JPH0652357A - Method and device for character recognition - Google Patents
Method and device for character recognitionInfo
- Publication number
- JPH0652357A JPH0652357A JP4224639A JP22463992A JPH0652357A JP H0652357 A JPH0652357 A JP H0652357A JP 4224639 A JP4224639 A JP 4224639A JP 22463992 A JP22463992 A JP 22463992A JP H0652357 A JPH0652357 A JP H0652357A
- Authority
- JP
- Japan
- Prior art keywords
- character
- area
- image
- character recognition
- binarization level
- 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
【０００１】[0001]
【産業上の利用分野】本発明は、画像処理による文字認
識方法及びそのための装置に関する。特に、計測領域内
の文字の色が互いに違っていたり、照明が不均一だった
りした場合であっても、適正な文字認識を行うことがで
きる文字認識方法及びそのための装置に関する。BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a character recognition method by image processing and an apparatus therefor. In particular, the present invention relates to a character recognition method and an apparatus therefor capable of performing proper character recognition even when the colors of characters in the measurement area are different from each other or the illumination is uneven.
【０００２】[0002]
【従来の技術】断面円形の部材（例えば、円筒形のコイ
ル等の鋼材）の表面にマーキングされた文字は、にじみ
やかすみを含む場合があり、また、曲面上に描かれてい
るがために均一に照明され得ないので、こうした文字の
自動認識を可能とするためには、文字を読み取るため
の、したがって濃淡を判別するための２値化レベルをど
のように検出し設定するかが重要になる。2. Description of the Related Art Characters marked on the surface of a member having a circular cross section (for example, a steel material such as a coil having a cylindrical shape) may include bleeding or haze and are drawn on a curved surface. Since it is not possible to illuminate uniformly, it is important to detect and set the binarization level for reading the character and thus for distinguishing the light and shade in order to enable such automatic recognition of the character. Become.
【０００３】従来の文字認識方法及び装置は、読み取る
べき文字の濃淡に基づいて２値化レベルが適正に設定さ
れていることを前提にしたものであった。このような文
字認識方法及び装置として、例えば、特開平１−２２９
３８８号公報は、文字の線素のなす閉ループ、交差、端
点等の特徴パラメータによって、１文字ずつ基準文字と
比較し、最近似値によって文字認識を行う「数字認識方
法及び装置」という名称の発明を開示している。また、
特開平２−６８６８３号公報により開示された「文字認
識方法及び装置」という名称の発明においては、従来か
らのＸ−Ｙ方向での文字毎の画素濃度（出現頻度）をパ
ターン化した基準値と対象文字との比較を行い、近似値
により文字を決定する場合、ノイズ部分を取り出して該
ノイズ部分を除いた残りの部分を照合するパターンマッ
チング法が採用されている。The conventional character recognition method and apparatus have been premised on that the binarization level is properly set based on the lightness and darkness of the character to be read. As such a character recognition method and apparatus, for example, Japanese Patent Laid-Open No. 1-229 is known.
Japanese Patent No. 388 is an invention named "Numerical Recognition Method and Device" which compares character by character with a reference character according to characteristic parameters such as closed loops, intersections, and end points of character line elements, and performs character recognition based on the closest approximation. Is disclosed. Also,
In the invention named “character recognition method and device” disclosed in Japanese Patent Laid-Open No. 2-68683, a conventional reference value is obtained by patterning pixel density (appearance frequency) of each character in the XY directions. When a character is compared with a target character and a character is determined by an approximate value, a pattern matching method is adopted in which a noise portion is taken out and the remaining portion excluding the noise portion is collated.
【０００４】一方、２値化法に関して、ＣＡＭＰ−ＩＳ
ＩＪ Ｖｏｌ４（１９９１）−３２０（２５０）におけ
る論文「ＵＯ鋼管用板マーキング文字読取装置の開発」
は、仮のしきい値で２値化を行い、１文字毎の比較的ラ
フなウィンドウを設定することにより、１文字毎に文字
を切り出し、文字内の濃度差を考慮してＹ軸方向を３分
割してＰタイル法２値化手段（各領域に対し、背景と対
象物との面積の比が既知の値Ｐとなるような濃度値をし
きい値とする方法）を採用して最適なしきい値を決定
し、最適なしきい値で２値化された画像と基本文字パタ
ーンとのマッチングを取る文字読み取りを開示してい
る。また、本願の出願人に係る特願平３−１８７４３４
号の発明「文字照合装置及びそのための方法」において
は、文字照合の過程で鮮明な２値化画像が得られない場
合、各領域毎に最適な２値化レベルを設定する手法が採
用されている。On the other hand, regarding the binarization method, CAMP-IS
IJ Vol 4 (1991) -320 (250) Paper "Development of UO Steel Pipe Plate Marking Character Reader"
Is binarized with a temporary threshold value and a relatively rough window is set for each character to cut out the character for each character, and the Y-axis direction is taken into consideration in consideration of the density difference within the character. Optimum by dividing into three and adopting the P-tile method binarization means (method of setting the density value as a threshold value such that the ratio of the area between the background and the object is a known value P for each region) It discloses a character reading in which an appropriate threshold value is determined and the binarized image and the basic character pattern are matched at the optimum threshold value. In addition, Japanese Patent Application No. 3-187434 relating to the applicant of the present application
In the invention "Character matching device and method therefor", a method of setting an optimum binarization level for each area is adopted when a clear binary image cannot be obtained in the process of character matching. There is.
