JP6630912B1 - Inspection device and inspection method - Google Patents
Inspection device and inspection method Download PDFInfo
- Publication number
- JP6630912B1 JP6630912B1 JP2019003969A JP2019003969A JP6630912B1 JP 6630912 B1 JP6630912 B1 JP 6630912B1 JP 2019003969 A JP2019003969 A JP 2019003969A JP 2019003969 A JP2019003969 A JP 2019003969A JP 6630912 B1 JP6630912 B1 JP 6630912B1
- Authority
- JP
- Japan
- Prior art keywords
- determination unit
- inspection
- defect
- image data
- absence
- 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.)
- Active
Links
- 238000007689 inspection Methods 0.000 title claims abstract description 57
- 238000000034 method Methods 0.000 title claims abstract description 21
- 230000007547 defect Effects 0.000 claims abstract description 64
- 230000002950 deficient Effects 0.000 claims abstract description 47
- 238000003384 imaging method Methods 0.000 claims abstract description 19
- 239000011800 void material Substances 0.000 claims description 17
- 238000010801 machine learning Methods 0.000 claims description 12
- 238000010586 diagram Methods 0.000 abstract description 4
- 239000004065 semiconductor Substances 0.000 description 36
- 238000011179 visual inspection Methods 0.000 description 9
- 238000005259 measurement Methods 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
- G01N2021/8854—Grading and classifying of flaws
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
- G01N2021/8854—Grading and classifying of flaws
- G01N2021/8874—Taking dimensions of defect into account
Abstract
【課題】予め決められた規格を満たした良品を不良品と誤判定する可能性が低減された検査装置及び検査方法を提供する。【解決手段】検査装置１０は、被検査対象物６０を撮像する撮像部２０と、撮像部２０によって得られた被検査対象物６０の撮像データを処理し、処理画像データを生成する画像処理部３０と、処理画像データ及び予め規定された欠陥のデータに基づいて被検査対象物６０の欠陥の有無を判定する第１の判定部４０と、処理画像データ及び予め構築された学習モデルに基づいて被検査対象物６０の欠陥の有無を判定する第２の判定部５０と、を備え、第２の判定部５０が欠陥があると判定した被検査対象物６０に対して第１の判定部４０が欠陥の有無を判定する。【選択図】図１Provided is an inspection apparatus and an inspection method that reduce the possibility of erroneously determining a non-defective product that satisfies a predetermined standard as a defective product. An inspection apparatus includes: an imaging unit configured to capture an image of an inspection target; and an image processing unit configured to process imaging data of the inspection target obtained by the imaging unit and generate processed image data. 30, a first determination unit 40 that determines the presence / absence of a defect in the inspection object 60 based on the processed image data and data of a predetermined defect, and based on the processed image data and a pre-constructed learning model. A second judging unit 50 for judging the presence / absence of a defect in the inspected object 60, wherein the first judging unit 40 Determines the presence or absence of a defect. [Selection diagram] Fig. 1
Description
本発明は、検査装置及び検査方法に関する。 The present invention relates to an inspection device and an inspection method.
特許文献１には、対象物の外観から良品と不良品を判定する外観検査装置が記載されている。この外観検査装置は、撮影照明装置により製品を撮影し、制御部において、撮影制御部にて画像を取得し、前処理部にて製品の位置決め及び背景除去を行い、色空間処理部により製品の傷や打痕等の不具合が強調される色空間処理を行うことで色空間処理画像を生成し、判定部にて、色空間処理画像に基づき製品の良否判定を機械学習した学習済みモデルを用いて、製品の良否判定を行う。 Patent Literature 1 describes an appearance inspection device that determines a non-defective product and a defective product based on the appearance of an object. This visual inspection device captures an image of a product by a shooting illumination device, acquires an image by a shooting control unit in a control unit, performs positioning and background removal of the product in a preprocessing unit, and performs product positioning by a color space processing unit. A color space processing image is generated by performing color space processing in which defects such as scratches and dents are emphasized, and a judgment unit uses a learned model that has machine-learned the quality of the product based on the color space processing image. Then, the quality of the product is determined.
本発明は、予め決められた規格を満たした良品を不良品と誤判定する可能性が低減された検査装置及び検査方法を提供することを目的とする。 An object of the present invention is to provide an inspection apparatus and an inspection method in which the possibility of erroneously determining a non-defective product that satisfies a predetermined standard as a defective product is reduced.
