JPS61193282A - Hand-written character recognizing device - Google Patents

Hand-written character recognizing device

Info

Publication number
JPS61193282A
JPS61193282A JP60031818A JP3181885A JPS61193282A JP S61193282 A JPS61193282 A JP S61193282A JP 60031818 A JP60031818 A JP 60031818A JP 3181885 A JP3181885 A JP 3181885A JP S61193282 A JPS61193282 A JP S61193282A
Authority
JP
Japan
Prior art keywords
character
pattern
input
writer
code
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
JP60031818A
Other languages
Japanese (ja)
Inventor
Hiroki Onishi
宏樹 大西
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.)
Sanyo Electric Co Ltd
Original Assignee
Sanyo 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 Sanyo Electric Co Ltd filed Critical Sanyo Electric Co Ltd
Priority to JP60031818A priority Critical patent/JPS61193282A/en
Publication of JPS61193282A publication Critical patent/JPS61193282A/en
Pending legal-status Critical Current

Links

Abstract

PURPOSE:To enable correct recognition regarding character input thereafter by freshly registering input pattern in case of mis-recognition when a character is inputted. CONSTITUTION:When 'OO' 'SAKA' are written in the character writing area T1 and when the order of stroke of the input character is wrong, or when the character is of style peculiar to the writer causing mis-recognition, e.g. in case 'SAKA' is recognized as 'NUKU', a character pattern of input character 'SAKA' mis-read as this writer's character 'NUKU' is freshly registered. This procedure is designated with a pen in the area T2 of the tablet 1 designating registration mode, and after setting to registration mode, character code in inputted by designating with a pen, hexadecimal number from O-F, the character code to be designated in the character code designation area T0. When the JIS code is not matched therewith, the writer is informed by display and re-inputted with the character code.

Description

【発明の詳細な説明】 イ)産業上の利用分野 本発明は手書きの文字を認識する事のできる文字認識装
置に関する。
DETAILED DESCRIPTION OF THE INVENTION A) Field of Industrial Application The present invention relates to a character recognition device capable of recognizing handwritten characters.

口)従来の技術 日本語ワードプロセサやパーソナル・コンピュータが身
近になってきている今日、キーボード操作に慣れない人
、キーボード操作を好まない人もなかにはいる。そんな
人に向けた、はとんど熟練を必要としない日本語入力装
置として手書き文字認識装置が最近製品化され始め、研
究も活発になっており、斯様な装置については日経エレ
クトロニクス1983年12月5日号の記事1くずし文
字など筆記制限を緩和する方向に進むオンライン手書き
漢字認識」に詳しい。
口)Conventional TechnologyJapanese Today, as word processors and personal computers become more commonplace, there are some people who are not accustomed to keyboard operations or who do not like keyboard operations. Handwritten character recognition devices have recently begun to be commercialized as Japanese input devices for such people that do not require much skill, and research has become active. For details, see article 1 in the May 5th issue, ``Online handwritten kanji recognition moving toward easing handwriting restrictions such as broken characters''.

第213Uに従来の手書き文字認識装置の構成図を示す
。同図に於いて(1)は文字入力の為のタブレット部で
ペンの圧力や電磁誘導等でその座標位置を検出する。(
2)はタブレット部(1)からの入力を受ける入力部、
(3)は入力座標データを一時的に格納する入力バッフ
ァである。この入力座標データは、特徴抽出部(4)で
特徴抽出され、その文字パターンが入力パターンバッフ
ァ(5)に格納される。(6)は識別部であり、入力パ
ターンと参照パターンメモリ(7)内の番手書き文字に
対応する標準パターンと比較し、最も類似した標準バタ
−ンに対応する文字フードを出力する、即ちパターン認
識する。(8)は表示部であり、識別部(6)からの文
字コードに対しその文字を表示し、筆者に知らせるもの
である。
213U shows a configuration diagram of a conventional handwritten character recognition device. In the figure, (1) is a tablet section for inputting characters, and its coordinate position is detected by pen pressure, electromagnetic induction, etc. (
2) is an input unit that receives input from the tablet unit (1);
(3) is an input buffer that temporarily stores input coordinate data. This input coordinate data is subjected to feature extraction by a feature extraction section (4), and its character pattern is stored in an input pattern buffer (5). (6) is an identification unit which compares the input pattern with the standard pattern corresponding to the number handwriting character in the reference pattern memory (7) and outputs the character hood corresponding to the most similar standard pattern, that is, the pattern recognize. (8) is a display section, which displays the character in response to the character code from the identification section (6) and informs the writer.

