JPH06100551B2 - Defect detection method - Google Patents
Defect detection methodInfo
- Publication number
- JPH06100551B2 JPH06100551B2 JP2017141A JP1714190A JPH06100551B2 JP H06100551 B2 JPH06100551 B2 JP H06100551B2 JP 2017141 A JP2017141 A JP 2017141A JP 1714190 A JP1714190 A JP 1714190A JP H06100551 B2 JPH06100551 B2 JP H06100551B2
- Authority
- JP
- Japan
- Prior art keywords
- image
- edge
- defect detection
- edge image
- defect
- 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.)
- Expired - Lifetime
Links
Landscapes
- Image Analysis (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
- Image Processing (AREA)
Description
【発明の詳細な説明】 [産業上の利用分野] 本発明は光沢のある物体表面の平面部分上の疵、クラッ
ク等の欠陥を外観より自動的に検出する欠陥検出方法に
関するものである。TECHNICAL FIELD The present invention relates to a defect detection method for automatically detecting defects such as flaws and cracks on a flat portion of a glossy object surface from the appearance.
[従来の技術] 従来、光沢のある物体、例えばプラスチックや、金属に
おいて、表面上の疵やクラックを検査する場合、TVカメ
ラで撮像して得られる画像を処理することにより行われ
てきた。[Prior Art] Conventionally, in the case of inspecting a flaw or a crack on the surface of a glossy object such as plastic or metal, it has been performed by processing an image captured by a TV camera.
大きな疵や、隙間が開くような大きなクラックに対して
は、2値画像を用いる簡便な面積判定処理が使われる。For a large flaw or a large crack that opens a gap, a simple area determination process using a binary image is used.
一方隙間が開かず段差のみを生じるようなクラックに対
しては、濃淡画像、空間微分画像、エッジ画像を用いる
複雑な処理が行われる。On the other hand, for a crack that does not open a gap and produces only a step, a complicated process using a grayscale image, a spatial differential image, and an edge image is performed.
[発明が解決しようとする課題] ところで濃淡画像、空間微分画像、エッジ画像を使用す
る処理としては、例えばエッジ画像上を探索してエッジ
を探し、見付け出したエッジ像周辺の微分値や濃度値の
変化具合、エッジ形状を検査することにより欠陥検出を
行う方法があるが、多くの処理を行うため、処理時間が
かかるという問題があった。[Problems to be Solved by the Invention] By the way, as the processing using the grayscale image, the spatial differential image, and the edge image, for example, a differential value and a density value around the edge image found by searching the edge image for an edge are found. Although there is a method of detecting defects by inspecting the change state of the edge shape and the edge shape, there is a problem that it takes a lot of processing time because a lot of processing is performed.
また実際の処理においては、検査時間が制限されるた
め、2値画像を使い大きな疵やクラックのみを検出する
か、又は演算速度が超高速なプロセッサを使用して濃淡
画像処理を行う必要があるが、超高速なプロセッサは高
価であるという問題があった。In the actual processing, since the inspection time is limited, it is necessary to use binary images to detect only large flaws or cracks, or to perform grayscale image processing using a processor with an extremely high calculation speed. However, there is a problem that the super high speed processor is expensive.
本発明は上述の問題点に鑑みて為されたもので、その目
的とするところは濃淡画像を用い、光沢を持つ物体表面
上の外観検査を高速に且つ精度良く行うことができる欠
陥検出方法を提供するにある。The present invention has been made in view of the above problems, and an object of the present invention is to provide a defect detection method capable of performing a visual inspection on a glossy object surface at high speed and with high accuracy using a grayscale image. To provide.
