WO2021166441A1 - Dispositif d'inspection et programme - Google Patents

Dispositif d'inspection et programme Download PDF

Info

Publication number
WO2021166441A1
WO2021166441A1 PCT/JP2020/048381 JP2020048381W WO2021166441A1 WO 2021166441 A1 WO2021166441 A1 WO 2021166441A1 JP 2020048381 W JP2020048381 W JP 2020048381W WO 2021166441 A1 WO2021166441 A1 WO 2021166441A1
Authority
WO
WIPO (PCT)
Prior art keywords
inspection
unit
trained model
trained
detection unit
Prior art date
Application number
PCT/JP2020/048381
Other languages
English (en)
Japanese (ja)
Inventor
幸寛 中川
Original Assignee
株式会社システムスクエア
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 株式会社システムスクエア filed Critical 株式会社システムスクエア
Publication of WO2021166441A1 publication Critical patent/WO2021166441A1/fr

Links

Images

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
    • G01N21/85—Investigating moving fluids or granular solids
    • 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
    • G01N21/89—Investigating the presence of flaws or contamination in moving material, e.g. running paper or textiles
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N23/00—Investigating or analysing materials by the use of wave or particle radiation, e.g. X-rays or neutrons, not covered by groups G01N3/00 – G01N17/00, G01N21/00 or G01N22/00
    • G01N23/02—Investigating or analysing materials by the use of wave or particle radiation, e.g. X-rays or neutrons, not covered by groups G01N3/00 – G01N17/00, G01N21/00 or G01N22/00 by transmitting the radiation through the material
    • G01N23/04—Investigating or analysing materials by the use of wave or particle radiation, e.g. X-rays or neutrons, not covered by groups G01N3/00 – G01N17/00, G01N21/00 or G01N22/00 by transmitting the radiation through the material and forming images of the material
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N23/00—Investigating or analysing materials by the use of wave or particle radiation, e.g. X-rays or neutrons, not covered by groups G01N3/00 – G01N17/00, G01N21/00 or G01N22/00
    • G01N23/02—Investigating or analysing materials by the use of wave or particle radiation, e.g. X-rays or neutrons, not covered by groups G01N3/00 – G01N17/00, G01N21/00 or G01N22/00 by transmitting the radiation through the material
    • G01N23/06—Investigating or analysing materials by the use of wave or particle radiation, e.g. X-rays or neutrons, not covered by groups G01N3/00 – G01N17/00, G01N21/00 or G01N22/00 by transmitting the radiation through the material and measuring the absorption
    • G01N23/18—Investigating the presence of flaws defects or foreign matter
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D30/00—Reducing energy consumption in communication networks
    • Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks

Definitions

  • the present invention relates to an inspection device that irradiates an object to be inspected with an electromagnetic wave and inspects a foreign substance or the like based on the amount of transmission.
  • the inspection target When the inspection target is irradiated with electromagnetic waves such as X-rays, ultraviolet rays, visible rays, infrared rays, or microwaves and the transmitted amount is detected, the strength of the transmitted amount due to the presence or absence of foreign matter in the inspection object and the material of the foreign matter is determined. A distribution is obtained. By generating a two-dimensional image in which this intensity distribution is expressed by the color tone of each pixel (tone of color tone, shade, intensity, etc.), the situation inside the inspection object that cannot be known from the outside is visualized. be able to.
  • the strength of the permeation amount depends on the presence or absence of foreign matter in the inspection object, the material of the foreign matter, etc., but depending on the combination of the foreign matter and the non-foreign matter, the difference in the permeation amount between the two is very small. It may be difficult to distinguish between the two from the two-dimensional image.
  • a method using a machine-learned trained model has recently been proposed.
  • a trained model for collecting a large amount of data such as an image showing the distribution of the permeation amount of an inspection object including a foreign substance to be detected and using this as learning data to detect the foreign substance contained in the inspection object.
  • Patent Document 1 discloses an inspection device capable of suppressing a decrease in inspection accuracy of a product containing a plurality of articles.
  • the inspection by the inspection device When reflecting a new trained model whose judgment accuracy has been improved by repeating re-learning using the newly obtained training data on the inspection device, the inspection by the inspection device has been stopped so far and the trained model It was necessary to carry out the replacement work. While the inspection by the inspection device is stopped, the inspection cannot be executed, and the productivity of the production line in which the inspection device is installed is lowered. In addition, for example, the X-ray source in the X-ray inspection device does not have stable output X-ray intensity for a while after the start, so even if the inspection device is stopped and the trained model is changed or replaced, it is immediate. In some cases, the inspection cannot be resumed and the productivity of the production line is further reduced.
  • the present invention has been made in view of the above problems, and an object of the present invention is to provide an inspection device capable of efficiently replacing a learned model for inspecting an inspection object.
  • the inspection device of the present invention inspects from a transport unit that conveys an inspection object, an image generation unit that sequentially generates an image of the inspection object conveyed by the transport unit, and an image of the inspection object generated by machine learning.
  • the storage unit that stores at least one trained model for detecting anomalies in the object and the image generated by the image generation unit, perform an AI inspection that detects anomalies in the inspection object using the trained model.
  • the trained model can be replaced by overwriting the trained model that is not used for the inspection by the abnormality detection unit with a new trained model among the trained model stored in the storage unit and the abnormality detection unit. Equipped with a trained model replacement unit to perform
  • the storage unit may store only one trained model.
  • the abnormality detection unit may be configured to be able to perform a non-AI test that does not use the trained model in addition to the AI test that uses the trained model. Then, when the trained model replacement unit replaces the trained model, the abnormality detection unit may continue to perform the non-AI inspection while invalidating the AI inspection.
  • the storage unit may be configured to be able to store two or more trained models.
  • the anomaly detection unit may use the selected one of the two or more trained models for the AI test
  • the trained model replacement unit may use the trained model replacement unit for the AI test among the two or more trained models. It is advisable to replace the trained model by overwriting the unused one with a new trained model.
  • the inspection device further includes an electromagnetic wave irradiation unit that irradiates an electromagnetic wave to an inspection object transported by the transport unit, and a transmission amount detection unit that detects the transmission amount of the electromagnetic wave irradiated by the electromagnetic wave irradiation unit. It is good. Then, the image generation unit generates an image based on the transmission amount of the electromagnetic wave detected by the transmission amount detection unit, and the trained model replacement unit learns regardless of whether or not the electromagnetic wave irradiation unit irradiates the electromagnetic wave. It would be nice to be able to replace the finished model.
  • the program according to another embodiment of the present invention includes a transport unit that conveys the inspection object, an image generation unit that sequentially generates an image of the inspection object conveyed by the transport unit, and an inspection object generated by machine learning.
  • a storage unit that stores at least one trained model for detecting anomalies in the inspection target from the image of This is a program that causes a computer to execute a process of replacing a learned model stored in a storage unit in an inspection device including an abnormality detection unit that executes an AI inspection.
  • the program replaces the trained model by overwriting the trained model stored in the storage unit with a new trained model that is not used for the inspection by the abnormality detection unit.