ここで、参照パターンメモリ(7)の−例を第3図に示
す。同図のメモリ(7)内には、多数の人間の手書き文
字を分析して得た座標データ等の標準パターンを表すデ
ータと、文字フードとが対応付けられて格納されている
Here, an example of the reference pattern memory (7) is shown in FIG. In the memory (7) in the figure, data representing a standard pattern, such as coordinate data obtained by analyzing handwritten characters of a large number of people, and character hoods are stored in association with each other.

ハ)発明が解決しようとする問題点 上述の如き従来の手書き文字認識装置は、入力文字をパ
ターン化した入力パターンを予め格納されている標準パ
ターンと比較し、その中のどのパターンに対応している
かのみを識別・判定しているので、筆順は違ったり、筆
者特有のクセのある字は、標準文字パターンとの誤差が
大きくなり、この結果、誤認識を招く欠点があった。
C) Problems to be Solved by the Invention The conventional handwritten character recognition device as described above compares an input pattern of input characters with pre-stored standard patterns, and determines which pattern among them corresponds to the input pattern. Since only dolphins are identified and judged, if the stroke order is different or if the characters have a unique character, there will be a large error with the standard character pattern, resulting in erroneous recognition.

二)問題点を解決する為の手段 本発明の手書き文字認識装置は、標準パターンの他に筆
者の文字パターンを登録する文字パターン登録手段設け
、この登録きれた筆者自身の文字パターンを参照して入
力パターンを認識するものである。
2) Means for solving the problem The handwritten character recognition device of the present invention is provided with a character pattern registration means for registering the writer's own character pattern in addition to the standard pattern, and refers to the writer's own character pattern that has been registered. It recognizes input patterns.

ホ) 作用 本発明に依れば、文字入力した時に誤認識した場合の入
力パターンを新たに登録しておくことに依り、以後のそ
の文字入力に対しては正しく認識される事になる。
(e) Effect: According to the present invention, by newly registering an input pattern in the case where a character is erroneously recognized when inputting a character, the character will be correctly recognized in subsequent inputs.

へ) 実施例 第1図に本発明の手書き文字認識装置の一実施例のブロ
ック図を示す。同図に於いて、第2図の従来装置と同じ
ブロックには第2図と同じ符号を付している。同図の本
発明装置が従来装置と異なる所は、参照パターンメモリ
(70)にある、即ち、該メモリ(70)のROM構成
の標準パターン格納領域(71)には、従来装置の参照
パターンメモリ(7)と同様に手書き文字の標準パター
ンが文字コードと対応付けられて格納されているが、該
メモリ(70)のRAM構成の登録パターン格納領域(
72)には、入力パターンバッファ(5)の入力パター
ンがそのまま参照されるべき登録パターンとしてその時
の入力文字の文字コードと対応付けられて格納されるの
である時開、この時の参照パターンとして登録する。入
力パターンに対応する文字フードの指定方法としては、
斯る実施例の場合、後述する如くタブレット部(1)か
らの数字入力に依って行なわれる。
f) Embodiment FIG. 1 shows a block diagram of an embodiment of the handwritten character recognition device of the present invention. In this figure, blocks that are the same as those of the conventional device in FIG. 2 are given the same reference numerals as in FIG. 2. The difference between the device of the present invention shown in the figure and the conventional device is in the reference pattern memory (70), that is, the standard pattern storage area (71) of the ROM configuration of the memory (70) contains the reference pattern memory of the conventional device. Similar to (7), the standard pattern of handwritten characters is stored in association with the character code, but the registered pattern storage area of the RAM configuration of the memory (70)
72), the input pattern in the input pattern buffer (5) is stored as a registered pattern that should be referred to as it is, in association with the character code of the input character at that time. do. To specify the character hood corresponding to the input pattern,
In the case of such an embodiment, the operation is performed by inputting numbers from the tablet unit (1), as will be described later.

