JPH05280958A - Defect inspection device - Google Patents
Defect inspection deviceInfo
- Publication number
- JPH05280958A JPH05280958A JP4103542A JP10354292A JPH05280958A JP H05280958 A JPH05280958 A JP H05280958A JP 4103542 A JP4103542 A JP 4103542A JP 10354292 A JP10354292 A JP 10354292A JP H05280958 A JPH05280958 A JP H05280958A
- Authority
- JP
- Japan
- Prior art keywords
- defect
- image
- pixel
- area
- threshold value
- 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
- 230000007547 defect Effects 0.000 title claims abstract description 112
- 238000007689 inspection Methods 0.000 title claims description 26
- 230000008602 contraction Effects 0.000 claims description 12
- 238000000605 extraction Methods 0.000 claims description 8
- 238000001514 detection method Methods 0.000 claims description 5
- 238000001914 filtration Methods 0.000 claims description 2
- 238000000034 method Methods 0.000 description 43
- 238000010586 diagram Methods 0.000 description 23
- 230000002950 deficient Effects 0.000 description 17
- 238000013528 artificial neural network Methods 0.000 description 5
- 238000005429 filling process Methods 0.000 description 5
- 238000006243 chemical reaction Methods 0.000 description 3
- 238000003066 decision tree Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000004804 winding Methods 0.000 description 3
- 229910000831 Steel Inorganic materials 0.000 description 2
- 230000015572 biosynthetic process Effects 0.000 description 2
- 239000002985 plastic film Substances 0.000 description 2
- 229920006255 plastic film Polymers 0.000 description 2
- 239000010959 steel Substances 0.000 description 2
- 230000007423 decrease Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 238000012887 quadratic function Methods 0.000 description 1
Landscapes
- Length Measuring Devices By Optical Means (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
Description
【０００１】[0001]
【産業上の利用分野】本発明は、検査対象を走査して得
られる画像信号を用いて欠陥の有無を自動で検査し、こ
の欠陥の種類等を高精度に判別し得るように欠陥の領域
を抽出する欠陥検査装置に関するものである。BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention automatically detects the presence or absence of a defect by using an image signal obtained by scanning an object to be inspected, and determines the defect area so that the type of the defect can be determined with high accuracy. The present invention relates to a defect inspection device for extracting
【０００２】[0002]
【従来の技術】鋼板やプラスチックフィルムや紙などの
表面を光学的に走査して、その表面のキズあるいは内部
の欠陥等を検出する欠陥検査装置が公知である。ここに
従来は検査対象を走査して得た画像信号を予め設定され
たしきい値と比較し、画像信号がこのしきい値を超える
時又は小さい時に欠陥であると判断していた（例えば特
願平２−２４５４１３号等参照）。2. Description of the Related Art A defect inspection apparatus is known which optically scans the surface of a steel plate, plastic film, paper or the like to detect scratches on the surface or internal defects. Here, conventionally, an image signal obtained by scanning an inspection object is compared with a preset threshold value, and when the image signal exceeds or is smaller than this threshold value, it is determined to be a defect (for example, See Japanese Patent Application No. 2-245413, etc.).
【０００３】また検出した欠陥の形状や種類などを判別
できるようにした装置も提案されている。例えば、検査
対象から得られた画像信号を一定のレベルで波高弁別
し、これによって得られた欠陥領域から特徴量を求め、
欠陥の種類を判別するものである。There has also been proposed an apparatus capable of discriminating the shape and type of the detected defect. For example, the image signal obtained from the inspection object is wave height discriminated at a constant level, and the feature amount is obtained from the defect area obtained by this,
The type of defect is determined.
【０００４】[0004]
【従来技術の問題点】しかしこの場合には、例えば欠陥
領域内の濃度レベルが低い部分では欠陥が十分に抽出で
きず、よって欠陥の特徴量を精度よく求めることが非常
に困難になる。However, in this case, the defects cannot be sufficiently extracted, for example, in a portion having a low density level in the defect region, and thus it becomes very difficult to accurately obtain the feature amount of the defect.
【０００５】[0005]
【発明の目的】従って本発明は、自動で欠陥を検出して
その形状や種類等を高精度に判別できるようにこの欠陥
を含む領域を抽出するようにした画像欠陥検出装置を提
供することを目的とする。SUMMARY OF THE INVENTION It is, therefore, an object of the present invention to provide an image defect detecting apparatus for automatically detecting a defect and extracting a region including this defect so that the shape, type, etc. of the defect can be accurately determined. To aim.
【０００６】[0006]
【発明の構成】本発明によればこの目的は、検査対象を
走査して得た画像信号を微分フィルタを含む空間フィル
タリング処理した微分画像を用いて検査対象の欠陥を含
む領域を抽出する欠陥検査装置において、前記微分画像
に基づいて前記欠陥のアドレスを求める欠陥アドレス検
出手段と、前記微分画像の全体の濃度分布から決まる第
１の閾値（ＴＨ１）と注目画素の近傍領域の濃度平均値
から得られる第２の閾値（ＴＨ２）とを用いて各注目画
素ごとに得られる第３の閾値（ＴＨ３）を用いて各注目
画素を２値化する２値化処理手段と、前記２値化処理手
段で得られた領域のうち前記欠陥のアドレスを含む領域
の画そを全て欠陥を示す値に塗り潰す欠陥領域抽出手段
とを備えることを特徴とする欠陥検査装置により達成さ
れる。According to the present invention, an object of the present invention is to perform a defect inspection for extracting an area including a defect of an inspection object by using a differential image obtained by subjecting an image signal obtained by scanning the inspection object to spatial filtering processing including a differential filter. In the apparatus, the defect address detection means for obtaining the address of the defect based on the differential image, the first threshold value (TH1) determined from the density distribution of the entire differential image, and the density average value of the neighborhood area of the target pixel are obtained. And a binarization processing means for binarizing each pixel of interest using a third threshold value (TH3) obtained for each pixel of interest using the second threshold value (TH2) The defect inspection apparatus is characterized by further comprising defect area extracting means for filling all of the image of the area including the address of the defect to a value indicating the defect in the area obtained in (1).
