WO2023223579A1 - 目視検査補助装置 - Google Patents
目視検査補助装置 Download PDFInfo
- Publication number
- WO2023223579A1 WO2023223579A1 PCT/JP2022/043352 JP2022043352W WO2023223579A1 WO 2023223579 A1 WO2023223579 A1 WO 2023223579A1 JP 2022043352 W JP2022043352 W JP 2022043352W WO 2023223579 A1 WO2023223579 A1 WO 2023223579A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- tablet
- image data
- tablets
- defective
- imaging
- 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.)
- Ceased
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61J—CONTAINERS SPECIALLY ADAPTED FOR MEDICAL OR PHARMACEUTICAL PURPOSES; DEVICES OR METHODS SPECIALLY ADAPTED FOR BRINGING PHARMACEUTICAL PRODUCTS INTO PARTICULAR PHYSICAL OR ADMINISTERING FORMS; DEVICES FOR ADMINISTERING FOOD OR MEDICINES ORALLY; BABY COMFORTERS; DEVICES FOR RECEIVING SPITTLE
- A61J3/00—Devices or methods specially adapted for bringing pharmaceutical products into particular physical or administering forms
- A61J3/06—Devices or methods specially adapted for bringing pharmaceutical products into particular physical or administering forms into the form of pills, lozenges or dragees
-
- 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
-
- 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
- G01N21/8803—Visual inspection
-
- 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
- G01N21/95—Investigating the presence of flaws or contamination characterised by the material or shape of the object to be examined
- G01N21/9508—Capsules; Tablets
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
- G06T7/001—Industrial image inspection using an image reference approach
-
- 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
- G01N2021/845—Objects on a conveyor
-
- 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
- G01N2021/888—Marking defects
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N2201/00—Features of devices classified in G01N21/00
- G01N2201/10—Scanning
- G01N2201/104—Mechano-optical scan, i.e. object and beam moving
- G01N2201/1042—X, Y scan, i.e. object moving in X, beam in Y
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
Definitions
- the present invention relates to a visual inspection auxiliary device used for visually inspecting tablets.
- the present invention has been made in view of the above circumstances, and its purpose is to more reliably prevent the occurrence of oversights, omissions, and mistakes as described above, and to further improve the reliability of visual inspection.
- the purpose of the present invention is to provide a visual inspection auxiliary device that enables visual inspection.
- a visual inspection auxiliary device comprising a conveying means for conveying tablets to be visually inspected, an imaging means for taking an image of the tablet transported by the transporting means; Judgment means for judging whether the tablet is good or bad based on image data obtained by the imaging means; a display means arranged at a position visible to an inspector who visually inspects the tablet; In the display means, a display control means capable of displaying at least information for specifying a tablet determined to be defective by the determination means;
- a visual inspection auxiliary device comprising: a conveyance control means for temporarily stopping conveyance of the tablet by the conveyance means when the determination means determines that the tablet is defective.
- the determination means determines the quality of the tablets conveyed by the conveyance means (that is, to be subjected to visual inspection). If the determining means determines that the tablet is defective, the conveyance of the tablet is temporarily stopped. Further, the display means displays information for specifying the tablets determined to be defective by the determination means.
- an inspector who performs a visual inspection can easily identify defective tablets (defective tablets) from among a plurality of tablets by effectively utilizing the display contents on the display means. Further, when a defective tablet is determined, since the tablet is temporarily stopped, the inspector can more accurately remove the identified defective tablet. This more reliably prevents overlooking defective tablets, failing to remove defective tablets, or mistakenly removing good tablets other than defective tablets. As a result, defective tablets can be more reliably prevented from flowing downstream, and the reliability of visual inspection can be further improved.
- the display control means is configured to allow the display means to display at least information for identifying a tablet determined to be defective by the determination means in association with the image data regarding the tablet.