【０００５】[0005]
【発明が解決しようとする課題】こうした従来の文字認
識装置及び文字認識方法は、２値化が正しく行われてい
ることを前提としているうえ、別原因で発生した大きな
ノイズ等の除去分離を対象としているにすぎず、文字の
にじみや色の変化、濃淡の変化に追従するものではなか
った。しかも、文字認識を行う全領域を単一且つ同一の
レベルで２値化しているため、このような装置によっ
て、円筒形の鋼材等の外面等の曲面にマーキングされた
文字を認識しようとしても、照明のむらやレンズのシェ
ーディングの影響を受け易く、全領域から鮮明な２値化
画像を得ることが不可能なことが多いという課題があっ
た。Such a conventional character recognition device and character recognition method are based on the premise that binarization is performed correctly, and are intended for removal and separation of large noise or the like caused by another cause. It does not follow the blurring of characters, the change of color, and the change of shading. Moreover, since the entire area for character recognition is binarized at a single and the same level, even if it is attempted to recognize a character marked on a curved surface such as an outer surface of a cylindrical steel material by such a device, There is a problem that it is often impossible to obtain a clear binarized image from the entire area because it is easily affected by uneven illumination and shading of the lens.
【０００６】また、先に挙げたＣＡＭＰ−ＩＳＩＪ Ｖ
ｏｌ４（１９９１）ー３２０に開示されたような、Ｙ軸
方向を３分割してＰタイル法２値化手段を採用する方法
は、分割によるメッシュが粗いので、場合によっては、
縦方向に対して滑らかな２値化画像を得ることが困難で
あるという課題があった。In addition, the above-mentioned CAMP-ISIJ V
In the method of dividing the Y-axis direction into three parts and adopting the P-tile method binarizing means as disclosed in ol4 (1991) -320, the mesh due to the division is coarse, and therefore, in some cases,
There is a problem that it is difficult to obtain a binary image that is smooth in the vertical direction.
【０００７】一般に、複数の文字領域を検査する場合、
各文字領域に存在する文字の形状、色、光沢度等が違っ
ていても、各領域毎に２値化レベルを設定することがで
きなければならない。Generally, when inspecting a plurality of character areas,
Even if the shape, color, glossiness, etc. of characters existing in each character area are different, it is necessary to be able to set the binarization level for each area.
【０００８】本発明は、全領域に対して単一且つ同一の
２値化レベルで文字認識する、又は、領域を分割して２
値化を行うという従来の方法及び装置の欠点を除去する
ためになされたもので、濃度分布から求めた濃度差に基
づいて各領域毎に最適な２値化レベルを設定することが
できる文字認識方法及びそのための装置を提供すること
を目的とするものである。According to the present invention, character recognition is performed with a single and same binarization level for all areas, or areas are divided into two areas.
The purpose of this is to eliminate the drawbacks of the conventional method and apparatus for performing the digitization, and it is possible to set the optimum binarization level for each area based on the density difference obtained from the density distribution. It is an object to provide a method and a device therefor.
【０００９】[0009]
【課題を解決するための手段】上記の目的を達成するた
めに、本発明は、マーキングされた文字の画像を２値化
して認識する文字認識方法において、認識すべき文字の
画像を含む文字範囲を得る段階と、該文字範囲を第１の
方向に走査し、（イ）この走査方向に沿う前記画像の濃
度分布を求め、（ロ）求めた濃度分布に基づいて、所要
の濃度差を持つ領域と該領域に対する２値化レベルとを
設定し、（ニ）設定された２値化レベルにより前記領域
を２値化する段階と、前記文字範囲に対する前記第１の
方向の走査を第２の方向に位置を変えて行い、それぞれ
の走査毎に前記（イ）−（ニ）の処理を行う段階とを含
み、それぞれの領域毎に最適な２値化レベルを設定する
ことを特徴とする文字認識方法を提供する。In order to achieve the above object, the present invention is a character recognition method for binarizing and recognizing an image of a marked character, and a character range including an image of a character to be recognized. And (b) obtain the density distribution of the image along the scanning direction, and (b) obtain the required density difference based on the obtained density distribution. Setting a region and a binarization level for the region, and (d) binarizing the region according to the set binarization level; and scanning the character range in the first direction in a second direction. A step of changing the position in each direction, and performing the processing of (a)-(d) for each scanning, and setting an optimum binarization level for each area. Provide a recognition method.