【０００７】ここに２値化処理に用いる第３の閾値ＴＨ
３は、微分画像全体の濃度分布から決まる第１の閾値Ｔ
Ｈ１と、注目画素の近傍領域の平均濃度から決まる第２
の閾値ＴＨ２とを用いて、例えばＴＨ３＝ＴＨ１−ｋ
（ＴＨ１−ＴＨ２）により求めることができる。なお２
値化処理の直後に走査方向または副走査方向に収縮処理
を行ったり、抽出した欠陥領域に対して膨張処理、穴埋
め処理、収縮処理などを施して形状補正を行ってもよ
い。Here, a third threshold value TH used in the binarization process
3 is a first threshold T determined from the density distribution of the entire differential image
H1 and the second density determined from the average density of the region near the pixel of interest
, And TH3 = TH1-k
It can be determined by (TH1-TH2). 2
Immediately after the binarization process, contraction processing may be performed in the scanning direction or the sub-scanning direction, or expansion processing, hole filling processing, contraction processing, or the like may be performed on the extracted defect area to perform shape correction.
【０００８】[0008]
【実施例】図１は本発明の第１の実施例のブロック図、
図２はその一部の詳細ブロック図、図３は閾値ＴＨ１の
決定法の一例の説明図、図４は閾値ＴＨ２の決定法の一
例の説明図、図５は閾値ＴＨ３の説明図である。FIG. 1 is a block diagram of a first embodiment of the present invention,
2 is a detailed block diagram of a part thereof, FIG. 3 is an explanatory diagram of an example of a method of determining the threshold TH1, FIG. 4 is an explanatory diagram of an example of a method of determining the threshold TH2, and FIG. 5 is an explanatory diagram of the threshold TH3.
【０００９】図１において、符号１０は鋼板、紙、プラ
スチックフィルムなどの検査対象であり、この検査対象
１０は供給ロール１２から巻取りロール１４に送られ
る。この巻取りロール１４は巻取りモータ１６により駆
動される。この検査対象１０の送り中にフライングスポ
ット方式による画像検出手段１８によって表面の画像が
読取られる。In FIG. 1, reference numeral 10 is an inspection object such as a steel plate, paper, or a plastic film, and this inspection object 10 is sent from a supply roll 12 to a winding roll 14. The winding roll 14 is driven by a winding motor 16. While the inspection object 10 is being fed, the image on the surface is read by the image detecting means 18 of the flying spot type.
【００１０】この画像検出手段１８は、レーザー光源２
０から射出されるレーザ光からなる走査ビームＬを、モ
ータ２２により回転される回転ミラー（ポリゴナルミラ
ー）２４によって検査対象１０の幅方向に一定の速度で
走査（主走査）する一方、検査対象１０の表面による反
射光を受光ロッド２６によって一対の受光器２８（２８
ａ、２８ｂ）に導いて受光するものである。すなわち受
光ロッド２６は走査ビームＬの主走査ライン３０に近接
してこれに平行に配設され、反射光を受光すると受光ロ
ッド２６の内面で全反射させてその両端に導き、フォト
マルチプライヤ（光電子増倍管）などの受光器２８によ
り受光量が検出される。The image detecting means 18 is a laser light source 2
The scanning beam L composed of the laser light emitted from 0 is scanned (main scanning) in the width direction of the inspection object 10 at a constant speed by the rotating mirror (polygonal mirror) 24 rotated by the motor 22 while the inspection object is inspected. The light reflected by the surface of 10 is received by a pair of light receivers 28 (28).
a, 28b) to receive light. That is, the light-receiving rod 26 is disposed close to and parallel to the main scanning line 30 of the scanning beam L, and when the reflected light is received, the light-receiving rod 26 is totally reflected by the inner surface of the light-receiving rod 26 and guided to both ends thereof, and the photomultiplier (photoelectron The amount of light received is detected by a light receiver 28 such as a multiplier.
【００１１】各受光器２８が出力する画像信号はプリア
ンプ、メインアンプで増幅され、また波形整形されてア
ナログ画像信号ａ1 、ａ2 となる。各信号ａ1 、ａ2 に
は、連続する異なる主走査ライン３０に対応する信号
が、時間軸方向に一定時間毎に現れている。The image signal output from each light receiver 28 is amplified by a preamplifier and a main amplifier, and waveform-shaped into analog image signals a 1 and a 2 . In each of the signals a 1 and a 2 , signals corresponding to successive different main scanning lines 30 appear at regular time intervals in the time axis direction.
【００１２】各信号ａ1 、ａ2 は走査ビームＬの主走査
ライン３０上の走査位置から遠くなるとレベルが低下
し、反対に走査位置に近くなるとレベルが上昇するよう
に変化する。そこでこの実施例では、両信号ａ1 、ａ2
は加算手段３２で加算され、主走査ライン３０上の走査
位置の変化による影響が除去されて信号ａ3 となる。こ
の信号ａ3 は、走査ビームＬの検査対象１０表面への入
射角度の変化、検査対象１０表面のむらや地合い、導光
ロッド２６内での減衰などのために、その出力レベルが
大きく湾曲していることがある。The levels of the signals a 1 and a 2 change such that the level thereof decreases when the scanning beam L is far from the scanning position on the main scanning line 30, and conversely the level increases when the scanning beam L is close to the scanning position. Therefore, in this embodiment, both signals a 1 and a 2
Are added by the adding means 32, and the influence due to the change of the scanning position on the main scanning line 30 is removed to become the signal a 3 . The output level of the signal a 3 is greatly curved due to a change in the incident angle of the scanning beam L on the surface of the inspection object 10, unevenness or texture of the surface of the inspection object 10, attenuation in the light guide rod 26, and the like. Sometimes
【００１３】この加算された信号ａ3 はＡ／Ｄ変換手段
３４においてデジタル信号ａ4 に変換される。例えば２
５６階調の濃度信号ａ4 に変換される。そしてラインメ
モリ（図示せず）にメモリされる。The added signal a 3 is converted into a digital signal a 4 in the A / D conversion means 34. Eg 2
It is converted into a density signal a 4 of 56 gradations. Then, it is stored in a line memory (not shown).