- the visual inspection auxiliary device according to means 1, characterized in that:
- the display means information for identifying a tablet determined to be defective by the determination means is displayed in association with image data related to the tablet (for example, information for identifying the tablet as being defective) (e.g., a specific mark shown is displayed overlapping a defective tablet in the image data). Therefore, defective tablets can be more easily identified, and the effort and time required to remove defective tablets can be further reduced.
- Means 3 comprising an imaging control means for controlling the imaging means,
- the imaging control means controls the imaging means to image at least a range corresponding to a tablet determined to be defective while the tablet is temporarily stopped, and obtain re-determination image data that is the image data for re-determination.
- the determining means determines whether the tablet is good or bad based on the re-judgment image data
- the conveyance control means is capable of restarting the conveyance means so that conveyance of the temporarily stopped tablet is resumed, Means 1 characterized in that the restart of the conveyance means by the conveyance control means is performed on the condition that the determination result by the determination means based on the re-judgment image data is good.
- the visual inspection aid described in .
- the above means 3 when a defective tablet is found and the transportation of the tablet is temporarily stopped, at least the range corresponding to the tablet determined to be defective (that is, the range including the location where the defective tablet was located) is imaged. Only when the judgment result based on the obtained image data for re-judgment is good, it is possible to restart the conveyance means and restart the conveyance of the tablet. Therefore, even if a defective tablet is forgotten or mistakenly taken, it is possible to more reliably prevent the defective tablet from being erroneously discharged downstream. As a result, the reliability of visual inspection can be further improved.
- the tablet has a front side and a back side
- the conveyance means is provided with a transparent or translucent transmission part at least in part, and has a tablet support part that supports the tablet to be conveyed
- the imaging means includes a first imaging means and a second imaging means arranged vertically sandwiching the tablet support part, one of the front and back surfaces of the tablet is imaged by the first imaging means, and the other of the front and back surfaces of the tablet is imaged by the second imaging means through the transparent portion;
- the determining means is configured to determine the quality of the tablet based on image data regarding the front and back surfaces of the tablet obtained by the first imaging means and the second imaging means.
- a visual inspection auxiliary device according to any one of means 1 to 3.
- the second imaging means since it is possible to determine the quality of the front and back surfaces of the tablet using the transparent part, it is possible to determine the quality of the tablet with higher accuracy. Furthermore, since the second imaging means is disposed below the tablet support, it is possible to prevent the second imaging means from interfering with visual inspection.
- the start-up of the conveying means is permitted only when both the front and back sides of the tablet are determined to be good by the pass/fail judgment based on the re-judgment image data. You can make it happen. Therefore, even if a defective tablet is missed and turned over, the conveying means will not be allowed to start. Therefore, it is possible to very effectively prevent defective tablets from flowing downstream.
- the conveying means conveys the tablets in a plurality of rows
- the determining means is A neural network that has an encoding unit that extracts feature quantities from input image data and a decoding unit that reconstructs image data from the feature quantities is provided with only learning data based on image data related to good quality tablets.
- the learning data is image data obtained by extracting a region corresponding to a plurality of tablets in one row from image data related to non-defective tablets for each row of tablets, Means 1 characterized in that the original image data is image data obtained by extracting a region corresponding to a plurality of tablets in one row for each row of tablets from the image data obtained by the imaging means.
- image data related to non-defective tablets may include image data related to non-defective tablets accumulated in previous pass/fail judgments, virtual non-defective image data generated using such image data, etc. can.
- neural network includes, for example, a convolutional neural network having multiple convolutional layers.
- learning includes, for example, deep learning.
- identification means (generation model) includes, for example, an autoencoder (self-encoder), a convolutional autoencoder (convolutional autoencoder), and the like.
- the identification means is generated by learning only the image data related to good tablets, the reconstructed image data generated when the original image data related to defective tablets is input to the identification means. almost matches the original image data from which noise portions (defective portions) have been removed. That is, when a tablet has a defective portion, virtual image data regarding the tablet assuming that there is no defective portion is generated as reconstructed image data.