請求項１に記載の発明は、被検査対象物を撮像する撮像部と、前記撮像部によって得られた前記被検査対象物の撮像データを処理し、処理画像データを生成する画像処理部と、前記処理画像データ及び予め規定されたボイド及び傷のデータに基づいて、前記被検査対象物の欠陥の有無を判定する第１の判定部と、前記処理画像データ及び予め構築された機械学習モデルに基づいて、前記被検査対象物の前記ボイド及び前記傷の有無を判定する第２の判定部と、を備え、前記第２の判定部が前記ボイドがあり不良であると判定した前記被検査対象物に対して、前記第１の判定部が該ボイドの有無を判定し、前記第２の判定部が前記傷があり不良であると判定した前記被検査対象物に対して、前記第１の判定部が該傷の有無を判定する検査装置である。 The invention according to claim 1, an imaging unit that captures an image of an object to be inspected, an image processing unit that processes image data of the object to be inspected obtained by the image capturing unit, and generates processed image data. the processed image data and based on predefined voids and scratches data, said a first determination unit for determining the presence or absence of a defect of the object to be inspected, the processed image data and the pre-built machine learning models based on the said voids and second determination unit for determining the presence or absence of the flaw in the object to be inspected, wherein the second judging unit is the inspection subject it is determined to be defective has the void For the object , the first determination unit determines the presence or absence of the void , and the second determination unit determines the presence or absence of the scratch and the inspection target object determined to be defective. in determining portion inspection device determines the presence or absence of該傷 That.
請求項２に記載の発明は、被検査対象物を撮像する撮像部と、前記撮像部によって得られた前記被検査対象物の撮像データを処理し、処理画像データを生成する画像処理部と、前記処理画像データ及び予め規定された複数の種類の欠陥のデータに基づいて、前記被検査対象物の欠陥の有無を判定する第１の判定部と、前記処理画像データ及び予め構築された機械学習モデルに基づいて、前記被検査対象物について前記複数の種類の欠陥の有無を判定する第２の判定部と、を備え、前記第２の判定部が、前記被検査対象物について特定の種類の前記欠陥があり不良品と判定した場合に、前記第１の判定部が、該不良品について該特定の種類の欠陥の有無を判定する検査装置である。 The invention according to claim 2, an imaging unit that captures an image of the object to be inspected, an image processing unit that processes image data of the object to be inspected obtained by the imaging unit, and generates processed image data. the processed image data and based on predefined data of a plurality of kinds of defects are, machine learning wherein a first determination unit for determining the presence or absence of a defect of the object to be inspected, was constructed the processed image data and the pre A second determining unit that determines the presence or absence of the plurality of types of defects with respect to the inspection target based on a model , wherein the second determination unit determines that the inspection target has a specific type of defect . An inspection apparatus in which , when it is determined that the defective product is defective, the first determination unit determines the presence or absence of the specific type of defect in the defective product .
請求項３に記載の発明は、被検査対象物を撮像する撮像部と、前記撮像部によって得られた前記被検査対象物の撮像データを処理し、処理画像データを生成する画像処理部と、前記処理画像データ及び予め規定された複数の種類の欠陥のデータに基づいて、前記被検査対象物の欠陥の有無を判定する第１の判定部と、前記処理画像データ及び予め構築された機械学習モデルに基づいて、前記被検査対象物について前記複数の種類の欠陥の有無を判定する第２の判定部と、を備えた検査装置を用いて前記被検査対象物を検査する検査方法であって、前記第２の判定部が、前記被検査対象物について特定の種類の前記欠陥があり不良品と判定した場合に、前記第１の判定部が、該不良品について該特定の種類の欠陥の有無を判定する検査方法である。 The invention according to claim 3, an imaging unit that captures an image of the object to be inspected, an image processing unit that processes image data of the object to be inspected obtained by the imaging unit, and generates processed image data. the processed image data and based on predefined data of a plurality of kinds of defects are, machine learning wherein a first determination unit for determining the presence or absence of a defect of the object to be inspected, was constructed the processed image data and the pre A second determination unit that determines the presence or absence of the plurality of types of defects on the inspection target based on a model , wherein the inspection method includes inspecting the inspection target using an inspection device including: When the second determination unit determines that the inspection object has the specific type of defect and is defective, the first determination unit determines the specific type of defect with respect to the defective product . Inspection method to determine presence .