  • the storage unit stores only one trained model, and the abnormality detection unit can perform a non-AI inspection using the trained model in addition to the AI inspection using the trained model. It is good to do it. Then, while disabling the AI test by the abnormality detection unit, the trained model may be replaced while the non-AI test is continuously performed.
  • the storage unit is configured to be able to store two or more trained models, and the abnormality detection unit uses a selected one of the two or more trained models for the AI inspection. May be good. Then, it is preferable to replace the trained models by overwriting the two or more trained models that are not used in the AI test with a new trained model.
  • FIG. 1 shows an example of a functional configuration of the inspection device 100 according to the first embodiment of the present invention.
  • the inspection device 100 includes an inspection unit 110, a control unit 120, a storage unit 130, a display unit 140, an input unit 150, and a communication unit 160.
  • the inspection unit 110 includes an electromagnetic wave irradiation unit 111, a transport unit 112, a transmission amount detection unit 113, an image generation unit 114, and an abnormality detection unit 115.
  • FIG. 2 shows an example of the physical configuration of the inspection unit 110.
  • the electromagnetic wave irradiation unit 111 irradiates the inspection object W mounted on the transport unit 112 and transported in the Y-axis direction in FIG. 2 with electromagnetic waves such as X-rays, ultraviolet rays, visible rays, and infrared rays. It is the source.
  • the transport unit 112 is an arbitrary transport mechanism that transports the inspection object W placed on the transport surface forming the XY plane in the three-dimensional Cartesian coordinate system at a predetermined speed in the Y-axis direction. It is desirable that the transport unit 112 has high electromagnetic wave transmission so that the electromagnetic wave transmitted through the inspection object W reaches the transmission amount detection unit 113 without being attenuated as much as possible.
  • the transmission amount detection unit 113 detects and outputs the transmission amount of the electromagnetic wave transmitted through the inspection object W.
  • the method of configuring the transmission amount detection unit 113 is arbitrary, and may be configured as, for example, a line sensor composed of a plurality of detection elements arranged in the X-axis direction.
  • the transmission amount detection unit 113 may be configured as an area sensor in which a plurality of detection elements are arranged in a matrix in the X-axis direction and the Y-axis direction.
  • the image generation unit 114 generates an image that visualizes the distribution of the transmission amount for the entire inspection object W based on the information on the transmission amount of the electromagnetic wave output by the transmission amount detection unit 113. Then, the image generation unit 114 stores the generated image in the storage unit 130. For example, when the transmission amount detection unit 113 is a line sensor, the image generation unit 114 periodically outputs the color tone of the pixels arranged in the X-axis direction according to the distribution of the transmission amount in the X-axis direction. By determining (the tone of color brightness, shading, intensity, etc.) and arranging the pixel sequences arranged in the X-axis direction in the Y-axis direction along the time series, the inspection object passed over the transmission amount detection unit 113. Obtain the distribution of the amount of transmission for the entire W. The image generation unit 114 determines the brightness and color of the pixel according to the transmission amount at the position corresponding to the pixel, and obtains an image in which the distribution of the transmission amount is visualized.
  • the abnormality detection unit 115 detects a predetermined abnormality in the inspection object W based on the image generated by the image generation unit 114 and stored in the storage unit 130.
  • Predetermined abnormalities include, for example, when the inspection target W is a package, the presence of foreign matter inside, an abnormality in the shape of the contents, biting of the contents into the seal portion, and the like, as well as the inspection target W. Abnormal shape of itself, cracks, chips, etc. can be mentioned.
  • the abnormality detection unit 115 of the present embodiment includes an AI inspection unit 116 that performs an inspection using the trained model M (hereinafter referred to as an AI inspection) and a non-AI inspection unit 117 that performs an inspection other than the AI inspection.
  • An example of an inspection other than the AI inspection carried out by the non-AI inspection unit 117 is an inspection in which the user defines various conditions such as image processing and determination threshold without using machine learning.
  • the learned model M used in the AI inspection executed by the AI inspection unit 116 is stored in the storage unit 130.
  • the trained model M uses the training data in advance, and when the image of the inspection object W having an abnormality is input, there is an abnormality (NG), and when the image of the inspection object W without an abnormality is input, there is no abnormality (no abnormality). It is generated by executing learning so that it is determined as OK).
  • an inference value indicating the possibility of an abnormality is output by inputting an image to the trained model M, and the output inference value and a predetermined threshold value are used. It may be done by the comparison of.
  • an inference value is defined in the range of 0 to 1 in which the case where the possibility of abnormality is extremely low is 0 and the case where the possibility of abnormality is extremely high is 1, and the output inference value is When it is a certain value (for example, 0.6) or more, it may be determined that there is an abnormality.
  • the abnormality detection unit 115 detects a predetermined abnormality in the inspection object W by inputting the image of the inspection object W generated by the image generation unit 114 into the AI inspection unit 116 and the non-AI inspection unit 117.
