JPS6054087A - On-line character recognizing device - Google Patents
On-line character recognizing deviceInfo
- Publication number
- JPS6054087A JPS6054087A JP58160772A JP16077283A JPS6054087A JP S6054087 A JPS6054087 A JP S6054087A JP 58160772 A JP58160772 A JP 58160772A JP 16077283 A JP16077283 A JP 16077283A JP S6054087 A JPS6054087 A JP S6054087A
- Authority
- JP
- Japan
- Prior art keywords
- character
- time
- threshold value
- section
- held
- 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.)
- Granted
Links
Landscapes
- Character Input (AREA)
- Character Discrimination (AREA)
Abstract
Description
【発明の詳細な説明】
本発明は、実時間で入力された筆跡をもと4こ文字をg
#lII!するオンライン文字認識装置の改良に関する
ものである。[Detailed Description of the Invention] The present invention is capable of converting four characters to g based on handwriting input in real time.
#lII! This invention relates to improvements to online character recognition devices.
従来、実時間で入力された筆跡をもとに文字を認識する
手法は既に存在する。たとえは特願昭56−19417
4号明細書(以下文献(1)と祢丁)に述べらねでいる
ようなバタンマツチングを用いた手法が有効である。文
字を認識する場合、書かれた筆跡の内どこからどこまで
が一つの文字であるかを決定しなければならない。この
決定方法には従来いくつかの方法があった。Conventionally, there is already a method for recognizing characters based on handwriting input in real time. Parable is patent application 1986-19417
A method using slam matching that is not described in the specification of No. 4 (hereinafter referred to as Document (1) and Necho) is effective. When recognizing characters, it is necessary to determine which part of a written handwriting constitutes a single character. There have been several methods for this determination.
代表的な一つの方法は、文字ヲ書りペンのペン先がタブ
レット上から離れてから次にタブレット上に清くまでの
時間、すなわち文字のストローク間の時間を計測して、
この時間がある定められた閾値をこえた時一つの文字の
書き終わりとする方法である。この方法では一つの文字
を書きはじめたら、その文字の終わりまでは休ますに−
気に書いてしまわなけれはならない。ひらがな、かたか
な、アルファベット、数字などのように画数の少ない文
字では文字を書いている途中でペンを休めることはほと
んどないが、画数の多い漢字などでは文字ヲ書いている
途中でペンを休めることも頻繁にある。この方法では、
まだ一つの文字を沓いている途中でも、ペンを休めると
誤って文字の終りと判断されてしまうため、文字を正し
く入力できないという欠点があった。One typical method is to measure the time from when the tip of a writing pen leaves the tablet to when it comes to rest on the tablet, that is, the time between character strokes.
This is a method in which one character is finished writing when this time exceeds a certain threshold. With this method, once you start writing a letter, you can rest until the end of the letter.
I have to write it down carefully. For characters with a small number of strokes, such as hiragana, katakana, alphabets, numbers, etc., it is almost impossible to rest the pen while writing them, but for characters with many strokes, such as kanji, it is possible to rest the pen while writing them. This happens often. in this way,
This had the disadvantage that if you stopped the pen while you were still in the middle of writing a single character, it would be mistakenly determined to be the end of the character, making it impossible to input the character correctly.
本発明の目的は、字種に応じてストローク間の時間の閾
値を替えることにより文字の省き終わり検出精度を上げ
て、オンライン文字認識装置の使い勝手を改善すること
にある。An object of the present invention is to improve the usability of an online character recognition device by increasing the accuracy of detecting the end of omitted characters by changing the threshold value of the time between strokes according to the character type.
本発明によれは、文字を書く際のあるストロ−クラ畏き
終わってから次のストロークを書き始めるまでの時間を
4測する計時部と、前記計時部において計測された時間
があらかじめ定めらfした閾値を越えたとき一文字の書
き終イつりと判断してその文字を認識する文字認識部と
、文字の字種を判別する字種判別部と、前記閾値の値を
複数個保持し前記字種の判別結果により前記閾値の値を
変更する閾値メモリ部とを有して成るオンライン文字認
識装置が得られる。According to the present invention, there is provided a timing unit that measures the time from the end of one stroke when writing a character to the start of writing the next stroke, and the time measured by the timing unit is predetermined. a character recognition unit that determines that a character has been written and recognizes the character when the threshold value exceeds the threshold; a character type discrimination unit that determines the character type of the character; There is obtained an online character recognition device comprising a threshold value memory section that changes the value of the threshold value according to the result of type discrimination.
