JPH05303662A - Stereo character reading method using light-section method - Google Patents
Stereo character reading method using light-section methodInfo
- Publication number
- JPH05303662A JPH05303662A JP4107894A JP10789492A JPH05303662A JP H05303662 A JPH05303662 A JP H05303662A JP 4107894 A JP4107894 A JP 4107894A JP 10789492 A JP10789492 A JP 10789492A JP H05303662 A JPH05303662 A JP H05303662A
- Authority
- JP
- Japan
- Prior art keywords
- image
- character
- slit
- dimensional
- range finder
- 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
Landscapes
- Character Input (AREA)
Abstract
Description
【０００１】[0001]
【産業上の利用分野】本発明は光切断法により得られた
高さ情報をもとに立体文字を認識する立体文字読取方法
に関するものである。BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a three-dimensional character reading method for recognizing a three-dimensional character based on height information obtained by a light cutting method.
【０００２】[0002]
【従来の技術】本発明に係る従来技術としては特開昭６
２−２９３４９２号の公報がある。2. Description of the Related Art As a prior art relating to the present invention, Japanese Patent Laid-Open No.
There is a gazette of 2-293492.
【０００３】このものは、通常のＴＶカメラからの映像
信号を、あるしきい値レベルで“０”と“１”の２値に
変換して図８に示すように文字パターンを読取り、操作
線を文字外周全面捜走して、文字凹凸部の位置を検出し
文字認識を行う方法である。This device converts a video signal from a normal TV camera into a binary value of "0" and "1" at a certain threshold level and reads a character pattern as shown in FIG. Is to search the entire outer circumference of the character, detect the position of the uneven portion of the character, and perform character recognition.
【０００４】[0004]
【発明が解決しようとする課題】前記方法を立体文字認
識に使用する場合に、その立体文字の断面形状が図９に
示すＢ〜Ｂ断面形状を（ｂ）、Ｃ〜Ｃ断面形状を（Ｃ）
に示すような「１」の文字の場合に、一様平面光下で文
字と背景とのコントラストがある場合に、通常のカメラ
で読取ろうとする時、コントラストを出すためにリング
照明を利用するが、文字高さが変化しているので文字の
各部分における散乱角が一定でなくなり、文字と背景の
輝度値が同じになる所が存在して、あるしきい値でもっ
て２値画像にすると図９（ｄ）に示すように文字かすれ
が起き、文字を読取ることが出来なくなるという問題点
がある。When the above method is used for recognizing a three-dimensional character, the three-dimensional character has a cross-sectional shape shown in FIG. )
In the case of the character "1" as shown in Fig. 2, when there is a contrast between the character and the background under uniform plane light, when reading with a normal camera, the ring illumination is used to provide the contrast. Since the height of the character changes, the scattering angle at each part of the character is not constant, and there is a place where the brightness value of the character and the background are the same. As shown in FIG. 9 (d), there is a problem that a character is faint and the character cannot be read.
【０００５】本発明は画像処理装置を使用して刻印文字
等の立体文字の認識において、２次元画像を用いて２値
画像を得る場合に、文字かすれが起きない立体文字の認
識方法を技術的課題とするものである。The present invention is technically directed to a method of recognizing a three-dimensional character such that a character is not blurred when a binary image is obtained using a two-dimensional image in recognizing a three-dimensional character such as a stamped character using an image processing device. This is an issue.
【０００６】[0006]
【課題を解決するための手段】課題を解決するための技
術的手段は次のようである。The technical means for solving the problems are as follows.
【０００７】ＸＹテーブル上に文字認識対象物を載置
し、レンジファインダーよりレーザースリット光を照射
し、そのスリット画像はレンジファインダーに内蔵した
カメラで取り込み後、画像処理装置に記憶し、パソコン
本体に記憶されたプログラムによりモータのドライバに
モータの回転角を指示し、ＸＹテーブルを移動させ、画
像処理装置にそのつど画像を取り込み、光切断法を用い
た取り込まれたスリット画像より文字の高さを計測し、
その情報を基に２次元の文字パターンを構成させ、文字
を読み取る立体文字読取方法である。A character recognition object is placed on an XY table, laser slit light is emitted from a range finder, and the slit image is captured by a camera incorporated in the range finder, stored in an image processing device, and stored in a personal computer body. The stored program instructs the motor driver to indicate the rotation angle of the motor, moves the XY table, captures an image each time in the image processing device, and displays the character height from the captured slit image using the optical cutting method. Measure,
It is a three-dimensional character reading method for reading a character by forming a two-dimensional character pattern based on the information.
