JPH03250278A - Table-formed document reader - Google Patents

Table-formed document reader

Info

Publication number
JPH03250278A
JPH03250278A JP2045547A JP4554790A JPH03250278A JP H03250278 A JPH03250278 A JP H03250278A JP 2045547 A JP2045547 A JP 2045547A JP 4554790 A JP4554790 A JP 4554790A JP H03250278 A JPH03250278 A JP H03250278A
Authority
JP
Japan
Prior art keywords
line
ruled
lines
ruled line
dotted
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
Application number
JP2045547A
Other languages
Japanese (ja)
Other versions
JP2813600B2 (en
Inventor
Katsumi Hosokawa
勝美 細川
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.)
Fuji Electric Co Ltd
Fuji Facom Corp
Original Assignee
Fuji Electric Co Ltd
Fuji Facom 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 Fuji Electric Co Ltd, Fuji Facom Corp filed Critical Fuji Electric Co Ltd
Priority to JP2045547A priority Critical patent/JP2813600B2/en
Publication of JPH03250278A publication Critical patent/JPH03250278A/en
Application granted granted Critical
Publication of JP2813600B2 publication Critical patent/JP2813600B2/en
Anticipated expiration legal-status Critical
Expired - Fee Related legal-status Critical Current

Links

Landscapes

  • Character Input (AREA)
  • Character Discrimination (AREA)
  • Image Analysis (AREA)

Abstract

PURPOSE:To reproduce ruled lines similar to those of an input document by discriminating the types of ruled lines such as a continuous line and a dotted line, etc. CONSTITUTION:When a table-formed document 1 is inputted, at first, a ruled line recognition part 6 extracts a ruled line candidate 21. On this stage, since even something not to be the continuous ruled line but to be dirt or folding, etc., like a code 22 are included as ruled line candidate 21A, it is decided whether this is the ruled line or not, and therefore, the respective feature patterns of the ruled line candidate 21 are extracted. Namely, the ruled line is provided with a feature pattern for each type of the line and according to the feature pattern, the type of the line is classified. For example, the continuous line is equipped with the feature of being basically continuous and equipped with fixed width and the ruled line of the dotted line or an alternate long and short dash line, etc., is equipped with fixed periodical regularity. Thus, the ruled lines are reproduced similarly to those of the input document.

Description

【発明の詳細な説明】 〔産業上の利用分野〕 本発明は、表を含む文書(表形式文書)の読取装置に関
する。
DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a reading device for a document including a table (table format document).

〔従来の技術〕[Conventional technology]

従来、表形式文書の罫線を認識する方法として投影法を
用いる方法、追跡法を用いる方法などがあるが、表形式
文書における罫線認識は本来の目的が表内に書かれた文
字を認識することであるため、罫線が実線であっても点
線であっても単なる罫線として認識しているのが普通で
ある。
Conventionally, methods for recognizing ruled lines in tabular documents include methods that use projection methods and methods that use tracing methods, but the original purpose of recognizing ruled lines in tabular documents is to recognize characters written in tables. Therefore, whether a ruled line is a solid line or a dotted line, it is normally recognized as a simple ruled line.

また、第8図のように手書き文書や印刷物で飾りが付い
た特殊な罫線26(罫線の内部が塗り潰されずにハンチ
ングになっている)の場合や、第9図のように何かの文
字(ここではNJ、r+J「−」を使用している)を羅
列することによって罫線26A、26Bの意味を持たせ
ている場合については、従来は認識ができなかった。
Also, as shown in Figure 8, there are special ruled lines 26 with decorations in handwritten documents or printed matter (the inside of the ruled line is not filled in but has a hunting pattern), and as shown in Figure 9, some characters ( In the past, it was not possible to recognize cases where the ruled lines 26A and 26B were given meaning by listing NJ, r+J (here, "-" is used).

〔発明が解決しようとする課題〕[Problem to be solved by the invention]

つまり、従来の方法では罫線の線種(種別)が不明にな
るという問題がある。また、手書き文書や印刷物で飾り
が付いた特殊な罫線や、何かの文字を羅列することによ
って罫線の意味を持たせている場合は従来は認識ができ
ず、罫線を文字と認識したりしてしまい、罫線によって
区切られている表の構造を正しく理解出来ないという問
題もある。
In other words, the conventional method has a problem in that the line type (type) of the ruled line is unclear. Additionally, in handwritten documents and printed matter, special ruled lines with decorations, or cases where the meaning of the ruled line is given by a list of characters, cannot be recognized in the past, and the ruled lines may be recognized as characters. There is also the problem that the structure of a table separated by ruled lines cannot be correctly understood.

