JPH09185683A - Method and device for processing image for character checking device - Google Patents

Method and device for processing image for character checking device

Info

Publication number
JPH09185683A
JPH09185683A JP7343290A JP34329095A JPH09185683A JP H09185683 A JPH09185683 A JP H09185683A JP 7343290 A JP7343290 A JP 7343290A JP 34329095 A JP34329095 A JP 34329095A JP H09185683 A JPH09185683 A JP H09185683A
Authority
JP
Japan
Prior art keywords
data
character
image
image memory
memory
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
JP7343290A
Other languages
Japanese (ja)
Inventor
Aritomo Kikuchi
有朋 菊池
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.)
Advantest Corp
Original Assignee
Advantest 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 Advantest Corp filed Critical Advantest Corp
Priority to JP7343290A priority Critical patent/JPH09185683A/en
Publication of JPH09185683A publication Critical patent/JPH09185683A/en
Pending legal-status Critical Current

Links

Landscapes

  • Character Discrimination (AREA)
  • Character Input (AREA)

Abstract

PROBLEM TO BE SOLVED: To surely prepare standard character data even when the row or column of characters of a device to be tested (DUT) to be defined as a reference is inclined. SOLUTION: The description plane of DUT, where the characters to be defined as a standard are described, is photographed, provided picture element values (the luminance of picture elements) are stored in an image memory M1, these data are converted into binary data by performing threshold processing and stored in an image memory M2, noises are reduced by thinning these data, these data are stored in an image memory M3, and these data are returned into original thickness by reforming thickening processing and stored in an image memory M4. A circumscribed rectangle is found for each character from these data, its coordinate positions are stored in a circumscribed rectangle memory M6, the image data in the image memory M1 are punched out by these circumscribed rectangle data, and the image data for each character are segmented and stored in an image memory M5 as standard character data. Further, the inclining angle of row line/column line is found and the inclination corrected standard character data can be provided as well.

Description

【発明の詳細な説明】Detailed Description of the Invention

【0001】[0001]

【発明の属する技術分野】この発明は、DUT(被試験
デバイス)のパッケージに表記された文字を検査する装
置において、リファレンスとなる標準文字データを得る
ための画像処理方法及びその装置の改良に関する。
BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing method for obtaining standard character data as a reference in an apparatus for inspecting characters written on a DUT (device under test) package and an improvement in the apparatus.

【0002】[0002]

【従来の技術】DUTパッケージの文字検査装置では、
DUTパッケージの表面の表記文字をカメラで撮影し、
その画像データ(画素の輝度を表す画素値のマトリクス
データ)を画像メモリに記憶し、標準文字メモリに格納
されている標準文字データと1字づつ比較して良/不良
の判定を行っている。
2. Description of the Related Art In a character inspection device for a DUT package,
Take a picture of the inscription on the surface of the DUT package with a camera,
The image data (matrix data of pixel values representing the brightness of the pixel) is stored in the image memory, and the character data is compared with the standard character data stored in the standard character memory one by one to judge whether the image is good or bad.

【0003】リファレンスとなる標準文字データを作成
するのに図4の画像処理装置が用いられる。以下に画像
処理装置による標準文字データの作成方法を説明する。 (a)DUTの内から文字データの標準モデルとすべき
DUTを選び、文字の表記面をカメラで撮影して、その
画像データ(画素値)を画像メモリM1に格納する。
The image processing apparatus shown in FIG. 4 is used to create standard character data serving as a reference. The method of creating standard character data by the image processing apparatus will be described below. (A) A DUT to be used as a standard model of character data is selected from the DUTs, a character notation surface is photographed by a camera, and the image data (pixel value) is stored in the image memory M1.

【0004】(b)複数行の表記文字が図5Aに示すよ
うにX軸にほぼ平行に印刷されている場合、演算部C1
では画像メモリM1のデータからY軸上での画素値(輝
度)の総和L1(y)(図5B)を求め、その輝度特性
の変化より各行の文字のY軸上の位置を求める。 (c)演算部C2では、各行についてX軸上での画素値
の総和L2(x)(図5C)を求め、その輝度特性の変
化より各文字のX軸上の位置を求める。
(B) When plural lines of written characters are printed substantially parallel to the X-axis as shown in FIG. 5A, the arithmetic unit C1
Then, the sum L1 (y) of pixel values (luminance) on the Y axis (FIG. 5B) is obtained from the data in the image memory M1, and the position on the Y axis of the character of each line is obtained from the change in the luminance characteristic. (C) The calculation unit C2 obtains the sum L2 (x) of pixel values on the X axis for each row (FIG. 5C), and obtains the position of each character on the X axis from the change in the luminance characteristic.