【００１４】Ａ／Ｄ変換された信号ａ4 は濃度変換手段
３６で濃度変換され、微分フィルタ３８に入力されて空
間フィルタリング処理される。濃度変換手段３６は所定
の変換テーブルにより信号ａ4 を例えば２５６階調の範
囲内で適切な濃度分布に変換する。微分フィルタ３８は
注目画素を中心とする例えば３×３画素領域に所定の重
み係数をもった空間フィルタを重ね、対応する画素同志
の積を求め、これらの総和を出力とするものである。The A / D-converted signal a 4 is density-converted by the density converting means 36, is input to the differential filter 38, and is spatially filtered. The density converting means 36 converts the signal a 4 into an appropriate density distribution within a range of 256 gradations by using a predetermined conversion table. The differential filter 38 superimposes a spatial filter having a predetermined weighting coefficient on, for example, a 3 × 3 pixel area centered on the pixel of interest, calculates the product of the corresponding pixels, and outputs the sum of these.
【００１５】これら濃度変換手段３６および微分フィル
タ３８については、特願平２−２４５４１３号等に詳細
に説明されているから、ここでは繰り返さない。なお濃
度変換手段３６を省いて信号ａ4 を直接微分フィルタ３
８に入力してもよい。微分フィルタ３８で空間フィルタ
リング処理をした信号ａ5 は、信号ａ4 にあった低周波
成分が除去されて欠陥の輪郭が強調されている。そして
この信号ａ5 は微分画像Ａ0 を形成するものである。The density converting means 36 and the differential filter 38 have been described in detail in Japanese Patent Application No. 2-245413, and will not be repeated here. The density conversion means 36 is omitted and the signal a 4 is directly differentiated by the differential filter 3
You may enter in 8. The signal a 5 spatially filtered by the differential filter 38 has the low-frequency component in the signal a 4 removed to enhance the contour of the defect. This signal a 5 forms the differential image A 0 .
【００１６】４０は欠陥アドレス検出手段であり、例え
ばこの信号ａ5 を閾値ＴＨ０と比較し、ａ5 ＞ＴＨ０ま
たはａ5 ＜ＴＨ０の時に欠陥と判断する。そしてこの欠
陥を検出すると、この欠陥のアドレスＡｄが求められメ
モリされる。このアドレスＡｄは速度検出器（パルスジ
ェネレータ）４２、モータ２２の回転角から求める。な
お、信号ａ5 を欠陥と判断する方法は、ここで述べた方
法に限定されないものとする。Reference numeral 40 denotes a defective address detecting means, which compares this signal a 5 with a threshold value TH0 and judges that it is defective when a 5 > TH0 or a 5 <TH0. When this defect is detected, the address Ad of this defect is obtained and stored. This address Ad is obtained from the rotation angle of the speed detector (pulse generator) 42 and the motor 22. The method of determining the signal a 5 as a defect is not limited to the method described here.
【００１７】以上のようにして欠陥のアドレスＡｄを求
めるが、この過程は単に欠陥の有無を判定すればよいの
でここに用いる閾値ＴＨ０は後記の閾値ＴＨ３などに比
べて高く、すなわち欠陥の信号レベルに接近したレベル
に設定され、ノイズによる誤検出を防止している。The defect address Ad is obtained as described above. In this process, the presence or absence of a defect may be simply determined. Therefore, the threshold value TH0 used here is higher than the threshold value TH3 described later, that is, the defect signal level. It is set to a level close to to prevent erroneous detection due to noise.
【００１８】[0018]
【２値化処理手段】一方微分フィルタ３８の出力信号ａ
5 が形成する画像すなわち微分画像Ａ0 は２値化処理手
段４２に入力される。この手段４２は図２に示すように
第１の閾値ＴＨ１と第２の閾値ＴＨ２とから、第３の閾
値ＴＨ３を求め、この第３の閾値ＴＨ３を用いて信号ａ
5 を２値化するものである。[Binarization processing means] Output signal a of the differential filter 38
The image formed by 5, that is, the differential image A 0 is input to the binarization processing means 42. As shown in FIG. 2, this means 42 obtains a third threshold TH3 from the first threshold TH1 and the second threshold TH2, and uses the third threshold TH3 to obtain the signal a.
Binarize 5
【００１９】第１の閾値ＴＨ１はこの微分画像の全体の
濃度分布から決めるものであり、例えば図３に示す濃度
のヒストグラムを用いて決めることができる。この図３
で濃度の分布幅がｘであれば、これを用いて度数の閾値
Ｎ（ＴＨ１）を Ｎ（ＴＨ１）＝（全度数）／ｆ（ｘ） により求める。The first threshold value TH1 is determined from the density distribution of the entire differential image, and can be determined using, for example, the density histogram shown in FIG. This Figure 3
If the distribution width of density is x, the frequency threshold N (TH1) is calculated by using N (TH1) = (total frequency) / f (x).
【００２０】ここにｆ（ｘ）は分布幅ｘの関数であり、
ｘの２次関数などで決める。そして度数の閾値Ｎ（ＴＨ
１）となるヒストグラムの濃度ＴＨ１を第１の閾値とす
れば、これは微分画像全体の濃度分布が反映された閾値
となる。なお、このＮ（ＴＨ１）の値は予め実験的に求
めておき、欠陥検出処理の過程では定数として扱っても
よい。Where f (x) is a function of the distribution width x,
It is determined by a quadratic function of x or the like. The frequency threshold N (TH
If the density TH1 of the histogram 1) is used as the first threshold value, this is a threshold value that reflects the density distribution of the entire differential image. The value of N (TH1) may be obtained experimentally in advance and treated as a constant in the process of defect detection processing.
【００２１】第２の閾値ＴＨ２は、注目画素の近傍領域
の濃度平均値から得られるものである。例えば図４に示
すように、注目画素を中心とする３×３の画素の濃度ｄ
の総和Σｄｎを求め、これを画素数９で割ることにより
求めることができる。The second threshold value TH2 is obtained from the density average value of the area near the target pixel. For example, as shown in FIG. 4, the density d of a 3 × 3 pixel centered on the pixel of interest
Can be obtained by obtaining the total sum Σdn of the above and dividing this by the number of pixels 9.
【００２２】第３の閾値ＴＨ３はこれら第１、第２の閾
値ＴＨ１およびＴＨ２を用いて決める。例えばｋを定数
として ＴＨ３＝ＴＨ１−ｋ（ＴＨ１−ＴＨ２） により決める。この定数ｋは地合いの程度などを考慮し
て予め実験的に決めておく。The third threshold TH3 is determined by using these first and second thresholds TH1 and TH2. For example, with k as a constant, TH3 = TH1-k (TH1-TH2) is determined. This constant k is experimentally determined in advance in consideration of the degree of texture.