したがって、本発明の目的は罫線をその種別とともに認
識できるようにし、飾りが付いた特殊な罫線や、文字等
を羅列した罫線等についても認識できるようにすること
にある。
Therefore, an object of the present invention is to make it possible to recognize ruled lines along with their types, and also to make it possible to recognize special ruled lines with decorations, ruled lines with enumerated characters, etc.

〔課題を解決するための手段〕[Means to solve the problem]

表形式文書を画像処理して表の構造を認識し、表内にあ
る文字を読取る表形式文書読取装置において、前記表の
罫線が実線1点線、破線、二重線。
In a tabular document reading device that performs image processing on a tabular document to recognize the structure of the table and read characters in the table, the ruled lines of the table may be solid single dot lines, broken lines, or double lines.

一点鎖線を含むいずれの線種からなるかを判別して罫線
を認識する罫線認識部を設け、認識結果を出力するに当
たっては罫線の線種情報も併せて出力可能にする。また
、前記表形式文書の或る罫線が指定されたときその罫線
の特徴データを抽出して登録する罫線登録部を設け、登
録された特徴データにもとづき以後の罫線の認識を可能
にし、さらにはその登録時に罫線を実線1点線、破線。
A ruled line recognition unit is provided to recognize a ruled line by determining which line type it is made of, including a dashed-dotted line, and when outputting the recognition result, it is also possible to output line type information of the ruled line. Further, when a certain ruled line of the tabular document is specified, a ruled line registration section is provided which extracts and registers characteristic data of the ruled line, and it is possible to recognize subsequent ruled lines based on the registered characteristic data. When registering, the ruled lines are one solid line, one dotted line, and one broken line.

重線、一点鎖線を含むいずれの線種にて出力するかの指
定も併せてできるようにする。
It is also possible to specify which line type to output, including double lines and dash-dot lines.

C作用〕 表形式文書の罫線を認識するに当たり、実線や点線等の
特徴を予め登録しておくことによって罫線の線種を判別
し、その結果を位置情報などと共に出力できるようにす
る。また、手書き文書や印刷物で飾りが付いた特殊な罫
線や、何かの文字を羅列することによって罫線の意味を
持たせている場合は、そのパターンを登録することによ
り、罫線として認識できるようにする。このとき、かか
る罫線がCRTやプリンタ等の装置で出力できないとき
は、実線や点線に置き換えて出力することができるよう
にする。
C Effect] When recognizing the ruled lines of a tabular document, the line type of the ruled line can be determined by registering features such as solid lines and dotted lines in advance, and the result can be output together with position information and the like. In addition, if a handwritten document or printed matter has special decorated ruled lines or a list of characters to give the meaning of the ruled line, you can register the pattern so that it can be recognized as a ruled line. do. At this time, if such ruled lines cannot be output by a device such as a CRT or a printer, it is possible to output them by replacing them with solid lines or dotted lines.

〔実施例〕〔Example〕

第1図は本発明の実施例を示すブロック図で、1は表形
式文書、2はホス1−CPU、3は画像入力部、4は画
像メモリ、5は文字認識部、6は罫線認識部、7は罫線
登録部、8は罫線特徴チーフル、9は補助記憶部、10
はプリンタ、11はCRTである。
FIG. 1 is a block diagram showing an embodiment of the present invention, in which 1 is a tabular document, 2 is a host 1-CPU, 3 is an image input section, 4 is an image memory, 5 is a character recognition section, and 6 is a ruled line recognition section. , 7 is a ruled line registration section, 8 is a ruled line characteristic feature, 9 is an auxiliary storage section, 10
is a printer, and 11 is a CRT.

以下、第1図の動作について、第2図ないし第9図を参
照して説明する。
The operation shown in FIG. 1 will be explained below with reference to FIGS. 2 to 9.

ホストCPU2により制御される画像入力部3によって
表形式文書1を入力すると、画像データは2値化され画
像メモリ4に格納される。いま、例えば第2図の如き表
形式文書1が入力されたとすると、罫線認識部6ではま
ず罫線候補21を抽出する。なお、この段階では符号2
2のように汚れ、折り目等で実際には罫線ではないもの
21Aも含まれている。そこで、罫線であるか否かを判
定するため、罫線候補21の各々の特徴パターンを抽出
する。すなわち、罫線は各線種毎に特徴パターンを持っ
ており、その特徴パターンによって線種を分類すること
ができる。例えば、実線であれば基本的には連続であり
、かつ一定の幅を有するという特徴を持っている。また
、点線や一点鎖線等の罫線は、一定の周期的な規則性を
持っている。
When the tabular document 1 is input by the image input unit 3 controlled by the host CPU 2, the image data is binarized and stored in the image memory 4. For example, if a tabular document 1 as shown in FIG. 2 is input, the ruled line recognition unit 6 first extracts ruled line candidates 21. Note that at this stage, the code 2
21A, which is not actually a ruled line due to stains, creases, etc., is also included. Therefore, in order to determine whether or not the line is a ruled line, characteristic patterns of each of the ruled line candidates 21 are extracted. That is, each ruled line has a characteristic pattern for each line type, and the line types can be classified based on the characteristic pattern. For example, a solid line is basically continuous and has a constant width. Furthermore, ruled lines such as dotted lines and dashed-dotted lines have a certain periodic regularity.

