JPS61100876A - Method of segmenting graphic in graphic recognizing device - Google Patents

Method of segmenting graphic in graphic recognizing device

Info

Publication number
JPS61100876A
JPS61100876A JP59222519A JP22251984A JPS61100876A JP S61100876 A JPS61100876 A JP S61100876A JP 59222519 A JP59222519 A JP 59222519A JP 22251984 A JP22251984 A JP 22251984A JP S61100876 A JPS61100876 A JP S61100876A
Authority
JP
Japan
Prior art keywords
cutting
graphic
cut out
lower ends
figures
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
JP59222519A
Other languages
Japanese (ja)
Inventor
Koichi Fujimoto
幸一 藤本
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.)
Omron Corp
Original Assignee
Omron Tateisi Electronics Co
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 Omron Tateisi Electronics Co filed Critical Omron Tateisi Electronics Co
Priority to JP59222519A priority Critical patent/JPS61100876A/en
Publication of JPS61100876A publication Critical patent/JPS61100876A/en
Pending legal-status Critical Current

Links

Landscapes

  • Character Input (AREA)

Abstract

PURPOSE:To make graphic recognition at high accuracy without influenced by outer form frame by segmenting upper and lower ends and right and left ends of whole graphic, right and left ends and upper and lower ends of each graphic, and further cutting out upper and lower ends and right and left ends of each cut out graphic. CONSTITUTION:An object of reading 2 is scanned in the direction of arrangement of graphics 1, and the first projection patterns 3 are made on Y-axis, and upper and lower ends as a whole are segmentted by a first segmenting device. Each segmented column is scanned in the direction at right angles to the direction of arrangement, and the second projection patterns 5- are made on X-axis. Right and left ends of each column are segmentted and the outer form frame 7 of the graphic 1 is substantially segmented. Outer frames 7- of the cut out graphic 1 is further scanned by a scanning device, and lower end and left and right ends of the graphic 1 are segmentted.

Description

【発明の詳細な説明】 (発明の分野) 本発明は、例えば、馬券や車券などの投票券が的中券か
どうかを判別するとか、入場券や会葬などを読み取ると
いったために、数字や記号、あるいは、英文字などの図
形を読み取るのに用いられる図形認識装置において、前
記投票券などの読み取り対象から各図形を切り出す図形
認識装置における図形切り出し方法に関する。
[Detailed Description of the Invention] (Field of the Invention) The present invention is useful for determining whether a betting ticket such as a horse racing ticket or a car ticket is a winning ticket, or for reading an entrance ticket, a funeral, etc. Alternatively, the present invention relates to a figure cutting method in a figure recognition apparatus used for reading figures such as English characters, in which each figure is cut out from an object to be read, such as the voting ticket.

(発明のlI!要) 本発明は、各図形ごとの標準パターンとのマツチングを
良好に行なえるようにするために、各図形の外形枠を正
確に切り出せるようにしようとするものである。
(lI! Essentials of the Invention) The present invention attempts to accurately cut out the outer frame of each figure in order to ensure good matching of each figure with a standard pattern.

(従来技術とその問題点) 従来のこの種の方法は、第8図に示すように、通常の印
字手段により、夫々が連続線で印刷構成された図形OI
を並列して成る読み取り対象02に対しては、図形を認
識する際に、イメージセンサなどの走査手段によって図
形並列方向(X軸方向)およびそれIこ直交する方向(
Y軸方向)夫々に走査し、夫々の方向で投影パターン0
3,04を清算してそれをスライスレベル05,06で
2値化し、上記X軸方向の走査によって配列された図形
01・・全体としての上下端を切り出すとともに、Y軸
方向の走査によって上下端を切り出された図形01・・
・に配列された1個づつの図形01の左右端を切り出し
、第9図に示すように、各図形01・・・ごとの外形枠
07を切り出すか、あるいは、逆に図形全体としての左
右端を切り出した後に1個づフの図形OIの上下端を切
り出して各図形01・・ごとの外形枠07を切り出すか
していた。
(Prior art and its problems) As shown in FIG.
For the object 02 to be read, which consists of parallel figures, when recognizing the figures, a scanning means such as an image sensor is used to scan the figures in the parallel direction (X-axis direction) and in the direction perpendicular thereto (I).
(Y-axis direction), and the projection pattern is 0 in each direction.
3 and 04, and binarize it at slice levels 05 and 06, and cut out the upper and lower ends of the entire figure 01 arranged by scanning in the X-axis direction, and cutting out the upper and lower ends by scanning in the Y-axis direction. Figure 01 cut out...
・Cut out the left and right ends of the figures 01 arranged in After cutting out the figure OI, the upper and lower ends of each figure OI were cut out to cut out the outer frame 07 for each figure 01, . . . .