以下具体的な一実施例に基づいて本発明の詳細な説明す
る。図は本発明の一実施例のブロック構成図である。タ
ブレット1より入力された文字のる。同時にタブレット
1よりペンのオン・オフ情報Pが計時部3に入力される
。これをもとに計時部3ではペンがオフになってから次
にオンになるまでの時間、すなわちストローク間の時間
Tが計測される。このストローク間の時間Tは比較部4
において閾値メモリ¥1IX5に保持されている閾値T
Tと比較される。閾値メモリ部5iこは各字種に対して
異なる閾値TTが保持されている。字種選択スイッチ6
ζこよりこわらの閾値の中から一つが献択され比較部4
に出力される。ストローク間の時間Tが閾値TTよりも
大きくなった時、比較部4からは認識開始信号Sが出力
される。この信号Sにより文字認識部2で(ま、丁でに
保持さ石ている筆跡情報Cをもとに文字が認識される。The present invention will be described in detail below based on a specific example. The figure is a block diagram of an embodiment of the present invention. Letters entered from Tablet 1. At the same time, pen on/off information P is input from the tablet 1 to the clock section 3. Based on this, the timer 3 measures the time from when the pen is turned off to when it is next turned on, that is, the time T between strokes. The time T between these strokes is determined by the comparing section 4.
Threshold T held in threshold memory ¥1IX5 in
Compared to T. The threshold value memory section 5i stores different threshold values TT for each character type. Character type selection switch 6
ζ One of the stiffer thresholds is selected and the comparison unit 4
is output to. When the time T between strokes becomes larger than the threshold value TT, the comparison unit 4 outputs a recognition start signal S. Based on this signal S, the character recognition unit 2 recognizes the character based on the handwriting information C held in the handwriting.
この文字認識部2には、文献(1)に述べられているよ
うなバタンマツチングを用いた手法i−使つ。This character recognition unit 2 uses a method i using slam matching as described in document (1).
たとえは、選択される字種として漢字さυ・らがなの2
柚を考えるものとす6゜この時、漢字の閾値としては1
秒、ひらがなの閾イ1自としては172秒と下る。これ
により、ひらがなの入力中は認識結果が早<’Jられ、
漢字入力中は多少ペンを休めても4(jiって文字の終
わりと判断されることはなくなるという利点が得られる
。For analogy, the selected character types are kanji sa υ and ragana.
Let's consider yuzu 6゜In this case, the threshold for kanji is 1
Seconds, the threshold for hiragana is 172 seconds. As a result, the recognition results will be faster when inputting hiragana.
The advantage is that even if you take a break from the pen while inputting kanji, 4 (ji) will not be judged as the end of the character.
以上、本発明による一実施例に関して述べたが本発明は
この実施例にのみ限定されるものではない。Although one embodiment of the present invention has been described above, the present invention is not limited to this embodiment.
字種の分けがたは、ここに述べた分は方の側にfこ、字
種の判別方法としてここではスイッチによる方法lこつ
いて述べたが、タブIy、)ト上の文字を病く位置イこ
より判別するという方法も使える。As for the classification of character types, the ones mentioned here are on one side, and as for the method of distinguishing character types, here I have described the method using a switch, but the characters on tabs Iy, ) and You can also use a method of determining based on position.
さらに本発明は文字枠により文字の書き終わりを検出す
る方法と共に用いることもできる。Tなわち、タブレッ
ト上にいくつかの文字枠をもうけておき、次の文字枠の
なかに文字を書き始めたら前の文字が完結したとする方
法である。この時ストロークlb1の時間がある一定の
時間を越えた場合は、まだ次の文字枠内に文字が書かれ
なくても文字の書き終わりとする。Furthermore, the present invention can also be used in conjunction with a method of detecting the end of a written character using a character frame. In other words, this is a method in which several character frames are created on the tablet, and when a character begins to be written in the next character frame, the previous character is completed. At this time, if the time of the stroke lb1 exceeds a certain fixed time, it is assumed that the writing of the character has ended even if the character has not yet been written in the next character frame.