[課題を解決するための手段] 本発明は上記目的を達成するために、被検査物の平面部
分に、撮像素子に被検査物平面部の正反射光が入射する
ように照明を行って撮像し、該撮像によって得られた濃
淡画像を空間微分、2値化、細線化、エッジ延長して得
られるエッジ画像において、第1番目の欠陥検出処理と
して所定検査エリア内で一定長以上のエッジ像の有無の
判定を行って、該当するエッジ像が有れば第1番目の欠
陥検出処理のみで欠陥検出を終了し、一定長以上のエッ
ジ像が無ければ、第2番目の欠陥検出処理として、撮像
素子に被検査物の平面部分に、拡散反射光が入射するよ
うに照明を切り換えて第1番目の欠陥検出処理と同一部
分を撮像し、該撮像によって得られた濃淡画像を空間微
分、2値化、細線化、エッジ延長して得られるエッジ画
像において、所定の検査エリア内のエッジ像についてエ
ッジ像の長さ、大きさ、エッジ像周辺の濃度値、微分値
の大きさにより欠陥の有無を判定をする第2番目の欠陥
検出処理を行うものである。[Means for Solving the Problems] In order to achieve the above-mentioned object, the present invention illuminates a plane portion of an inspection object so that specular reflection light of the inspection object flat surface portion is incident on an image sensor. Then, in the edge image obtained by spatial differentiation, binarization, thinning, and edge extension of the grayscale image obtained by the image pickup, an edge image having a certain length or more in a predetermined inspection area as the first defect detection process. If there is a corresponding edge image, the defect detection is ended only by the first defect detection process, and if there is no edge image of a certain length or more, as the second defect detection process, The illumination is switched so that the diffuse reflection light is incident on the plane portion of the object to be inspected on the image pickup element, the same portion as the first defect detection processing is imaged, and the grayscale image obtained by the image pickup is spatially differentiated, 2 Obtained by digitizing, thinning, and extending edges Second defect detection processing for determining the presence / absence of a defect based on the length and size of the edge image, the density value around the edge image, and the size of the differential value of the edge image in the predetermined inspection area Is to do.
[作用] 而して光沢のある被検査物の平面部分に正反射照明を行
うと、平面状態の崩れているところ、即ちクラックが生
じて、亀裂、段差のある所や、光沢面が削り取れられて
疵となっている所は、濃淡画像上で正常面に比べて暗く
なっているため、本発明のように濃淡画像を空間微分、
2値化、細線化、エッジ延長して得られるエッジ画像は
クラック、疵の欠陥をコントラスト良く表すものとな
る。[Operation] When specular illumination is applied to the flat surface of the glossy inspection object, the flat surface is broken, that is, a crack is generated, and a crack, a stepped portion, or a glossy surface can be scraped off. Since the spots that have been damaged are darker than the normal surface on the grayscale image, the grayscale image is spatially differentiated as in the present invention,
The edge image obtained by binarization, thinning, and edge extension shows cracks and flaws with good contrast.
よって大きな疵や、クラックは正反射照明下での撮像に
より得られたエッジ画像上において、一定長以上のエッ
ジ像の有無を判定するだけで検出することができ、その
ため第1番目の欠陥検出処理が高速に行える。Therefore, large flaws and cracks can be detected by simply determining the presence or absence of an edge image of a certain length or longer on the edge image obtained by imaging under regular reflection illumination. Therefore, the first defect detection process Can be done at high speed.
一方上記の正反射照明下により得られる画像上では小さ
な疵やクラックは検出できないため、上記照明を拡散反
射照明に切り換えて再度撮像する。On the other hand, since small flaws and cracks cannot be detected on the image obtained under the regular reflection illumination, the illumination is switched to the diffuse reflection illumination and the image is taken again.
そこで第1番目の欠陥検出処理で、欠陥検出が無かった
場合に照明を拡散反射照明に切り換えて撮像を行う。こ
の拡散反射照明下による撮像で得られた濃淡画像を空間
微分、2値化、細線化、エッジ延長して得られるエッジ
画像は、小さな疵やクラックの欠陥像ともにノイズ成分
である、埃、擦り疵などの像が混在しているため、エッ
ジ像の有無だけでは欠陥の有無の判定ができないが、本
発明では第2番目の欠陥検出処理として、所定の検査エ
リア内のエッジ像についてエッジ像の長さ、大きさ、エ
ッジ像周辺の濃度値、微分値の大きさにより欠陥の有無
を判定するので、ノイズ成分と欠陥との分離を行って欠
陥を検出することができるのである。Therefore, in the first defect detection process, when no defect is detected, the illumination is switched to the diffuse reflection illumination and imaging is performed. The edge image obtained by spatial differentiation, binarization, thinning, and edge extension of the grayscale image obtained by imaging under this diffuse reflection illumination is a noise component in both the defect image of small flaws and cracks, dust, rubbing, etc. Since images such as flaws are mixed, the presence / absence of a defect cannot be determined only by the presence / absence of an edge image. However, in the present invention, as the second defect detection process, an edge image of an edge image in a predetermined inspection area is detected. Since the presence / absence of a defect is determined based on the length, the size, the density value around the edge image, and the size of the differential value, the defect can be detected by separating the noise component and the defect.
