JPS58203578A - character recognition device - Google Patents
character recognition deviceInfo
- Publication number
- JPS58203578A JPS58203578A JP57070989A JP7098982A JPS58203578A JP S58203578 A JPS58203578 A JP S58203578A JP 57070989 A JP57070989 A JP 57070989A JP 7098982 A JP7098982 A JP 7098982A JP S58203578 A JPS58203578 A JP S58203578A
- Authority
- JP
- Japan
- Prior art keywords
- character
- recognition
- output frequency
- similarity
- recognision
- 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
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/24—Character recognition characterised by the processing or recognition method
- G06V30/248—Character recognition characterised by the processing or recognition method involving plural approaches, e.g. verification by template match; Resolving confusion among similar patterns, e.g. "O" versus "Q"
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Character Discrimination (AREA)
Abstract
Description
【発明の詳細な説明】
（８）発明の技術分野
本発明は特に認識対象として漢字等を含む文字認識装置
に関す。DETAILED DESCRIPTION OF THE INVENTION (8) Technical Field of the Invention The present invention particularly relates to a character recognition device that includes Chinese characters and the like as objects to be recognized.
（Ｂ） 技術の背景
漢字を認識対象として含む文字認識装置は、認識対象文
字種が多いため相互に類似する文字種が多く、Ｌｆ７Ｌ
がって認識が非常に困難でらＱｌいまだ完全な実用化段
階には至っていない。(B) Technical background Character recognition devices that recognize kanji have many character types that are similar to each other, and Lf7L
Therefore, recognition is extremely difficult, and Ql has not yet reached the stage of complete practical use.
Ｃ）従来技術と問題点
従来の文字認識装置に２いては、例えば入力文字画像の
％ａｔ−抽出し比較基準として辞書メモリに記憶する文
字種毎の標準特徴との類似度によって入力文字画像の認
識をおこなっているが、認識対象文字種が多い場合には
認識が非常に困難なため誤認識する場合がしばしば生じ
、修正手段によって修正せざるを得す、したがってオペ
レータの負荷が大きいという問題があった。C) Prior art and problems Conventional character recognition devices recognize input character images based on their similarity to standard features for each character type, which are extracted from the input character image and stored in a dictionary memory as a comparison standard. However, when there are many types of characters to be recognized, recognition is extremely difficult and errors in recognition often occur, forcing corrections to be made using correction methods, resulting in a heavy burden on the operator. .
■ 発明の目的
本発明は既述従来例に２ける誤認識を減少しオペレータ
の負荷を径減することを目的とする。(2) Purpose of the Invention It is an object of the present invention to reduce the number of false recognitions that occur in the prior art described above, thereby reducing the burden on the operator.
四）発明の構成
本発明になる文字認識装置は、入力文字画像を比較基準
との４似度に二って認識する認識部と前記認識部に２・
′・する誤っ友認識を修正する修正手段とｔ−備え入力
文字画像を認識し誤りた認識に対しては修正上２こ：：
って出力する装置において、文字種毎の一力頬夏を記録
するテーブル全役け、入力文字誦；オを前記類似度と前
記出力頻度とによって認識するようにしたもので多る。4) Structure of the Invention The character recognition device according to the present invention includes a recognition section that recognizes an input character image based on 4 degrees of similarity with a comparison standard, and 2.
'・Correcting means for correcting erroneous friend recognition and T-equipped with two correction methods for recognizing input character images and erroneous recognition:
There are many devices that output a table that records the number of characters for each character type, and recognize input characters based on the similarity and the output frequency.