【０００８】[0008]
【作用】レーザスリット光は、高強度赤外光であるため
に外乱光に強く、かつ局所的であるためにコンスラスト
調整も容易である。The laser slit light is a high-intensity infrared light, so it is strong against ambient light, and since it is local, it is easy to adjust the contrast.
【０００９】従ってレンジファインダーから測定された
高さ情報をもとに２次元の文字パターンを構成すれば、
従来のような文字かすれが起きない２値画像が得られる
ものである。Therefore, if a two-dimensional character pattern is constructed based on the height information measured from the range finder,
It is possible to obtain a binary image in which the conventional character blurring does not occur.
【００１０】[0010]
【実施例】以下実施例について説明する。EXAMPLES Examples will be described below.
【００１１】従来より使用されているレンジファインダ
（Ｒ．Ｆ．）１からレーザスリット光２を認識対象物３
上に照射する。The object 3 to recognize the laser slit light 2 from the conventionally used range finder (RF) 1.
Irradiate on top.
【００１２】そのスリット画像は、レンジファインダ１
に内蔵されているカメラで取り込まれた後、白黒モニタ
６に表示され画像処理装置で画像データとして記憶処理
される。The slit image is the range finder 1
After being taken in by a camera built in, the image is displayed on the black and white monitor 6 and stored as image data in the image processing apparatus.
【００１３】キーボード１０からパソコン本体９にコマ
ンドを入力し、パソコン本体９はモータドライバ５にモ
ータの回転角を指示し、ＸＹテーブルを移動させ画像処
理装置６にそのつど、画像取込み、各種処理を行うよう
に指示しその結果をＣＲＴ８に表示させる。A command is input from the keyboard 10 to the personal computer body 9, the personal computer body 9 instructs the motor driver 5 about the rotation angle of the motor, the XY table is moved, and the image processing device 6 takes in an image and performs various processes. An instruction is given to display the result on the CRT 8.
【００１４】読取文字は、認識対象物表面上にある図２
のような本体的なものであり、刻印文字のような凹文字
１１や、浮出し文字のような凸文字１２である。The read character is on the surface of the object to be recognized as shown in FIG.
Such a main body is a concave character 11 such as a stamped character or a convex character 12 such as an embossed character.
【００１５】次に具体的な文字の読取手順について図３
〜図７を用いて説明する。Next, a specific character reading procedure will be described with reference to FIG.
~ It demonstrates using FIG.
【００１６】図３に示すように立体文字列に照射したス
リット画像をもとに三角測量の原理を用いて計測した結
果得られた立体文字列のＹＺ断面を図４に示す。FIG. 4 shows a YZ cross section of the three-dimensional character string obtained as a result of measurement using the principle of triangulation based on the slit image irradiated on the three-dimensional character string as shown in FIG.
【００１７】立体文字を認識するためには図３に示すよ
うな立体文字の中心線１３を抽出すればよい。In order to recognize a three-dimensional character, the center line 13 of the three-dimensional character as shown in FIG. 3 may be extracted.
【００１８】立体文字中心線１３は、図４に示すように
凹文字の場合、谷形の最低点１４の集合があり、凸文字
の場合、山形の最高点１５の集合である。As shown in FIG. 4, the solid character center line 13 has a set of valley-shaped lowest points 14 in the case of a concave character, and a set of mountain-shaped highest points 15 in the case of a convex character.
【００１９】この最低点あるいは最高点の決定方法には
いろいろあるが、例えば最低点を求める場合、Ｙ＝０か
ら順次Ｚ座標を比べて今のＺの値が基準高さ１６よりも
小さく、前のＺの値よりも小さくなければ、そのＸ、Ｙ
座標を記憶するという一般によく使われる最低値抽出ア
ルゴリズムを応用して決定できる。There are various methods for determining the lowest point or the highest point. For example, when obtaining the lowest point, the Z coordinates are sequentially compared from Y = 0, and the current Z value is smaller than the reference height 16, If it is not smaller than the Z value of
It can be determined by applying a commonly used minimum value extraction algorithm of storing coordinates.
