JPS6160187A - Character recognizer - Google Patents
Character recognizerInfo
- Publication number
- JPS6160187A JPS6160187A JP59182198A JP18219884A JPS6160187A JP S6160187 A JPS6160187 A JP S6160187A JP 59182198 A JP59182198 A JP 59182198A JP 18219884 A JP18219884 A JP 18219884A JP S6160187 A JPS6160187 A JP S6160187A
- Authority
- JP
- Japan
- Prior art keywords
- line
- center position
- character
- row
- virtual
- 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)
Abstract
Description
【発明の詳細な説明】
〔発明の技術分野〕
本発明は文字認識装置に係り、特に帳票上の文字行を確
実に検出、切出しをすることができる文字認識装置に関
する。DETAILED DESCRIPTION OF THE INVENTION [Technical Field of the Invention] The present invention relates to a character recognition device, and more particularly to a character recognition device that can reliably detect and cut out character lines on a form.
帳票に記録された文字を認識する文字認識装置の認識率
を悪化させる原因には種々のものがある。There are various causes for deteriorating the recognition rate of a character recognition device that recognizes characters recorded on a form.
特に認識率に重要な影響を及ぼすものとして、印字品質
の相違による字形の変化があるが、印字精度不良や帳票
の裁断精度不良による印字位置のずれも認識率に多大の
影響を及ぼす。すなわち印字位置のずれがあると行切出
し、文字切出しのエラ−を生じ、正しい文字パターンを
得ることができず、誤認識の原因となっていた。In particular, changes in character shape due to differences in print quality have a particularly important effect on the recognition rate, but misalignment of print positions due to poor printing accuracy or poor cutting accuracy of forms also has a significant impact on the recognition rate. That is, if there is a misalignment in the printing position, errors occur in line cutting and character cutting, making it impossible to obtain a correct character pattern and causing erroneous recognition.
このような印字位置のずれによる誤認識を防止するため
、従来、帳票全体の読取パターンを記憶することができ
るシートバッファを用い、フリーフA−マット方式どす
る方法があるが、この方法は、記憶容量の増大を招くと
ともに、切出し方式が複雑で時間がかかるという問題が
あった。また−竹のずつを記憶するラインバッファの容
量を増して視野を拡大するいう方法もあるが、視野を拡
大すると帳票の行ピッチもそれに応じて拡大しなければ
ならず、帳票に記録できる文字数が少なくなるという問
題があった。In order to prevent such misrecognition due to misalignment of the print position, there is a conventional method such as the Freef A-Mat method using a sheet buffer that can store the reading pattern of the entire document. There were problems in that the capacity increased and the cutting method was complicated and time consuming. There is also a method of expanding the field of view by increasing the capacity of the line buffer that stores each bamboo, but if the field of view is expanded, the line pitch of the form must also be increased accordingly, which reduces the number of characters that can be recorded on the form. The problem was that it was getting smaller.
本発明は上記事情を考慮してなされたもので、印字位置
がずれても簡単な制御で確実に行切出し、文字切出しを
おこなうことができる文字認識装置を提供することを目
的とする。The present invention has been made in consideration of the above circumstances, and it is an object of the present invention to provide a character recognition device that can reliably cut out lines and cut out characters with simple control even if the printing position shifts.
上記目的を達成するために本発明による文字認識装置は
、記憶部に記憶された読取パターンの行方向の射影像を
とり、この射影像から1行目の文字行中心位置を検出し
、この1行目の文字行中心位置と仮想行中心位百との差
分を求め、この差分に基づいて2行目以後の仮想付中心
位置を修正し、この修正された仮想付中心位置に基づい
て文字行を切出すことを特徴どづる。In order to achieve the above object, the character recognition device according to the present invention takes a projected image in the row direction of the reading pattern stored in the storage section, detects the center position of the first character line from this projected image, and detects the center position of the first character line from this projected image. Find the difference between the character line center position of the first line and the virtual line center position, correct the virtual center position for the second and subsequent lines based on this difference, and adjust the character line position based on the corrected virtual line center position. It is characterized by cutting out.