このように本発明欠陥検出方法では大きな欠陥について
は大まかに簡単に検出を行って高速処理を図り、小さな
欠陥については、撮像時の照明を拡散反射照明に切り換
えて詳細な検出処理を行うことにより精度の高い欠陥検
出が行えるのである。Thus, in the defect detection method of the present invention, a large defect is roughly and easily detected for high-speed processing, and for a small defect, illumination at the time of imaging is switched to diffuse reflection illumination to perform detailed detection processing. It is possible to detect defects with high accuracy.
[実施例] 以下本発明を実施例装置により説明する。[Example] The present invention will be described below with reference to an example device.
第1図は本発明の実施例装置を示しており、この装置の
動作を第2図に示すフローチャートに沿って説明する。
まず検査開始時に制御部15から照明切り換えコントロー
ラ10に対して、正反射照明となるように指令を出し、照
明切り換えコントローラ10により正反射照明の設定を行
って正反射照明用ライト2を点灯させる。FIG. 1 shows an apparatus according to an embodiment of the present invention, and the operation of this apparatus will be described with reference to the flow chart shown in FIG.
First, at the start of the inspection, the control unit 15 issues a command to the illumination switching controller 10 to set the regular reflection illumination, and the illumination switching controller 10 sets the regular reflection illumination to turn on the regular reflection illumination light 2.
この点灯後において、TVカメラ2に被検査物1を撮像さ
せる。After this lighting, the TV camera 2 is made to image the inspection object 1.
このTVカメラ2の撮像信号はA/D変換器5に入力してA/D
変換され、A/D変換器5は変換出力を分岐させて一方を
原画像フレームメモリ13に送って濃淡画像を作成し、も
う一方を空間微分回路6へ送り込む。The image pickup signal of this TV camera 2 is input to the A / D converter 5 and A / D
After being converted, the A / D converter 5 branches the converted output, sends one to the original image frame memory 13 to create a grayscale image, and sends the other to the spatial differentiation circuit 6.
空間微分回路6は送り込まれてデジタル化された画像信
号を空間微分した後、2値化回路7と空間微分画像フレ
ームメモリ12とに分岐出力しており、2値化回路7へ送
られた画像信号は2値化された後、更に細線化回路8、
エッジ延長処理回路9を経て、明るさが変化する部分を
示すエッジ画像となってエッジ画像フレームメモリ11に
蓄積され、また空間微分画像フレームメモリ12に送られ
た画像信号が空間微分画像となる。The spatial differentiating circuit 6 spatially differentiates the image signal that has been sent and digitized, and then branches and outputs it to the binarizing circuit 7 and the spatial differential image frame memory 12, and the image sent to the binarizing circuit 7 After the signal is binarized, the thinning circuit 8,
The image signal, which has passed through the edge extension processing circuit 9 and is stored in the edge image frame memory 11 as an edge image showing a portion where the brightness changes, is sent to the spatial differential image frame memory 12 as a spatial differential image.
このようにして、原画像フレームメモリ13、空間微分画
像フレームメモリ12、エッジ画像フレームメモリ11の総
てに画像が作成された時点で制御部15は判定処理部14に
対して第1番目の欠陥検出処理の開始を指令する。In this way, when the images are created in all of the original image frame memory 13, the spatial differential image frame memory 12, and the edge image frame memory 11, the control unit 15 tells the determination processing unit 14 the first defect. Command the start of the detection process.
判定処理部14はエッジ画像フレームメモリ11上の第3図
(a)に示す検査エリアa内に一定長以上のエッジ像イ
が存在するかどうかの探索を行って、エッジ像イに該当
するものがあれば欠陥ありと判定するもので、欠陥あり
と判定した場合には処理を終了し、エッジ像イが存在し
なかった場合には第1番目の欠陥検出処理ではエッジ像
イが無かったことを制御部15へ通知する。The determination processing unit 14 searches the edge image frame memory 11 in the inspection area “a” shown in FIG. If there is a defect, it is determined that there is a defect. If it is determined that there is a defect, the process is terminated. If there is no edge image a, there is no edge image a in the first defect detection process. To the control unit 15.