【００２０】このようにして求めた最低点あるいは最高
点のＸ−Ｙ座標値を配列に割りあてる。図５に示すよう
に、例えば最低点の座標が（Ｘ、Ｙ）＝（０．１５ｍ
ｍ、０．２１ｍｍ）とすると、Ｘ座標は０．１≦×＜
０．２であるからｉ＝２とし、Ｙ座標は０．２≦Ｙ＜
０．３であるからＪ＝３として、（ｉ，ｊ）＝（２，
３）の配列を黒色にする。（ただし、初期配列の色はす
べて白とする）。次に図６に示すようにスリット光２を
スリット走査方向（Ｍ）１８と（Ｎ）１９（もしくは
（Ｐ）２０と（Ｑ）２１）というような直交する２方向
で走査して最低点あるいは最高点を２次元配列にプロッ
トすれば図７のような文字のドットパターンが得られ、
このパターンを画像処理装置のＲＡＭに転送すれば、従
来から使われている２値画像の文字特徴抽出法を用いて
文字を認識することができる。The X-Y coordinate values of the lowest point or the highest point thus obtained are assigned to the array. As shown in FIG. 5, for example, the coordinates of the lowest point are (X, Y) = (0.15 m
m, 0.21 mm), the X coordinate is 0.1 ≦ × <
Since it is 0.2, i = 2 and the Y coordinate is 0.2 ≦ Y <
Since it is 0.3, J = 3 and (i, j) = (2
Make the array of 3) black. (However, the initial array color is all white). Next, as shown in FIG. 6, the slit light 2 is scanned in two orthogonal directions such as slit scanning directions (M) 18 and (N) 19 (or (P) 20 and (Q) 21) to obtain the lowest point or If you plot the highest points in a two-dimensional array, you will get a dot pattern of characters as shown in Figure 7.
If this pattern is transferred to the RAM of the image processing apparatus, the character can be recognized by using the conventionally used character feature extraction method of the binary image.
【００２１】[0021]
【発明の効果】本発明は次の効果を有する。すなわち、 （１）従来技術では照明方法を考慮しても、局所的に文
字と背景の輝度値が同じになり、通常のＴＶカメラでは
認識するために必要な文字パターンを十分に読み取れな
いような立体文字でも本実施例の方法では読取り可能で
ある。The present invention has the following effects. That is, (1) In the prior art, even if the illumination method is taken into consideration, the brightness values of the character and the background are locally the same, and a character pattern necessary for recognition cannot be sufficiently read by a normal TV camera. Even a three-dimensional character can be read by the method of this embodiment.
【００２２】（２）本手法は局所的な３次元情報を得る
ことができるため、図１０のような立体文字の文字高さ
ｈ、ｈ2 及び文字幅Ｓ1 、Ｓ2 等の測定ができ、従来法
ではできなかった文字の“品質”を評価できる。(2) Since this method can obtain local three-dimensional information, it is possible to measure the character heights h and h 2 and the character widths S 1 and S 2 of a three-dimensional character as shown in FIG. , It is possible to evaluate the "quality" of characters that was not possible with the conventional method.
【図面の簡単な説明】[Brief description of drawings]
【図１】文字認識装置の全体構成図である。FIG. 1 is an overall configuration diagram of a character recognition device.
【図２】立体文字例の説明図である。FIG. 2 is an explanatory diagram of an example of three-dimensional characters.
【図３】スリット照射例の説明図である。FIG. 3 is an explanatory diagram of an example of slit irradiation.
【図４】立体文字列のＹＺ断面図である。FIG. 4 is a YZ sectional view of a three-dimensional character string.
【図５】実平面座標と配列の対応図である。FIG. 5 is a correspondence diagram of real plane coordinates and arrays.
【図６】スリット走査方向の説明図である。FIG. 6 is an explanatory diagram of a slit scanning direction.
【図７】文字のドットパターン例である。FIG. 7 is an example of a dot pattern of a character.
【図８】従来例の捜査方法の説明図である。FIG. 8 is an explanatory diagram of a conventional investigation method.
【図９】従来法による立体文字の読取例の説明図であ
る。FIG. 9 is an explanatory diagram of an example of reading a three-dimensional character by a conventional method.
【図１０】立体文字の規格に関する説明図である。FIG. 10 is an explanatory diagram related to a standard for three-dimensional characters.