【0005】(d)文字切り出し部C3では、各文字の
X,Y座標を用いて、画像メモリM1のデータより、1
字ごとのドットマトリクスより成る画像データを切り出
して、リファレンスとなる標準文字データとして画像メ
モリM2に格納する。
(D) In the character slicing section C3, the X and Y coordinates of each character are used to extract 1 from the data in the image memory M1.
Image data composed of a dot matrix for each character is cut out and stored in the image memory M2 as standard character data serving as a reference.

【0006】[0006]

【発明が解決しようとする課題】従来の画像処理装置で
は、DUTの例えば表面の一辺を例えば座標のX軸上に
とって、各行のY軸上での画素値の和L1(y)を求め
ているため、図6に示すようにリファレンスとすべきD
UTの文字の行が傾いている場合には、画素値の和L1
(y)は図6に示すように行と行との間の谷が現れず、
各行のY座標位置を検出できなくなり、標準文字データ
が得られないと言う問題があった。
In the conventional image processing apparatus, for example, one side of the surface of the DUT is set on the X axis of the coordinates, and the sum L1 (y) of the pixel values on the Y axis of each row is obtained. Therefore, as shown in FIG.
If the line of characters of UT is tilted, the sum L1 of pixel values
(Y), as shown in FIG. 6, the valley between the rows does not appear,
There is a problem that the Y coordinate position of each line cannot be detected and standard character data cannot be obtained.

【0007】この発明の目的は、リファレンスとすべき
DUTの文字の行(または列)が傾いていても、確実に
標準文字データを格納できるようにするものである。
An object of the present invention is to ensure that standard character data can be stored even if the row (or column) of the character of the DUT to be referenced is tilted.

【0008】[0008]

【課題を解決するための手段】[Means for Solving the Problems]

(1)請求項1、2、3の発明はそれぞれ次の(2)、
(3)、(4)項で述べる画像処理方法を採用した画像
処理装置であり、以下の記載と重複するところが多いの
で詳しい説明を省略する。 (2)請求項4の文字検査装置の画像処理方法は、標準
にしようとする文字が表記されているDUTの表記面を
撮影して得られた画素値(画素の輝度)を画像メモリ1
に格納し、その画像メモリ1のデータをしきい値処理し
て2値データに変換し、画像メモリ2に格納し、その画
像メモリ2のデータを細線化処理して、画像メモリ3に
格納し、その画像メモリ3のデータを太線化処理して、
画像メモリ4に格納する。
(1) The inventions of claims 1, 2 and 3 are as follows (2),
The image processing apparatus adopts the image processing method described in the items (3) and (4), and since there is much overlap with the following description, detailed description thereof will be omitted. (2) In the image processing method of the character inspection apparatus according to claim 4, the pixel value (luminance of pixel) obtained by photographing the notation surface of the DUT on which the characters to be standardized are photographed is stored in the image memory 1.
The image data in the image memory 1 is thresholded to be converted into binary data, stored in the image memory 2, the data in the image memory 2 is thinned, and stored in the image memory 3. , The data in the image memory 3 is thickened,
It is stored in the image memory 4.

【0009】その画像メモリ4のデータより一文字ごと
に外接四角形を求めて、その座標位置を外接四角形メモ
リに格納し、その外接四角形メモリに格納された各外接
四角形(データ)で前記画像メモリ1の画像データを打
ち抜いて、1字ごとの画像データを切り出し、標準文字
データとして画像メモリ5に格納するものである。 (3)請求項5の発明は、請求項4において、一字ごと
の外接四角形の面積を求めて、しきい値と比較し、しき
い値以下のものは一文字に対応する外接四角形より除外
するものである。
The circumscribed quadrangle is obtained for each character from the data in the image memory 4, the coordinate position is stored in the circumscribed quadrangle memory, and the circumscribed quadrangle (data) stored in the circumscribed quadrangle memory is used to store the circumscribed quadrangle in the image memory 1. The image data is punched out, the image data for each character is cut out, and stored in the image memory 5 as standard character data. (3) In the invention of claim 5, in claim 4, the area of the circumscribing quadrangle for each character is calculated and compared with a threshold value, and those less than or equal to the threshold value are excluded from the circumscribing quadrangle corresponding to one character. It is a thing.