次に、第4図乃至第6図を参照しながら、第1図の本発
明実施例装置の動作を詳述する。先ず第4図は表示部(
8)を備えたタブレット部(1)の指示エリアを模式的
に示しており、入力モードに於て第5図に示す如く、文
字を書くエリア(T1)にr大」1阪」と書いた時、そ
の入力座標データが入力部(2′)から入力バッファ(
3)に格納きれ、さらに特徴抽出部(4)で抽出された
文字パターンが入力パターンバッファ(5)に格納され
る。そして、参照パターンメモリ(70)の全ての参照
パターン、ただし始めは登録パターン格納領域(72)
には何も登録されていないので、実質的にその標準パタ
ーン格納領域(71)の各標準パターンに基づいて識別
部がパターン認識する。ところが、この時の入力文字の
筆順が違ったり、筆者特有のクセのある字であったりす
る事に起因して、誤認識が生じ、例えばr阪、を「抜」
と誤認識した場合、第5図に示す如く、表示部(8)に
1大」r抜、と表示される事となる。尚、同図の表示部
(8)の一部には入力モードを示す図表示がなされてい
る。
Next, the operation of the apparatus according to the embodiment of the present invention shown in FIG. 1 will be described in detail with reference to FIGS. 4 to 6. First, Figure 4 shows the display section (
This diagram schematically shows the instruction area of the tablet unit (1) equipped with When the input coordinate data is transferred from the input section (2') to the input buffer (
3), and the character pattern extracted by the feature extraction unit (4) is stored in the input pattern buffer (5). Then, all the reference patterns in the reference pattern memory (70), however, the first one is in the registered pattern storage area (72).
Since nothing is registered in the standard pattern storage area (71), the identification unit performs pattern recognition substantially based on each standard pattern in the standard pattern storage area (71). However, due to the stroke order of the input characters being different or the characters having a peculiar character unique to the author, erroneous recognition occurs, and for example, r is not written as "excluded".
If it is erroneously recognized as such, as shown in FIG. 5, the display section (8) will display ``1 large'' r omitted. It should be noted that a part of the display section (8) in the same figure is graphically displayed to indicate the input mode.