本発明によれば、予め決められた規格を満たした良品を不良品と誤判定する可能性が低減された検査装置及び検査方法を提供できる。 According to the present invention, it is possible to provide an inspection apparatus and an inspection method in which the possibility of erroneously determining a good product satisfying a predetermined standard as a defective product is reduced.
続いて、添付した図面を参照しつつ、本発明を具体化した実施の形態につき説明し、本発明の理解に供する。なお、図において、説明に関連しない部分は図示を省略する場合がある。 Next, embodiments of the present invention will be described with reference to the accompanying drawings to provide an understanding of the present invention. In the drawings, portions not related to the description may be omitted.
本発明の一実施の形態に係る外観検査装置（検査装置の一例）１０は、図１に示すように、撮像部２０、画像処理部３０、第１の判定部４０及び第２の判定部５０を備え、半導体（被検査対象物の一例）６０の外観上の欠陥を検査することができる。
なお、この欠陥の例として、図２（Ａ）〜図２（Ｅ）にそれぞれ示すような傷、ボイド（気泡により生じた穴）、汚れ、異物（異物の混入又は付着）及び未充填が挙げられる。
As shown in FIG. 1, a visual inspection device (an example of an inspection device) 10 according to an embodiment of the present invention includes an imaging unit 20, an image processing unit 30, a first determination unit 40, and a second determination unit 50. And inspects a defect on the appearance of the semiconductor (an example of the object to be inspected) 60.
Examples of such defects include scratches, voids (holes formed by air bubbles), dirt, foreign matter (mixing or adhesion of foreign matter), and unfilling as shown in FIGS. 2A to 2E. Can be
撮像部２０は、半導体６０の外観を撮像できるカメラであり、半導体６０の上方に配置されている。
画像処理部３０は、撮像部２０によって得られた半導体６０の撮像データを処理し、処理画像データを生成できる。
The image capturing unit 20 is a camera that can capture an image of the external appearance of the semiconductor 60, and is disposed above the semiconductor 60.
The image processing unit 30 can process image data of the semiconductor 60 obtained by the image capturing unit 20 and generate processed image data.
第１の判定部４０は、ルールベースにより半導体６０の外観上の欠陥の有無を判定できる。より具体的には、第１の判定部４０は、画像処理部３０が生成した処理画像データ及び予め規定された欠陥のデータに基づいて、半導体６０の外観上の欠陥の有無を判定できる。
第１の判定部４０は、図３（Ａ）に示す半導体６０の画像に対し、図３（Ｂ）に示すように、設定された検査領域（検査対象としない不問エリア以外の領域）において、欠陥を規定するデータとして予め設定された長さや面積等を計測することにより、半導体６０に欠陥があるか否かを判定する。
なお、欠陥として計測する対象は、検出対象とする欠陥の種類によって異なる。例えば、ボイドを規定するデータとして径が０．２ｍｍ以上と設定されると、径が０．２ｍｍ未満のボイドが計測された半導体６０は良品と判定され、それ以外の半導体６０は不良品と判定される。また例えば、傷を規定するデータとして長さ０．２ｍｍ以上と設定されると、長さ０．２ｍｍ未満の傷が計測された半導体６０は良品と判定され、それ以外の半導体６０は不良品と判定される。
従って、第１の判定部４０により、検査結果として保証すべき絶対的な外観上の規格を満たすか否かが判定される。
The first determination unit 40 can determine the presence or absence of a defect in the appearance of the semiconductor 60 based on a rule base. More specifically, the first determination unit 40 can determine the presence / absence of a defect in the appearance of the semiconductor 60 based on the processed image data generated by the image processing unit 30 and data of a predetermined defect.
As shown in FIG. 3B, the first determination unit 40 sets the image of the semiconductor 60 shown in FIG. 3A in a set inspection area (an area other than an unquestionable area not to be inspected). It is determined whether the semiconductor 60 has a defect by measuring a length, an area, or the like set in advance as data defining the defect.
The target to be measured as a defect differs depending on the type of the defect to be detected. For example, if the diameter is set to 0.2 mm or more as data defining the void, the semiconductor 60 in which the void having a diameter of less than 0.2 mm is measured is determined to be good, and the other semiconductors 60 are determined to be defective. Is done. Further, for example, when the length defining the flaw is set to 0.2 mm or more, the semiconductor 60 in which the flaw having a length of less than 0.2 mm is measured is determined to be good, and the other semiconductors 60 are determined to be defective. Is determined.
