JPS5835678A - Character recognizing device - Google Patents
Character recognizing deviceInfo
- Publication number
- JPS5835678A JPS5835678A JP56133545A JP13354581A JPS5835678A JP S5835678 A JPS5835678 A JP S5835678A JP 56133545 A JP56133545 A JP 56133545A JP 13354581 A JP13354581 A JP 13354581A JP S5835678 A JPS5835678 A JP S5835678A
- Authority
- JP
- Japan
- Prior art keywords
- category
- character
- candidate
- displayed
- resemblance
- 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
- 150000003839 salts Chemical class 0.000 claims description 9
- 241000282412 Homo Species 0.000 description 3
- XEEYBQQBJWHFJM-UHFFFAOYSA-N Iron Chemical compound [Fe] XEEYBQQBJWHFJM-UHFFFAOYSA-N 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000000034 method Methods 0.000 description 2
- 238000013139 quantization Methods 0.000 description 2
- 235000016496 Panda oleosa Nutrition 0.000 description 1
- 240000000220 Panda oleosa Species 0.000 description 1
- CWFOCCVIPCEQCK-UHFFFAOYSA-N chlorfenapyr Chemical compound BrC1=C(C(F)(F)F)N(COCC)C(C=2C=CC(Cl)=CC=2)=C1C#N CWFOCCVIPCEQCK-UHFFFAOYSA-N 0.000 description 1
- 229910052742 iron Inorganic materials 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
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
- G06V30/18—Extraction of features or characteristics of the image
- G06V30/1801—Detecting partial patterns, e.g. edges or contours, or configurations, e.g. loops, corners, strokes or intersections
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】 この56W14は文字認識装置に関する。[Detailed description of the invention] This 56W14 relates to a character recognition device.
大学的に文字記入@票を走査して、文字を自動認識する
、いわ(少る0Ca(光学的文字認識装置)は英字、数
字、カタカナ、特殊記号迄認識出来るようになった。し
かし、日本人が日常用いる漢字を、含んだ文書を日本語
ワードプロセッサなどに入力するために用いる段渚には
まだ至っていない。A university-based system that automatically recognizes characters by scanning character entry @ slips, has become able to recognize letters, numbers, katakana, and even special symbols. It has not yet reached the stage where people use kanji that people use on a daily basis to enter documents that contain them into Japanese word processors.
タブレットに向ってベンを動して文字を書き、認識する
オンライン文字認識が、新しい入力手段として注目され
つつあるが、能力的にも不十分で実用はまだである〇
一方漢字キーボードの代りに1かな漢字変換による方法
が普及しつつあるが、不十分な変換率、。Online character recognition, which allows users to write and write characters by moving a pen toward a tablet, is attracting attention as a new input method, but its capabilities are insufficient and it has yet to be put into practical use. Methods using 1-kana-kanji conversion are becoming popular, but the conversion rate is insufficient.
かなキーの探車、制御用のキーボードの操作などのため
まだ問題は多い〇
このような事情からOCRによる日本語の入力装置に依
然として大会なルJ待が寄せられている0曽遮のように
、OCaでは、100−正しく入力出来る技術はまだ開
発されていないものの、入力文字に対して複数個の候補
文字を挙げれば、候補の中に正しい解がある割合は、非
常に高くなって来つつある。There are still many problems due to the search for the car with the kana key, the operation of the keyboard for control, etc. Due to these circumstances, Japanese input devices using OCR are still being asked for a long time. , in OCa, although the technology that allows 100-correct input has not yet been developed, if multiple candidate characters are listed for an input character, the percentage of correct answers among the candidates is becoming extremely high. be.