- the learning data used for learning the identification means and the original image data used for comparison (that is, pass/fail judgment) by the comparison means for example, image data corresponding to the entire field of view of the imaging means (region data), and individual It is conceivable to use image data (tablet individual data) obtained by extracting the area occupied by each tablet one by one. However, when using whole area data, the amount of data required for learning becomes enormous, which may lead to an increase in learning costs. In addition, when attempting to use individual tablet data, since the shape and size of each tablet is slightly different, it is necessary to perform extremely troublesome settings to extract individual tablet data. The processing may take a lot of time, and as a result, the extraction processing may become difficult.
- regions corresponding to a plurality of tablets in one row are extracted for each row of tablets as learning data and original image data.
- Image data is used. Therefore, the amount of data required for learning can be made relatively small, and the cost associated with learning can be reduced. Further, the extraction process of learning data and original image data can be made easier.
- the original image data is compared with the reconstructed image data that is reconstructed by inputting the original image data to the identification means, and based on the comparison result, the quality of the tablet is determined. ing. Therefore, both image data to be compared relate to the same tablet. Therefore, since the shape and appearance of the tablets are almost the same in both image data to be compared, there is no need to set relatively lenient judgment conditions to prevent false detections due to differences in shape and appearance, and stricter judgment Conditions can be set. Furthermore, it is possible to match the imaging conditions (for example, the arrangement position and arrangement angle of the tablet, the brightness and darkness state, the viewing angle of the camera, etc.) in both image data to be compared. As a result, it is possible to determine the quality of tablets with high accuracy.
- the imaging conditions for example, the arrangement position and arrangement angle of the tablet, the brightness and darkness state, the viewing angle of the camera, etc.
- the tablet has a front surface, a back surface, and a side surface located between the front surface and the back surface
- the conveyance means has an inversion part that inverts the front and back of the tablet to be conveyed
- the imaging means includes side imaging means for imaging a side surface of the tablet located in the reversal part, 2.
- the visual inspection auxiliary device according to claim 1, wherein the determining means is configured to be able to determine the quality of the side surface of the tablet based on the image data obtained by the side image capturing means.
- the inversion section can be used to determine the quality of the side surface of the tablet. Therefore, the accuracy of determining the quality of tablets can be further improved. Further, defective tablets can be more reliably prevented from flowing downstream, and the reliability of visual inspection can be further improved.
- FIG. 2 is a schematic diagram showing a schematic configuration of a visual inspection auxiliary device and the like.
- FIG. 2 is a schematic perspective view showing a conveyance device and the like. It is a side view of a tablet.
- FIG. 2 is a block diagram showing the functional configuration of a determination unit.
- FIG. 2 is a schematic plan view showing a tablet to be transported, an area to be inspected, and the like.
- FIG. 3 is a schematic diagram showing image data obtained by a camera.
- FIG. 3 is a schematic diagram showing original image data.
- FIG. 3 is a schematic diagram showing learning data.
- FIG. 2 is a schematic diagram for explaining the structure of a neural network. 3 is a flowchart showing the flow of learning processing of a neural network.
- FIG. 3 is a schematic diagram showing an example of information displayed on the display unit when a tablet is determined to be defective.
- it is a schematic diagram showing the schematic structure of a visual inspection auxiliary device etc. in which a transmission part was provided.
- it is a perspective schematic diagram showing a conveyance device etc. in which a transmission part was provided.
- it is a perspective schematic diagram showing a side camera etc. which image the side of a tablet.
- FIG. 7 is a schematic diagram showing an example of information displayed on a display unit in another embodiment.
- FIG. 7 is a schematic diagram showing an example of information displayed on a display unit in another embodiment.
- FIG. 7 is a schematic diagram showing an example of information displayed on a display unit in another embodiment.
- FIG. 7 is a schematic diagram showing an example of information displayed on a display unit in another embodiment.
- the visual inspection auxiliary device is a device used to visually inspect the tablets 9 (see FIG. 3) being transported.
- the tablet 9 is a flat tablet having a front surface 9a, a back surface 9b, and a side surface 9c located between these surfaces 9a and 9b (see FIG. 3).