しかしながら、このような従来の方法ては、同しく第9
図で示されるように、図形OIの配列に上下方向あるい
は左右方向でずれがある場合に、切り出された各図形0
1・・・ごとの外形枠07の上′上長さあるいは左右長
さが各図形01の上下長さあるいは左右長さ夫々よりし
長くなってしまい、11Mパターンとのパターンマツチ
ングにおいて、切り出された外形枠07が標準パターン
と対応される結果、図形OIが互いに対応すべき乙ので
あっても、それを認識不能と判断してしまうなど認識精
度が低下する欠点があった。
However, such conventional methods also
As shown in the figure, if there is a vertical or horizontal shift in the arrangement of the figures OI, each cut out figure 0
The top length or left and right length of the outer frame 07 for each figure 01 is longer than the top and bottom length or the left and right length of each figure 01, and when pattern matching with the 11M pattern, it is cut out. As a result of the outer shape frame 07 being made to correspond to the standard pattern, there is a drawback that the recognition accuracy is lowered, such as determining that it is unrecognizable even if the figure OIs should correspond to each other.

また、第1O図に示すように、図形01の近辺に、印字
の際のインクのはねやにじみなどのノイズ08・・があ
った場合に、それが図形01と不連続であるにもかかわ
らず、それらノイズ08を含んだ扶植で外形枠07が切
り出され、上述同様に認!l精度が低下する欠点があっ
た。
Furthermore, as shown in Figure 1O, if there is noise 08, such as ink splashing or blurring during printing, in the vicinity of figure 01, even though it is discontinuous with figure 01, First, the outer frame 07 is cut out with the fuss that includes those noises 08, and it is recognized as above! There was a drawback that accuracy decreased.

(発明の目的) 本発明は、このような事情に鑑みてなされたものであっ
て、各図形ごとの外形枠を正確に切り出せるようにする
ことを目的とする。
(Object of the Invention) The present invention has been made in view of the above circumstances, and an object of the present invention is to enable accurate cutting out of the outer frame of each figure.

(発明の燐酸と効果) 本発明図形認識装置の図形切り出し方法は、このような
目的を4成するために、夫々が連続線で構成された図形
を並列して成る読み取り対象に対゛ して走査手段によ
り図形並列方向に走査し、第1切り出し手段によって前
記図形の上下端を切り出すとともに、前記読み取り対象
に対して前記走査手段により図形並列方向に直交する方
向に走査し、第2切り出し手段によって前記図形の左右
端を切り出し、それにより各図形ごとの外形枠を切り出
し、次いで、前記各図形ごとの外形枠に対して、前記第
1および第2切り出し手段による切り出し処理を行なう
ようにする。
(Phosphoric Acid and Effects of the Invention) In order to achieve the above four purposes, the figure cutting method of the figure recognition device of the present invention applies to a reading target consisting of parallel figures, each of which is composed of continuous lines. The scanning means scans in the figure parallel direction, the first cutting means cuts out the upper and lower ends of the figure, the scanning means scans the object to be read in a direction perpendicular to the figure parallel direction, and the second cutting means cuts out the upper and lower ends of the figure. The left and right ends of the figure are cut out, thereby cutting out an external frame for each figure, and then the first and second cutting means perform a cutting process on the outer frame for each figure.

つまり、先ず、配列された図形全体としての上下端ある
いは左右端を切り出すとともに、それから各図形の左右
端あるいは上下端を切り出し、しかる後に、切り出した
各図形ごとに更に上下端および左右端を切り出していく
のである。
In other words, first, cut out the top and bottom or left and right ends of the entire array of figures, then cut out the left and right or top and bottom ends of each figure, and then cut out the top and bottom and left and right ends of each figure. I'm going.