ここで、この筆者の文字「抜」と誤認識きれる入力文字
「阪」の文字パターンを新たに登録する。その手順は、
登録モードを指定するタブレット部(1)のエリア(T
2)をペンで指定し、登録モードに設定してからO−F
までの16進数に文字コードを指定する文字フード指定
エリア〈T3)をペンで指定する事に依って文字フード
具体的にはJISフードを入力する。このJISフード
は入力部(2)から表示部(8)に与えられ第6図に示
す如く、表示部(8)にて1阪」なる入力文字フード呻
対応した表示がなされるが、このJISコードが違って
いた場合は、筆者は表示に依ってそれを知り、文字フー
ドの再入力を行なうのである。尚、第6図の表示部(8
)の面表示は登録モードを示している。このようにして
正しい文字フードが入力きれた事を確認できたなら、タ
ブレット部(1〉の実行エリア(T4)をペンで指示す
る。これに依って、入力部(2′)は参照パターンメモ
リ(70)の登録パターン格納領域(72)に書き込み
信号を与え、上記入力パターンバッファ〈5)に格納さ
れているこの時の入力パターンと指示きれた文字フード
とが互いに対応づけられて登録パターン格納領域(72
)に格納される。このようにして文字パターンの登録が
終了すると、これと同時に入力モードに復帰するのであ
る。
Here, a new character pattern for the input character ``Saka'' that can be mistakenly recognized as the author's character ``Nuki'' is registered. The procedure is
Area (T) of the tablet section (1) that specifies the registration mode
2) Specify with a pen, set to registration mode, and then O-F
Input the character hood, specifically the JIS hood, by specifying the character hood designation area (T3) in hexadecimal digits with a pen. This JIS hood is applied from the input section (2) to the display section (8), and as shown in Fig. 6, a display corresponding to the input character hood "1" is displayed on the display section (8) as shown in Fig. 6. If the code is incorrect, I will know from the display and re-enter the code. In addition, the display part (8
) indicates the registration mode. Once you have confirmed that the correct character hood has been input in this way, point the execution area (T4) of the tablet section (1>) with the pen.This allows the input section (2') to move into the reference pattern memory. A write signal is given to the registered pattern storage area (72) of (70), and the input pattern at this time stored in the input pattern buffer (5) and the specified character hood are associated with each other and the registered pattern is stored. Area (72
). When the character pattern registration is completed in this way, the input mode is returned to at the same time.

従って、この後の入力モードに於いては、参照パターン
メモリ(70)の標準パターン格納領域(71)のパタ
ーンに加えてその登録パターン格納領域(72)のパタ
ーンをも参照してパターン認識できるので、上述の如く
して一度登録した文字については誤認識する惧れはなく
、新たに誤認識する文字については夫々最初に上述の如
き登録処理を行なえばよい。
Therefore, in the subsequent input mode, pattern recognition can be performed by referring to the pattern in the registered pattern storage area (72) in addition to the pattern in the standard pattern storage area (71) of the reference pattern memory (70). There is no risk of erroneously recognizing characters that have been registered once as described above, and it is sufficient to first perform the registration process as described above for each newly erroneously recognized character.

ト〉 発明の効果 本発明の手書き文字認識装置は、以上の説明から明らか
な如く、標準パターンの他に筆者自身の文字パターンを
用いて、入力文字をパターン化した入力パターンに対し
てパターン認識処理できるので、誤認識が解消きれ、こ
れに依って手書き文字の入力が確実に行なえ、しかもそ
の入力速度も向上する。
G> Effects of the Invention As is clear from the above description, the handwritten character recognition device of the present invention uses the writer's own character pattern in addition to the standard pattern to perform pattern recognition processing on an input pattern obtained by patterning input characters. As a result, misrecognition can be eliminated and handwritten characters can be input reliably, and the input speed can also be improved.

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

第1図は本発明の手書き文字認識装置の一実施例の構成
を示すブロック図、第2図は従来装置の構成を示すブロ
ック図、第3図は従来装置のメモリ図、第4図は乃至第
6図は本発明の要部模式図である。 (1)・・・タブレット部、(2)(2’)・・・入力
部、(3)・・・入力バッファ、く4)・・・特徴抽出
部、(5)・・・入力パターンバッファ、(6〉・・・
識5J+Jm、(7)(70)・・・参照パターンメモ
リ、(8)・・・表示部、(71)・・・標準パターン
格納領域、(72)・・・登録パターン格納領域。 第1図 第2図 第4図 第6図
FIG. 1 is a block diagram showing the configuration of an embodiment of the handwritten character recognition device of the present invention, FIG. 2 is a block diagram showing the configuration of a conventional device, FIG. 3 is a memory diagram of the conventional device, and FIG. 4 is a block diagram showing the configuration of a conventional device. FIG. 6 is a schematic diagram of the main part of the present invention. (1)...Tablet part, (2)(2')...Input part, (3)...Input buffer, 4)...Feature extraction part, (5)...Input pattern buffer , (6〉...
ID5J+Jm, (7) (70)...Reference pattern memory, (8)...Display section, (71)...Standard pattern storage area, (72)...Registered pattern storage area. Figure 1 Figure 2 Figure 4 Figure 6