【００１０】また、本発明は、マーキングされた文字の
画像を２値化して認識する文字認識装置において、認識
すべき文字の画像を含む文字範囲を得る手段と、該文字
範囲を第１の方向に走査し、（イ）この走査方向に沿う
前記画像の濃度分布を求め、（ロ）求めた濃度分布に基
づいて、所要の濃度差を持つ領域と該領域に対する２値
化レベルとを設定し、（ニ）設定された２値化レベルに
より前記領域を２値化する手段と、前記文字範囲に対す
る前記第１の方向の走査を第２の方向に位置を変えて行
わせ、それぞれの走査毎に前記（イ）−（ニ）の処理を
行う手段とを具備し、それぞれの領域毎に最適な２値化
レベルを設定することを特徴とする文字認識装置を提供
する。Further, according to the present invention, in a character recognition device for binarizing and recognizing an image of a marked character, a means for obtaining a character range including an image of the character to be recognized, and the character range in a first direction. And (b) the density distribution of the image along the scanning direction is obtained, and (b) the area having a required density difference and the binarization level for the area are set based on the obtained density distribution. (D) Means for binarizing the area according to the set binarization level, and causing the character range to be scanned in the first direction by changing the position in the second direction. And a means for performing the above processes (a) to (d), and an optimum binarization level is set for each area.
【００１１】一実施例においては、前記濃度差は、前記
濃度分布における隣り合う極大値と極小値との差であ
り、前記領域がこれら極大値と極小値との前記第１の方
向での座標値の差の絶対値により画定され、前記最適な
２値化レベルが前記極大値と前記極小値との差の１／２
である。また、前記文字はコイル等の曲面にマーキング
されている。In one embodiment, the density difference is a difference between the maximum value and the minimum value adjacent to each other in the density distribution, and the region has coordinates of the maximum value and the minimum value in the first direction. The optimum binarization level is defined by the absolute value of the difference between the values, and the optimum binarization level is 1/2 of the difference between the maximum value and the minimum value.
Is. Further, the characters are marked on a curved surface such as a coil.
【００１２】[0012]
【作用】本発明においては、認識すべき文字の画像を含
む文字範囲が第１の方向に走査され、この走査方向に沿
う前記画像の濃度分布が求められ、求められた濃度分布
に基づいて、所要の濃度差を持つ領域と該領域に対する
２値化レベルとが設定され、設定された２値化レベルに
より前記領域が２値化される。前記文字範囲に対する前
記第１の方向の走査は第２の方向に位置を変えて行わ
れ、それぞれの走査毎に前記の処理が行われる。これに
より、それぞれの領域毎に最適な２値化レベルが設定さ
れる。In the present invention, the character range including the image of the character to be recognized is scanned in the first direction, the density distribution of the image along this scanning direction is obtained, and based on the obtained density distribution, An area having a required density difference and a binarization level for the area are set, and the area is binarized by the set binarization level. The scanning in the first direction with respect to the character range is performed by changing the position in the second direction, and the above process is performed for each scanning. As a result, the optimum binarization level is set for each area.
【００１３】[0013]
【実施例】以下、図面を参照して本発明の一実施例を詳
細に説明する。図１は、本発明に係る文字認識装置の概
略的な構成と他の装置との接続関係を示す図である。図
１において、円筒形のコイル１の湾曲した外面に、コイ
ル識別のための数字列２がマーキングされている。数字
列２は照明装置３によって照明される。しかし、数字列
２が湾曲した面にマーキングされているため、それぞれ
の文字領域が均一に照明されることはほとんどない。DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described in detail below with reference to the drawings. FIG. 1 is a diagram showing a schematic configuration of a character recognition device according to the present invention and a connection relationship with another device. In FIG. 1, a number string 2 for identifying a coil is marked on the curved outer surface of a cylindrical coil 1. The number string 2 is illuminated by the illumination device 3. However, since the number string 2 is marked on the curved surface, the respective character areas are hardly uniformly illuminated.