したがって、配列された図形全体を切り出すことに起因
して、切り出された各図形ごとの外形枠の上下長さある
いは左右長さが凸図形の上F長さあるいは左右長さ夫々
よりら長くなってしまってら、その長くなった部分を各
図形ごとの切り出しによって除去でき、また、ばねやに
じみなどのノイズがあっても、それが図形配列方向とそ
れに直交する方向の両方にまたがって連続していない隔
り除去でき、各図形自体の外縁に近接!7た扶植で、外
形枠を正確に切り出すことができ、標県パターンとのマ
ツチングに際し、外形枠に影響されずに図形どうしを比
較して、図形認識を精度良く行なえるようになった。
Therefore, due to cutting out the entire array of figures, the vertical length or horizontal length of the outer frame of each cut out figure becomes longer than the upper F length or the left and right lengths of the convex figure, respectively. If it is fixed, the long part can be removed by cutting out each shape, and even if there is noise such as springs or bleeding, it is not continuous across both the shape arrangement direction and the direction perpendicular to it. The gap can be removed and each shape is close to its outer edge! With the 7-fold support, the external frame can be accurately cut out, and when matching with the Shibei pattern, figures can be compared without being affected by the external frame, allowing highly accurate figure recognition.

(実施例の説明) 以下、本発明の図形認ra装置の図形切り出し方法を図
面に示す実施例に基づいて詳細に説明する。
(Description of Embodiments) Hereinafter, the figure cutting method of the figure recognition RA apparatus of the present invention will be described in detail based on the embodiments shown in the drawings.

第1図は、本発明の実施例の動作を示すフローチャート
である 先ず、夫々が連続線で構成された英文字や数字などの図
形rhtx軸方向に並列された読み取り対!J!2に対
し、第2図に示すように、イメージセンサなどの走査手
段(図示せず)によって図形lの配列方向に走査し、そ
の第1投影パターン3・・・をY軸上に作り、かつ、そ
の投影パターン3・を第1スライスレベル4で2値化し
、配列された図形1全体としての上下端を第1切り出し
手段によって切り出す(ステップNl)。
FIG. 1 is a flowchart showing the operation of the embodiment of the present invention. First, read pairs of figures such as alphabetic characters and numbers, each composed of continuous lines, arranged in parallel in the rhtx axis direction. J! 2, as shown in FIG. 2, a scanning means (not shown) such as an image sensor scans in the arrangement direction of the figures l, and creates a first projection pattern 3 on the Y axis, and , the projection pattern 3 is binarized at a first slice level 4, and the upper and lower ends of the arranged figures 1 as a whole are cut out by the first cutting means (step Nl).

その後、第3図に示すように一1前記走査手段により、
切り出された列ごとに配列方向と直交rる方向に走査し
、その第2投彰パターン5・・・をX軸上に作り、かつ
、その投影パターン5 を第2スライスレベル6で2f
k化し、第2切り出し手段により各列ごとて各図形1の
左右端を切り出して、実質的に図形lの外形枠7を切り
出す(ステップN2)。
Thereafter, as shown in FIG.
Each cut out column is scanned in the direction perpendicular to the arrangement direction, and its second projection pattern 5... is created on the X-axis, and the projection pattern 5 is scanned at the second slice level 6 by 2f.
Then, the left and right ends of each figure 1 are cut out for each row by the second cutting means, thereby cutting out substantially the outer frame 7 of the figure 1 (step N2).

次いで、第4図に示すように、上述のようにして切り出
された図1ニlの外形枠7・・に対し、更に、航記走査
手段により、図形配列方向に走査し、その第3投影パタ
ーン8をY軸上に作り、かつ、その投影パターン8を第
3スライスレベル9で2値化し、第1切り出し手段によ
り図形!の上下端を切り出す(ステップN3)。
Next, as shown in FIG. 4, the outer frame 7 of FIG. A pattern 8 is created on the Y axis, and the projected pattern 8 is binarized at the third slice level 9, and the first cutting means generates a figure! Cut out the upper and lower ends of (Step N3).