以下本発明を図示の一実施例に基づいて説明する。本実
施例ににる文字認識装置では第1図に示すように帳票1
上の像は光電変換部10にJ:り光電変換され、光電変
換された電気信号は量子化部20で量子化される。量子
化された読取パターンは前処理部30でノイズ除去され
てラインバッファ40に格納される。ラインバッファに
記憶された読取パターンから、行検出切出部50にJ:
り文字行パターンが検出し、切出される。文字行パター
ンが切出されると、文字検出切出部60により各文字ご
とに文字パターンが切出され、文字認識部70に送られ
る。文字認識部70は入力した文字パターンを認識する
。The present invention will be explained below based on an illustrated embodiment. In the character recognition device according to this embodiment, as shown in FIG.
The upper image is photoelectrically converted by the photoelectric converter 10, and the photoelectrically converted electrical signal is quantized by the quantizer 20. The quantized reading pattern is subjected to noise removal by the preprocessing section 30 and stored in the line buffer 40 . From the reading pattern stored in the line buffer, the line detection cutting unit 50
A character line pattern is detected and cut out. When the character line pattern is cut out, the character detection and cutting section 60 cuts out the character pattern for each character and sends it to the character recognition section 70. The character recognition unit 70 recognizes the input character pattern.
本実施例による文字認識装置は、文字行を検出して切出
す行検出切出部50に特徴がある。この行検出切出部5
0は、ラインバッファ40の行方向の射影像をとる射影
手段51と、文字行の中心位置を検出する行位置検出手
段52と、検出された行中心位置と仮想性中心位置バッ
ファ54に格納された仮想付中心位置との差分を演算す
る演算手段53と、仮想性中心位置バッファ54の値を
修正する修正手段55と、行切出しをする行切出手段5
6を有している。The character recognition device according to this embodiment is characterized by a line detection and cutting section 50 that detects and cuts out character lines. This line detection cutting section 5
0 is stored in the projection means 51 that takes a projected image in the line direction of the line buffer 40, the line position detection means 52 that detects the center position of a character line, and the detected line center position and virtual center position buffer 54. a calculation means 53 for calculating the difference between the virtual center position and the virtual center position, a correction means 55 for correcting the value of the virtual center position buffer 54, and a line cutting means 5 for cutting out lines.
6.
この行検出切出部50による文字行の検出切出し方法を
第2図を用いてさらに詳しく説明する。The method of detecting and cutting out character lines by the line detection and cutting section 50 will be explained in more detail with reference to FIG.
まずラインバッファ40に帳票1の上端からLだけの幅
の読取パターンを格納する。射影手段51によりライン
バッファ40の行方向(水平方向)の射影像をとり、第
2図に示すような射影像を得る。この射影像から、行中
心位置検出手段52により行中心位置を検出する。この
行中心位置と、仮想性中心位置バッファ54の1行目の
行中6位置との差分αを演算手段53ににり求める。次
に修正手段55により2行目以後の仮想付中心位置を修
正する。n行目の仮想行中心位1がA、であれば、差分
αを考慮して、n行目の仮想付中心位置をAn+αと修
正する。2行目以後については、この修正された仮想性
中心位置バッファ54の値に基づいて、ラインバッファ
40の視点と視野を定め、ラインバッファ40を効率的
に用いて行切出しをおこなう。First, a reading pattern having a width L from the top of the form 1 is stored in the line buffer 40. The projection means 51 takes a projected image of the line buffer 40 in the row direction (horizontal direction) to obtain a projected image as shown in FIG. From this projected image, the row center position detection means 52 detects the row center position. The difference α between this row center position and the 6th position in the first row of the virtual center position buffer 54 is determined by the calculation means 53. Next, the correction means 55 corrects the virtual attached center position for the second and subsequent lines. If the virtual row center position 1 of the n-th row is A, the virtual center position 1 of the n-th row is corrected to An+α in consideration of the difference α. For the second and subsequent lines, the viewpoint and field of view of the line buffer 40 are determined based on the corrected values of the virtual center position buffer 54, and the line buffer 40 is used efficiently to perform line extraction.