【0010】(4)請求項6の発明は、請求項4におい
て、外接四角形メモリのデータから表記文字の各行線
(または列線)に平行な直線を求め、更にその直線の水
平線(または垂直線)となす角度θを求め、画像メモリ
5の一字ごとに切り出した画像データに対して、一θだ
け傾けた画像データを演算して、画像メモリ6に格納す
るものである。
(4) According to the invention of claim 6, in claim 4, a straight line parallel to each row line (or column line) of the notation character is obtained from the data of the circumscribed quadrangle memory, and the horizontal line (or vertical line) of the straight line is obtained. ) Is formed, and the image data cut out for each character of the image memory 5 is calculated by inclining the image data by one θ and stored in the image memory 6.

【0011】[0011]

【発明の実施の形態】図1〜図3の実施例を参照して発
明の実施の形態を説明する。DUTの中から、リファレ
ンスにしようとする文字(例えば黒地に白)で表記され
ているDUTの表記面(図2A)を撮影して得られた画
素値(画素の輝度)が画像メモリM1に格納される。画
像メモリM1のデータはしきい値処理部2に入力され、
しきい値処理されて、“1”または“0”の2値データ
に変換されて画像メモリM2に格納される。
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described with reference to FIGS. The pixel value (pixel brightness) obtained by photographing the notation surface (FIG. 2A) of the DUT, which is described by the character to be used as a reference (for example, white on a black background) from the DUT, is stored in the image memory M1. To be done. The data in the image memory M1 is input to the threshold processing unit 2,
It is thresholded, converted into binary data of "1" or "0", and stored in the image memory M2.

【0012】DUTの表記面にほこりや傷等があると、
画像メモリM2のデータを表示した場合、図2Bに示す
ようにノイズNが画面に現れる。このノイズNを消すた
め、画像メモリM2のデータは細線化処理部(収縮処理
部とも言う)3に入力され、文字の線を細くする処理が
行われ(図2C)、ノイズの消された処理後のデータが
画像メモリM3に格納される。画像メモリM3のデータ
は太線化処理部(膨張処理部とも言う)4に入力され、
文字の太さが細線化処理を行う前の太さに戻され(図2
D)、画像メモリM4に格納される。上述の細線化処理
や太線化処理それぞれの技術は画像処理の分野ではよく
知られていることである。
If there is dust or scratches on the written surface of the DUT,
When the data in the image memory M2 is displayed, noise N appears on the screen as shown in FIG. 2B. In order to erase the noise N, the data in the image memory M2 is input to the thinning processing unit (also referred to as contraction processing unit) 3, and the processing for thinning the character lines is performed (FIG. 2C), and the noise-erased processing is performed. The subsequent data is stored in the image memory M3. The data in the image memory M3 is input to the thickening processing unit (also called expansion processing unit) 4,
The thickness of the characters is returned to the thickness before the thinning process (Fig. 2
D), and stored in the image memory M4. The techniques of the thinning processing and the thickening processing described above are well known in the field of image processing.

【0013】画像メモリM4のデータはラベリング処理
部5に入力され、各文字の外接四角形Sqが演算され
(図2E)、そのX,Y座標が外接四角形メモリ6に格
納される。外接四角形Sqの各辺はそれぞれ画素の位置
座標の基準となるX軸またはY軸に平行とされる。上述
の細線化処理及び太線化処理で消すことのできなかった
ノイズがある場合には、ラベリング処理部5で各外接四
角形Sqの面積を求めて、しきい値と比較し、しきい値
以下のものは外接四角形より除外するのが望ましい。
The data in the image memory M4 is input to the labeling processing unit 5, the circumscribed quadrangle Sq of each character is calculated (FIG. 2E), and the X and Y coordinates thereof are stored in the circumscribed quadrangle memory 6. Each side of the circumscribing quadrangle Sq is parallel to the X-axis or the Y-axis that serves as a reference for the pixel position coordinates. If there is noise that cannot be eliminated by the above-described thinning processing and thickening processing, the labeling processing unit 5 calculates the area of each circumscribing quadrangle Sq, compares it with a threshold value, and It is desirable to exclude objects from the circumscribed rectangle.