その後、第5図に示すように、前記走査手段により、図
形1の外形枠7 に対し、図形配列方向と直交する方向
に走査し、その投影パター710をX軸上に作り、かつ
、その第4投影パターン10を第4スライスレベル11
で2値化し、第2切り出し手段により図形1の左右端を
切り出す(ステップN4)。
Thereafter, as shown in FIG. 5, the scanning means scans the outer frame 7 of the figure 1 in a direction perpendicular to the figure arrangement direction to create the projection pattern 710 on the X axis, and 4 projection pattern 10 at the 4th slice level 11
Then, the left and right ends of figure 1 are cut out by the second cutting means (step N4).

上記ステップN3およびステップN4において、投影パ
ターン8.IO夫々の切り出し幅の挾い部分8a、IO
aを除去することにより、前述のノイズ12を除去する
のである。 更に、第6図および第7図に示すように、
ステップN5およびステ1プN6において、上下端の切
り出し、および、左右端の切り出しを行ない、ノイズ1
2を除去して図形lの外形枠を正確に切り出していく。
In step N3 and step N4, the projection pattern 8. The cutting width part 8a of each IO, IO
By removing a, the noise 12 mentioned above is removed. Furthermore, as shown in FIGS. 6 and 7,
In step N5 and step N6, the upper and lower ends and the left and right ends are cut out, and the noise 1
2 is removed to accurately cut out the outer frame of figure l.

本発明としては、上記上下端の切り出しおよび左右端の
切り出しを3回以上繰り返すものでも良い。また、凌敗
回繰り返す場合においては、例えば、その時点の切り出
し結果と、前回の切り出し結果とを比較手段によって比
較し、両方の切り出し結果が同じかどうかを判断さけ、
同しで無い場合にのみ切り出しを繰り返させるようにし
、ノイズが無い場合には、切り出しを繰り返さずに処理
速度を高め、かつ、ノイズが多い場合には、それらに対
応してより確実にノイズを除去するようにしても良い。
According to the present invention, the cutting of the upper and lower ends and the cutting of the left and right ends may be repeated three or more times. In addition, in the case of repeating defeats and defeats, for example, compare the cutting result at that point with the cutting result from the previous time using a comparing means, and judge whether or not both cutting results are the same.
By repeating the extraction only when they are not the same, if there is no noise, the processing speed is increased without repeating the extraction, and if there is a lot of noise, the noise can be suppressed more reliably. It may be removed.

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

第1図は、本発明の図形認識装置における図形切り出し
方法の実施例を示すフローチャート、第2図ないし第7
図は、順にステップNlないしステップN6における夫
々の作用状態を示す正面図、第8図ないし第10図は、
夫々従来例における作用状態を示す正面図である。 1・図形、2 読み取り対象。
FIG. 1 is a flowchart showing an embodiment of the figure cutting method in the figure recognition device of the present invention, and FIGS.
The figures are front views showing the operating states of steps Nl to N6 in order, and FIGS. 8 to 10 are
FIG. 6 is a front view showing the operating state of each conventional example. 1. Figure, 2. Reading target.

Claims (3)