制御部15は第1番目の欠陥検出処理で欠陥が見付からな
かった通知を受け取ると、照明切り換えコントローラ10
に指令を出して正反射照明用ライト4を消灯させ、この
消灯後に拡散反射照明用ライト3を点灯させ、TVカメラ
2により被検査物1を撮像する。TVカメラ2で得られた
画像信号は、第1番目の欠陥検出処理の時と同様にして
A/D変換器5、空間微分回路6、2値化回路7、細線化
回路8、エッジ延長処理回路9の回路を経て、エッジ画
像フレームメモリ11、空間微分画像フレームメモリ12、
原画像フレームメモリ13上に画像を作成する。各フレー
ムメモリ11,12,13上に画像作成が終了すると、制御部15
は判定処理部14に第2番目の欠陥検出処理開始を指令す
る。When the control unit 15 receives the notification that no defect is found in the first defect detection process, the lighting switching controller 10
Command to turn off the specular reflection illumination light 4, after which the diffuse reflection illumination light 3 is turned on, and the TV camera 2 captures an image of the inspection object 1. The image signal obtained by the TV camera 2 is processed in the same manner as in the first defect detection process.
An edge image frame memory 11, a spatial differential image frame memory 12, through an A / D converter 5, a spatial differentiation circuit 6, a binarization circuit 7, a thinning circuit 8, and an edge extension processing circuit 9.
An image is created on the original image frame memory 13. When the image creation on each frame memory 11, 12, 13 is completed, the control unit 15
Instructs the determination processing unit 14 to start the second defect detection processing.
判定処理部14はエッジ画像フレームメモリ11上の第3図
(b)に示す所定の検査エリアa内を探索し、小さなク
ラックのエッジ像ロ、擦り疵のエッジ像ハ、埃によるエ
ッジ像ニ等の小さなエッジ像が見付かると、そのエッジ
像を追跡し、エッジ像の長さ、大きさが規定の範囲内か
どうかを調べ、範囲内であれば欠陥無しと判定し、残り
の検査エリア内において次のエッジ像の探索を行う。規
定の範囲を越える場合には、エッジ周辺の濃度値、微分
値を調べて規定の範囲内であれば欠陥無しと判定し、残
りの検査エリア内において次のエッジ像の探索を行う。
ここで規定の範囲を越える場合には欠陥有りと判定して
処理を終了する。The determination processing unit 14 searches for a predetermined inspection area a shown in FIG. 3B on the edge image frame memory 11, and detects an edge image b of a small crack, an edge image c of a scratch, an edge image d of dust, and the like. When a small edge image of is found, the edge image is traced and it is checked whether the length and size of the edge image are within the specified range. If it is within the range, it is determined that there is no defect, and the remaining inspection area The next edge image is searched. If it exceeds the specified range, the density value and differential value around the edge are examined, and if it is within the specified range, it is determined that there is no defect, and the next edge image is searched in the remaining inspection area.
If it exceeds the specified range, it is determined that there is a defect, and the process ends.
このようにして第2番目の欠陥検出処理では詳細にエッ
ジ像を検定するため、擦り疵によるエッジ像ハや、埃に
よるエッジ像ニ等のノイズ成分と欠陥である小さなクラ
ックによるエッジ像ロとを識別して確実に欠陥を検出す
ることができるのである。In this way, in the second defect detection processing, the edge image is inspected in detail. Therefore, the edge image C due to the scratches, the noise component such as the edge image D due to the dust, and the edge image B due to the small crack which is the defect are detected. It is possible to identify and reliably detect the defect.