- the visual inspection auxiliary device 1 includes a tablet feeding device 2, a feeding tram 3, a conveying device 4, and a chute 5.
- the transport device 4 constitutes a "transport means”.
- a counter 6a and a bottling device 6b are provided downstream of the chute 5. The counter 6a counts the tablets 9 that have undergone visual inspection, and the bottling device 6b fills the tablets 9 into containers. It is about to be done.
- the tablet feeding device 2 is, for example, a hopper, and feeds the tablets 9 to the feeding tram 3 intermittently.
- the supply tram 3 aligns and transports the tablets 9 toward the transport device 4.
- the supply tram 3 is equipped with a predetermined first vibrator 3a, and the tablets 9 on the supply tram 3 are sent to the conveying device 4 by the operation of the first vibrator 3a.
- the transport device 4 transports the tablets 9 to be visually inspected in a plurality of rows.
- the conveying device 4 includes an upstream conveying section 41, a reversing section 42, a downstream conveying section 43, and a second vibrator 44.
- one inspector who performs the visual inspection is placed on each side of the upstream conveyance section 41 and the downstream conveyance section 43.
- the inspector may be placed on both sides of the upstream conveyance section 41 and the downstream conveyance section 43 so as to sandwich them therebetween. In other words, a total of four or more inspectors may be assigned.
- FIG. 2 etc. only some of the many tablets 9 conveyed are schematically shown.
- the upstream conveyance section 41 continuously conveys the tablet 9 to the downstream side with one of the front surface 9a and the back surface 9b facing upward.
- the upstream conveyance section 41 is provided with a plurality of parallel upstream conveyance grooves 41a for conveying the tablets 9 in an aligned state to the downstream side. During transportation by the upstream transportation section 41, one of the front surface 9a and back surface 9b of the tablet 9 is visually inspected.
- the reversing unit 42 is for reversing the tablet 9 from front to back while continuously conveying the tablet 9.
- the reversing section 42 includes a plurality of intermediate conveyance grooves 42a that are continuous with the upstream conveyance groove 41a, and these intermediate conveyance grooves 42a have a shape that gradually twists toward the downstream side.
- the tablet 9 transported in the upstream transport section 41 with one of the front surface 9a and the back surface 9b facing upward passes through the intermediate transport groove 42a.
- the paper is sent to the downstream conveyance section 43 with the other side facing upward.
- the downstream transport section 43 continuously transports the tablet 9 downstream with the surface of the front surface 9a and the back surface 9b, which was the bottom surface in the upstream transport section 41, set as the top surface.
- the downstream conveyance section 43 is provided with a plurality of parallel downstream conveyance grooves 43a that are continuous with the intermediate conveyance groove 42a and are configured to convey the tablets 9 in an aligned state to the chute 5 side.
- the downstream transportation section 43 of the front surface 9a and back surface 9b of the tablet 9, the surface opposite to the surface visually inspected by the upstream transportation section 41 is visually inspected.
- the tablets 9 that have passed through the downstream conveyance section 43 flow down the chute 5 and are sent to the counter 6a.
- the tablet support part 45 which constitutes the bottom of the conveyance grooves 41a, 42a, and 43a and supports the lower surface of the tablet 9 from below, has a sloped upper surface that gradually descends toward the downstream. It is considered a thing.
- the second vibrator 44 is a device for applying vibration to the upstream conveying section 41, the reversing section 42, and the downstream conveying section 43.
- the second vibrator 44 By applying vibrations from the second vibrator 44 to the upstream conveyance section 41 and the like, the tablets 9 located in the respective grooves 41a, 42a, and 43a are conveyed downstream.
- the conveyance of the tablet 9 is stopped.
- the visual inspection auxiliary device 1 includes a transport control device 7 and a determination unit 8, as shown in FIGS. 1 and 4.
- the transport control device 7 constitutes a "transport control means".
- the transport control device 7 has a function of starting (resuming) or temporarily stopping transport of the tablets 9 in the transport device 4 by controlling the operation of the transport device 4 (particularly the second vibrator 44).