  • the AI inspection unit 116 and the non-AI inspection unit 117 individually detect an abnormality in the inspection object W, and when at least one of the AI inspection unit 116 and the non-AI inspection unit 117 detects an abnormality, the abnormality detection unit 115 It is determined that the inspection object W has an abnormality.
  • the abnormality detection unit 115 may be realized as hardware as a single functional unit, or a program in which the above functions are described is stored in advance in the storage unit 130, and the control unit 120 executes the program. It may be realized by.
  • the control unit 120 is realized by a CPU (Central Processing Unit) or the like.
  • the control unit 120 reads various programs for operating the inspection device 100 from the storage unit 130 and executes them.
  • the storage unit 130 is realized by, for example, a storage medium such as an HDD or a flash memory, a non-volatile memory, a volatile memory, or a combination thereof. A part or all of the storage unit 130 may be provided outside the inspection device 100 by a cloud storage or the like connected via a communication unit provided in the inspection device 100.
  • the storage unit 130 stores the images generated by the image generation unit 114. Further, the storage unit 130 stores a program executed by the control unit 120 and various data used in the program.
  • the storage unit 130 is, for example, a control program that controls each component of the inspection unit 110 (electromagnetic wave irradiation unit 111, transport unit 112, transmission amount detection unit 113, etc.), and has been learned to be used in the AI inspection by the AI inspection unit 116.
  • the model M, the inspection program that regulates the non-AI inspection in the non-AI inspection unit 117, the trained model replacement program that regulates the control at the time of replacement of the trained model M, and the like are stored.
  • the storage unit 130 stores only one trained model M, and the AI inspection unit 116 applies the only trained model M stored by the storage unit 130 to the AI inspection.
  • the trained model replacement unit 122 is realized by the control unit 120 executing the trained model replacement program.
  • the display unit 140 is a display means for displaying an input interface for inputting various instructions in the inspection device 100, an inspection status, a foreign matter detection result, and the like under the control of the control unit 120.
  • the display unit 140 may be built in the main body of the inspection device 100 or may be externally attached.
  • the input unit 150 is an input means such as a pointing device and a keyboard for the device user to input information as needed.
  • a touch panel display may be adopted as the display unit 140, and this may be used as the input unit 150.
  • the communication unit 160 is a communication interface that communicates with an external computer, cloud storage, server, or the like.
  • the communication method of the communication unit 160 may be either wireless communication or wired communication.
  • a start program automatically executed at the time of start-up is read from the storage unit 130 and executed by the control unit 120, and an input interface screen including a selectable operation menu of the inspection device 100 is displayed. It is displayed in the unit 140.
  • menus such as setting inspection conditions, executing inspections, and replacing learned models are displayed.
  • the user operates the input unit 150 to perform operations such as selecting a menu item.
  • the inspection condition setting menu provides an operation interface in which the user sets various inspection conditions (for example, electromagnetic wave intensity, transport speed, image size, non-AI inspection conditions, etc.) prior to execution of the inspection.
  • various inspection conditions for example, electromagnetic wave intensity, transport speed, image size, non-AI inspection conditions, etc.
  • the inspection execution menu provides an operation interface in which the user sets on / off of electromagnetic wave irradiation, start / stop of inspection, and the like.
  • the inspection by the inspection device 100 is started by turning on the electromagnetic wave irradiation and instructing the start of the inspection in this inspection execution menu.
  • the inspection unit 110 sequentially generates an electromagnetic wave transmission image of the inspection object W by the transmission amount detection unit 113 and the image generation unit 114 while the inspection object W is conveyed by the transportation unit 112.
  • the non-AI inspection unit 117 applies the AI inspection by the AI inspection unit 116 and the non-AI inspection by the non-AI inspection unit 117 to the generated image, and determines the presence or absence of the above in the inspection object W.
  • the inspection target object is removed by a sorter provided downstream of the transport unit 112.
  • the trained model replacement menu provides an operation interface for executing the trained model replacement program.
  • the trained model replacement unit 122 realized by the control unit 120 executing the trained model replacement program is a storage unit regardless of whether or not the electromagnetic wave is being irradiated and whether or not the inspection is being executed. The control described below is performed so that the trained model M stored in 130 can be replaced.
  • FIG. 3 is a flowchart showing a procedure for replacing the trained model in the first embodiment.
  • the trained model replacement unit 122 accepts the user to select a new trained model M (step S10).
  • the trained model replacement unit 122 can access the storage means of the external computer or cloud storage via the communication unit 160, acquire the list of the trained models M stored in the storage means, and select the trained model M. It is advisable to present it to the user and accept the user's selection.
  • it is determined whether or not the inspection is being executed (step S20). If the inspection is not being executed (step S20; No), the process is moved to step S40.