Therefore, the first determination unit 40 determines whether or not an absolute appearance standard to be guaranteed as an inspection result is satisfied.
第２の判定部５０は、機械学習により半導体６０の外観上の欠陥の有無を判定できる。より具体的には、第２の判定部５０は、画像処理部３０が生成した処理画像データ及び予め構築された機械学習モデルに基づいて、半導体６０の外観上の欠陥の有無を判定できる。
第２の判定部５０は、図４に示すように、予め複数の良品を撮像した画像群とボイド、傷、未充填等の各種の不良品を撮像した画像群とを例えばディープラーニング（教師あり学習の一例）により学習し、構築された機械学習モデルに基づいて、半導体６０の外観上の欠陥の有無を判定できることに加え、不良品をボイド、傷、未充填等の欠陥の種類に応じて仕分けできる。
なお、第２の判定部５０及び第１の判定部４０がそれぞれ単独で半導体６０の全数を検査する場合、機械学習による第２の判定部５０による判定処理は、ルールベースよる第１の判定部４０による判定処理よりも高速である。
The second determination unit 50 can determine whether there is a defect in the appearance of the semiconductor 60 by machine learning. More specifically, the second determination unit 50 can determine whether there is a defect in the appearance of the semiconductor 60 based on the processed image data generated by the image processing unit 30 and a machine learning model constructed in advance.
As shown in FIG. 4, the second determination unit 50 performs, for example, deep learning (supervised) with a group of images of a plurality of non-defective products and a group of images of various defective products such as voids, scratches, and unfilled. learned by the learning an example of), based on the constructed machine learning models, especially addition Ru can determine the presence or absence of a defect in appearance of the semiconductor 60, the void defective products, wound, the type of defect unfilled like We can sort according to.
In the case where the second determination unit 50 and the first determination unit 40 each independently test the total number of the semiconductors 60, the determination processing by the second determination unit 50 by machine learning is performed by the first determination unit based on a rule base. 40 is faster than the determination process by 40.
次に、外観検査装置１０の動作（半導体６０の外観検査方法）について、図５に基づいて説明する。外観検査装置１０は、以下のステップに従って半導体６０の欠陥の有無を検査できる。ただし、可能な場合には、各ステップは順番を入れ替えて実施されてもよいし、並行して実施されてもよい。
以下、説明を単純化するために、規定のボイド及び規定の傷の有無を検査し、これら欠陥がない良品を出荷する出荷検査の例について説明する。
Next, an operation of the appearance inspection apparatus 10 (a method of inspecting the appearance of the semiconductor 60) will be described with reference to FIG . The visual inspection apparatus 10 can inspect the semiconductor 60 for defects according to the following steps. However, where possible, each step may be performed in a different order, or may be performed in parallel.
Hereinafter, in order to simplify the description, an example of a shipping inspection in which a specified void and a specified scratch are inspected and a non-defective product having no such defects is shipped will be described.
（ステップＳ１）
撮像部２０が出荷前の半導体６０を撮像し、画像処理部３０が撮像データを処理して処理画像を生成する。その後、第２の判定部５０が、機械学習モデルに基づいて、半導体６０の全数について外観を検査する。
(Step S1)
The imaging unit 20 captures an image of the semiconductor 60 before shipping, and the image processing unit 30 processes the captured data to generate a processed image. Thereafter, the second determination unit 50 inspects the appearance of all the semiconductors 60 based on the machine learning model.
（ステップＳ２）
前ステップＳ１の検査にて半導体６０に欠陥がない（ボイドも傷もない）と判定された場合には、良品として扱われる。
欠陥があると判定された場合には、不良品として扱われる。
(Step S2)
If it is determined in the inspection in the previous step S1 that the semiconductor 60 has no defect (no void or scratch), it is treated as a non-defective product.
If it is determined that there is a defect, it is treated as a defective product.