【００２３】２値化処理手段４２では、この微分画像Ａ
0 の各画素ごとに決まる第３の閾値ＴＨ３を微分画像Ａ
0 の各画素の濃度と比較することにより（比較手段４
４）、２値画像Ａ1 を求める。この２値画像Ａ1 は、微
分画像Ａ0 （信号ａ5 ）の画素濃度がＴＨ３以上の領域
が例えば“１”に、それ以外の領域が例えば“０”にな
った画像であり、本当の欠陥の領域だけでなくノイズに
より欠陥と判定された領域も含むものである。In the binarization processing means 42, this differential image A
The third threshold TH3 determined for each pixel of 0 is set to the differential image A
By comparing with the density of each pixel of 0 (comparing means 4
4) Obtain the binary image A 1 . This binary image A 1 is an image in which the area of the differential image A 0 (signal a 5 ) in which the pixel density is TH3 or more is, for example, “1”, and the other area is, for example, “0”. It includes not only the defective area but also the area determined to be defective due to noise.
【００２４】この２値化処理手段４２の作用を図５によ
ってさらに説明する。図５の（Ａ）は全体の閾値ＴＨ１
に対し、注目画素近傍の濃度平均値ＴＨ２を重ねて表し
たものであり、両者の差（ＴＨ１−ＴＨ２）を示す。図
５の（Ｂ）はＴＨ１からこの差に定数ｋを掛けた曲線ｋ
（ＴＨ１−ＴＨ２）を減算して得たＴＨ３を示す。図５
の（Ｃ）は微分画像Ａ0 とこの第３の閾値ＴＨ３との関
係を示すものである。The operation of the binarization processing means 42 will be further described with reference to FIG. FIG. 5A shows the overall threshold TH1.
On the other hand, the density average value TH2 in the vicinity of the pixel of interest is shown in an overlapping manner, and the difference between them (TH1−TH2) is shown. FIG. 5B shows a curve k obtained by multiplying this difference from TH1 by a constant k.
The TH3 obtained by subtracting (TH1-TH2) is shown. Figure 5
(C) shows the relationship between the differential image A 0 and this third threshold value TH3.
【００２５】この（Ｃ）から明らかなように、微分画像
Ａ0 を２値化する第３の閾値ＴＨ３は、画像濃度の大局
的な変化に追従して変動し、微分画像Ａ0 の微少な濃度
変化も取り出すことができる。なおここに用いる定数ｋ
は、０に近いほどＴＨ１の影響が大きく、１に近いほど
ＴＨ２の影響が大きくなる。従って地合いの汚い工程で
はｋを０に近く、地合いのきれいな工程ではｋを１に近
く設定するなどのように、工程・対象部材に応じてこの
２値化処理を変えることが可能である。As is apparent from (C), the third threshold value TH3 for binarizing the differential image A 0 changes in accordance with the global change in the image density, and the differential image A 0 has a small amount. The change in concentration can also be taken out. The constant k used here
Is closer to 0, the effect of TH1 is greater, and the closer to 1, the effect of TH2 is greater. Therefore, it is possible to change this binarization processing depending on the process and target member, such as setting k close to 0 in a process with a dirty formation and k close to 1 in a process with a beautiful formation.
【００２６】[0026]
【垂直収縮処理】図２において４６は垂直収縮処理手段
である。この手段４６は２値画像Ａ1 を垂直方向、すな
わち副走査方向（検査対象１０の送り方向）に収縮し、
縦方向の線状欠陥の形状を損なうことなく欠陥部分と背
景ノイズとの連結を除去するものである。[Vertical Shrinkage Processing] In FIG. 2, reference numeral 46 is a vertical shrinkage processing means. This means 46 contracts the binary image A 1 in the vertical direction, that is, the sub-scanning direction (the feeding direction of the inspection object 10),
The connection between the defective portion and the background noise is removed without impairing the shape of the linear defect in the vertical direction.
【００２７】図６はこの手段４６のアルゴリズムを説明
する図、図７はその処理の前後の画像を示す図である。
この処理では、処理対象画素Ｘ0 の上下２近傍画素Ｘ
1 、Ｘ2 の値を用いて、処理後の画素Ｘ0 ′を、Ｘ0 、
Ｘ1 、Ｘ2 のいずれかが０の時には０に、全てが１の時
に１にする。ここで“１”は欠陥がある画素を、“０”
は背景となる画素とする。この処理を全ての画素に対し
てラスタ走査順に行う。この結果図７に示すように縦方
向に連続する領域のみが残り、横方向にのびる領域や点
状のノイズが消去される。FIG. 6 is a diagram for explaining the algorithm of this means 46, and FIG. 7 is a diagram showing images before and after the processing.
In this processing, the upper and lower two neighboring pixels X of the processing target pixel X 0 are
Using the values of 1 and X 2 , the processed pixel X 0 ′ is converted into X 0 ,
Set to 0 when either X 1 or X 2 is 0, and set to 1 when all are 1. Here, "1" indicates a defective pixel and "0"
Is a background pixel. This process is performed for all pixels in raster scan order. As a result, as shown in FIG. 7, only the vertically continuous region remains, and the horizontally extended region and dot noise are erased.
【００２８】なおこの垂直収縮処理は、前記２値化処理
手段のｋの値が０に近い場合や欠陥が縦方向だけでなく
横方向にも多く発生する場合には省いてもよい。また以
上の説明では縦方向に連続する欠陥を残すものとしてい
るが、横方向（主走査方向）に連続する欠陥を残す場合
には図６の上下２近傍Ｘ1 、Ｘ2 に代えて、左右２近傍
の画素を用いて同様な処理を行えばよい。The vertical contraction processing may be omitted when the value of k of the binarization processing means is close to 0 or when many defects occur not only in the vertical direction but also in the horizontal direction. Further, in the above description, it is assumed that defects that are continuous in the vertical direction are left, but when defects that are continuous in the horizontal direction (main scanning direction) are left, the left and right sides of the upper and lower two neighborhoods X 1 and X 2 in FIG. Similar processing may be performed using pixels in the vicinity of two.