  • step S20 when the inspection is being executed (step S20; Yes), the AI inspection by the AI inspection unit 116 is invalidated (step S30), and the process is moved to step S40. While the AI inspection is invalidated, the inspection unit 110 continues the inspection of the inspection target W, but during this time, the abnormality detection unit 115 of the inspection target W is performed only by the non-AI inspection by the non-AI inspection unit 117. Judge anomalies.
  • the trained model replacement unit 122 overwrites the trained model M stored in the storage unit 130 with the new trained model M selected in step S10 (step S40). Then, when the test is being executed (step S50; Yes), the AI test is enabled (step S60), the AI test using the new learned model M is applied to the subsequent tests, and the process is terminated. On the other hand, if the inspection is not being executed (step S50; No), the process is terminated as it is.
  • the trained model can be replaced without stopping the irradiation of electromagnetic waves and the inspection being performed by the control by the trained model replacement unit 122 described above.
  • an inspection device that takes time from the start of electromagnetic wave irradiation to the stabilization of the output of the electromagnetic wave source, such as an X-ray inspection device that uses X-rays as electromagnetic waves, even if the trained model is replaced in a short time. Once the irradiation of the electromagnetic wave is stopped, it takes more time than the replacement time of the learned model M to restart the inspection, but according to the configuration of the present embodiment, such an inconvenience can be avoided.
  • the storage unit 130 can store a plurality of learned models M, and the learned model replacement unit 122 for replacing the learned model M stored in the storage unit 130.
  • the control and the like are different from those of the inspection device 100 in the first embodiment.
  • the parts common to the inspection device 100 in the first embodiment will be omitted, and the parts different from the first embodiment will be described in detail.
  • the inspection device 100A in the present embodiment is the same as the inspection device 100 in the first embodiment shown in FIG. 1, the inspection unit 110, the control unit 120, the storage unit 130, the display unit 140, the input unit 150, and the communication unit 160. To be equipped.
  • the storage unit 130 is configured to store a plurality of trained models M. Which of the plurality of trained models M to be used for the AI inspection by the AI inspection unit 116 can be set in the inspection condition setting menu.
  • the AI inspection unit 116 stores the learned model M used for the inspection in the storage unit 130 as another learned model. It is preferable to configure it so that it can be changed to M.
  • the AI inspection unit 116 is configured to specify the learned model M to be used for inspection by a pointer indicating a storage location in the storage unit 130 of the learned model, and the pointer is rewritten by control by the control unit 120. It is preferable that the trained model M used for the inspection can be switched instantly.
  • the trained model replacement unit 122 realized by the control unit 120 executing the trained model replacement program is a storage unit regardless of whether or not the electromagnetic wave is being irradiated and whether or not the inspection is being executed. The control described below is performed so that the trained model M stored in 130 can be replaced.
  • FIG. 4 is a flowchart showing the procedure for replacing the trained model in the second embodiment.
  • the trained model replacement unit 122 accepts the user to select a new trained model M (step S110).
  • the trained model replacement unit 122 can access the storage means of the external computer or cloud storage via the communication unit 160, acquire the list of the trained models M stored in the storage means, and select the trained model M. It is advisable to present it to the user and accept the user's selection.
  • the trained model replacement unit 122 specifies the trained model M stored in the storage unit 130 that is not set to be used for the AI test as the replacement target model (step S120).
  • the trained model replacement unit 122 overwrites the replacement target model specified in step S120 with the new trained model selected in step S110 (step S130), and ends the process. If the inspection is being executed when the training model replacement process described above is performed, the inspection unit 110 will perform all inspections including the AI inspection by the trained model M to which the inspection target W is being applied. To continue.
  • the trained model M can be replaced without stopping the irradiation of electromagnetic waves and the inspection being performed by the control by the trained model replacement unit 122 described above.
  • an inspection device that takes time from the start of electromagnetic wave irradiation to the stabilization of the output of the electromagnetic wave source, such as an X-ray inspection device that uses X-rays as electromagnetic waves, it is assumed that the trained model M is replaced in a short time.
  • the learned model M used in the AI inspection by the AI inspection unit 116 is not subject to replacement, so that the inspection including the AI inspection can be performed even while the replacement processing is being performed. ..
  • the trained model replacement program is executed by the input interface screen and the input unit 150 displayed on the display unit 140 of the inspection device 100, but the operation is performed from an external computer. May be configured to execute the trained model replacement program by accepting the above via the communication unit 160.
  • Inspection device 110 Inspection unit 111 Electromagnetic wave irradiation unit 112 Transport unit 113 Transmission amount detection unit 114 Image generation unit 115 Abnormality detection unit 116 AI inspection unit 117 Non-AI inspection unit 120 Control unit 122 Learned model replacement unit 130 Storage unit 140 Display unit 150 Input unit 160 Communication unit W Inspection target