ω）発明の実施例
次に不発明の要旨を山示災彪例によって具体的に説明す
、Ｓｏ
第１図に本発明の第１の実施例の得成金示し、１は入力
文半画１象の特徴を佃出し比較基準として後記辞書メモ
リに記憶される又字種毎の標準特徴との類似度と後記テ
ーブルに記録される文字種毎の出力頻度こに＝って入力
又字画像を認識する認識部、２は前記類似Ｘｋ得る九め
の比較基準として文字種毎の標準特徴？記憶する辞書メ
モ１ハ ３は文字種毎の出力頻度を記録するテーブル、
４は認識部１における誤った認識を１５正する修正手段
として用いる表示部、５は認識出力全記録する出力媒体
でるるっ
例凡ばめる入力文字ａｉ像に対し、この入力文字画像に
対する類似度が全く等しい複数の又字種が得られ之場合
、出力頻度の多い又字種が認識確度：び礪ハこ見ること
ができる０したで八って、認識部１において文字種毎の
出力頻度を考慮することにより、このような場合にも認
識が容易となり誤認識全減少することができる。このた
め表示部４によって修正せざるを得ないような場合が大
幅に減少する。な２、テーブル３の内容は認識結果もし
くは誤認識があっ反爾今、・ては傷正、砧果によって更
新される。ω) Embodiment of the invention Next, the gist of the non-invention will be concretely explained using an example of Yamashiba Biao. Recognizes the input or character image based on the similarity with standard features for each character type stored in the dictionary memory (described later) and the output frequency for each character type recorded in the table described later. Is the recognition unit 2 the standard feature for each character type as the ninth comparison standard to obtain the similarity Xk? Dictionary memo 1c 3 is a table that records the output frequency of each character type,
4 is a display unit used as a correction means for correcting erroneous recognition in the recognition unit 1; 5 is an output medium for recording all recognition output; and 5 is an output medium for recording all recognition outputs. If multiple character types with exactly the same frequency are obtained, the recognition accuracy for the character type with the highest output frequency is 0 and 8. By taking this into consideration, recognition becomes easier even in such cases, and misrecognition can be completely reduced. Therefore, the number of cases in which corrections must be made using the display section 4 is greatly reduced. 2. If there is a recognition result or a misrecognition, the contents of Table 3 will be updated based on the results.
第２崗は不発明の第２の実施例の構成を示し、第１図と
共通の符号は同一対家物金示すほか、１１は入力文字画
家の一トチ徴？：、！出し比較基準として辞書メモリ２
に記憶される文字種毎の標準特徴との類似度、／こよっ
て入力文字画１象全認識し類似度の大きい文字種金須似
度ととも１こ田刀する第１０館識部、１２に第１の認識
部１１．ｊＣよって得られた文字１Ｍが碌妃のｆ％Ｃで
チーフル３に記録される文字種毎り出力゛嗅裳を番１４
．ζしレニ識確度の高い順位に出力する第２の認識部で
める。The second diagram shows the structure of the second embodiment of the invention, in which the same symbols as those in FIG. :,! Dictionary memory 2 as a comparison standard
The degree of similarity with standard features for each character type stored in 1 recognition unit 11. The character 1M obtained by jC is output for each character type recorded in Chiful 3 with f%C of Rokuhi.
．． A second recognition unit outputs the results in the order of high recognition accuracy.
上記実施例はいずれも特徴抽出方式による文子認識装置
に本発明全通用した例であるが、重合わせ方式の文子認
識装置に対しても不発明を適用することかでさる。Although the above embodiments are all examples in which the present invention is fully applicable to a sentence recognition device using a feature extraction method, the present invention can also be applied to a sentence recognition device using a superimposition method.
１Ｇ）発明の詳細
な説明し足ようＶＣ不発明によルば既述従来例に２しす
る：Ａ醪譲上大横に減少レオベンータの負荷？軽減しく
４る文字認識装置を傅ωこにができ、う。1G) I would like to explain the invention in detail, but based on the VC invention, I would like to add 2 to the previously described conventional example: Is the load on the rheoventor significantly reduced in A-melting? It is possible to use a character recognition device that is easy to use.