【００１４】数字列２に対向して、数字列２を撮影する
テレビカメラ４が配置され、テレビカメラ４に本発明に
係る文字認識装置５が接続されている。文字認識装置５
は、テレビカメラ４の出力をＡ／Ｄ変換するＡ／Ｄコン
バータ５−１と、Ａ／Ｄコンバータ５−１の出力を記憶
する第１のメモリ５−２と、標準の文字パターンを記憶
する第２のメモリ５−３と、文字認識結果を記憶する第
３のメモリ５−４と、第１のメモリの出力と第２のメモ
リの出力とを順次取り出して比較し、数字列２の数字を
１字ずつ標準の文字パターンと照合してその結果を第３
のメモリに記憶させる制御部５−５とを具備する。制御
部５−５は、本発明にしたがって文字認識のための独特
の動作を行う。A television camera 4 for photographing the number sequence 2 is arranged opposite to the number sequence 2, and the character recognition device 5 according to the present invention is connected to the television camera 4. Character recognition device 5
Stores an A / D converter 5-1 for A / D converting the output of the television camera 4, a first memory 5-2 for storing the output of the A / D converter 5-1 and a standard character pattern. The second memory 5-3, the third memory 5-4 that stores the character recognition result, and the output of the first memory and the output of the second memory are sequentially taken out and compared, and the numbers in the numeral string 2 are compared. Character by character against the standard character pattern and the result is
And a control unit 5-5 for storing it in the memory. The control unit 5-5 performs a unique operation for character recognition according to the present invention.
【００１５】次に、制御部５−５の動作を中心に、本発
明に係る文字認識装置で行われる文字認識の手順を図２
及び図３により説明する。Next, focusing on the operation of the control unit 5-5, the procedure of character recognition performed by the character recognition apparatus according to the present invention will be described with reference to FIG.
3 and FIG.
【００１６】 まず、テレビカメラ４により被写体で
ある数字列２の画像を取り込んで入力画像６（図２の
Ａ）を得る。続いて、入力画像６に対して、濃淡ヒスト
グラム計測により仮のしきい値を求め、これによって入
力画像６を２値化し、２値化された入力画像を第１のメ
モリ５−２に記憶させる。この処理により、数字列２の
入力画像は、仮のしきい値よりも濃度の高い画像と低い
画像とによって表示されることになるので、濃淡にむら
のある入力画像は、濃度の高低という２種類の濃度の画
素から構成されることになる。First, the television camera 4 captures an image of the number string 2 that is a subject to obtain an input image 6 (A in FIG. 2). Subsequently, a temporary threshold value is obtained for the input image 6 by measuring the density histogram, the input image 6 is binarized, and the binarized input image is stored in the first memory 5-2. . By this processing, the input image of the number string 2 is displayed as an image having a density higher than the temporary threshold and an image having a density lower than the tentative threshold value. It will be composed of pixels of different densities.
【００１７】 次に、仮のしきい値で２値化された入
力画像６の縦方向及び横方向の周辺分布、即ち、画素の
値を縦方向及び横方向に１列ずつ合計した値を求め、比
較的ラフな文字列範囲を決定する。この処理は、横方向
及び縦方向に画素の値を投射したヒストグラムを求め、
この濃度分布の交差部分から文字列範囲を決める処理で
ある。既に入力画像６は上記において２値化されてい
るため、この処理により、数字列２における数字の位置
が鮮明化される。Next, the vertical and horizontal peripheral distributions of the input image 6 binarized with a temporary threshold value, that is, the values obtained by summing the pixel values one column at a time in the vertical and horizontal directions are obtained. , Determine a relatively rough string range. This process obtains a histogram in which the pixel values are projected in the horizontal and vertical directions,
This is a process of determining the character string range from the intersection of this density distribution. Since the input image 6 has already been binarized in the above, the position of the numeral in the numeral string 2 is sharpened by this processing.
【００１８】 これに続いて、で求めた文字列範囲
（図２のＢ）内の数字列の画像を入力画像６から抽出す
る。即ち、入力画像６から文字列範囲以外の画像を消去
し、文字列範囲内にある数字列の画像のみを残す。Subsequently to this, the image of the numeral string within the character string range (B in FIG. 2) obtained in step S is extracted from the input image 6. That is, the image other than the character string range is erased from the input image 6, and only the image of the number string within the character string range is left.
【００１９】 抽出された文字列範囲の画像に対し
て、数字の配列された方向にＸ軸を、それに直交する方
向にＹ軸を設定し、Ｙ軸方向の異なる全ての位置におい
てＸ軸方向に走査を行い（図２のＢ）、各回のＸ軸方向
の走査毎に以下の処理を行う。With respect to the image of the extracted character string range, the X axis is set in the direction in which the numbers are arranged, and the Y axis is set in the direction orthogonal thereto, and the X axis direction is set at all different positions in the Y axis direction. Scanning is performed (B in FIG. 2), and the following processing is performed for each scanning in the X-axis direction.
【００２０】（イ）Ｘ軸方向の濃度分布を求める（図２
のＣ）。(B) Obtaining the concentration distribution in the X-axis direction (Fig. 2
C).