（ステップＳ３ａ）
前ステップＳ２にて不良品と判定され、その欠陥がボイド不良の場合には、その不良品について第１の判定部４０がボイドの有無を検査する。
（ステップＳ３ｂ）
第１の判定部４０が規定のボイドが存在しないと判定した場合は、その半導体６０は良品として扱われる。一方、第１の判定部４０が規定のボイドが存在すると判定した場合は、その半導体６０は不良品として扱われる。
なお、第１の判定部４０により判定されたボイド不良は、その具体的な測定結果が検査データとして記録される（ボイド不良が計数値化される）。
(Step S3a)
If it is determined in the previous step S2 that the defective product is defective and the defect is a void defect, the first determining unit 40 inspects the defective product for a void.
(Step S3b)
When the first determination unit 40 determines that the prescribed void does not exist, the semiconductor 60 is treated as a non-defective product. On the other hand, when the first determination unit 40 determines that the prescribed void exists, the semiconductor 60 is treated as a defective product.
For the void defect determined by the first determination unit 40, a specific measurement result is recorded as inspection data (the void defect is converted into a count value).
（ステップＳ４ａ）
前ステップＳ２にて不良品と判定され、欠陥が傷不良の場合には、その不良品について第１の判定部４０が傷の有無を検査する。
（ステップＳ４ｂ）
第１の判定部４０が規定の傷が存在しないと判定した場合は、その半導体６０は良品として扱われる。一方、第１の判定部４０が規定の傷が存在すると判定した場合は、その半導体６０は不良品として扱われる。
なお、第１の判定部４０により判定された傷不良は、その具体的な測定結果が検査データとして記録される（傷不良が計数値化される）。
(Step S4a)
If the defect is determined to be defective in the previous step S2 and the defect is defective, the first determination unit 40 inspects the defective product for the presence or absence of a defect.
(Step S4b)
If the first determination unit 40 determines that there is no specified flaw, the semiconductor 60 is treated as a non-defective product. On the other hand, when the first determination unit 40 determines that the specified scratch is present, the semiconductor 60 is treated as a defective product.
For the flaws determined by the first determination unit 40, the specific measurement results are recorded as inspection data (the flaws are converted into a count value).
（ステップＳ５）
ステップＳ３ｂ及びステップＳ４ｂにて不良品と判定された半導体６０は、不良品として所定の処理がなされる。
ステップＳ１にて不良品と判定された半導体６０のうち、欠陥がボイド不良かつ傷不良である半導体６０は、不良品として所定の処理がなされる。欠陥がボイド不良でも傷不良でもない欠陥（その他不良）である半導体６０も、不良品として所定の処理がなされる。
(Step S5)
The semiconductor 60 determined to be defective in steps S3b and S4b is subjected to predetermined processing as a defective.
Among the semiconductors 60 determined to be defective in step S1, the semiconductor 60 whose defect is void defect and scratch defect is subjected to predetermined processing as defective. The semiconductor 60 whose defect is not a void defect or a scratch defect (other defect) is also subjected to a predetermined process as a defective product.
（ステップＳ６）
ステップＳ２にて良品と判定された半導体６０が出荷される。
ステップＳ３ｂ及びステップＳ４ｂにて良品と判定された半導体６０も出荷される。
(Step S6)
The semiconductor 60 determined to be non-defective in step S2 is shipped.
The semiconductor 60 determined to be non-defective in steps S3b and S4b is also shipped.
このように、外観検査装置１０によれば、第２の判定部５０が機械学習により欠陥がある不良品と判定した半導体６０に対して、第１の判定部４０が再度ルールベースにより欠陥の有無を判定し、欠陥がないと判断した半導体６０を良品として取り扱うことによって、予め決められた規格を満たした良品を不良品と誤判定する可能性が低減される。
また、単独での判定処理に関しては、第２の判定部５０の方が第１の判定部４０よりも高速で良否を判定できるため、第１の判定部４０のみで判定する場合よりも、検査に要する時間は全体として短縮される。
As described above, according to the appearance inspection apparatus 10, for the semiconductor 60 that the second determination unit 50 has determined as a defective product having a defect by machine learning, the first determination unit 40 determines again whether there is a defect based on the rule base. Is determined, and the semiconductor 60 determined to be free from defects is handled as a non-defective product, thereby reducing the possibility of erroneously determining a non-defective product that satisfies a predetermined standard.
In addition, as for the determination processing by itself, the second determination unit 50 can determine the quality at a higher speed than the first determination unit 40. The time required is reduced as a whole.
