に存
在する前記検索された単語内の文字を、認識不能に置き
換えて読み取り結果として出力することにより、単語を
読み取る様にしたことを特徴とする単語読み取り装置。
In a word reading device that recognizes and reads words made up of multiple characters, a character recognition means that recognizes each character constituting a word, a result recognized by this character recognition means, and a character position within the word are provided. a recognition result storage means for storing recognition result storage means, a word dictionary storing words including both recognition target characters (internal characters) and non-recognition target characters (external characters), and a recognition result storage means for storing a specific internal character from this word dictionary. comprising a word search means for searching for a word containing at a specific position, and when the character recognition means recognizes at least one of the characters constituting the word to be read and at least one becomes unrecognizable, Using the word search means, search the word dictionary for a word in which all of the recognized characters are in the corresponding character position, and search for a word in the searched word that exists in the unrecognized character position. A word reading device characterized in that a word is read by replacing it with an unrecognizable word and outputting it as a reading result.
JP59195527A 1984-09-18 1984-09-18 Word reading device Pending JPS6174087A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP59195527A JPS6174087A (en) 1984-09-18 1984-09-18 Word reading device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP59195527A JPS6174087A (en) 1984-09-18 1984-09-18 Word reading device

Publications (1)

Publication Number Publication Date
JPS6174087A true JPS6174087A (en) 1986-04-16

Family

ID=16342570

Family Applications (1)

Application Number Title Priority Date Filing Date
JP59195527A Pending JPS6174087A (en) 1984-09-18 1984-09-18 Word reading device

Country Status (1)

Country Link
JP (1) JPS6174087A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5729630A (en) * 1990-05-14 1998-03-17 Canon Kabushiki Kaisha Image processing method and apparatus having character recognition capabilities using size or position information

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPS5953986A (en) * 1982-09-20 1984-03-28 Toshiba Corp Character recognizing device
JPS59106084A (en) * 1982-12-09 1984-06-19 Comput Basic Mach Technol Res Assoc Character reader

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPS5953986A (en) * 1982-09-20 1984-03-28 Toshiba Corp Character recognizing device
JPS59106084A (en) * 1982-12-09 1984-06-19 Comput Basic Mach Technol Res Assoc Character reader

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5729630A (en) * 1990-05-14 1998-03-17 Canon Kabushiki Kaisha Image processing method and apparatus having character recognition capabilities using size or position information

Similar Documents

Publication Publication Date Title
JPS63182793A (en) Character segmenting system
JPS6174087A (en) Word reading device
JPS6262388B2 (en)
JPH0247788B2 (en)
JPS5842904B2 (en) Handwritten kana/kanji character recognition device
JPS6239794B2 (en)
JP3419251B2 (en) Character recognition device and character recognition method
JP2538543B2 (en) Character information recognition device
JP2784004B2 (en) Character recognition device
KR920006874A (en) Recognition method of document with predetermined format
JP2947832B2 (en) Word matching method
JPS6095689A (en) Optical character reader
JPH0475556B2 (en)
JP2886690B2 (en) Character recognition method for optical character reader
JP2851865B2 (en) Character recognition device
JP2839515B2 (en) Character reading system
JPS6330991A (en) Character recognizing device
JPS6293776A (en) Information recognizing device
JPH07325888A (en) Dictionary registering method, character recognizing method and character recognizing device
JPH02193281A (en) Character recognizing device
Shufelt Martial Arts of the World: an Encyclopedia
JPS63268082A (en) Pattern recognizing device
JPS5878276A (en) Optical character reader
JPH03163681A (en) Character recognizing system
JPH0340079A (en) Post-processing method for character recognition in character reader