第3図に点線の例を示す。同図からも明らかなように、
線部分DIと間隔部分D2の長さ、DIとD2の比、D
lとD2の出現回数および線幅等のデータがあり、他の
線種も同様に固有の特徴パターンを持っているので、こ
れらの罫線特徴パターンの中の最も近い線として線種を
認識する。そのために、罫線特徴テーブル8には予め実
線1点線などの基本的な線種に関するデータを格納して
おき、罫線認識部6で抽出した罫線データをテーブル8
の内容と比較することにより、罫線をその線種も含めて
認識することが可能となる。この段階で、22は罫線と
しての特徴を持たないことから、不適合として除去され
ることになり、したがって第2図の如き文書からは第4
図のような罫線情報23が得られ、補助記憶部9に保存
される。
FIG. 3 shows an example of dotted lines. As is clear from the figure,
Length of line portion DI and interval portion D2, ratio of DI and D2, D
Since there is data such as the number of occurrences of I and D2 and line widths, and other line types have unique characteristic patterns as well, the line type is recognized as the closest line among these ruled line characteristic patterns. To this end, the ruled line feature table 8 stores data related to basic line types such as solid lines and single dotted lines in advance, and the ruled line data extracted by the ruled line recognition unit 6 is stored in the ruled line feature table 8.
By comparing the contents with the contents, it becomes possible to recognize the ruled line including its line type. At this stage, since 22 does not have the characteristics of a ruled line, it is removed as nonconforming, and therefore, from the document shown in Figure 2, the 4th line is removed.
Ruled line information 23 as shown in the figure is obtained and stored in the auxiliary storage section 9.

第4図からは罫線候補21Aのデータは除かれているの
が分かる。
It can be seen from FIG. 4 that the data for the ruled line candidate 21A is excluded.

次に、第4図の如き情報は文字認識部5に送られ、ここ
で第5図に符号24で示すような文字領域が算出され、
この領域24毎に文字認識を行ない、第6図の如き文字
データ25を補助記憶部9に保存する。保存されたデー
タは必要に応じて読み出され、プリンタ10またはCR
TIIに出力される。
Next, the information as shown in FIG. 4 is sent to the character recognition section 5, where a character area as shown by reference numeral 24 in FIG. 5 is calculated.
Character recognition is performed for each area 24, and character data 25 as shown in FIG. 6 is stored in the auxiliary storage section 9. The saved data is read out as needed and sent to the printer 10 or CR.
Output to TII.

また、第8図、第9図のような特殊な罫線がある場合は
、ホストCPU2で罫線登録モードを選択すると、画像
入力部3に入力され画像メモリ4に格納された表形式文
書の画像データがCRT 11に表示されるので、カー
ソル等によりその特殊な罫線部分の範囲27を第8図の
ように指示すると、その範囲の情報が罫線登録部7に与
えられる。
In addition, when there are special ruled lines as shown in FIGS. 8 and 9, when the ruled line registration mode is selected on the host CPU 2, the image data of the tabular document input to the image input section 3 and stored in the image memory 4 is displayed on the CRT 11, and when a range 27 of the special ruled line portion is indicated with a cursor or the like as shown in FIG. 8, information on that range is given to the ruled line registration section 7.

罫線登録部7は指定された範囲27内の画像データから
種々の特徴パターンを抽出し、そのデータをテーブル8
に追加登録する。なお、この追加登録された罫線を出力
するときの罫線種別を、例えば実線とか点線のように指
定することができる。
The ruled line registration unit 7 extracts various characteristic patterns from the image data within the specified range 27 and stores the data in the table 8.
Register additionally. Note that the type of ruled line when outputting this additionally registered ruled line can be specified, for example, as a solid line or a dotted line.