【0014】文字切り出し部7では、画像メモリM1の
画像を外接四角形メモリ6の各文字に対応する外接四角
形の縦線及び横線にそれぞれ2〜3ピクセルの余裕をと
った外接四角形で打ち抜いた画像データが作成され、リ
ファレンスとすべき標準文字データとして画像メモリM
5に格納される。上述の外接四角形Sqを求めて画像メ
モリM1の画像より各文字を切り出す方法は、図3
(a)に示すように複数行がX軸方向に対して傾いてい
ても、各文字の外接四角形Sqを求めることができるの
で、従来のような問題は生じない。列線が傾いている場
合も同様である。
In the character slicing section 7, the image data of the image memory M1 is punched out by a circumscribed quadrangle with a margin of 2 to 3 pixels on each of the vertical and horizontal lines of the circumscribed quadrangle corresponding to each character of the circumscribed quadrangle memory 6. Is created, and the image memory M is used as standard character data to be used as a reference.
Stored in 5. A method for obtaining the circumscribed quadrangle Sq and cutting out each character from the image in the image memory M1 is described in FIG.
Even if a plurality of lines are inclined with respect to the X-axis direction as shown in (a), since the circumscribing quadrangle Sq of each character can be obtained, the conventional problem does not occur. The same applies when the column lines are inclined.

【0015】場合によっては、仮想的に各行(または
列)の傾斜角θだけ各行(または列)を逆方向に回転さ
せて傾斜を補正し、そして得られた各文字の画像データ
をリファレンスにしたい場合がある。その場合には行線
/列線傾斜角演算部8で各行(または列)の各外接四角
形Sqの例えば上辺の左端のPi点の座標(x1
(i),y2(i))を用いて、最小自乗法により行
(または列)に対応する直線LNを演算し、更に直線L
NのX(またはY)軸とのなす角θを演算する。傾斜角
補正部9では、画像メモリM5の1字ごとの画像データ
に対して角度θだけ逆方向に傾けて(回転させて)傾き
を補正した文字(図3(b))の画像データを演算し、
画像メモリM6に格納するようにしてもよい。
In some cases, it is desired to virtually rotate each row (or column) in the opposite direction by the tilt angle θ of each row (or column) to correct the tilt, and use the obtained image data of each character as a reference. There are cases. In that case, the row line / column line inclination angle calculation unit 8 calculates, for example, the coordinates (x1
(I), y2 (i)) is used to calculate the straight line LN corresponding to the row (or column) by the least squares method, and then the straight line LN
The angle θ between N and the X (or Y) axis is calculated. The inclination angle correction unit 9 calculates the image data of the character (FIG. 3B) whose inclination is corrected by tilting (rotating) the image data of each character in the image memory M5 in the opposite direction by the angle θ. Then
It may be stored in the image memory M6.

【0016】[0016]

【発明の効果】以上述べたように、この発明では各文字
に対応する外接四角形を求め、画像メモリに格納された
リファレンスとすべきDUTに表記された各文字の画像
(データ)をその外接四角形で打ち抜いて、標準文字デ
ータとして切り出して、画像メモリM5に格納してい
る。これによって図3(a)に示したように、各行線
(または列線)が多少傾斜していても、外接四角形を求
めることができるので、各文字データを確実に切り出す
ことができる。
As described above, according to the present invention, the circumscribed quadrangle corresponding to each character is obtained, and the image (data) of each character written in the DUT to be used as the reference stored in the image memory is circumscribed quadrangle. Punched out, cut out as standard character data, and stored in the image memory M5. As a result, as shown in FIG. 3A, even if each row line (or column line) is slightly inclined, the circumscribed quadrangle can be obtained, so that each character data can be reliably cut out.

【0017】またこの発明では必要に応じ各行線(また
は列線)の傾斜角を補正した標準文字データを得ること
もできるので、利便性が頗るよい。
Further, according to the present invention, standard character data in which the inclination angle of each row line (or column line) is corrected can be obtained if necessary, which is very convenient.