【特許請求の範囲】[Claims] (1)夫々が連続線で構成された図形を並列して成る読
み取り対象に対して走査手段により図形並列方向に走査
し、第1切り出し手段によって前記図形の上下端を切り
出すとともに、前記読み取り対象に対して前記走査手段
により図形並列方向に直交する方向に走査し、第2切り
出し手段によって前記図形の左右端を切り出し、それに
より各図形ごとの外形枠を切り出し、次いで、前記各図
形ごとの外形枠に対して、前記第1および第2切り出し
手段による切り出し処理を行なうことを特徴とする図形
認識装置における図形切り出し方法。
(1) A scanning device scans a reading target consisting of parallel figures, each of which is composed of continuous lines, in the figure parallel direction, and a first cutting unit cuts out the upper and lower ends of the figure, and The scanning means scans the figure in a direction perpendicular to the figure parallel direction, the second cutting means cuts out the left and right ends of the figure, thereby cutting out an outer frame for each figure, and then an outer frame for each figure. A method for cutting out a figure in a figure recognition apparatus, characterized in that the first and second cutting means perform a cutting process on the figure.
(2)前記第1および第2切り出し手段による前記各図
形ごとの切り出し処理を複数回繰り返すことを特徴とす
る特許請求の範囲第1項に記載の図形切り出し方法。
(2) The figure cutting method according to claim 1, characterized in that the cutting process for each of the figures by the first and second cutting means is repeated a plurality of times.
(3)前記第1および第2切り出し手段による各図形ご
との切り出し処理において、その時点の切り出し処理結
果と前回の切り出し処理結果とを比較し、両切り出し処
理結果が同じで無い場合にのみ、切り出しを繰り返すこ
とを特徴とする特許請求の範囲第2項に記載の図形切り
出し方法。
(3) In the cutting process for each figure by the first and second cutting means, the cutting process result at that point and the previous cutting process result are compared, and only when the two cutting process results are not the same, the cutting process is performed. 3. The figure cutting method according to claim 2, wherein the method repeats the following steps.
JP59222519A 1984-10-23 1984-10-23 Method of segmenting graphic in graphic recognizing device Pending JPS61100876A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP59222519A JPS61100876A (en) 1984-10-23 1984-10-23 Method of segmenting graphic in graphic recognizing device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP59222519A JPS61100876A (en) 1984-10-23 1984-10-23 Method of segmenting graphic in graphic recognizing device

Publications (1)

Publication Number Publication Date
JPS61100876A true JPS61100876A (en) 1986-05-19

Family

ID=16783697

Family Applications (1)

Application Number Title Priority Date Filing Date
JP59222519A Pending JPS61100876A (en) 1984-10-23 1984-10-23 Method of segmenting graphic in graphic recognizing device

Country Status (1)

Country Link
JP (1) JPS61100876A (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPS6278686A (en) * 1985-10-02 1987-04-10 Sumitomo Electric Ind Ltd Optical character reader
JPS6320637U (en) * 1986-07-22 1988-02-10
JPS63113689A (en) * 1986-10-31 1988-05-18 Mazda Motor Corp Reading and discriminating method for bar code
JPS6433687A (en) * 1987-07-29 1989-02-03 Toppan Printing Co Ltd Pattern collation device

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPS57113182A (en) * 1980-12-29 1982-07-14 Fujitsu Ltd Segmenting method for character

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPS57113182A (en) * 1980-12-29 1982-07-14 Fujitsu Ltd Segmenting method for character

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPS6278686A (en) * 1985-10-02 1987-04-10 Sumitomo Electric Ind Ltd Optical character reader
JPS6320637U (en) * 1986-07-22 1988-02-10
JPS63113689A (en) * 1986-10-31 1988-05-18 Mazda Motor Corp Reading and discriminating method for bar code
JPS6433687A (en) * 1987-07-29 1989-02-03 Toppan Printing Co Ltd Pattern collation device

Similar Documents

Publication Publication Date Title
JPH0519753B2 (en)
JPS61100876A (en) Method of segmenting graphic in graphic recognizing device
EP0375352A1 (en) Method of searching a matrix of binary data
JPH07105312A (en) Method and device for eliminating dirt from character image in optical character reader
JPH076187A (en) Device for analysis of information carrier
JP2613959B2 (en) Fingerprint pattern classification device
JPH07220081A (en) Segmenting method for graphic of image recognizing device
JPH09128480A (en) Pattern area segmenting system and pattern extraction device
JP4254008B2 (en) Pattern detection apparatus and method
JP2531800B2 (en) Character cutting method in character reading device
JP3400154B2 (en) Document image area extraction method and apparatus
JP2888885B2 (en) Character extraction device
JPH04311283A (en) Line direction discriminating device
JPH0540846A (en) Japanese/european language discrimination method for document image
JPH05282487A (en) Character recognizing device
JPS61206088A (en) Character reading device
JPH04109381A (en) Character reader
JPH04267494A (en) Character segmenting method and character recognizing device
JPS60142480A (en) Printed character reader
JPH10233930A (en) Image processor
JPS62169286A (en) Character segmenting system
JPH05274472A (en) Image recognizing device
JPH09167205A (en) Method for picking-up character pattern
JPH04291691A (en) Method for recognizing character
JPH01145787A (en) Character segmenting method