第１Ａは不発明の第１の実１例、第２讃は、＝発明の第
２、つ央、：ｖａ Ｍ を示し、とｎらの図ＱＣ＞いて
１は２跪部、３，２テーブル、４は表示部、１１は第１
の認識部、１２に第２の認識部である。
代理人 弁理士 松 岡 云閣町つ
芒、御
嗜ｉ孟
第１図
第 ２ 図1A shows the first instance of non-invention, the second example shows = the second point of invention, : va M, and the figure QC of n et al. table, 4 is the display section, 11 is the first
12 is a second recognition unit. Agent: Patent Attorney Matsuoka, Yunkakucho Tsuho, Goshu Imeng Figure 1 Figure 2
Claims (1)
識部と前記認識部における誤った認識？、き正する１曝
正手段と全備え人力文字画家を認識し誤った認識に対し
ては修正？おこなって出力する装置において、文字種毎
の出力頻度を記録するテーブルを設け、人力文字画像を
前記類似度と前記出力頻度とによって認識すること全＋
ｆ黴とする文字認識装置。A recognition unit that recognizes an input character image based on its similarity to a comparison standard, and an incorrect recognition in the recognition unit? , Is there a way to correct 1 exposure and fully equipped human-powered character painters to recognize and correct erroneous recognition? In a device that performs and outputs characters, a table is provided to record the output frequency of each character type, and human character images are recognized based on the similarity and the output frequency.
A character recognition device that uses f mold.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP57070989A JPS58203578A (en) | 1982-04-27 | 1982-04-27 | character recognition device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP57070989A JPS58203578A (en) | 1982-04-27 | 1982-04-27 | character recognition device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| JPS58203578A true JPS58203578A (en) | 1983-11-28 |
Family
ID=13447450
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| JP57070989A Pending JPS58203578A (en) | 1982-04-27 | 1982-04-27 | character recognition device |
Country Status (1)
| Country | Link |
|---|---|
| JP (1) | JPS58203578A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0221383A (en) * | 1988-01-04 | 1990-01-24 | Sumitomo Electric Ind Ltd | Optical character reader |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPS5023258A (en) * | 1973-06-29 | 1975-03-12 | ||
| JPS5699583A (en) * | 1980-01-09 | 1981-08-10 | Nippon Telegr & Teleph Corp <Ntt> | Character decision processing system |
-
1982
- 1982-04-27 JP JP57070989A patent/JPS58203578A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPS5023258A (en) * | 1973-06-29 | 1975-03-12 | ||
| JPS5699583A (en) * | 1980-01-09 | 1981-08-10 | Nippon Telegr & Teleph Corp <Ntt> | Character decision processing system |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0221383A (en) * | 1988-01-04 | 1990-01-24 | Sumitomo Electric Ind Ltd | Optical character reader |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JPS58203578A (en) | character recognition device | |
| JPH0388062A (en) | Device for preparing document | |
| JP2000082110A (en) | Ruled line deletion device, character picture extraction device, ruled line deletion method, character picture extraction method and storage medium | |
| JPH0554072A (en) | Digital translation device | |
| JPS6272085A (en) | Character recognizing device | |
| JPH05298487A (en) | Alphabet recognizing device | |
| JPS61272882A (en) | Information recognizing device | |
| JPS5926974B2 (en) | Online handwritten character input device | |
| JPS5977579A (en) | Pattern recognizer | |
| JPH08180064A (en) | Document retrieval method and document filing device | |
| JPH02118786A (en) | Method for recognizing character | |
| JPS6175978A (en) | recognition device | |
| JP2891368B2 (en) | Post-processing method of character recognition result | |
| JP3476872B2 (en) | Character recognition device | |
| JPS61224088A (en) | Character pattern recognizing device | |
| Zheng et al. | A new method for baseline extraction of Manchu word | |
| JPH0520492A (en) | Document recognizing/correcting device | |
| JPS61198376A (en) | Optical character reader | |
| JP3725206B2 (en) | Character recognition device | |
| JPH0418352B2 (en) | ||
| JPS6320584A (en) | Document preparing device | |
| JPH04289989A (en) | Roman letter recognizing device | |
| JPH03217994A (en) | Document processor | |
| JPH0554089A (en) | Graphic recognizing device and recognition error correcting device | |
| JPS62134765A (en) | Electronic retrieving method for dictionary of chinese character explained in japanese |