- the conveyance control device 7 is configured to be able to communicate with a control system 85 (described later) of the determination unit 8, and is capable of grasping the results of the quality determination of the tablets 9 performed by the control system 85 (particularly by the inspection section 858 described later). has been done. Then, when the control system 85 determines that the tablet 9 is defective, the transport control device 7 temporarily stops the transport of the tablet 9 by the transport device 4 . To be more specific, in this embodiment, one tablet 9 is imaged at least twice by each camera 83, 84, which will be described later. When it is determined that the tablet 9 is defective based on the obtained image data, the transport device 4 is controlled so that the transport of the tablet 9 is temporarily stopped at the timing when the second image of the tablet 9 is taken. .
- a predetermined transport operation unit (not shown) is connected to the transport control device 7, and when an inspector who performs a visual inspection operates the transport operation unit, basically, the transport control device 7 It is possible to start (restart) or temporarily stop the conveyance of the tablets 9 by switching between starting (restarting) and pausing. However, when the conveyance of the tablet 9 is temporarily stopped due to the determination of the tablet 9 as being defective as described above, the conveyance control device 7 transfers the tablet 9 that has been determined to be defective [hereinafter simply referred to as "defective tablet 9x" (FIG. 13).
- the transport operation section It is configured to allow restarting of the transport device 4 by operation. Therefore, when the conveyance of the tablets 9 is temporarily stopped due to the determination that the tablets 9 are defective, the conveyance of the tablets 9 cannot be restarted simply by operating the conveyance operation section.
- the determination unit 8 includes a first illumination device 81 , a second illumination device 82 , a first camera 83 , a second camera 84 and a control system 85 .
- the first camera 83 constitutes a "first imaging means”
- the second camera 84 constitutes a "second imaging means”.
- the first camera 83 and the second camera 84 constitute an "imaging means”.
- the first illumination device 81 and the second illumination device 82 apply a predetermined light (for example, white light, etc.) from above to a predetermined inspection area KA (see FIG. 5) that includes at least all rows of tablets 9 to be transported. irradiate.
- the first illumination device 81 irradiates the inspection area KA in the upstream transport section 41 with light
- the second illumination device 82 irradiates the inspection area KA in the downstream transport section 43 with light.
- the first camera 83 and the second camera 84 image the inspection area KA from almost directly above.
- Each of the cameras 83 and 84 includes an image sensor, such as a CCD (Charge Coupled Device) image sensor or a CMOS (Complementary Metal Oxide Semiconductor) image sensor, and an optical system ( (lens unit, aperture, etc.).
- an image sensor other than the above-mentioned image sensor may be used.
- each of the cameras 83 and 84 is arranged such that its optical axis is perpendicular to the upper surface of the tablet support section 45.
- Each camera 83, 84 is driven and controlled by a control system 85 (particularly a camera control unit 853, which will be described later).
- a control system 85 particularly a camera control unit 853, which will be described later.
- area data AD (see FIG. 6), which is image data indicating the entire field of view of the cameras 83 and 84, is obtained.
- the entire field of view of the cameras 83 and 84 corresponds to the inspection area KA.
- the image data captured and generated by each camera 83, 84 is converted into a digital signal inside each camera 83, 84, and then sent to a control system 85 (particularly an image acquisition unit 855, which will be described later) in the form of a digital signal. transferred and stored. Then, the control system 85 performs various image processing, arithmetic processing, etc., which will be described later, based on the image data.
- the control system 85 includes a CPU (Central Processing Unit) that executes predetermined arithmetic processing, a ROM (Read Only Memory) that stores various programs, fixed value data, etc., and a ROM (Read Only Memory) that temporarily stores various data when executing various arithmetic processing. It consists of a computer including a RAM (Random Access Memory) and peripheral circuits thereof, an input/output device, a display device, etc.