[発明の効果] 本発明は被検査物の平面部分に、撮像素子に被検査物平
面部の正反射光が入射するように照明を行って撮像し、
該撮像によって得られた濃淡画像を空間微分、2値化、
細線化、エッジ延長して得られるエッジ画像において、
第1番目の欠陥検出処理として所定検査エリア内で一定
長以上のエッジ像の有無の判定を行って、該当するエッ
ジ像が有れば第1番目の欠陥検出処理のみで欠陥検出を
終了し、一定長以上のエッジ像が無ければ、第2番目の
欠陥検出処理として、撮像素子に被検査物の平面部分
に、拡散反射光が入射するように照明を切り換えて第1
番目の欠陥検出処理と同一部分を撮像し、該撮像によっ
て得られた濃淡画像を空間微分、2値化、細線化、エッ
ジ延長して得られるエッジ画像において、所定の検査エ
リア内のエッジ像についてエッジ像の長さ、大きさ、エ
ッジ像周辺の濃度値、微分値の大きさにより欠陥の有無
を判定をする第2番目の欠陥検出処理を行うので、第1
番目の欠陥検出処理では高速で大きな欠陥の検出処理が
行なえ、また第2番目の欠陥処理では第1番目の欠陥検
出処理では検出できない小さな欠陥の検出を精度良く行
え、結果被検査物の外観検査処理が超高速プロセッサを
用いることなく高速に精度良く行えることができるとい
う効果を奏する。EFFECTS OF THE INVENTION The present invention illuminates a plane portion of an inspection object so that specular reflection light of the inspection object flat surface portion is incident on the image sensor, and images the
The grayscale image obtained by the imaging is spatially differentiated, binarized,
In the edge image obtained by thinning and edge extension,
As the first defect detection process, the presence or absence of an edge image of a certain length or more is determined in a predetermined inspection area, and if there is a corresponding edge image, the defect detection is finished only by the first defect detection process, If there is no edge image of a certain length or longer, as the second defect detection processing, the illumination is switched so that the diffuse reflected light is incident on the flat portion of the inspection object in the image sensor.
In the edge image in the predetermined inspection area, an edge image obtained by imaging the same portion as the second defect detection process and spatially differentiating, binarizing, thinning, and edge extending the grayscale image obtained by the imaging Since the second defect detection process for determining the presence / absence of a defect based on the length and size of the edge image, the density value around the edge image, and the size of the differential value is performed,
In the second defect detection process, a large defect can be detected at high speed, and in the second defect process, a small defect that cannot be detected by the first defect detection process can be accurately detected. The processing can be performed at high speed with high accuracy without using an ultra-high speed processor.
第1図は本発明の実施例装置の回路構成図、第2図は同
上のフローチャート、第3図(a)は正反射照明による
エッジ画像例図、第3図(b)は拡散反射照明によるエ
ッジ画像例図である。 1は被検査物、2はTVカメラ、3は拡散照明用ライト、
4は正反射照明用ライト、5はA/D変換器、6は空間微
分回路、7は2値化回路、8は細線化回路、9はエッジ
延長処理回路、10は照明切り換えコントローラ、11はエ
ッジ画像フレームメモリ、12空間微分画像フレームメモ
リ、13は原画像フレームメモリ、14は判定処理部、15は
制御部、aは検査エリア、イ〜ニはエッジ像である。FIG. 1 is a circuit configuration diagram of an apparatus according to an embodiment of the present invention, FIG. 2 is a flowchart of the above, FIG. 3 (a) is an edge image example diagram by regular reflection illumination, and FIG. 3 (b) is diffuse reflection illumination. It is an edge image example figure. 1 is an object to be inspected, 2 is a TV camera, 3 is a light for diffuse illumination,
4 is a light for regular reflection illumination, 5 is an A / D converter, 6 is a spatial differentiation circuit, 7 is a binarization circuit, 8 is a thinning circuit, 9 is an edge extension processing circuit, 10 is an illumination switching controller, 11 is An edge image frame memory, a 12-spatial differential image frame memory, 13 is an original image frame memory, 14 is a determination processing unit, 15 is a control unit, a is an inspection area, and a to i are edge images.