このように本実施例によれば1行目の行検出に基づいて
2行目以後の行検出をおこなうようにしているので、確
実に行検出することが□できる。As described above, according to this embodiment, since the second and subsequent lines are detected based on the detection of the first line, lines can be detected reliably.
この実施例では1行目を読取る場合と2行目以後を読取
る場合とでラインバッファの視野を変えているが、2行
分のラインバッファを用意し、1行目を読取る場合には
2行分のラインバッファを用いて視野を広くし、2行目
以後は1行分のラインバッファを交互に用いるようにし
てもよい。これにより位置マージを向上させることがで
きるとともに高速処理が可能となる。In this embodiment, the field of view of the line buffer is changed depending on when reading the first line and when reading the second and subsequent lines, but a line buffer for two lines is prepared, and when reading the first line, two lines The field of view may be widened by using line buffers for one line, and from the second line onwards, line buffers for one line may be used alternately. This makes it possible to improve positional merging and enable high-speed processing.
またラインバッファのかわりに帳票全体の読取パターン
を格納できるシートバッファを用いてもよい。シートバ
ッファを用いた場合でも本発明ににれば高速処理が可能
である。Further, instead of the line buffer, a sheet buffer that can store the reading pattern of the entire form may be used. Even when a sheet buffer is used, high-speed processing is possible according to the present invention.
さらにラインバッファ、シートバッファを複数個用い、
並列処理するようにすれば、さらに高速化できる。Furthermore, multiple line buffers and sheet buffers are used,
You can speed it up even more by performing parallel processing.
なお、本発明は文字パターンの検出、切出しにも同様に
適用できる。すなわち、1文字目を検出して、その文字
位置から2文字目以後の仮想文字位置を修正するように
する。これにより高速文字切出しが可能となる。Note that the present invention can be similarly applied to detecting and cutting out character patterns. That is, the first character is detected and the virtual character positions of the second and subsequent characters are corrected from that character position. This enables high-speed character extraction.
以上の通り本発明によれば、印字位置がずれても簡単な
制御で確実にかつ高速で行切出しが可能であり、その結
果、文字TM認率を向上さけることができる。As described above, according to the present invention, even if the printing position is shifted, line cutting can be performed reliably and at high speed with simple control, and as a result, it is possible to avoid improving the character TM recognition rate.
第1図は本発明の一実施例による文字認識装置のブロッ
ク図、第2図は同文字認識装置の動作の説明図である。
10・・・光電変換部、20・・・量子化部、30・・
・前処理部、40・・・ラインバッファ、50・・・行
検出切出部、60・・・文字検出切出部、70・・・文
字認識部。FIG. 1 is a block diagram of a character recognition device according to an embodiment of the present invention, and FIG. 2 is an explanatory diagram of the operation of the character recognition device. 10... Photoelectric conversion section, 20... Quantization section, 30...
- Preprocessing unit, 40... line buffer, 50... line detection cutting unit, 60... character detection cutting unit, 70... character recognition unit.
Claims (1)
読取部により読取られた帳票の読取パターンを記憶する
記憶部と、この記憶部に記憶された読取パターンの文字
行を検出して切出す行検出切出部と、この行検出切出部
により切出された文字行パターンの各文字を検出して1
文字ずつ切出す文字検出切出部と、この文字検出切出部
により切出された文字パターンを認識する認識部とを備
えた文字認識装置において、 前記行検出切出部は、前記記憶部に記憶された読取パタ
ーンの行方向の射影像をとる射影手段と、この射影手段
による行方向の射影像から1行目の文字行中心位置を検
出する行中心位置検出手段と、この行中心位置検出手段
により検出された1行目の文字行中心位置と予め定めら
れた1行目の仮想行中心位置との差分を演算する演算手
段と、この演算手段により演算された差分に基づいて2
行目以後の仮想行中心位置を修正する修正手段とを有し
、この修正手段により修正された仮想行中心位置に基づ
いて文字行を切出すことを特徴とする文字認識装置。[Scope of Claims] A reading unit that reads a form in which a plurality of character lines are recorded, a storage unit that stores the reading pattern of the form read by the reading unit, and characters of the reading pattern stored in the storage unit. A line detection and cutting section that detects and cuts out a line, and a line detection and cutting section that detects each character of the character line pattern cut out by this line detection and cutting section.