このようなレベルの認識技術を生かすために、人間が機
械に多少は歩み寄っても人間か持つ文章解読力を利用出
来るようにすれば人間の負担は、認識装置以外の入力装
置を用いる場合の負担よりもはるかに着くすることが出
来る〇
本発明ては、人力文字に対する類似度を調べ、同一であ
る可能性の多い文字カテゴリーを可能性の多いj14c
複数個選択し、しかる後、人間のオペレータが、正しい
文字カテゴリーを指示し易いように入力文字の順に従っ
て、候補カテゴリー塩を豪数文字分表示し、正しい文字
カテゴリーが第冨位以下にある時にオペレータが指示す
るだけて正しい文字コードが入力されるようにした文字
g鐵装置を提供する@
以下、図面に従って詳細に説明する。In order to take advantage of this level of recognition technology, if humans are able to make use of the text deciphering ability that humans have, even if they are a little closer to machines, the burden on humans will be reduced from the burden of using input devices other than recognition devices. 〇The present invention examines the degree of similarity to human-powered characters, and selects character categories that are likely to be the same.
After selecting multiple items, the human operator displays candidate categories for several characters in the order of input characters so that the human operator can easily indicate the correct character category, and when the correct character category is below the maximum character category, To provide a character g iron device in which a correct character code can be input simply by an operator's instructions.
第1図は本発明の一実施例を示すブロック図である0
文字■處I11は、文書の文字を光学的に走査し量子化
することで得られた入力文字の量子化パターンと、予め
登録されたカテゴリー名付きの標準パターン(または暉
鐵用辞書)と解合して#A似縦を求め、一定値以上の大
きさの類似度の大きさに応じてカテゴリー塩をN個以下
選び候補カテゴリー塩りストとして信号線11を通じて
表示部4に表示する。FIG. 1 is a block diagram showing an embodiment of the present invention.Character I11 contains a quantization pattern of input characters obtained by optically scanning and quantizing characters of a document, and a quantization pattern registered in advance. The standard pattern with the category name (or the dictionary for Kotetsu) is combined to find the #A similar length, and N or less category salts are selected as candidates depending on the degree of similarity that is greater than a certain value. It is displayed on the display unit 4 through the signal line 11 as a category list.
例えば、第2図のような文書の一行がii識された時′
、第3図のように候補カテゴリ御名リストが1行文字分
表示される0 「天」以外は全て正しい力゛テゴリー名
が候補カテゴリ御名リストの第1位に表示されているが
「天」だけがjlllK2位に示されている〇
文書上の1行分の文字がamされると1信号隷■に信号
“loが送られ、制御部3は、文字I[識を停止し、オ
ペレータの指示待ちの状態となるような信号を信号−蕊
に出力する。For example, when a line of a document like the one shown in Figure 2 is recognized,
, as shown in Figure 3, the list of candidate category names is displayed for one line of characters.0 All of the names except "Heaven" are correct.The category names are displayed in the first place in the candidate category name list, but only "Heaven" is displayed. is shown in the second position of jlllK. When one line of characters on the document is amended, the signal "lo" is sent to the 1st signal slave, and the control unit 3 stops recognizing the character I and responds to the operator's instructions. Outputs a signal that causes a wait state to the signal pin.
オペレータは、指示待ちの状態になると、指示部Sを用
いて、正しいカテゴリー塩が第り位(L−2,3、・・
・・・・N)以下に表示されているならば、その第1位
を指示する。指示された第1位のカテゴリー塩は、第1
位に移動する。When the operator is waiting for an instruction, the operator uses the instruction section S to determine the correct category salt (L-2, 3,...
...N) If it is displayed below, indicate the first place. The indicated first category salt is
Move to position.
指示部5は、種々の実現の手段がある0キーボードで実
現した場合、キーボードの「5」とrBJを順にキーイ
ンすると、「天」を「夫」のある場所に移動させること
は容易である。When the instruction unit 5 is implemented using a 0 keyboard, which has various implementation means, it is easy to move "Ten" to the location where "Husband" is located by keying in "5" and rBJ on the keyboard in sequence.