- a CPU Central Processing Unit
- ROM Read Only Memory
- ROM Read Only Memory
- the control system 85 includes a main control section 851, a lighting control section 852, a camera control section 853, a display control section 854, an image acquisition section 855, a data processing section 856, and a learning section by the CPU operating according to various programs. It functions as various functional units such as a section 857 and an inspection section 858.
- control system 85 includes an input section 85a composed of a keyboard, a mouse, a touch panel, etc., a first display section 85b and a second display section 85b, each equipped with a display screen that is composed of a liquid crystal display, etc. and can display various information.
- a storage section 85d capable of storing various data, programs, calculation results, test results, etc.
- a communication section 85f capable of transmitting and receiving various data to and from the outside are provided.
- a re-judgment operation section 85e is provided at a position that can be operated by the examiner.
- the first display section 85b and the second display section 85c each constitute a "display means".
- the main control section 851 is a functional section that controls the entire determination unit 8, and is configured to be able to send and receive various signals to and from other functional sections such as a lighting control section 852 and a camera control section 853.
- the lighting control unit 852 is a functional unit that controls the lighting devices 81 and 82 based on command signals from the main control unit 851.
- the camera control unit 853 is a functional unit that controls the cameras 83 and 84, and controls the imaging timing of the cameras 83 and 84 based on the command signal from the main control unit 851.
- the camera control unit 853 controls each camera 83 and 84 so that the same tablet 9 is imaged at least twice in the upstream conveyance unit 41 and the same tablet 9 is imaged at least twice in the downstream conveyance unit 43. Control.
- the camera control section 853 controls the area corresponding to the defective tablet 9x (that is, the location where the defective tablet 9x was located) while the tablet 9 is temporarily stopped.
- Each camera 83, 84 is controlled to capture an image of a range including That is, the camera control unit 853 controls the cameras 83 and 84 to obtain re-determination image data that is image data for re-determination.
- the camera control unit 853 constitutes "imaging control means".
- the display control unit 854 controls the display content on each display unit 85b, 85c based on the information stored in the storage unit 85d.
- the display control unit 854 constitutes a "display control means.”
- the image acquisition unit 855 is a functional unit for capturing image data captured and acquired by each of the cameras 83 and 84.
- the data processing unit 856 is a functional unit that performs predetermined image processing on the image data captured by the image acquisition unit 855.
- the data processing unit 856 extracts a region corresponding to one row of a plurality of tablets 9 for each row of tablets 9 from the entire area data AD (image data indicating the entire field of view of the cameras 83 and 84). By doing so, processing to obtain the original image data XD and learning data SD (see FIGS. 7 and 8) is performed.
- the original image data XD is used to determine the quality of the tablet 9, and the learning data SD is used for learning of the neural network 90, which will be described later.
- a plurality of original image data XD and learning data SD (in this embodiment, the same number as the number of rows of tablets 9) can be acquired from one area data AD.
- the learning unit 857 uses the learning data SD to train the deep neural network 90 (hereinafter simply referred to as the "neural network 90"; see FIG. 9), and uses the AI (Artificial Intelligence) model 100 as the “identification means”. This is the functional part to be constructed.
- the AI model 100 in this embodiment was constructed by performing deep learning on the neural network 90 using only the image data related to the inspection area KA of the non-defective tablet 9 as the learning data SD, as will be described later. It is a generative model and has the structure of a so-called autoencoder.
- FIG. 9 is a schematic diagram conceptually showing the structure of the neural network 90.
- the neural network 90 includes an encoder section 91 as an "encoding section” that extracts a feature amount (latent variable) TA from input image data GA, and an encoder section 91 that extracts image data GB from the feature amount TA. It has a convolutional auto-encoder (CAE) structure including a decoder section 92 as a "decoding section” for reconfiguring.
- CAE convolutional auto-encoder
- the encoder section 91 has a plurality of convolution layers 93, and each convolution layer 93 applies a plurality of filters to input data.
- the result of the convolution operation using the (kernel) 94 is output as input data for the next layer.
- the decoder unit 92 has a plurality of deconvolution layers 95, and in each deconvolution layer 95, the result of deconvolution operation using a plurality of filters (kernels) 96 on input data is obtained. is output as the input data of the next layer.