〔１〕レンジファインダー 〔２〕レーザースリット光 〔３〕文字認識対象物 〔４〕ＸＹテーブル 〔５〕モータドライバ 〔７〕画像処理装置 〔８〕ＣＲＴ [1] Range finder [2] Laser slit light [3] Character recognition target [4] XY table [5] Motor driver [7] Image processing device [8] CRT
〔９〕パソコン 〔１０〕キーボード[9] PC [10] Keyboard
Claims (1)
し、レンジファインダーよりレーザースリット光を照射
し、そのスリット画像はレンジファインダーに内蔵した
カメラで取り込み後、画像処理装置に記憶し、パソコン
本体に記憶されたプログラムにより、モータのドライバ
にモータの回転角を指示し、ＸＹテールを移動させ、画
像処理装置にそのつど画像を取り込み、光切断法を用い
て取り込まれたスリット画像より文字の高さを計測しそ
の情報を基に２次元の文字パターンを構成させ、文字を
読み取る立体文字読取方法。1. A character recognition target object is placed on an XY table, a laser slit light is emitted from a range finder, and the slit image is captured by a camera incorporated in the range finder, and then stored in an image processing device, and is stored in a personal computer. The program stored in the main body instructs the motor driver to indicate the rotation angle of the motor, moves the XY tail, and captures an image each time in the image processing device. A three-dimensional character reading method that measures the height, forms a two-dimensional character pattern based on that information, and reads characters.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP4107894A JPH05303662A (en) | 1992-04-27 | 1992-04-27 | Stereo character reading method using light-section method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP4107894A JPH05303662A (en) | 1992-04-27 | 1992-04-27 | Stereo character reading method using light-section method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| JPH05303662A true JPH05303662A (en) | 1993-11-16 |
Family
ID=14470769
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| JP4107894A Pending JPH05303662A (en) | 1992-04-27 | 1992-04-27 | Stereo character reading method using light-section method |
Country Status (1)
| Country | Link |
|---|---|
| JP (1) | JPH05303662A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH07270142A (en) * | 1994-03-30 | 1995-10-20 | Yamada Mach Tool Kk | Method and apparatus for electronic rubbing of engraved character |
| JP2007219943A (en) * | 2006-02-17 | 2007-08-30 | Arefu Net:Kk | Casted character recognition device |
| JP2009301411A (en) * | 2008-06-16 | 2009-12-24 | Kobe Steel Ltd | Image processing method and image processing device for sampling embossed characters |
-
1992
- 1992-04-27 JP JP4107894A patent/JPH05303662A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH07270142A (en) * | 1994-03-30 | 1995-10-20 | Yamada Mach Tool Kk | Method and apparatus for electronic rubbing of engraved character |
| JP2007219943A (en) * | 2006-02-17 | 2007-08-30 | Arefu Net:Kk | Casted character recognition device |
| JP2009301411A (en) * | 2008-06-16 | 2009-12-24 | Kobe Steel Ltd | Image processing method and image processing device for sampling embossed characters |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR100792106B1 (en) | Captured image projection apparatus and captured image correction method | |
| JP4445454B2 (en) | Face center position detection device, face center position detection method, and program | |
| JP3143819B2 (en) | Eyelid opening detector | |
| JPH11244261A (en) | Iris recognition method and device thereof, data conversion method and device thereof | |
| CN109345597B (en) | Camera calibration image acquisition method and device based on augmented reality | |
| JP2893078B2 (en) | Shading correction method and device | |
| JP2007025902A (en) | Image processor and image processing method | |
| US6915022B2 (en) | Image preprocessing method capable of increasing the accuracy of face detection | |
| JP6904624B1 (en) | Club head measurement mark and image processing device | |
| JP6898150B2 (en) | Pore detection method and pore detection device | |
| JP2017012384A (en) | Wrinkle state analysis device and wrinkle state analysis method | |
| JPH05303662A (en) | Stereo character reading method using light-section method | |
| EP0265769A1 (en) | Method and apparatus for measuring with an optical cutting beam | |
| JP2519445B2 (en) | Work line tracking method | |
| JP3186308B2 (en) | Method and apparatus for monitoring sintering state of ceramic substrate | |
| JPH06109437A (en) | Measuring apparatus of three-dimensional shape | |
| JPH0933227A (en) | Discrimination method for three-dimensional shape | |
| JP4454075B2 (en) | Pattern matching method | |
| JP2733170B2 (en) | 3D shape measuring device | |
| JP4852454B2 (en) | Eye tilt detection device and program | |
| JPH0560518A (en) | Three-dimensional coordinate measurement device | |
| JPH0534117A (en) | Image processing method | |
| JP2521156Y2 (en) | Position measurement target | |
| JPH1091785A (en) | Check method and device using pattern matching | |
| JP4465911B2 (en) | Image processing apparatus and method |