【００２９】[0029]
【欠陥領域抽出手段】図１、２において４８は欠陥領域
抽出手段である。この手段４８は、２値化処理で得られ
た２値画像Ａ1 から、前記欠陥アドレス検出手段４０で
得た欠陥のアドレスＡｄが含まれる連続領域を抽出す
る。そして２値画像Ａ1 のうち、このアドレスＡｄが含
まれていない領域は欠陥でないとして除去するものであ
る。[Defective Area Extracting Means] In FIGS. 1 and 2, reference numeral 48 is a defective area extracting means. This means 48 extracts a continuous area containing the address Ad of the defect obtained by the defect address detecting means 40 from the binary image A 1 obtained by the binarization process. Then, the area of the binary image A 1 which does not include the address Ad is removed because it is not a defect.
【００３０】図８はこの処理のアルゴリズム説明図、図
９はこの処理の前後の画像を示す図である。この処理は
次のように行われる。まず欠陥アドレス検出手段４０に
より欠陥アドレスＡｄで指定された画素、すなわち欠陥
信号が発生した画素（欠陥画素）を着目点とし、 この着目点の座標を記録し、対応する画素にラベルを
付ける。 図８に示した探索順に２値画像Ａ1 内の欠陥画素でか
つまだラベルが付いていない画素を見つける。 見つかったら着目点をその画素に移し、へ戻る。 見つからなかったら着目点を一つ前の着目点に戻し、
へ戻る。FIG. 8 is a diagram for explaining the algorithm of this process, and FIG. 9 is a diagram showing images before and after this process. This process is performed as follows. First, the pixel designated by the defect address Ad by the defect address detecting means 40, that is, the pixel (defective pixel) in which the defect signal is generated is set as a point of interest, the coordinates of this point of interest are recorded, and the corresponding pixel is labeled. A defective pixel in the binary image A 1 which is not yet labeled is found in the search order shown in FIG. When found, the point of interest is moved to the pixel and the process returns to. If not found, return the point of interest to the previous point of interest,
Return to.
【００３１】〜の手順を、領域中にラベルが付いて
いない画素が見つからなくなるまで繰り返す。この処理
を、すでにラベル付けされた領域以外の領域のアドレス
（Ａｄ）による欠陥信号全てについて行う。この結果２
値画像Ａ1 の複数の領域のうち、欠陥信号を含む全ての
領域を抽出し、他の領域を除去することができる（図
９）。The procedure from to is repeated until no unlabeled pixels are found in the area. This process is performed for all defective signals due to the address (Ad) of the area other than the already labeled area. This result 2
Of the plurality of areas of the value image A 1 , all the areas including the defect signal can be extracted and the other areas can be removed (FIG. 9).
【００３２】[0032]
【形状補正】次にこのように抽出した欠陥を含む領域に
対して、形状補正手段５０（図２）による形状補正処理
が施される。この処理は、抽出した領域を後記するよう
に欠陥形状や種類、等級などを認識する際に、処理をし
易くし、認識精度を向上させるために行うものであり、
欠陥領域抽出手段４８の出力形式によってはこの手段５
０による処理を省いてもよい。[Shape Correction] Next, shape correction processing is performed by the shape correction means 50 (FIG. 2) on the region including the defect thus extracted. This process is performed in order to facilitate the process and improve the recognition accuracy when recognizing the defect shape, type, grade, etc. as described later on the extracted area.
Depending on the output format of the defect area extraction means 48, this means 5
The processing by 0 may be omitted.
【００３３】この形状補正手段５０は、図２に示すよう
に膨張処理５２と、穴埋め処理５４と、収縮処理５６と
を含む。図１０、１１、１２はそれぞれ膨張処理５２、
穴埋め処理５４、収縮処理５６のアルゴリズムの説明
図、図１３はこれら各処理による画像の変化を示す図で
ある。The shape correction means 50 includes an expansion process 52, a filling process 54, and a contraction process 56 as shown in FIG. 10, 11 and 12 show expansion processing 52,
FIGS. 13A and 13B are explanatory diagrams of algorithms of the hole filling process 54 and the contraction process 56, and FIGS.
【００３４】[0034]
【膨張処理】膨張処理５２は、断片化された近接する複
数の欠陥領域を連続させたり、途切れた輪郭を連結させ
るものである。すなわち図１０に示すように、処理対象
画素Ｘ0 を中心とする例えば８近傍の画素Ｘ1 〜Ｘ8 と
の関係から、処理後の画素Ｘ0 ′を、Ｘ0 〜Ｘ8 のいず
れかが１の時に１に、Ｘ1 〜Ｘ8 の全てが０の時に０に
変換する。[Expansion process] The expansion process 52 is for continuing a plurality of fragmented adjacent defect regions and for connecting discontinuous contours. That is, as shown in FIG. 10, the relationship between the target pixel X 0 pixel X 1 to X 8 for example of eight neighbors around the, after processing the pixels X 0 ', one of X 0 to X 8 is When it is 1, it is converted to 1, and when all of X 1 to X 8 are 0, it is converted to 0.
【００３５】なおここで“１”は欠陥がある画素を、
“０”は背景となる画素とする。この処理を各画素ごと
にラスタ走査順に繰り返す。この処理は連続して２回繰
り返すのが望ましい。この結果図１３（Ａ）に示す処理
前の画像を（Ｂ）に示すように１つの連続した領域にま
とめることができる。Here, "1" indicates a defective pixel,
“0” is a background pixel. This process is repeated for each pixel in raster scan order. It is desirable to repeat this process twice consecutively. As a result, the unprocessed image shown in FIG. 13A can be combined into one continuous area as shown in FIG. 13B.
【００３６】[0036]
【穴埋め処理】穴埋め処理５４は、図１３の（Ｂ）に示
すように画像に穴がある場合に、この穴を埋めるもので
ある。すなわち図１１に示すように、背景と同じ“０”
の画素で形成され画像の縁に連続しない領域を穴の領域
と判定し、この穴の領域を欠陥と同じ“１”の画素に変
えるものである。この処理により図１３の（Ｂ）に示す
穴は（Ｃ）に示すように埋められる。[Blank Filling Process] The hole filling process 54 fills a hole in an image when the hole is present in the image as shown in FIG. That is, as shown in FIG. 11, the same "0" as the background
The area formed by the pixels of # 1 and # 2 that is not continuous with the edge of the image is determined as a hole area, and this hole area is changed to the pixel of "1" which is the same as the defect. By this processing, the hole shown in FIG. 13B is filled in as shown in FIG.