ライトペンを用いるなら、ライトペンを「天」の所で押
すだけで良い。カーソルを用いるなら、カーソルを「天
」の所に移動し、 指示した乙とを示すキーを押す・
また、特にマンマシン的に優れたデバイスとしてCRT
ディスプレイ上に透明パネルを重ね、 オペレータが指
などで軽く押すと押した場所を検出することの出来るタ
ッチセンシティブデバイスが開発されているが、この場
合は、オペレータが因の部分を指などで−すだけで「天
コの場所を検知し、その場所にある文字「天」を昧」の
所に移動することか出来る〇
表示部のA欄の文字が正しいカテゴリー塩になった段階
で、オペレータが、予め定められた指示終了キーを押せ
ば信号線s3を通じて終了状態が制御l!3に知られる
。If you use a light pen, just press the light pen at the ``heaven'' point. If you use a cursor, move the cursor to ``heaven'' and press the key that indicates the specified item.
Touch-sensitive devices have been developed in which a transparent panel is layered on top of a display and the operator can detect the location of the touch when the operator presses lightly with a finger. It is possible to "detect the location of the sky and move to the location of the character 'Ten'" at that location by just , when a predetermined instruction end key is pressed, the end state is controlled through the signal line s3! Known to 3.
この後、信号線34に信号“1′″が送られると表示部
A―の文字コードを、日本語ワードプウセシ量などの外
部の機器かフロッピーディスクか、この入力装置が持っ
ているメモリに信号fg40 を通じて、(113図
の場合は左から)贋に移動記帰専せる〇移動配置が終了
すると、信号線31を通じて制御部3は、次の行の5文
字の認識を指令する。After that, when the signal "1'" is sent to the signal line 34, the character code on the display section A is transferred to an external device such as a Japanese word processor or a floppy disk, or to the memory of this input device using the signal fg40. When the movement arrangement is completed, the control unit 3 issues a command to recognize five characters in the next line through the signal line 31 (from the left in the case of FIG. 113).
以上説明したように、本発明による文字認識装置によれ
ば、文字&!11部によって、入力文字に対するa偏度
の大きさkよってN@以下の候補カテゴリーが選択され
て候補カテゴリ御名リストが形成され、表示部に、入力
文字に対応した候補カテゴリ一本リストが1個以上彼数
個分表示され、表示部に表示された候補カテゴリ一本リ
ストのうち正解が2位以下にある時に正解文字を指示す
るだけで入力すべ会文字を正しく入力する仁とが出来る
ので、漢字キーボードなどの入力装置の場合のように、
キーボード上の文字を探す必要もなく速く入力するため
の特別の訓練も不必要である0説明では、文字認識部は
光学的なi!織部としたが、オンライン文字認識として
も良いことは明らかである。As explained above, according to the character recognition device according to the present invention, the characters &! In the 11th section, candidate categories of N@ or less are selected according to the magnitude k of the a bias against the input characters to form a candidate category name list, and a list of one candidate category corresponding to the input characters is displayed on the display section. As mentioned above, when the correct answer is in the second place or lower in the list of one candidate category displayed on the display section, you can input all the characters correctly by simply indicating the correct answer. As in the case of input devices such as kanji keyboards,
There is no need to search for characters on the keyboard and no special training is required to type quickly.In the 0 explanation, the character recognition unit is an optical i! Oribe is used here, but it is clear that online character recognition may also be used.