Landscapes

  • Analytical Chemistry (AREA)
  • General Health & Medical Sciences (AREA)
  • Physics & Mathematics (AREA)
  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Chemical & Material Sciences (AREA)
  • Pathology (AREA)
  • Biochemistry (AREA)
  • Immunology (AREA)
  • General Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Textile Engineering (AREA)
  • Analysing Materials By The Use Of Radiation (AREA)
  • Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)

Abstract

L'invention concerne un dispositif d'inspection permettant le remplacement efficace d'un modèle appris pour inspecter un objet à inspecter. Le dispositif d'inspection de l'invention comprend : une unité de transport, qui transporte un objet à inspecter; une unité de production d'images, qui génère séquentiellement des images de l'objet à inspecter transporté par l'unité de transport; une unité de stockage, qui mémorise au moins un modèle appris pour détecter, à partir des images de l'objet à inspecter, une anomalie de l'objet à inspecter, le modèle appris étant généré selon un apprentissage automatique; une unité de détection d'anomalies, qui exécute une inspection par IA sur les images générées par l'unité de production d'images, l'inspection par IA impliquant la détection d'une anomalie de l'objet à inspecter à l'aide du modèle appris; et une unité de remplacement de modèle appris qui, parmi les modèles appris mémorisés dans l'unité de stockage, écrase un modèle appris inutilisé lors de l'inspection par l'unité de détection d'anomalies par un nouveau modèle appris, ce qui permet de remplacer le modèle appris.
PCT/JP2020/048381 2020-02-21 2020-12-24 Dispositif d'inspection et programme WO2021166441A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP2020-028236 2020-02-21
JP2020028236A JP7016179B2 (ja) 2020-02-21 2020-02-21 検査装置およびプログラム

Publications (1)

Publication Number Publication Date
WO2021166441A1 true WO2021166441A1 (fr) 2021-08-26

Family

ID=77390719

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2020/048381 WO2021166441A1 (fr) 2020-02-21 2020-12-24 Dispositif d'inspection et programme

Country Status (2)

Country Link
JP (1) JP7016179B2 (fr)
WO (1) WO2021166441A1 (fr)