【００２１】（ロ）上記（イ）で求めた濃度分布に基づ
いて、Ｘ軸方向の濃度値の極大値及び極小値を順次求め
ると共に、それぞれのＸ座標を求める。ここで、極大値
とは濃度分布の山の頂上での値であり、極小値とは濃度
分布の谷での値である。(B) Based on the concentration distribution obtained in (a) above, the maximum value and the minimum value of the density value in the X-axis direction are sequentially obtained, and the respective X coordinates are obtained. Here, the maximum value is the value at the top of the peak of the concentration distribution, and the minimum value is the value at the valley of the concentration distribution.
【００２２】（ハ）こうして求められた値のうち、任意
の隣り合う極大値Ｑｔ（Ｘ座標をＸｔとする）及び極小
値Ｑｂ（Ｘ座標をＸｂとする）とに着目し、│Ｘｔ−Ｘ
ｂ│によって定まる領域を適宜の２値化レベルにより２
値化する。この場合、２値化レベルは（Ｑｔ−Ｑｂ）／
ｋ（ｋは任意の正の整数）に設定される（図２のＤ）。
例えば図３に示すような濃度分布の場合、第１の隣り合
う極大値Ｑｔ1と極小値Ｑｂ1との組については、それら
のＸ座標をそれぞれＸｔ1、Ｘｂ1とすると、│Ｘｔ1−
Ｘｂ1│の領域に対する２値化レベルは（Ｑｔ1−Ｑ
ｂ1）／２に設定される。また、第２の隣り合う極大値
Ｑｔ2と極小値Ｑｂ2との組の場合には、それらのＸ座標
の差の絶対値│Ｘｔ2−Ｘｂ2│により画定される領域に
対して（Ｑｔ2−Ｑｂ2）／２を２値化レベルとして２値
化を行う。(C) Focusing on any adjacent maximum value Qt (where X coordinate is Xt) and minimum value Qb (where X coordinate is Xb) among the values thus obtained, | Xt-X
The area defined by b│ is set to 2 by an appropriate binarization level.
Quantify. In this case, the binarization level is (Qt-Qb) /
k (k is an arbitrary positive integer) (D in FIG. 2).
For example, in the case of the concentration distribution as shown in FIG. 3, for the pair of the first adjacent maximum value Qt 1 and minimum value Qb 1 , if their X coordinates are Xt 1 and Xb 1 , respectively, then | Xt 1 −
The binarization level for the region of Xb 1 | is (Qt 1 −Q
b 1 ) / 2 is set. Further, in the case of the set of the second maximum value Qt 2 and the minimum value Qb 2 which are adjacent to each other, (Qt 2) is defined for the area defined by the absolute value | Xt 2 −Xb 2 | Binarization is performed with 2- Qb 2 ) / 2 as the binarization level.
【００２３】（ニ）こうして、文字列範囲を縦横に細か
い領域に細分化し、各領域毎に好適な２値化レベルを設
定しながら全領域を２値化する。(D) In this way, the character string range is subdivided vertically and horizontally into fine regions, and the entire region is binarized while setting a suitable binarization level for each region.
【００２４】 上記において数字列の２値化画像を
得ることができ、この２値化画像に対して画素のＸ軸方
向及びＹ軸方向の周辺分布を求める。こうして１つずつ
の数字の位置を鮮明にすることにより、それぞれの数字
の画像を切り出し、切り出された画像のそれぞれについ
て、数字列２とテレビカメラ４との距離の違いによる画
像の大きさの変動を補正し、パターンマッチングさせる
べき標準の文字パターンの大きさと一致させる。In the above, a binarized image of a number string can be obtained, and the peripheral distribution of pixels in the X-axis direction and the Y-axis direction is obtained for this binarized image. By sharpening the position of each number in this way, the image of each number is cut out, and for each of the cut-out images, the variation of the image size due to the difference in the distance between the number string 2 and the television camera 4 is cut. To match the size of the standard character pattern to be pattern matched.
【００２５】 上記で切り出された数字の画像を１
つずつ、第２のメモリ５−３に記憶された標準の文字パ
ターンとパターンマッチングにより一致度を計算し、最
大一致度を示す標準パターンのフォントを認識結果とす
る。The numbered image cut out above is 1
The matching degree is calculated by pattern matching with the standard character pattern stored in the second memory 5-3 one by one, and the font of the standard pattern showing the maximum matching degree is used as the recognition result.