したがって、第8図で範囲27の罫線を登録し、出力種
別を点線と指定すると、範囲270部分は登録した罫線
として認識され、第7図のように点線として出力される
ことになる。
Therefore, if the ruled lines in the range 27 in FIG. 8 are registered and the output type is specified as dotted lines, the range 270 will be recognized as the registered ruled lines and will be output as dotted lines as shown in FIG. 7.

〔発明の効果〕〔Effect of the invention〕

本発明によれば、実線1点線等の罫線の種別を判別可能
とすることにより、入力文書と同様の罫線を再現するこ
とができる利点が得られる。また、特殊な罫線の場合は
その画像データを指定し特徴を抽出して登録することに
より、如何なる罫線をも識別することができ、表の構造
を正確に把握することが可能となる。
According to the present invention, by making it possible to determine the type of ruled line such as a solid line and one dotted line, it is possible to reproduce the same ruled line as that of the input document. Furthermore, in the case of special ruled lines, by specifying the image data, extracting and registering the characteristics, any ruled line can be identified, and the structure of the table can be accurately grasped.

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

第1図は本発明の実施例を示すブロック図、第2図は表
形式文書の一例を説明するための説明図、第3図は点線
の例を説明するための説明図、第4図は第2図の罫線情
報を説明するための説明図、第5図は第2図の文字領域
を説明するための説明図、第6図は第2図の文字データ
を説明するための説明図、第7図は罫線登録による認識
結果の例を説明するための説明図、第8図は特殊な罫線
を持つ表形式文書の一例を説明するための説明図、第9
図は文字を罫線として使用した例を説明するための説明
図である。 符号説明 1・・・表形式文書、2・・・ホストCPU、3・・・
画像入力部、4・・・画像メモリ、5・・・文字認識部
、6・・・罫線認識部、7・・・罫線登録部、8・・・
罫線特徴テーブル、9・・・補助記憶部、10・・・プ
リンタ、11・・・CRT、21.21A・・・罫線候
補、22・・・汚れ折れ等、23・・・罫線情報データ
、24・・・文字領域、25・・・文字データ、26.
26A、26B・・・特殊な罫線、27・・・罫線指定
範囲。
FIG. 1 is a block diagram showing an embodiment of the present invention, FIG. 2 is an explanatory diagram for explaining an example of a tabular document, FIG. 3 is an explanatory diagram for explaining an example of a dotted line, and FIG. 4 is an explanatory diagram for explaining an example of a tabular document. An explanatory diagram for explaining the ruled line information in FIG. 2; FIG. 5 is an explanatory diagram for explaining the character area in FIG. 2; FIG. 6 is an explanatory diagram for explaining the character data in FIG. 2; FIG. 7 is an explanatory diagram for explaining an example of a recognition result based on ruled line registration, FIG. 8 is an explanatory diagram for explaining an example of a tabular document with special ruled lines, and FIG.
The figure is an explanatory diagram for explaining an example of using characters as ruled lines. Code explanation 1...Tabular document, 2...Host CPU, 3...
Image input section, 4... Image memory, 5... Character recognition section, 6... Ruled line recognition section, 7... Ruled line registration section, 8...
Ruled line characteristic table, 9... Auxiliary storage unit, 10... Printer, 11... CRT, 21. 21A... Ruled line candidate, 22... Dirt, creases, etc., 23... Ruled line information data, 24 ...Character area, 25...Character data, 26.
26A, 26B... Special ruled lines, 27... Ruled line specified range.

Claims (1)

【特許請求の範囲】 1)表形式文書を画像処理して表の構造を認識し、表内
にある文字を読取る表形式文書読取装置において、 前記表の罫線が実線、点線、破線、二重線、一点鎖線を
含むいずれの線種からなるかを判別して罫線を認識する
罫線認識部を設け、認識結果を出力するに当たっては罫
線の線種情報も併せて出力可能にしてなることを特徴と
する表形式文書読取装置。 2)前記表形式文書の或る罫線が指定されたときその罫
線の特徴データを抽出して登録する罫線登録部を設け、
登録された特徴データにもとづき以後の罫線の認識を可
能にしてなることを特徴とする請求項1)に記載の表形
式文書読取装置。 3)前記罫線の特徴データ登録時に罫線を実線、点線、
破線、二重線、一点鎖線を含むいずれの線種にて出力す
るかの指定も併せて可能にしてなることを特徴とする請
求項2)に記載の表形式文書読取装置。
[Scope of Claims] 1) A tabular document reading device that performs image processing on a tabular document to recognize the structure of the table and read characters in the table, wherein the ruled lines of the table are solid lines, dotted lines, broken lines, double lines, etc. A ruled line recognition unit is provided to recognize a ruled line by determining whether it is made of a line or a line type including a dashed line, and when outputting the recognition result, it is also possible to output line type information of the ruled line. Table format document reading device. 2) providing a ruled line registration unit that extracts and registers characteristic data of a ruled line when a certain ruled line of the tabular document is specified;
2. The tabular document reading device according to claim 1, wherein subsequent ruled lines can be recognized based on registered feature data. 3) When registering the characteristic data of the ruled line, the ruled line can be set as a solid line, a dotted line,
3. The tabular document reading device according to claim 2, wherein it is also possible to specify which line type to output, including broken lines, double lines, and dashed-dotted lines.
JP2045547A 1990-02-28 1990-02-28 Tabular document reader Expired - Fee Related JP2813600B2 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP2045547A JP2813600B2 (en) 1990-02-28 1990-02-28 Tabular document reader