絡1図は、本発明の一実施例を示すブロック図、第2図
は文書上の文字列の1部、蕗3図は、候補カテゴリ御名
リストの表示列を夫々表わす。
図において、
1は文字Ig織部、 3は制御部、 4は表示部、
・ 5は指示部。
第1図
オ 2 図
t 3 図
+ 2 3 4 5 6 7 @ Q
IQ II505Figure 1 is a block diagram showing an embodiment of the present invention, Figure 2 is a part of a character string on a document, and Figure 3 is a display column of a list of candidate category names. In the figure, 1 is the character Ig Oribe, 3 is the control section, 4 is the display section,
・ 5 is the instruction section. Figure 1 O 2 Figure t 3 Figure + 2 3 4 5 6 7 @Q
IQ II505
Claims (1)
パターンとを照合してS偏度を求め、一定値以上の類似
度の大なる願に、N個以下のカテゴリー塩を入力文字に
対する候補カテゴリー49ストとして付与する文字認識
部と、各入力文字に付与されたN個以下の候補カテゴリ
ー49ス[を複数入力文手分、表示装置Elii1時に
表示する表示部と、該表示装置に表示された候補カテゴ
リ御名リストのうち、正しいカテゴリー塩が2位以下の
第i位にある時は、オペレータが311位の候補カテゴ
リー塩を第1位に入れるよう化指示し、表示中の候補カ
テゴリー塩の指示が終了したら、表示中の全ての第1位
のカテゴリー塩が、入力文字のカテゴリー塩として出力
されるようにした指示部とで構成されたことを**とす
る文字lI識懺装。The S bias is calculated by comparing the input character with the standard pattern of the category names registered in advance, and if the degree of similarity is greater than a certain value, N or less category salts are added as candidates for the input character. a character recognition unit that assigns the category 49 marks, a display unit that displays the N or less candidate categories 49 marks assigned to each input character for multiple input sentences; If the correct category salt is in the i-th position below the 2nd place in the list of candidate category names, the operator instructs the 311st candidate category salt to be placed in the first position, and then changes the displayed candidate category salt. When the instruction is completed, all the first category salts being displayed are configured with an instruction part that is output as the category salt of the input character.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP56133545A JPS5835678A (en) | 1981-08-26 | 1981-08-26 | Character recognizing device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP56133545A JPS5835678A (en) | 1981-08-26 | 1981-08-26 | Character recognizing device |
Publications (1)
Publication Number | Publication Date |
---|---|
JPS5835678A true JPS5835678A (en) | 1983-03-02 |
Family
ID=15107312
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
JP56133545A Pending JPS5835678A (en) | 1981-08-26 | 1981-08-26 | Character recognizing device |
Country Status (1)
Country | Link |
---|---|
JP (1) | JPS5835678A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPS63263589A (en) * | 1987-04-22 | 1988-10-31 | Hitachi Ltd | Character type mode selecting system for hand-written character recognizing device |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPS5635275A (en) * | 1979-08-31 | 1981-04-07 | Fujitsu Ltd | Monitor correcting method for optical character reading |
JPS5668869A (en) * | 1979-11-08 | 1981-06-09 | Toshiba Corp | Optical character reader |
-
1981
- 1981-08-26 JP JP56133545A patent/JPS5835678A/en active Pending
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPS5635275A (en) * | 1979-08-31 | 1981-04-07 | Fujitsu Ltd | Monitor correcting method for optical character reading |
JPS5668869A (en) * | 1979-11-08 | 1981-06-09 | Toshiba Corp | Optical character reader |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPS63263589A (en) * | 1987-04-22 | 1988-10-31 | Hitachi Ltd | Character type mode selecting system for hand-written character recognizing device |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
KR930000679B1 (en) | Individual dictionary making function part handwrite character recognition apparatus | |
US20070173240A1 (en) | Handwriting style data input via keys | |
JPH02289100A (en) | Kanji coding and decoding equipment | |
KR100298547B1 (en) | Character input apparatus | |
Brocklehurst | The NPL electronic paper project | |
JPS5835678A (en) | Character recognizing device | |
JP2503237B2 (en) | Character string information retrieval device | |
JPS592191A (en) | Recognizing and processing system of handwritten japanese sentence | |
JPH0574867B2 (en) | ||
JPS6224382A (en) | Method for recognizing handwritten character | |
JPS60217483A (en) | Recognizer of character | |
JPS61292774A (en) | Handwritten word processor | |
JPH03129459A (en) | Kana/kanji converter | |
JPH10320107A (en) | Handwritten character input device having handwritten character recognizing function | |
JP2659702B2 (en) | Character recognition device | |
JPH01103787A (en) | Device for recognizing handwritten character with personal dictionary preparing function | |
JPS61226883A (en) | Character recognizing device | |
JPS60189070A (en) | Character input device | |
JP2629946B2 (en) | Document processing device | |
JPH0737039A (en) | On-line handwritten character recognizing device | |
JPS61139828A (en) | Language input device | |
JP3710157B2 (en) | Kanji phrase processing method and apparatus | |
JPH06187502A (en) | On-line handwritten character recognizer | |
JPH09153116A (en) | Character string predictive method | |
JPH10254871A (en) | Document input method and its device |