【００２６】図４は、本発明の文字認識装置による２値
化結果を示す写真である。図４のＡ−１は、コイルの外
面にマーキングされた数字列９９５５６の原画像の写
真、図４のＡ−２はこの数字列を本発明の文字認識装置
により２値化した結果の写真である。図４のＢ−１は数
字列９１７８９の原画像の写真であり、図４のＢ−２
は、この数字列を本発明の文字認識装置により２値化し
た数字列の写真である。本発明にあっては、コイル外面
にマーキングされた数字列の左右、上下方向に明るさの
変動や文字のにじみがある場合であっても、その数字列
の鮮明な２値化画像を得ることができる。FIG. 4 is a photograph showing the result of binarization by the character recognition device of the present invention. A-1 of FIG. 4 is a photograph of the original image of the number string 99556 marked on the outer surface of the coil, and A-2 of FIG. 4 is a photograph of the result of binarizing this number string by the character recognition device of the present invention. is there. B-1 in FIG. 4 is a photograph of the original image of the number string 91789, and B-2 in FIG.
Is a photograph of the digit string digitized by the character recognition device of the present invention. According to the present invention, a clear binary image of a number string marked on the outer surface of the coil can be obtained even if there is a brightness variation or a character blur in the left and right and up and down directions of the number string. You can
【００２７】このように、コイルのような曲面にマーキ
ングされた数字列を認識する場合にも、２値化レベルを
細かな領域毎に個別に設定して２値化画像を求め、これ
に基づいて数字の認識を行うので、数字のにじみ、照明
の不均一等によるノイズに起因する誤認識を低減するこ
とができるので、コイルの誤搬送を防止することがで
き、監視に要する人員を省くことができる。In this way, also when recognizing a number string marked on a curved surface such as a coil, the binarization level is individually set for each fine area to obtain a binarized image, and based on this Since the numbers are recognized by using this method, it is possible to reduce erroneous recognition due to noise due to bleeding of numbers, uneven lighting, etc., so that it is possible to prevent erroneous conveyance of the coil and save personnel required for monitoring. You can
【００２８】また、マーキングされた数字の色が異なる
場合や、照明が不均一の場合にも、それぞれの領域毎に
最適な２値化レベルを設定することができるので、文字
認識精度が向上され、文字認識の適用分野を拡大するこ
とができるばかりでなく、種々の応用分野に柔軟に対応
することが可能になる。Further, even when the colors of the marked numbers are different or the illumination is uneven, the optimum binarization level can be set for each area, so that the character recognition accuracy is improved. Not only can the field of application of character recognition be expanded, but it is also possible to flexibly deal with various fields of application.
【００２９】以上、数字列の認識について説明してきた
が、本発明は数字の認識に限られるものではなく、アル
ファベットやカタカナ等の文字や記号等を含む一般的な
パターンの認識にも適用することができる。したがっ
て、本願の明細書において使用される「文字」という用
語は、文字、数字、記号等を含むパターン一般を意味す
るものと解すべきである。Although the recognition of the numeric character string has been described above, the present invention is not limited to the recognition of the numeric character, but can be applied to the recognition of a general pattern including characters and symbols such as alphabets and katakana. You can Therefore, the term "letter" as used in the specification of the present application should be understood to mean a general pattern including letters, numbers, symbols and the like.
【００３０】[0030]
【発明の効果】以上、実施例により本発明を詳細に説明
してきたところから明らかなとおり、本発明は、文字が
曲面上にマーキングされている場合、異なる色でマーキ
ングされた文字を認識する場合、文字が均一に照明され
ていない場合等、文字の認識が困難であったり、誤認識
が発生し易い場合であっても、抽出された文字を含む文
字範囲を第１の方向に位置をずらせながら第２の方向に
走査し、それぞれの走査において走査方向に沿う濃度分
布を求め、求めた濃度分布の濃度差に基づいて決定され
た領域毎に最適な２値化レベルを設定するようにしたの
で、認識すべき文字のマーキングされている区域の照明
の不均一にも拘わらず正確な文字認識を行うことがで
き、認識すべき文字の鮮明な画像を得ることができる。
したがって、本発明は、文字の認識精度を向上させるこ
とができるばかりでなく、文字の誤認識による誤処理を
防止することができ、更に、誤認識の発生の有無を監視
する人員を削減することができるという格別の効果を奏
する。As is apparent from the above description of the present invention in detail with reference to the embodiments, the present invention is applicable to the case where a character is marked on a curved surface and the character marked with a different color is recognized. , Even if the characters are difficult to recognize, such as when the characters are not evenly illuminated, or when erroneous recognition is likely to occur, the character range including the extracted characters is displaced in the first direction. While scanning in the second direction, the density distribution along the scanning direction is obtained in each scan, and the optimum binarization level is set for each area determined based on the density difference of the obtained density distribution. Therefore, accurate character recognition can be performed regardless of uneven illumination of the area where the character to be recognized is marked, and a clear image of the character to be recognized can be obtained.
Therefore, the present invention can not only improve the recognition accuracy of characters, but also prevent erroneous processing due to erroneous recognition of characters, and further reduce the number of personnel who monitor whether or not erroneous recognition has occurred. There is a special effect that you can.