Claims (1)

【特許請求の範囲】[Claims] 手書き文字をパターン化してなる入力パターンを予め参
照パターンメモリに格納された複数の参照パターンに基
づいてパターン認識する手書き文字認識装置に於いて、
上記参照パターンメモリは、標準的な複数の手書き文字
をパターン化してなる標準パターンが格納されている標
準パターン格納領域と、これとは別に書き込み可能な登
録パターン格納領域と、からなり、該登録パターン格納
領域に筆者の手書き文字をパターン化した登録パターン
を格納する文字パターン登録手段を設けた事を特徴とし
た手書き文字認識装置。
In a handwritten character recognition device that recognizes an input pattern formed by patterning handwritten characters based on a plurality of reference patterns stored in advance in a reference pattern memory,
The reference pattern memory is composed of a standard pattern storage area in which a standard pattern formed by patterning a plurality of standard handwritten characters is stored, and a registered pattern storage area that can be written separately from the standard pattern storage area. A handwritten character recognition device characterized in that a character pattern registration means is provided for storing a registered pattern of a writer's handwritten characters in a storage area.
JP60031818A 1985-02-20 1985-02-20 Hand-written character recognizing device Pending JPS61193282A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP60031818A JPS61193282A (en) 1985-02-20 1985-02-20 Hand-written character recognizing device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP60031818A JPS61193282A (en) 1985-02-20 1985-02-20 Hand-written character recognizing device

Publications (1)

Publication Number Publication Date
JPS61193282A true JPS61193282A (en) 1986-08-27

Family

ID=12341667

Family Applications (1)

Application Number Title Priority Date Filing Date
JP60031818A Pending JPS61193282A (en) 1985-02-20 1985-02-20 Hand-written character recognizing device

Country Status (1)

Country Link
JP (1) JPS61193282A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPS6381586A (en) * 1986-09-25 1988-04-12 Toshiba Corp Information processor
JPH05217025A (en) * 1991-10-31 1993-08-27 Internatl Business Mach Corp <Ibm> Apparatus and method for on-line handwritten- character recognition

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPS6381586A (en) * 1986-09-25 1988-04-12 Toshiba Corp Information processor
JPH05217025A (en) * 1991-10-31 1993-08-27 Internatl Business Mach Corp <Ibm> Apparatus and method for on-line handwritten- character recognition

Similar Documents

Publication Publication Date Title
JPS60217477A (en) Handwritten character recognizing device
JPH02266485A (en) Information recognizing device
JPS61193282A (en) Hand-written character recognizing device
JP4935459B2 (en) Character recognition method, character recognition program, and character recognition device
JP5455364B2 (en) Information processing apparatus and locus data recognition method
JPS62159291A (en) Sentence input device
JP2012234345A (en) Form reader, square seal detection method, and square seal detection program
JPH05225399A (en) Document processor
JPS6228889A (en) Information recognizing device
JPH0442316A (en) Electronic computer
JP2598894B2 (en) Handwritten character recognition device
JPH0527914B2 (en)
JPS6111891A (en) Recognizing device of hand-written character/picture
JPS608991A (en) Registering system of on-line handwriting character recognizing dictionary
JP2731394B2 (en) Character input device
JPH053631B2 (en)
JPS638513B2 (en)
JPS60217483A (en) Recognizer of character
JP3365537B2 (en) Online character recognition method and apparatus
JP2922512B2 (en) Character recognition device
JPS61120286A (en) Recognizer for handwritten character
JP2665488B2 (en) Personal dictionary registration method
JPS62286190A (en) Identification system for online voiced sound and semi-voiced sound
JP2011209814A (en) Character input device and character input method
JPH0731708B2 (en) Handwriting recognition device