Claims (1)
物平面部の正反射光が入射するように照明を行って撮像
し、該撮像によって得られた濃淡画像を空間微分、2値
化、細線化、エッジ延長して得られるエッジ画像におい
て、第1番目の欠陥検出処理として所定検査エリア内で
一定長以上のエッジ像の有無の判定を行って、該当する
エッジ像が有れば第1番目の欠陥検出処理のみで欠陥検
出を終了し、一定長以上のエッジ像が無ければ、第2番
目の欠陥検出処理として、撮像素子に被検査物の平面部
分に、拡散反射光が入射するように照明を切り換えて第
1番目の欠陥検出処理と同一部分を撮像し、該撮像によ
って得られた濃淡画像を空間微分、2値化、細線化、エ
ッジ延長して得られるエッジ画像において、所定の検査
エリア内のエッジ像についてエッジ像の長さ、大きさ、
エッジ像周辺の濃度値、微分値の大きさにより欠陥の有
無を判定する第2番目の欠陥検出処理を行うことを特徴
とする欠陥検出方法。1. A flat portion of an object to be inspected is illuminated so that specularly reflected light of the flat portion of the object to be inspected is incident on an image pickup element, and an image is taken, and a grayscale image obtained by the image pickup is spatially differentiated, 2 In the edge image obtained by binarization, thinning, and edge extension, the presence or absence of the corresponding edge image is determined as the first defect detection process by determining whether or not there is an edge image having a certain length or more within a predetermined inspection area. For example, if the defect detection is completed only by the first defect detection process and there is no edge image of a certain length or longer, then as the second defect detection process, diffuse reflection light is generated on the plane portion of the inspection object on the image sensor. In the edge image obtained by switching the illumination so as to be incident, capturing the same portion as in the first defect detection processing, and performing spatial differentiation, binarization, thinning, and edge extension on the grayscale image obtained by the imaging , An edge in a given inspection area The length of the edge image, the size,
A defect detection method characterized by performing a second defect detection process for determining the presence / absence of a defect based on the magnitude of the density value and the differential value around the edge image.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2017141A JPH06100551B2 (en) | 1990-01-26 | 1990-01-26 | Defect detection method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2017141A JPH06100551B2 (en) | 1990-01-26 | 1990-01-26 | Defect detection method |
Publications (2)
Publication Number | Publication Date |
---|---|
JPH03221849A JPH03221849A (en) | 1991-09-30 |
JPH06100551B2 true JPH06100551B2 (en) | 1994-12-12 |
Family
ID=11935729
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
JP2017141A Expired - Lifetime JPH06100551B2 (en) | 1990-01-26 | 1990-01-26 | Defect detection method |
Country Status (1)
Country | Link |
---|---|
JP (1) | JPH06100551B2 (en) |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP5148159B2 (en) * | 2007-04-20 | 2013-02-20 | 東芝Itコントロールシステム株式会社 | Tomography apparatus for subject and layer structure extraction method |
JP5206620B2 (en) * | 2009-08-05 | 2013-06-12 | 三菱電機株式会社 | Member position recognition device, positioning device, joining device, and member joining method |
JP5475167B1 (en) * | 2013-04-30 | 2014-04-16 | キヤノンマシナリー株式会社 | Work detection device and work detection method |
-
1990
- 1990-01-26 JP JP2017141A patent/JPH06100551B2/en not_active Expired - Lifetime
Also Published As
Publication number | Publication date |
---|---|
JPH03221849A (en) | 1991-09-30 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JPH0736001B2 (en) | Bottle defect inspection method | |
EP1277042B1 (en) | Directional lighting and method to distinguish three dimensional information | |
JPH06294749A (en) | Flaw inspection method for plat glass | |
JPH06100551B2 (en) | Defect detection method | |
JPH04147045A (en) | Surface inspection device | |
JPH0633368A (en) | Method for inspecting cloth and its device | |
JPH04265847A (en) | Surface defect inspecting apparatus | |
JPH10132754A (en) | Device for inspecting appearance of lead frame | |
JPS62211544A (en) | Inspecting method for egg | |
JPH04194701A (en) | Picture image inputting method and apparatus and appearance inspecting instrument | |
JPH04270951A (en) | Method for inspecting bottle | |
JPH043820B2 (en) | ||
JPH01214743A (en) | Optical apparatus for checking | |
JP3682249B2 (en) | Glass bottle thread inspection device | |
JPH0299806A (en) | Inspection of surface defect | |
JPH07146252A (en) | Visual inspection of noodle | |
JP2000105831A (en) | Device and method for surface defect inspection | |
JPS61283857A (en) | Surface defect detection | |
JP2756738B2 (en) | Apparatus for visual inspection of semiconductor devices | |
JP2818347B2 (en) | Appearance inspection device | |
JPH07151693A (en) | Appearance inspection method and appearance inspection apparatus used for the method | |
JPH0627037A (en) | Flaw inspecting apparatus | |
JP2021149587A (en) | Method and device for detecting defects | |
JP2024032350A (en) | Visual inspection device and visual inspection method | |
JPS63122229A (en) | Inspecting device for pattern of thick film ic |