以上、本発明の実施の形態を説明したが、本発明は、上記した形態に限定されるものでなく、要旨を逸脱しない条件の変更等は全て本発明の適用範囲である。
検査装置は、被検査対象の外観を検査する外観検査装置に限定されるものではなく、例えば、内部の欠陥を非接触で検査できる非接触検査装置であってもよい。
Although the embodiments of the present invention have been described above, the present invention is not limited to the above-described embodiments, and any changes in conditions that do not depart from the gist are within the scope of the present invention.
The inspection device is not limited to a visual inspection device that inspects the external appearance of the inspection target, and may be, for example, a non-contact inspection device that can inspect internal defects without contact.
１０ 外観検査装置
２０ 撮像部
３０ 画像処理部
４０ 第１の判定部
５０ 第２の判定部
６０ 半導体
Reference Signs List 10 visual inspection device 20 imaging unit 30 image processing unit 40 first determination unit 50 second determination unit 60 semiconductor
Claims (3)
前記撮像部によって得られた前記被検査対象物の撮像データを処理し、処理画像データを生成する画像処理部と、
前記処理画像データ及び予め規定されたボイド及び傷のデータに基づいて、前記被検査対象物の欠陥の有無を判定する第１の判定部と、
前記処理画像データ及び予め構築された機械学習モデルに基づいて、前記被検査対象物の前記ボイド及び前記傷の有無を判定する第２の判定部と、を備え、
前記第２の判定部が前記ボイドがあり不良であると判定した前記被検査対象物に対して、前記第１の判定部が該ボイドの有無を判定し、
前記第２の判定部が前記傷があり不良であると判定した前記被検査対象物に対して、前記第１の判定部が該傷の有無を判定する検査装置。 An imaging unit for imaging an object to be inspected;
An image processing unit that processes image data of the inspection object obtained by the imaging unit and generates processed image data;
And said processing the image data and based on predefined voids and flaws of the data, the first determination unit determines the presence or absence of a defect of the object to be inspected,
Based on the processed image data and the pre-built machine learning model, and a second determination unit for determining the presence or absence of the void and the flaw of the object to be inspected,
With respect to the inspection object to the second determination unit determines that a failure has said void, said first determination unit determines the presence of the voids,
An inspection apparatus in which the first determination unit determines the presence or absence of the scratch on the inspection target object that the second determination unit has determined to be defective due to the scratch .
前記撮像部によって得られた前記被検査対象物の撮像データを処理し、処理画像データを生成する画像処理部と、
前記処理画像データ及び予め規定された複数の種類の欠陥のデータに基づいて、前記被検査対象物の欠陥の有無を判定する第１の判定部と、
前記処理画像データ及び予め構築された機械学習モデルに基づいて、前記被検査対象物について前記複数の種類の欠陥の有無を判定する第２の判定部と、を備え、
前記第２の判定部が、前記被検査対象物について特定の種類の前記欠陥があり不良品と判定した場合に、前記第１の判定部が、該不良品について該特定の種類の欠陥の有無を判定する検査装置。 An imaging unit for imaging an object to be inspected;
An image processing unit that processes image data of the inspection object obtained by the imaging unit and generates processed image data;
A first determination unit configured to determine whether there is a defect in the inspection target based on the processed image data and data of a plurality of types of predetermined defects;
A second determination unit that determines the presence or absence of the plurality of types of defects on the inspection target based on the processed image data and a machine learning model constructed in advance ,
When the second determination unit determines that the inspection object has the specific type of defect and is defective, the first determination unit determines whether the specific type of defect is present in the defective product. Inspection device for determining
前記第２の判定部が、前記被検査対象物について特定の種類の前記欠陥があり不良品と判定した場合に、前記第１の判定部が、該不良品について該特定の種類の欠陥の有無を判定する検査方法。 An imaging unit that images the inspected object, an image processing unit that processes image data of the inspected object obtained by the imaging unit and generates processed image data, and the processed image data and a predefined image data. based on the data of a plurality of kinds of defects, the a first determination unit for determining the presence or absence of a defect in the inspected object, based on the processed image data and the pre-built machine learning model, the object to be inspected A second determination unit that determines the presence or absence of the plurality of types of defects on the object, and an inspection method for inspecting the inspection object using an inspection apparatus including:
When the second determination unit determines that the inspection object has the specific type of defect and is defective, the first determination unit determines whether the specific type of defect is present in the defective product. Inspection method to judge.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP2019003969A JP6630912B1 (en) | 2019-01-14 | 2019-01-14 | Inspection device and inspection method |
| CN202010025454.XA CN111435118A (en) | 2019-01-14 | 2020-01-10 | Inspection apparatus and inspection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP2019003969A JP6630912B1 (en) | 2019-01-14 | 2019-01-14 | Inspection device and inspection method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| JP6630912B1 true JP6630912B1 (en) | 2020-01-15 |
| JP2020112456A JP2020112456A (en) | 2020-07-27 |
Family
ID=69146709
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| JP2019003969A Active JP6630912B1 (en) | 2019-01-14 | 2019-01-14 | Inspection device and inspection method |
Country Status (2)
| Country | Link |
|---|---|
| JP (1) | JP6630912B1 (en) |
| CN (1) | CN111435118A (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP7407670B2 (en) | 2020-08-04 | 2024-01-04 | デンカ株式会社 | Evaluation system, evaluation method, and evaluation program |
| JP7257470B2 (en) * | 2021-09-01 | 2023-04-13 | 株式会社エフ・シー・シー | Defect inspection device, defect inspection method, and prediction model generation method |
| WO2023095416A1 (en) * | 2021-11-25 | 2023-06-01 | 株式会社ジャパンディスプレイ | Display device inspection method, and information processing device |
Family Cites Families (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH04364449A (en) * | 1991-06-11 | 1992-12-16 | Sumitomo Heavy Ind Ltd | Fruit defect detecting apparatus |
| JPH0633060U (en) * | 1992-10-02 | 1994-04-28 | 谷電機工業株式会社 | Visual inspection system with automatic inspection function |
| JPH09167383A (en) * | 1995-10-12 | 1997-06-24 | Toray Ind Inc | Inspecting system and manufacturing method for optical recording medium |
| JPH11211701A (en) * | 1998-01-26 | 1999-08-06 | Hitachi Constr Mach Co Ltd | Ultrasonic inspection device |
| JP4345930B2 (en) * | 2005-01-28 | 2009-10-14 | Ｙｋｋ株式会社 | Appearance inspection equipment for articles |
| JP2010038723A (en) * | 2008-08-05 | 2010-02-18 | Hitachi Chem Co Ltd | Flaw inspecting method |
| JP5168215B2 (en) * | 2009-04-10 | 2013-03-21 | 株式会社デンソー | Appearance inspection device |
| JP2012026982A (en) * | 2010-07-27 | 2012-02-09 | Panasonic Electric Works Sunx Co Ltd | Inspection device |
| JP5921190B2 (en) * | 2011-12-28 | 2016-05-24 | 株式会社キーエンス | Image processing apparatus and image processing method |
| JP5943722B2 (en) * | 2012-06-08 | 2016-07-05 | 三菱重工業株式会社 | Defect determination apparatus, radiation imaging system, and defect determination method |
| CN104458755B (en) * | 2014-11-26 | 2017-02-22 | 吴晓军 | Multi-type material surface defect detection method based on machine vision |
| US10591388B2 (en) * | 2015-04-27 | 2020-03-17 | Virtual Fluid Monitoring Services LLC | Fluid analysis and monitoring using optical spectroscopy |
| JP6218094B1 (en) * | 2016-04-15 | 2017-10-25 | Ｊｅインターナショナル株式会社 | Inspection method, inspection apparatus, inspection program, and recording medium |
| JP7185388B2 (en) * | 2016-11-21 | 2022-12-07 | 日東電工株式会社 | Inspection device and inspection method |
| JP6358351B1 (en) * | 2017-03-21 | 2018-07-18 | Ｊｆｅスチール株式会社 | Surface defect inspection method and surface defect inspection apparatus |
| JP6560707B2 (en) * | 2017-04-20 | 2019-08-14 | ファナック株式会社 | Machined surface quality evaluation device |
| JP6549644B2 (en) * | 2017-06-27 | 2019-07-24 | ファナック株式会社 | Machine learning apparatus, robot control system and machine learning method |
| JP6705777B2 (en) * | 2017-07-10 | 2020-06-03 | ファナック株式会社 | Machine learning device, inspection device and machine learning method |
| CN207636508U (en) * | 2017-12-14 | 2018-07-20 | 北京木业邦科技有限公司 | A kind of defect of veneer detecting system and equipment |
| CN108107051B (en) * | 2017-12-19 | 2020-03-31 | 无锡先导智能装备股份有限公司 | Lithium battery defect detection system and method based on machine vision |
| CN108257122A (en) * | 2018-01-09 | 2018-07-06 | 北京百度网讯科技有限公司 | Paper sheet defect detection method, device and server based on machine vision |