【００３７】[0037]
【収縮処理】収縮処理５６は膨張処理５２によって膨張
し肥大化した欠陥の画像を元の大きさに戻すものであ
る。この処理のアルゴリズムは、図１２に示すように処
理対象画素Ｘ0 を中心とする例えば８近傍の画素Ｘ1 〜
Ｘ8 との関係から、処理後の画素Ｘ0 ′を、Ｘ0 〜Ｘ8
のいずれかが０の時に０とし、Ｘ0 〜Ｘ8 の全てが１の
時に１とする変換を行うものである。[Shrinking Process] The shrinking process 56 is to restore the image of the defect expanded and enlarged by the expansion process 52 to the original size. The algorithm of the processing is centered on the target pixel X 0 as shown in FIG. 12, for example eight neighboring pixels X 1 ~
From the relationship with X 8 , the processed pixel X 0 ′ is converted into X 0 to X 8
Is converted to 0 when any one of them is 0, and is converted to 1 when all of X 0 to X 8 are 1.
【００３８】ここに“１”は欠陥のある画素を、“０”
は背景となる画素を示す。この処理を各画素ごとにラス
タ走査順に繰り返す。この処理は前記膨張処理５２の繰
り返し回数と同回数すなわち２回繰り返す。この結果図
１３の（Ｄ）のように、形状補正された欠陥を含む領域
（欠陥領域）だけが抽出された画像（欠陥画像Ａ2 ）が
出力される。Here, "1" indicates a defective pixel and "0".
Indicates a background pixel. This process is repeated for each pixel in raster scan order. This process is repeated the same number of times as the expansion process 52, that is, twice. As a result, as shown in (D) of FIG. 13, an image (defect image A 2 ) in which only an area (defect area) including a shape-corrected defect is extracted is output.
【００３９】以上のように欠陥領域だけが抽出された画
像Ａ2 は、画像計測処理手段５８に入力され、領域分割
処理６０によって欠陥ごとに領域分割され、さらに特徴
量計測処理６２によって分割されたそれぞれの欠陥の特
徴量が求められる。The image A 2 in which only the defective area is extracted as described above is input to the image measuring processing means 58, divided into areas by the area dividing processing 60, and further divided by the characteristic amount measuring processing 62. The feature amount of each defect is obtained.
【００４０】領域分割処理６０は、例えば欠陥画像Ａ2
のｘ軸方向およびｙ軸方向へ投影された周辺分布情報を
用いて欠陥領域の分布を調べることにより、複数の欠陥
領域を分割する。The area division processing 60 is performed by, for example, the defect image A 2
A plurality of defect areas are divided by examining the distribution of the defect areas by using the peripheral distribution information projected in the x-axis direction and the y-axis direction.
【００４１】特徴量計測処理６２は、分割された欠陥領
域ごとの面積、長さ、幅、方向、濃度平均値、極性（正
または負）等の特徴量を求める。ここに濃度平均値は欠
陥領域をマスク画像としてこれに対応する原画像の領域
を抜き出し、その濃度を求めることにより算出する。The characteristic amount measuring process 62 obtains the characteristic amount such as area, length, width, direction, density average value, polarity (positive or negative) for each of the divided defect areas. The density average value is calculated by extracting the area of the original image corresponding to the defective area as a mask image and calculating the density.
【００４２】[0042]
【画像認識処理】このように欠陥の特徴量が求められる
と、次に画像認識処理手段６４においてこの欠陥の形状
を求め、その欠陥の等級を判別して最終的に欠陥に対す
る総合判定を下す。[Image recognition processing] When the feature amount of the defect is obtained in this way, the shape of the defect is then obtained in the image recognition processing means 64, the grade of the defect is discriminated, and finally a comprehensive judgment is made for the defect.
【００４３】欠陥の形状は形状認識処理６６により行わ
れる。この処理６６は、例えばニューラルネットワーク
による認識法により欠陥形状の大分類を行い、さらに決
定木による認識法により欠陥形状ごとに細分類を行うよ
うに、２段階に処理する。The shape of the defect is determined by the shape recognition processing 66. This process 66 is performed in two steps such that the defect shape is roughly classified by a recognition method using a neural network, and the defect shape is finely classified by the recognition method using a decision tree.
【００４４】ここにニューラルネットワークは、例えば
「面積」と「幅／長さ」の２つの特徴量を入力とし、２
次元特徴空間上で３つの識別平面により点・面・線の３
つのカテゴリーの分類を行う。従ってこの場合のニュー
ラルネットワークは、入力層、中間層、出力層の３層構
造として、入力層のユニット数を特徴量すなわち特徴空
間の次元数２とし、中間層のユニット数を識別平面の数
３とし、出力層のユニット数を点・線・面のカテゴリー
数３とすることができる。Here, the neural network receives two feature quantities, for example, "area" and "width / length", as input.
3 points, 3 faces, and 3 lines by 3 identification planes in the dimensional feature space
Classify into one category. Therefore, the neural network in this case has a three-layer structure of an input layer, an intermediate layer, and an output layer, the number of units in the input layer is the feature quantity, that is, the dimension number of the feature space is 2, and the number of units in the intermediate layer is 3 in the identification plane. And the number of units in the output layer can be set to 3 categories of dots, lines, and planes.
【００４５】決定木を用いる方法では、ニューラルネッ
トワークにより求めた欠陥形状ごとに判断ツリーを作成
し、このツリーの各ノードに予め設定されている閾値と
特徴量とを比較し、その結果に基づいて進路を選択して
ゆき、終端に着いたらそこに対応するカテゴリーを分類
結果とする。In the method using the decision tree, a decision tree is created for each defect shape obtained by the neural network, the threshold value preset in each node of this tree is compared with the feature quantity, and based on the result. When the route is selected and the end is reached, the category corresponding to that is taken as the classification result.
【００４６】欠陥の等級は等級認識処理６８により行わ
れる。この処理６８は、例えば欠陥の「面積」、「濃度
平均値」の２つの特徴量を用い、ニューラルネットワー
クを用いて欠陥の等級を軽、中、重の３つのレベルに分
けることができる。Defect grading is performed by a grade recognition process 68. In this processing 68, for example, two feature quantities of the "area" and "density average value" of the defect are used, and the class of the defect can be divided into three levels of light, medium and heavy by using a neural network.