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

【図1】この発明の実施例を示すブロック図。FIG. 1 is a block diagram showing an embodiment of the present invention.

【図2】図1の画像メモリM1〜M4の各データを表示
した画像の一例を示す図。
FIG. 2 is a diagram showing an example of an image displaying each data in the image memories M1 to M4 of FIG.

【図3】(a)は図1の画像メモリM4のデータを表示
した画像の行線Laが傾いている場合の外接四角形と、
最小自乗法で求めた直線LNと、その傾斜角の一例を示
す図、(b)は(a)に対応して画像メモリM5の文字
データを−θだけ回転させて傾斜を補正した場合の画像
メモリM6の各文字データを示す図。
3A is a circumscribed quadrangle when a row line La of an image displaying data in the image memory M4 of FIG. 1 is inclined,
The figure which shows the straight line LN calculated | required by the least squares method, and an example of the inclination angle, (b) shows the image when the inclination is corrected by rotating the character data of the image memory M5 by -θ corresponding to (a). The figure which shows each character data of memory M6.

【図4】従来の文字検査装置に用いられる画像処理装置
のブロック図。
FIG. 4 is a block diagram of an image processing apparatus used in a conventional character inspection apparatus.

【図5】従来の画像処理方法を説明するための図で、A
はリファレンスとすべきDUTの表記文字の一例、Bは
AのY軸上での画素値の和L1(y)を示すグラフ、C
はAの第1行のX軸上での画素値の和L2(x)を示す
グラフ。
FIG. 5 is a diagram for explaining a conventional image processing method.
Is an example of a DUT notation character to be used as a reference, B is a graph showing the sum L1 (y) of pixel values of A on the Y axis, and C
Is a graph showing the sum L2 (x) of pixel values on the X axis of the first row of A.

【図6】従来の画像処理方法の他の例を説明するための
図で、AはリファレンスとすべきDUTの表記文字の行
線が傾斜している場合、BはAのY軸上での画素値の和
L1(y)を示すグラフ。
FIG. 6 is a diagram for explaining another example of the conventional image processing method, where A is a line in which a notation character of a DUT to be a reference is inclined, and B is a line on the Y axis of A. The graph which shows the sum L1 (y) of pixel values.

Claims (6)