- the weights (parameters) of each filter 94 and 96 will be updated.
- the inspection unit 858 is a functional unit that determines the quality of the tablet 9 conveyed by the conveyance device 4. For example, in the present embodiment, an inspection is performed to determine whether or not foreign matter or dirt is attached to the tablet 9, and whether or not the tablet 9 is damaged such as chipping or cracking. In addition, when the tablet 9 is provided with a printed part, the inspection part 858 may determine whether the printed part is good or bad. In this embodiment, as described above, since the same tablet 9 is imaged at least twice by each of the cameras 83 and 84, the quality determination for the surface 9a of one tablet 9 is performed at least twice, and the quality of the tablet 9 is determined at least twice. The quality determination for the back surface 9b will be performed at least twice. In this embodiment, the inspection unit 858 constitutes a "judgment means".
- the reason why the front surface 9a and back surface 9b of the same tablet 9 are imaged at least twice is basically to improve the accuracy of determining whether the tablet 9 is good or bad.
- the first imaging is to discover the defective tablet 9x
- the second imaging is to obtain image data of the defective tablet 9x in a temporarily stopped state. It can be said that each of these is being carried out.
- the image data of the defective tablet 9x in a temporarily stopped state is used to identify the defective tablet 9x in the determination result handling process of step S305, which will be described later. It becomes possible to substantially match the state of the tablet 9 indicated by the data (position, spacing, orientation, etc.) with the actual state of the tablet 9 during the paused state.
- the inspection section 858 re-judges the quality of the tablet 9 based on the above-mentioned re-judgment image data.
- the first display section 85b and the second display section 85c are arranged, for example, in the vicinity of the transport device 4, at a position that can be seen by an inspector who performs a visual inspection (see FIG. 2).
- the first display section 85b is provided corresponding to the upstream conveyance section 41
- the second display section 85c is provided corresponding to the downstream conveyance section 43.
- the display control unit 854 controls each of the display units 85b and 85c, so that the first display unit 85b displays the quality determination result based on the image data obtained by the first camera 83, and the second display unit 85c displays the quality determination result based on the image data obtained by the first camera 83. , the quality determination results based on the image data obtained by the second camera 84, etc. are displayed.
- each display section 85b, 85c may be stationary or may be portable.
- the number of display units installed may be changed as appropriate depending on, for example, the number of examiners.
- the storage unit 85d is composed of an HDD (Hard Disk Drive), an SSD (Solid State Drive), etc., and has a predetermined storage area for storing, for example, the AI model 100 (neural network 90 and learning information acquired by its learning). ing. Furthermore, the storage section 85d has a function of storing image data obtained by each of the cameras 83 and 84 and the results of the quality determination of the tablet 9 performed by the inspection section 858.
- HDD Hard Disk Drive
- SSD Solid State Drive
- the re-judgment operation section 85e is an operation device used when the inspection section 858 re-executes the quality determination of the tablet 9 in a state where the conveyance of the tablet 9 is temporarily stopped.
- the cameras 83 and 84 obtain re-evaluation image data, and the inspection section 858 analyzes the tablet 9 based on the re-evaluation image data. A pass/fail judgment is made.
- the communication unit 85f includes, for example, a wireless communication interface that conforms to communication standards such as a wired LAN (Local Area Network) or wireless LAN, and is configured to be able to transmit and receive various data to and from the outside. For example, the results of the inspection performed by the inspection section 858 are outputted to the outside (such as the transport control device 7) via the communication section 85f.
- a wireless communication interface that conforms to communication standards such as a wired LAN (Local Area Network) or wireless LAN, and is configured to be able to transmit and receive various data to and from the outside.
- the results of the inspection performed by the inspection section 858 are outputted to the outside (such as the transport control device 7) via the communication section 85f.
- the main control unit 851 When the learning process is started based on the execution of a predetermined learning program, the main control unit 851 first performs preprocessing for learning the neural network 90 in step S101.