【００４７】このようにして欠陥の形状、等級が判定さ
れると、次に総合判定処理７０により総合判定される。
この時複数の欠陥の形状が全て同一カテゴリーであれば
その旨の情報が付加される。さらに同一形状の欠陥が複
数ある場合には、その等級をそれぞれの単一の欠陥に対
する等級より高くすることにより、欠陥個数の影響を加
味するようにしてもよい。When the shape and the grade of the defect are determined in this way, the comprehensive determination processing 70 then makes a comprehensive determination.
At this time, if the shapes of a plurality of defects are all in the same category, information to that effect is added. Further, when there are a plurality of defects having the same shape, the grade may be set higher than the grade for each single defect, so that the influence of the number of defects may be taken into consideration.
【００４８】[0048]
【発明の効果】請求項１の発明は以上のように、欠陥ア
ドレス検出手段（４０）によって欠陥のアドレス（Ａ
ｄ）を求める一方、微分画像全体の濃度分布から決まる
第１の閾値ＴＨ１と注目画素の近傍領域の濃度平均値か
ら得られる第２の閾値ＴＨ２とを用いて各注目画素ごと
に第３の閾値ＴＨ３を求め、この第３の閾値ＴＨ３を用
いて微分画像を２値化する（２値化処理手段４２）。そ
してこの２値化した画像（２値画像Ａ1 ）から得た領域
のうち、前記欠陥アドレス（Ａｄ）を含む領域のみを抽
出して（領域抽出手段４８）欠陥を示す値に塗り潰すも
のである。As described above, according to the invention of claim 1, the defective address (A) is detected by the defective address detecting means (40).
While obtaining d), a third threshold value for each target pixel is obtained using the first threshold value TH1 determined from the density distribution of the entire differential image and the second threshold value TH2 obtained from the average density value of the neighboring region of the target pixel. TH3 is obtained, and the differential image is binarized using the third threshold TH3 (binarization processing means 42). Then, of the regions obtained from this binarized image (binary image A 1 ), only the region containing the defect address (Ad) is extracted (region extraction means 48) and filled to a value indicating a defect. is there.
【００４９】このため欠陥を無人で検出し、この欠陥を
含む欠陥領域の形状や大きさなどを明確にすることがで
きる。この結果その抽出した欠陥画像（Ａ2 ）を用いて
欠陥の等級などを高精度に判定するのに適する。Therefore, the defect can be detected unattended, and the shape and size of the defect area including the defect can be clarified. As a result, the defect image (A 2 ) thus extracted is suitable for highly accurately determining the grade of the defect and the like.
【００５０】ここに第３の閾値ＴＨ３は ＴＨ３＝ＴＨ１−ｋ（ＴＨ１−ＴＨ２） により求めることができる（請求項２）。この場合、定
数ｋを検査対象の状態や特性に対応して変えることによ
り、一層高精度な抽出が可能になる。Here, the third threshold value TH3 can be obtained by TH3 = TH1-k (TH1-TH2) (claim 2). In this case, by changing the constant k in accordance with the state and characteristics of the inspection target, it is possible to perform extraction with higher accuracy.
【００５１】なお２値化処理手段（４２）の次に、この
２値化した画像（Ａ1 ）を垂直方向（走査方向または副
走査方向）に収縮させる垂直収縮手段（４６）を介在さ
せれば、筋状の欠陥を検出する場合にこの垂直方向に直
交する方向の欠陥や点状の欠陥を除去して、目的とする
筋状の欠陥を精度良く抽出できる（請求項３）。After the binarization processing means (42), a vertical contraction means (46) for contracting the binarized image (A 1 ) in the vertical direction (scanning direction or sub-scanning direction) is interposed. For example, when a streak-like defect is detected, the defect in the direction orthogonal to the vertical direction or the dot-like defect can be removed, and the intended streak-like defect can be accurately extracted (claim 3).
【００５２】さらに欠陥領域抽出手段（４８）の出力で
ある欠陥画像（Ａ2 ）に対して、膨張、穴埋め、収縮な
どの処理を施せば、欠陥の形状は一層正確になり、それ
以後における形状認識などの処理に一層都合が良い（請
求項４）。Further, if the defect image (A 2 ) output from the defect area extracting means (48) is subjected to processing such as expansion, hole filling, and contraction, the shape of the defect becomes more accurate, and the shape after that. This is more convenient for processing such as recognition (claim 4).
【図１】本発明の第１の実施例のブロック図FIG. 1 is a block diagram of a first embodiment of the present invention.
【図２】その一部の詳細ブロック図FIG. 2 is a detailed block diagram of a part thereof.
【図３】閾値ＴＨ１の決定法の一例の説明図FIG. 3 is an explanatory diagram of an example of a method of determining a threshold TH1.
【図４】閾値ＴＨ２の決定法の一例の説明図FIG. 4 is an explanatory diagram of an example of a method of determining a threshold TH2.
【図５】閾値ＴＨ３の説明図FIG. 5 is an explanatory diagram of a threshold TH3.
【図６】垂直収縮のアルゴリズム説明図FIG. 6 is an explanatory diagram of a vertical contraction algorithm.
【図７】垂直収縮の概念説明図FIG. 7 is a conceptual explanatory diagram of vertical contraction.
【図８】領域抽出のアルゴリズム説明図FIG. 8 is an explanatory diagram of a region extraction algorithm.
【図９】領域抽出処理説明図FIG. 9 is an explanatory diagram of region extraction processing
【図１０】膨張処理のアルゴリズム説明図FIG. 10 is an explanatory diagram of an expansion processing algorithm.
【図１１】穴埋め処理の説明図FIG. 11 is an explanatory diagram of a filling process.
【図１２】収縮処理のアルゴリズム説明図FIG. 12 is an algorithm explanatory diagram of contraction processing.
【図１３】欠陥画像の形状補正過程を示す図FIG. 13 is a diagram showing a process of correcting the shape of a defect image.