【特許請求の範囲】[Claims] 【請求項1】 標準にしようとする文字が表記されてい
るDUTの表記面を撮影して得られた画素値(画素の輝
度)を格納する画像メモリ1と、 その画像メモリ1のデータをしきい値処理して2値化す
るしきい値処理手段と、 そのしきい値処理手段で得られた2値データを格納する
画像メモリ2と、 その画像メモリ2のデータに対して細線化処理(収縮処
理)を行ってノイズを除去する細線化処理手段と、 その細線化処理手段で得られたデータを格納する画像メ
モリ3と、 その画像メモリ3のデータに対して太線化処理(膨張処
理)を行って、文字の太さを元に戻す太線化処理手段
と、 その太線化処理手段で得られたデータを記憶する画像メ
モリ4と、 その画像メモリ4のデータを用いて、一字ごとに外接四
角形を求めるラベリング処理手段と、 そのラベリング処理手段で得られた一字ごとの外接四角
形の座標情報を格納する外接四角形メモリと、 その外接四角形メモリの各外接四角形で前記画像メモリ
1の画像データを打ち抜いて、1字ごとの画像データを
得る文字切り出し手段と、 その文字切り出し手段で得られた1字ごとの画像データ
を標準文字データとして格納する画像メモリ5と、を具
備することを特徴とする文字検査装置の画像処理装置。
1. An image memory 1 for storing a pixel value (luminance of a pixel) obtained by photographing an image surface of a DUT on which characters to be standardized are written, and data of the image memory 1 are stored. Threshold processing means for thresholding and binarizing, image memory 2 for storing the binary data obtained by the threshold processing means, and thinning processing for the data in the image memory 2 ( Shrinking processing) to remove noise to reduce noise, image memory 3 for storing data obtained by the thinning processing means, and thickening processing (expansion processing) for the data in the image memory 3. By using the thickening processing means for restoring the thickness of the character to the original, the image memory 4 for storing the data obtained by the thickening processing means, and the data in the image memory 4 for each character. Labeling process to find the circumscribed rectangle A column, a circumscribing quadrilateral memory that stores the coordinate information of the circumscribing quadrangle for each character obtained by the labeling processing unit, and the circumscribing quadrilateral of the circumscribing quadrangle memory punches out the image data of the image memory 1 to obtain one character. An image of a character inspection device, comprising: character cutout means for obtaining image data for each character; and image memory 5 for storing the image data for each character obtained by the character cutout means as standard character data. Processing equipment.
【請求項2】 請求項1において、前記ラベリング処理
手段は、前記外接四角形の面積を求めて、しきい値と比
較し、しきい値以下のものは一文字に対応する外接四角
形より除外することを特徴とする文字検査装置の画像処
理装置。
2. The labeling processing means according to claim 1, wherein the labeling processing means obtains an area of the circumscribing quadrangle and compares it with a threshold value, and excludes a value less than or equal to the threshold value from the circumscribing quadrangle corresponding to one character. Image processing device of character inspection device.
【請求項3】 請求項1において、 前記外接四角形メモリのデータから表記文字の各行線
(または列線)に平行な直線を求め、更にその直線の水
平線(または垂直線)となす角θを求める行線(列線)
傾斜角演算手段と、 前記画像メモリ5の一字ごとに切出した画像データに対
して、前記傾斜角演算手段で求めた傾斜角θだけ逆方向
に傾けた各文字の画像データを演算する傾斜角補正手段
と、 その傾斜角を補正した各文字の画像データを格納する画
像メモリ6と、を設けたことを特徴とする文字検査装置
の画像処理装置。
3. The straight line parallel to each row line (or column line) of a written character is obtained from the data of the circumscribed quadrilateral memory according to claim 1, and an angle θ formed between the straight line and a horizontal line (or vertical line) is obtained. Row line (column line)
The tilt angle calculating means and the tilt angle for calculating the image data of each character which is tilted in the opposite direction by the tilt angle θ obtained by the tilt angle calculating means with respect to the image data cut out for each character of the image memory 5. An image processing apparatus of a character inspection apparatus, comprising: a correction unit; and an image memory 6 for storing image data of each character whose inclination angle is corrected.
【請求項4】 標準にしようとする文字が表記されてい
るDUTの表記面を撮影して得られた画素値(画素の輝
度)を画像メモリ1に格納し、 その画像メモリ1のデータをしきい値処理して2値デー
タに変換し、画像メモリ2に格納し、 その画像メモリ2のデータを細線化処理して、画像メモ
リ3に格納し、 その画像メモリ3のデータを太線化処理して、画像メモ
リ4に格納し、 その画像メモリ4のデータより一字ごとに外接四角形を
求めて、その座標位置を外接四角形メモリに格納し、 その外接四角形メモリに格納された各外接四角形(デー
タ)で前記画像メモリ1の画像データを打ち抜いて、1
字ごとの画像データを切り出し、標準文字データとして
画像メモリ5に格納することを特徴とする文字検査装置
の画像処理方法。
4. A pixel value (pixel brightness) obtained by photographing a notation surface of a DUT on which characters to be standardized are recorded is stored in the image memory 1, and the data in the image memory 1 is stored. Threshold processing is performed to convert it to binary data, which is stored in the image memory 2. The data in the image memory 2 is thinned and stored in the image memory 3. The data in the image memory 3 is thickened. Then, the circumscribed quadrangle is stored in the image memory 4, the circumscribed quadrangle is obtained for each character from the data in the image memory 4, the coordinate position is stored in the circumscribed quadrangle memory, and each circumscribed quadrangle (data is stored in the circumscribed quadrangle memory (data ), The image data in the image memory 1 is punched out, and 1
An image processing method for a character inspection device, wherein image data for each character is cut out and stored as standard character data in the image memory 5.
【請求項5】 請求項4において、一字ごとの前記外接
四角形の面積を求めて、しきい値と比較し、しきい値以
下のものは一文字に対応する外接四角形より除外するこ
とを特徴とする文字検査装置の画像処理方法。
5. The area according to claim 4, wherein the area of the circumscribed quadrangle for each character is calculated and compared with a threshold value, and those below the threshold value are excluded from the circumscribed quadrangle corresponding to one character. Image processing method for character inspection device.
【請求項6】 請求項4において、前記外接四角形メモ
リのデータから表記文字の各行線(または列線)に平行
な直線を求め、更にその直線の水平線(または垂直線)
となす角度θを求め、 前記画像メモリ5の一字ごとに切り出した画像データに
対して、一θだけ傾けた画像データを演算して、画像メ
モリ6に格納することを特徴とする文字検査装置の画像
処理方法。
6. The straight line according to claim 4, wherein a straight line parallel to each row line (or column line) of the written characters is obtained from the data of the circumscribed quadrangle memory, and the horizontal line (or vertical line) of the straight line is obtained.
The character inspection device is characterized in that the angle θ formed by the image memory 5 is calculated, and the image data cut out for each character of the image memory 5 is calculated by inclining the image data by one θ and stored in the image memory 6. Image processing method.
JP7343290A 1995-12-28 1995-12-28 Method and device for processing image for character checking device Pending JPH09185683A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP7343290A JPH09185683A (en) 1995-12-28 1995-12-28 Method and device for processing image for character checking device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP7343290A JPH09185683A (en) 1995-12-28 1995-12-28 Method and device for processing image for character checking device