【図１】本発明に係る文字認識装置の構成と他の装置と
の接続関係とを示す概略図。FIG. 1 is a schematic diagram showing a configuration of a character recognition device according to the present invention and a connection relationship with another device.
【図２】本発明に係る文字認識装置における文字認識手
順を説明するための図。FIG. 2 is a diagram for explaining a character recognition procedure in the character recognition device according to the present invention.
【図３】本発明に係る文字認識装置において設定される
最適な２値化レベルを説明するための図。FIG. 3 is a diagram for explaining an optimum binarization level set in the character recognition device according to the present invention.
【図４】本発明に係る文字認識装置により２値化された
数字列の画像の写真である。FIG. 4 is a photograph of an image of a digit string digitized by the character recognition device according to the present invention.
１：コイル、 ２：数字列、 ３：照明装置、
４：テレビカメラ、５：文字認識装置、 Ｑｔ：極大
値、 Ｑｂ：極小値1: coil, 2: number string, 3: lighting device,
4: TV camera, 5: Character recognition device, Qt: maximum value, Qb: minimum value
Claims (6)
て認識する文字認識方法において、 認識すべき文字の画像を含む文字範囲を得る段階と、 該文字範囲を第１の方向に走査し、（イ）この走査方向
に沿う前記画像の濃度分布を求め、（ロ）求めた濃度分
布に基づいて、所要の濃度差を持つ領域と該領域に対す
る２値化レベルとを設定し、（ニ）設定された２値化レ
ベルにより前記領域を２値化する段階と、 前記文字範囲に対する前記第１の方向の走査を第２の方
向に位置を変えて行い、それぞれの走査毎に前記（イ）
−（ニ）の処理を行う段階とを含み、それぞれの領域毎
に最適な２値化レベルを設定することを特徴とする文字
認識方法。1. A character recognition method for binarizing and recognizing a marked character image, the method comprising: obtaining a character range including an image of a character to be recognized; scanning the character range in a first direction; (A) The density distribution of the image along the scanning direction is obtained, and (b) an area having a required density difference and a binarization level for the area are set based on the obtained density distribution, and (d) Binarizing the area according to the set binarization level, and scanning the character range in the first direction by changing the position in the second direction, and (a) for each scan.
-(D) performing a process, and setting an optimum binarization level for each area.
前記濃度差が、前記濃度分布における隣り合う極大値と
極小値との差であり、前記領域がこれら極大値と極小値
との前記第１の方向での座標値の差の絶対値により画定
され、前記最適な２値化レベルが前記極大値と前記極小
値との差の１／２である文字認識方法。2. The character recognition method according to claim 1, wherein
The density difference is a difference between an adjacent maximum value and a minimum value in the density distribution, and the region is defined by an absolute value of a difference between the maximum value and the minimum value in the coordinate value in the first direction. A character recognition method in which the optimum binarization level is 1/2 of the difference between the maximum value and the minimum value.
前記文字がコイル等の曲面にマーキングされている文字
認識方法。3. The character recognition method according to claim 2, wherein
A character recognition method in which the character is marked on a curved surface such as a coil.
て認識する文字認識装置において、 認識すべき文字の画像を含む文字範囲を得る手段と、 該文字範囲を第１の方向に走査し、（イ）この走査方向
に沿う前記画像の濃度分布を求め、（ロ）求めた濃度分
布に基づいて、所要の濃度差を持つ領域と該領域に対す
る２値化レベルとを設定し、（ニ）設定された２値化レ
ベルにより前記領域を２値化する手段と、 前記文字範囲に対する前記第１の方向の走査を第２の方
向に位置を変えて行わせ、それぞれの走査毎に前記
（イ）−（ニ）の処理を行う手段とを具備し、それぞれ
の領域毎に最適な２値化レベルを設定することを特徴と
する文字認識装置。4. A character recognition device for binarizing and recognizing an image of a marked character, means for obtaining a character range including an image of a character to be recognized, and scanning the character range in a first direction, (A) The density distribution of the image along the scanning direction is obtained, and (b) an area having a required density difference and a binarization level for the area are set based on the obtained density distribution, and (d) A unit that binarizes the area according to the set binarization level, and causes the scanning of the character range in the first direction to be performed while changing the position in the second direction. )-(D) processing means, and an optimal binarization level is set for each area.
前記濃度差が、前記濃度分布における隣り合う極大値と
極小値との差であり、前記領域がこれら極大値と極小値
との前記第１の方向での座標値の差の絶対値により画定
され、前記最適な２値化レベルが前記極大値と前記極小
値との差の１／２である文字認識装置。5. The character recognition device according to claim 4,
The density difference is a difference between an adjacent maximum value and a minimum value in the density distribution, and the region is defined by an absolute value of a difference between the maximum value and the minimum value in the coordinate value in the first direction. A character recognition device in which the optimum binarization level is 1/2 of a difference between the maximum value and the minimum value.