１０ 検査対象 ３８ 微分フィルタ ４０ 欠陥アドレス検出手段 ４２ ２値化処理手段 ４６ 垂直収縮処理手段 ４８ 欠陥領域抽出手段 ５０ 形状補正手段 ５２ 膨張処理 ５４ 穴埋め処理 ５６ 収縮処理 Ａ0 微分画素 Ａ1 ２値画像 Ａ2 欠陥画像10 inspection object 38 differential filter 40 defect address detection means 42 binarization processing means 46 vertical shrinkage processing means 48 defect area extraction means 50 shape correction means 52 expansion processing 54 hole filling processing 56 shrinkage processing A 0 differential pixel A 1 binary image A 2 defect image
Claims (4)
フィルタを含む空間フィルタリング処理した微分画像を
用いて検査対象の欠陥を含む領域を抽出する欠陥検査装
置において、前記微分画像に基づいて前記欠陥のアドレ
スを求める欠陥アドレス検出手段と、前記微分画像の全
体の濃度分布から決まる第１の閾値（ＴＨ１）と注目画
素の近傍領域の濃度平均値から得られる第２の閾値（Ｔ
Ｈ２）とを用いて各注目画素ごとに得られる第３の閾値
（ＴＨ３）を用いて各注目画素を２値化する２値化処理
手段と、前記２値化処理手段で得られた領域のうち前記
欠陥のアドレスを含む領域の画素を全て欠陥を示す値に
塗り潰す欠陥領域抽出手段とを備えることを特徴とする
欠陥検査装置。1. A defect inspection apparatus for extracting an area including a defect of an inspection object by using a differential image obtained by spatially filtering an image signal obtained by scanning an inspection object based on the differential image. Defect address detection means for obtaining the address of the defect, a first threshold value (TH1) determined from the entire density distribution of the differential image, and a second threshold value (T) obtained from the density average value in the region near the target pixel.
H2) and a binarization processing unit that binarizes each pixel of interest using a third threshold value (TH3) obtained for each pixel of interest, and a region obtained by the binarization processing unit. A defect inspection apparatus comprising: a defect area extraction unit that fills all pixels in an area including the address of the defect with a value indicating a defect.
次式 ＴＨ３＝ＴＨ１−ｋ（ＴＨ１−ＴＨ２） （ただしｋは定数） により求める請求項１の欠陥検査装置。2. The defect inspection apparatus according to claim 1, wherein the binarization processing means obtains the third threshold value TH3 by the following expression TH3 = TH1-k (TH1-TH2) (where k is a constant).
した画像を走査方向または副走査方向に収縮処理する垂
直収縮処理手段を設けた請求項１または２の欠陥検査装
置。3. The defect inspection apparatus according to claim 1, further comprising a vertical contraction processing unit for contracting the binarized image in the scanning direction or the sub-scanning direction, after the binarization processing unit.
領域に対し、膨張処理、穴埋め処理、収縮処理の少くと
も１つの処理を施す形状補正手段を有する請求項１また
は２または３の欠陥検査装置。4. The defect inspection apparatus according to claim 1, further comprising shape correction means for performing at least one of expansion processing, hole filling processing, and contraction processing on the defect area extracted by the defect area extraction means. ..
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP4103542A JPH05280958A (en) | 1992-03-30 | 1992-03-30 | Defect inspection device |
| EP93105276A EP0563897A1 (en) | 1992-03-30 | 1993-03-30 | Defect inspection system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP4103542A JPH05280958A (en) | 1992-03-30 | 1992-03-30 | Defect inspection device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| JPH05280958A true JPH05280958A (en) | 1993-10-29 |
Family
ID=14356734
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| JP4103542A Pending JPH05280958A (en) | 1992-03-30 | 1992-03-30 | Defect inspection device |
Country Status (1)
| Country | Link |
|---|---|
| JP (1) | JPH05280958A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6597805B1 (en) * | 1998-11-27 | 2003-07-22 | Nec Corporation | Visual inspection method for electronic device, visual inspecting apparatus for electronic device, and record medium for recording program which causes computer to perform visual inspecting method for electronic device |
| US11341616B2 (en) * | 2020-03-23 | 2022-05-24 | Ge Precision Healthcare | Methods and system for selective removal of streak artifacts and noise from images using deep neural networks |
-
1992
- 1992-03-30 JP JP4103542A patent/JPH05280958A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6597805B1 (en) * | 1998-11-27 | 2003-07-22 | Nec Corporation | Visual inspection method for electronic device, visual inspecting apparatus for electronic device, and record medium for recording program which causes computer to perform visual inspecting method for electronic device |
| US11341616B2 (en) * | 2020-03-23 | 2022-05-24 | Ge Precision Healthcare | Methods and system for selective removal of streak artifacts and noise from images using deep neural networks |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US6614924B1 (en) | Adaptive mask technique for defect inspection | |
| JP4416365B2 (en) | Automatic detection of scanned documents | |
| CN109377485A (en) | A kind of instant noodles packaging defect machine vision detection method | |
| CN113362326A (en) | Method and device for detecting welding spot defects of battery | |
| JPH05280960A (en) | Defect inspection device | |
| EP0563897A1 (en) | Defect inspection system | |
| CN115601359A (en) | Welding seam detection method and device | |
| EP0505729B1 (en) | Image binarization system | |
| JP3255562B2 (en) | Surface inspection equipment | |
| JPH05280959A (en) | Defect inspecton device | |
| CN117115095A (en) | Method and system for detecting tiny defects of ceramic tiles with complex textures | |
| JPH05280958A (en) | Defect inspection device | |
| JP4293653B2 (en) | Appearance inspection method | |
| JP3890844B2 (en) | Appearance inspection method | |
| JP2691794B2 (en) | Defect inspection equipment | |
| JP3367449B2 (en) | Electronic component appearance inspection method, appearance inspection apparatus, and recording medium storing a program for causing a computer to execute the appearance inspection process | |
| JPH08292158A (en) | Method and apparatus for detecting wrinkle of sheet or the like | |
| JPH0829145A (en) | Inspection of surface defect | |
| CN107092909B (en) | Angle detection algorithm based on triangle correspondence theorem | |
| JPH04157353A (en) | Cracked image data processing method | |
| JP2812619B2 (en) | Image processing device | |
| JP2710685B2 (en) | Defect detection method by visual inspection | |
| JP3715473B2 (en) | Address reader | |
| JP3523945B2 (en) | Surface inspection method | |
| JPH0319990B2 (en) |