Publications (1)

Publication Number Publication Date
JPH09185683A true JPH09185683A (en) 1997-07-15

Family

ID=18360376

Family Applications (1)

Application Number Title Priority Date Filing Date
JP7343290A Pending JPH09185683A (en) 1995-12-28 1995-12-28 Method and device for processing image for character checking device

Country Status (1)

Country Link
JP (1) JPH09185683A (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2001093188A1 (en) * 2000-06-02 2001-12-06 Japan Science And Technology Corporation Method for processing document, recorded medium on which document processing program is recorded and document processor
JP2007122500A (en) * 2005-10-28 2007-05-17 Ricoh Co Ltd Character recognition device, character recognition method and character data
JP2010101828A (en) * 2008-10-27 2010-05-06 Chuo Electronics Co Ltd Time measuring device and time measuring method

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2001093188A1 (en) * 2000-06-02 2001-12-06 Japan Science And Technology Corporation Method for processing document, recorded medium on which document processing program is recorded and document processor
US7221795B2 (en) 2000-06-02 2007-05-22 Japan Science And Technology Corporation Document processing method, recording medium having recorded thereon document processing program, document processing program, document processing apparatus, and character-input document
JP2007122500A (en) * 2005-10-28 2007-05-17 Ricoh Co Ltd Character recognition device, character recognition method and character data
JP2010101828A (en) * 2008-10-27 2010-05-06 Chuo Electronics Co Ltd Time measuring device and time measuring method

Similar Documents

Publication Publication Date Title
US5528290A (en) Device for transcribing images on a board using a camera based board scanner
US7342572B2 (en) System and method for transforming an ordinary computer monitor into a touch screen
JP2986383B2 (en) Method and apparatus for correcting skew for line scan images
US6016148A (en) Automated mapping of facial images to animation wireframes topologies
JP3515199B2 (en) Defect inspection equipment
CN111553870B (en) Image processing method based on distributed system
CN112419207A (en) Image correction method, device and system
JPH09185683A (en) Method and device for processing image for character checking device
JP3698867B2 (en) Circular pattern determination method, apparatus and recording medium
JP2846486B2 (en) Image input device
JP2775404B2 (en) Automatic target identification device
JP2993007B2 (en) Image area identification device
JP2000187705A (en) Document reader, document reading method and storage medium
JPS63113779A (en) Peak extraction method
JPH06259531A (en) Method for correcting inclination of dot pattern
JP3038714B2 (en) Document skew correction device
JPH0224323B2 (en)
US11900643B2 (en) Object detection method and object detection system
JP2003317107A (en) Method and device for ruled-line detection
JP3438974B2 (en) Method for storing and reading template images
JP2966448B2 (en) Image processing device
JP2856465B2 (en) Image generation processing method
JPH07190888A (en) Inspection apparatus of liquid-crystal display panel
CN115830134A (en) Camera calibration method and device, electronic equipment and medium
JPH064670A (en) Pattern matching device for gradation picture

Legal Events

Date Code Title Description
A02 Decision of refusal

Free format text: JAPANESE INTERMEDIATE CODE: A02

Effective date: 20030408