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP2045547A JP2813600B2 (en) 1990-02-28 1990-02-28 Tabular document reader

Publications (2)

Publication Number Publication Date
JPH03250278A true JPH03250278A (en) 1991-11-08
JP2813600B2 JP2813600B2 (en) 1998-10-22

Family

ID=12722391

Family Applications (1)

Application Number Title Priority Date Filing Date
JP2045547A Expired - Fee Related JP2813600B2 (en) 1990-02-28 1990-02-28 Tabular document reader

Country Status (1)

Country Link
JP (1) JP2813600B2 (en)

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPS5638683A (en) * 1979-09-05 1981-04-13 Hitachi Ltd Discriminating method of type of line in pattern recognizer
JPS62200472A (en) * 1986-02-28 1987-09-04 Mitsubishi Electric Corp Line detector
JPH01147786A (en) * 1987-12-04 1989-06-09 Fuji Electric Co Ltd Reading device for document containing table

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPS5638683A (en) * 1979-09-05 1981-04-13 Hitachi Ltd Discriminating method of type of line in pattern recognizer
JPS62200472A (en) * 1986-02-28 1987-09-04 Mitsubishi Electric Corp Line detector
JPH01147786A (en) * 1987-12-04 1989-06-09 Fuji Electric Co Ltd Reading device for document containing table

Also Published As

Publication number Publication date
JP2813600B2 (en) 1998-10-22

Similar Documents

Publication Publication Date Title
JPS61502495A (en) Cryptographic analysis device
KR19980042782A (en) Character recognition device and method
JPH07105308A (en) How to locate and identify monetary fields on documents
JPH03161891A (en) Table type document reader
JP3178483B2 (en) Document processing device
JP2003018393A (en) Image processing system
US5157767A (en) Apparatus and method for automatically outputting document having data dependent size
JP4738668B2 (en) Document management apparatus, document management method, and storage medium storing program for causing computer to perform processing in document management apparatus
JP4807618B2 (en) Image processing apparatus and image processing program
JP3868537B2 (en) Personal identification medium issuing system and data check method
Chen et al. Analysis, Understanding and Representation of Chinese newspaper with complex layout
JP2813600B2 (en) Tabular document reader
JP3732254B2 (en) Format information generation method and format information generation apparatus
JP4517822B2 (en) Image processing apparatus and program
JP2974346B2 (en) Kanji processing device
JPH02202689A (en) Character recognizing device
JP2008028716A (en) Image processing method and apparatus
KR0183779B1 (en) Bill registering method and apparatus thereof
JP3006294B2 (en) Optical character reader
JP2976990B2 (en) Character recognition device
JP2784004B2 (en) Character recognition device
JPH05266244A (en) Character reader
JPS6180477A (en) Document transcription device
JPH08194772A (en) Optical character reader
JPH04309B2 (en)

Legal Events

Date Code Title Description
R250 Receipt of annual fees

Free format text: JAPANESE INTERMEDIATE CODE: R250

R250 Receipt of annual fees

Free format text: JAPANESE INTERMEDIATE CODE: R250

R250 Receipt of annual fees

Free format text: JAPANESE INTERMEDIATE CODE: R250

FPAY Renewal fee payment (event date is renewal date of database)

Free format text: PAYMENT UNTIL: 20070814

Year of fee payment: 9

FPAY Renewal fee payment (event date is renewal date of database)

Free format text: PAYMENT UNTIL: 20080814

Year of fee payment: 10

FPAY Renewal fee payment (event date is renewal date of database)

Free format text: PAYMENT UNTIL: 20080814

Year of fee payment: 10

FPAY Renewal fee payment (event date is renewal date of database)

Free format text: PAYMENT UNTIL: 20090814

Year of fee payment: 11

LAPS Cancellation because of no payment of annual fees