In the character recognition device, the character recognition device includes a character detection and cutout section that cuts out each character, and a recognition section that recognizes the character pattern cut out by the character detection and cutout section, wherein the line detection and cutout section is configured to store information in the storage section. a projection means for taking a projected image in the row direction of the stored reading pattern; a line center position detection means for detecting the center position of the first character line from the projected image in the row direction by the projection means; and a line center position detection means for detecting the line center position. a calculation means for calculating the difference between the character line center position of the first line detected by the means and a predetermined virtual line center position of the first line;
1. A character recognition device comprising: a correction means for correcting a virtual line center position after the line, and cutting out a character line based on the virtual line center position corrected by the correction means.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP59182198A JPS6160187A (en) | 1984-08-31 | 1984-08-31 | Character recognizer |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP59182198A JPS6160187A (en) | 1984-08-31 | 1984-08-31 | Character recognizer |
Publications (2)
Publication Number | Publication Date |
---|---|
JPS6160187A true JPS6160187A (en) | 1986-03-27 |
JPH0529957B2 JPH0529957B2 (en) | 1993-05-06 |
Family
ID=16114062
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
JP59182198A Granted JPS6160187A (en) | 1984-08-31 | 1984-08-31 | Character recognizer |
Country Status (1)
Country | Link |
---|---|
JP (1) | JPS6160187A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH04344584A (en) * | 1991-05-21 | 1992-12-01 | Oki Electric Ind Co Ltd | Optical character reader |
JPH04344585A (en) * | 1991-05-21 | 1992-12-01 | Oki Electric Ind Co Ltd | Optical character reader |
-
1984
- 1984-08-31 JP JP59182198A patent/JPS6160187A/en active Granted
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH04344584A (en) * | 1991-05-21 | 1992-12-01 | Oki Electric Ind Co Ltd | Optical character reader |
JPH04344585A (en) * | 1991-05-21 | 1992-12-01 | Oki Electric Ind Co Ltd | Optical character reader |
Also Published As
Publication number | Publication date |
---|---|
JPH0529957B2 (en) | 1993-05-06 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP2822189B2 (en) | Character recognition apparatus and method | |
EP0103788B1 (en) | Method and apparatus for reading mark sheets | |
US4053737A (en) | Magnetic reader for bar encoded characters | |
US4878124A (en) | Image inclination detecting method and apparatus | |
US4083034A (en) | Plural tracker character detection and boundary location system | |
JPS6160187A (en) | Character recognizer | |
US5086481A (en) | Apparatus for converting outline data into dot data, having means for assuring continuity of visible representations reproduced by the dot data | |
JPS6343789B2 (en) | ||
JPS5824265A (en) | Mark read system using facsimile transmitter | |
JP2734538B2 (en) | Binary data reproducing method and apparatus, and recording medium | |
JPS59206989A (en) | Letter segmenting device | |
JPH0420507B2 (en) | ||
JPS61109181A (en) | Character detection system | |
JPH027183A (en) | Character segmenting device | |
JPH0573720A (en) | Optical character reader | |
JPH0782524B2 (en) | Optical character reader | |
JPS6027436B2 (en) | Character recognition correction method | |
JPS6325388B2 (en) | ||
JPS5840691A (en) | Pattern reader | |
JPS59157774A (en) | Character recognizing device | |
JPS6394385A (en) | Printed character pitch detection device | |
JPH036552B2 (en) | ||
JPS5864575A (en) | Optical character reader | |
JPH0417474B2 (en) | ||
JPH0244487A (en) | 7b character reading system for optical character reader |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
EXPY | Cancellation because of completion of term |