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2018112550A (ja) * 2017-01-12 2018-07-19 清華大学Ｔｓｉｎｇｈｕａ Ｕｎｉｖｅｒｓｉｔｙ 検査機器および銃器検出方法
JP2019106090A (ja) * 2017-12-14 2019-06-27 オムロン株式会社 識別装置、識別方法及びプログラム
JP2019159820A (ja) * 2018-03-13 2019-09-19 オムロン株式会社 検査装置、画像識別装置、識別装置、検査方法、及び検査プログラム
JP2019158684A (ja) * 2018-03-14 2019-09-19 オムロン株式会社 検査システム、識別システム、及び識別器評価装置
JP2019174421A (ja) * 2018-03-29 2019-10-10 日本電気株式会社 選別支援装置、選別支援システム、選別支援方法及びプログラム
CN110560376A (zh) * 2019-07-19 2019-12-13 华瑞新智科技（北京）有限公司 一种产品表面缺陷检测方法及装置
JP2020011182A (ja) * 2018-07-14 2020-01-23 株式会社京都製作所 商品検査装置、商品検査方法および商品検査プログラム

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP7294768B2 (ja) * 2017-08-04 2023-06-20 オムロン株式会社 画像処理システム
JP2020008481A (ja) * 2018-07-11 2020-01-16 オムロン株式会社 画像処理装置、画像処理方法及び画像処理プログラム

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2018112550A (ja) * 2017-01-12 2018-07-19 清華大学Ｔｓｉｎｇｈｕａ Ｕｎｉｖｅｒｓｉｔｙ 検査機器および銃器検出方法
JP2019106090A (ja) * 2017-12-14 2019-06-27 オムロン株式会社 識別装置、識別方法及びプログラム
JP2019159820A (ja) * 2018-03-13 2019-09-19 オムロン株式会社 検査装置、画像識別装置、識別装置、検査方法、及び検査プログラム
JP2019158684A (ja) * 2018-03-14 2019-09-19 オムロン株式会社 検査システム、識別システム、及び識別器評価装置
JP2019174421A (ja) * 2018-03-29 2019-10-10 日本電気株式会社 選別支援装置、選別支援システム、選別支援方法及びプログラム
JP2020011182A (ja) * 2018-07-14 2020-01-23 株式会社京都製作所 商品検査装置、商品検査方法および商品検査プログラム
CN110560376A (zh) * 2019-07-19 2019-12-13 华瑞新智科技（北京）有限公司 一种产品表面缺陷检测方法及装置

Also Published As

Publication number Publication date
JP2021131364A (ja) 2021-09-09
JP7016179B2 (ja) 2022-02-04

Similar Documents

Publication Publication Date Title
JP4317566B2 (ja) Ｘ線検査装置およびｘ線検査装置の画像処理手順の生成方法
EP1720005B1 (fr) Appareil d'inspection par rayons X
JP5415182B2 (ja) 画像処理装置及びプログラム作成支援装置並びに画像処理方法
JP6754155B1 (ja) 教師データ生成装置、検査装置およびコンピュータプログラム
EP2261645B1 (fr) Appareil d'inspection par rayons x
JP5469433B2 (ja) 画像処理装置及び画像処理方法
JP5156546B2 (ja) Ｘ線検査装置
JP4849805B2 (ja) 検査装置及びｐｔｐ包装機
JP2006220448A (ja) 検査装置及びｐｔｐ包装機
US20220318985A1 (en) Training data generation device, inspection device and program
WO2021166441A1 (fr) Dispositif d'inspection et programme
JP2017138901A (ja) 検査支援装置、検査支援方法及び検査支援プログラム
JP5696194B2 (ja) プログラム作成支援装置及び画像処理装置
JP2009080030A (ja) Ｘ線検査装置
JP6388826B2 (ja) Ｘ線検査装置
JP7373840B2 (ja) 検査装置
JP2022041718A (ja) 画像処理装置、その制御方法、及びプログラム
JP2021012107A (ja) 検査装置及び学習装置
JP6144584B2 (ja) 破損検査装置
JP3733586B2 (ja) 異物検査装置
JP7323177B2 (ja) 検査システム、検査装置、学習装置及びプログラム
JP2021012098A (ja) 検査装置
JP2021060913A (ja) 検査システム
JP2005210467A (ja) 画像表示装置およびそれを備えた異物検査装置
JP7250301B2 (ja) 検査装置、検査システム、検査方法、検査プログラム及び記録媒体

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20920717

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20920717

Country of ref document: EP

Kind code of ref document: A1