-
2019
- 2019-01-14 JP JP2019003969A patent/JP6630912B1/en active Active
-
2020
- 2020-01-10 CN CN202010025454.XA patent/CN111435118A/en active Pending
Also Published As
| Publication number | Publication date |
|---|---|
| JP2020112456A (en) | 2020-07-27 |
| CN111435118A (en) | 2020-07-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP6560220B2 (en) | Method and apparatus for inspecting inspection system for detection of surface defects | |
| JP6342046B2 (en) | Generation of wafer inspection processes using bit defects and virtual inspection | |
| JP6630912B1 (en) | Inspection device and inspection method | |
| JP6922539B2 (en) | Surface defect determination method and surface defect inspection device | |
| JP2018159705A (en) | Automatic optical inspection system and method for operation | |
| JP7053366B2 (en) | Inspection equipment and inspection method | |
| US8401272B2 (en) | Patterned wafer defect inspection system and method | |
| TWI526683B (en) | Method of setting parameters for common setting between optical inspection units | |
| US11280744B2 (en) | Appearance inspection apparatus and appearance inspection method | |
| CN110108713A (en) | A kind of Superficial Foreign Body defect fast filtering method and system | |
| JP2007047079A (en) | Seaming defect inspection method of can | |
| JP6953712B2 (en) | Tire visual inspection device | |
| JPH09264856A (en) | Article appearance inspection device | |
| US20230138331A1 (en) | Motion in images used in a visual inspection process | |
| JP2008304231A (en) | Inspection method for mounting state of shrink label | |
| TWI493177B (en) | Method of detecting defect on optical film with periodic structure and device thereof | |
| JP2008015930A (en) | Production management method and production management system of manufacturing line | |
| TWM585899U (en) | System for inspecting defects of semiconductor device | |
| JP2021056004A (en) | Image determination device and image determination method | |
| Devasena et al. | AI-Based Quality Inspection of Industrial Products | |
| WO2023002673A1 (en) | Analysis device, analysis system, analysis method, and analysis program | |
| JP3348059B2 (en) | Method for manufacturing semiconductor device | |
| JP2008011005A (en) | Defect inspection method and program of image sensor | |
| CN114627093A (en) | Quality inspection method and device, quality inspection system, electronic device and readable medium | |
| JP2023049527A (en) | Visual inspection device for products and method for visually inspecting products using the same |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| A621 | Written request for application examination |
Free format text: JAPANESE INTERMEDIATE CODE: A621 Effective date: 20190605 |
|
| A871 | Explanation of circumstances concerning accelerated examination |
Free format text: JAPANESE INTERMEDIATE CODE: A871 Effective date: 20190605 |
|
| A975 | Report on accelerated examination |
Free format text: JAPANESE INTERMEDIATE CODE: A971005 Effective date: 20190607 |
|
| A131 | Notification of reasons for refusal |
Free format text: JAPANESE INTERMEDIATE CODE: A131 Effective date: 20190621 |
|
| A601 | Written request for extension of time |
Free format text: JAPANESE INTERMEDIATE CODE: A601 Effective date: 20190819 |
|
| A521 | Request for written amendment filed |
Free format text: JAPANESE INTERMEDIATE CODE: A523 Effective date: 20190928 |
|
| TRDD | Decision of grant or rejection written | ||
| A01 | Written decision to grant a patent or to grant a registration (utility model) |
Free format text: JAPANESE INTERMEDIATE CODE: A01 Effective date: 20191015 |
|
| A61 | First payment of annual fees (during grant procedure) |
Free format text: JAPANESE INTERMEDIATE CODE: A61 Effective date: 20191109 |
|
| R150 | Certificate of patent or registration of utility model |
Ref document number: 6630912 Country of ref document: JP Free format text: JAPANESE INTERMEDIATE CODE: R150 |
|
| R250 | Receipt of annual fees |
Free format text: JAPANESE INTERMEDIATE CODE: R250 |
|
| R250 | Receipt of annual fees |
Free format text: JAPANESE INTERMEDIATE CODE: R250 |