前記文字がコイル等の曲面にマーキングされている文字
認識装置。6. The character recognition device according to claim 5,
A character recognition device in which the character is marked on a curved surface such as a coil.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP4224639A JP2998443B2 (en) | 1992-07-31 | 1992-07-31 | Character recognition method and device therefor |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP4224639A JP2998443B2 (en) | 1992-07-31 | 1992-07-31 | Character recognition method and device therefor |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| JPH0652357A true JPH0652357A (en) | 1994-02-25 |
| JP2998443B2 JP2998443B2 (en) | 2000-01-11 |
Family
ID=16816869
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| JP4224639A Expired - Lifetime JP2998443B2 (en) | 1992-07-31 | 1992-07-31 | Character recognition method and device therefor |
Country Status (1)
| Country | Link |
|---|---|
| JP (1) | JP2998443B2 (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2008097590A (en) * | 2006-09-13 | 2008-04-24 | Keyence Corp | Character excising apparatus, method, and program |
| JP2014153820A (en) * | 2013-02-06 | 2014-08-25 | Nidec Sankyo Corp | Character segmentation device and character segmentation method |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPS5411625A (en) * | 1977-06-27 | 1979-01-27 | Ricoh Co Ltd | Processing system for video signal |
| JPS55119781A (en) * | 1979-03-09 | 1980-09-13 | Fujitsu Ltd | Binary system |
| JPS6452785A (en) * | 1987-06-05 | 1989-02-28 | Donegani Guido Ist | Manufacture of silane or siloxane compound containing at least one cycloalkyl ring |
| JPH0468482A (en) * | 1990-07-10 | 1992-03-04 | Kawasaki Steel Corp | Method and device for recognizing character |
-
1992
- 1992-07-31 JP JP4224639A patent/JP2998443B2/en not_active Expired - Lifetime
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPS5411625A (en) * | 1977-06-27 | 1979-01-27 | Ricoh Co Ltd | Processing system for video signal |
| JPS55119781A (en) * | 1979-03-09 | 1980-09-13 | Fujitsu Ltd | Binary system |
| JPS6452785A (en) * | 1987-06-05 | 1989-02-28 | Donegani Guido Ist | Manufacture of silane or siloxane compound containing at least one cycloalkyl ring |
| JPH0468482A (en) * | 1990-07-10 | 1992-03-04 | Kawasaki Steel Corp | Method and device for recognizing character |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2008097590A (en) * | 2006-09-13 | 2008-04-24 | Keyence Corp | Character excising apparatus, method, and program |
| JP2014153820A (en) * | 2013-02-06 | 2014-08-25 | Nidec Sankyo Corp | Character segmentation device and character segmentation method |
Also Published As
| Publication number | Publication date |
|---|---|
| JP2998443B2 (en) | 2000-01-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108416766B (en) | Double-side light-entering type light guide plate defect visual detection method | |
| CN108596166A (en) | A kind of container number identification method based on convolutional neural networks classification | |
| CN109816644A (en) | A kind of bearing defect automatic checkout system based on multi-angle light source image | |
| JP3251918B2 (en) | Binarization method in optical character recognition system | |
| US7333656B2 (en) | Image processing method and image processing apparatus | |
| EP0831421B1 (en) | Method and apparatus for retouching a digital color image | |
| CN114486903B (en) | Gray-scale self-adaptive coiled material detection system, device and algorithm | |
| JPH0397086A (en) | Character recognizing method | |
| CN114387269A (en) | Fiber yarn defect detection method based on laser | |
| CN113591507A (en) | Robust two-dimensional code DataMatrix positioning method and system | |
| JP3890844B2 (en) | Appearance inspection method | |
| JPH0950527A (en) | Frame extracting device and rectangle extracting device | |
| CN109165650B (en) | Automatic detection method for minimum repeated unit of printed fabric image | |
| JPH0652357A (en) | Method and device for character recognition | |
| US5119441A (en) | Optical character recognition apparatus and method using masks operation | |
| JPH06223177A (en) | Improvement of image processing | |
| JP2000132684A (en) | External appearance inspecting method | |
| US5682440A (en) | Method of extracting pixels constituting intensity changes in an image and method of identifying collection of intensity changes | |
| JPH04350546A (en) | Detection of foreign matter | |
| JPH041867A (en) | Picture quality evaluating method | |
| JP2000193420A (en) | Eye position detecting device | |
| JP2001126027A (en) | Number plate recognition system | |
| Martín et al. | Automatic reading of digital instrumentation | |
| JPH0644404A (en) | Inspecting instrument for printed matter | |
| JPH11351835A (en) | Method for extracting contour line of cutting edge by processing taken image of tool |