文字認識部21こおける文字認識方式は現在知られてい
るすべての文字h3識方式を用いることかできる。As the character recognition method in the character recognition unit 21, all currently known character h3 recognition methods can be used.
図は本発明の一実施例のブロック構成図である。
図中、1はタブレ、)、2は文字認識部、3は計時部、
4は比1咬部、5(・ま閾値メモリ部、6は字種選択ス
イッチである。The figure is a block diagram of an embodiment of the present invention. In the figure, 1 is a tablet, ), 2 is a character recognition section, 3 is a timekeeping section,
4 is a ratio 1 bite part, 5 is a threshold value memory part, and 6 is a character type selection switch.
Claims (1)
ストロ−クラ看き始めるまでの時間ftl測する計時部
さ、前記計時部−こおいて計測された時間があらかじめ
定められた閾値そ越えたとき一文字の書き終わりと判断
してその文字を認識する文字認識部と、文字の字種全判
別する字種判別部と、前記閾値の値そり数個保持し前記
字種の判別結果により前記閾値の値を変更する閾値メモ
リ部とを有して成るオンライン文字認識装置。A timing unit that measures the time from the end of writing a certain stroke of a character to the start of the next stroke, and the time measured by the timing unit exceeds a predetermined threshold. a character recognition unit that recognizes the end of a character and recognizes that character; a character type discrimination unit that discriminates all the character types of the character; An online character recognition device comprising: a threshold memory unit for changing a threshold value.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP58160772A JPS6054087A (en) | 1983-09-01 | 1983-09-01 | On-line character recognizing device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP58160772A JPS6054087A (en) | 1983-09-01 | 1983-09-01 | On-line character recognizing device |
Publications (2)
Publication Number | Publication Date |
---|---|
JPS6054087A true JPS6054087A (en) | 1985-03-28 |
JPH0410670B2 JPH0410670B2 (en) | 1992-02-26 |
Family
ID=15722116
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
JP58160772A Granted JPS6054087A (en) | 1983-09-01 | 1983-09-01 | On-line character recognizing device |
Country Status (1)
Country | Link |
---|---|
JP (1) | JPS6054087A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH01270185A (en) * | 1988-04-22 | 1989-10-27 | Canon Inc | Character recognition device |
-
1983
- 1983-09-01 JP JP58160772A patent/JPS6054087A/en active Granted
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH01270185A (en) * | 1988-04-22 | 1989-10-27 | Canon Inc | Character recognition device |
Also Published As
Publication number | Publication date |
---|---|
JPH0410670B2 (en) | 1992-02-26 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP2009543204A (en) | Handwritten symbol recognition method and apparatus | |
US6369807B1 (en) | Online character entry device | |
JPS6054087A (en) | On-line character recognizing device | |
JPS592191A (en) | Recognizing and processing system of handwritten japanese sentence | |
CN100386774C (en) | Presenting recognised handwritten symbols | |
JP3391257B2 (en) | Kana-kanji automatic conversion method and kana-kanji automatic conversion device | |
Kotak et al. | An accelerometer based handwriting recognition of English alphabets using basic strokes | |
JPH04337890A (en) | Device for recognizing on-line handwritten character | |
JPS59188787A (en) | Input device of on-line manuscript character | |
JP7285018B2 (en) | Program, erroneous character detection device, and erroneous character detection method | |
JPS6197790A (en) | On-line character recognizing device | |
JPS6142083A (en) | Character recognition device | |
JPH046995B2 (en) | ||
JPS61150081A (en) | Character recognizing device | |
JPH0527916B2 (en) | ||
JP3051628B2 (en) | Character recognition method and device | |
JPS6133583A (en) | Character segmentation device | |
JPS6083185A (en) | On-line character recognizing system | |
JPH046996B2 (en) | ||
JPH0981668A (en) | Character recognition method and device | |
JPS6041394B2 (en) | Character reading method | |
JPH11126236A (en) | Device and method for on-line handwritten character recognition, and recording medium where the same method is recorded | |
KR100367580B1 (en) | Device for recognizing on-line character of stroke order independence | |
JPH023222B2 (en) | ||
JPS6293776A (en) | Information recognizing device |