- the data processing unit 856 generates learning data SD by extracting a region corresponding to one row of a plurality of tablets 9 from the image data (wide area data AD) for each row of tablets 9.
- the learning data SD has the same format as the original image data XD, and is used to train the neural network 90 as described above. Such processing is performed until the required number of learning data SD is acquired.
- virtual non-defective product image data generated using image data related to non-defective tablets 9 may be used as the learning data SD.
- the learning unit 857 prepares the unlearned neural network 90 based on a command from the main control unit 851.
- the neural network 90 stored in advance in the storage unit 85d or the like is read out.
- the neural network 90 is constructed based on network configuration information (for example, the number of layers of the neural network, the number of nodes in each layer, etc.) stored in the storage unit 85d or the like.
- step S103 reconstructed image data is acquired. That is, based on a command from the main control unit 851, the learning unit 857 supplies the learning data SD acquired in step S101 as input data to the input layer of the neural network 90, thereby causing the output layer of the neural network 90 to Obtain the output reconstructed image data.
- the reconstructed image data has the same format as the learning data SD and the original image data XD, and is image data indicating an area corresponding to one row of tablets 9.
- the learning unit 857 compares the input image data (learning data SD) with the reconstructed image data output by the neural network 90, and determines whether the error is sufficiently small (a predetermined threshold value (or not) is determined.
- step S106 the learning unit 857 determines whether the learning end condition is satisfied. For example, if an affirmative determination is made in step S104 without going through the process of step S105, which will be described later, a predetermined number of times in a row, or if learning using all of the prepared learning data SD is repeated a predetermined number of times. If so, it is determined that the termination condition is satisfied. If the termination condition is satisfied, the neural network 90 and its learning information (updated parameters, etc., which will be described later) are stored in the storage unit 85d as the AI model 100, and the present learning process is terminated.
- step S106 if the termination condition is not satisfied in step S106, the process returns to step S103 and the learning of the neural network 90 is performed again.
- step S104 network updating processing (learning of the neural network 90) is performed in step S105, and then the process returns to step S103 to repeat the series of processes described above.
- a known learning algorithm such as error backpropagation is used to minimize the loss function representing the difference between the learning data SD and the reconstructed image data.
- the weights (parameters) of each of the filters 94 and 96 in the neural network 90 are updated to more appropriate ones.
- the loss function for example, BCE (Binary Cross-entropy) can be used.
- the neural network 90 minimizes the error between the learning data SD and the reconstructed image data, and outputs more accurate reconstructed image data.
- image data in this embodiment, original image data
- original image data XD original image data
- the AI model 100 when the tablet 9 is defective, virtual image data regarding the tablet 9 assuming that there is no defective portion is generated as reconstructed image data regarding the tablet 9.
- This process is executed at a preset normal pass/fail judgment execution timing or when a predetermined operation is performed on the re-judgment operation section 85e.
- step S301 the tablet 9 is imaged by each camera 83, 84. That is, the camera control unit 853 controls the cameras 83 and 84, so that the inspection areas KA of the transport units 41 and 43 are imaged by the cameras 83 and 84, respectively. As a result, image data (all-area data AD) representing the entire area to be inspected KA is obtained.
- step S302 the data processing unit 856 performs a process of extracting (cutting out) a region corresponding to one row of a plurality of tablets 9 for each row of tablets 9 from the obtained area data AD.
- step S302 one column of original image data XD is acquired from the entire area data AD.
- step S303 a quality determination process is performed. Specifically, as shown in FIG. 12, first, in step S501, the inspection unit 858 inputs the original image data XD acquired in step S302 to the input layer of the AI model 100. As a result, the AI model 100 outputs reconstructed image data. In this embodiment, the inspection unit 858 constitutes a "reconstructed image data acquisition means.”
- the inspection unit 858 compares the original image data XD and the reconstructed image data, and calculates the difference between the two image data. For example, dots at the same coordinates in both image data are compared, and the area (number of dots) of a cluster of dots for which the difference in brightness is equal to or greater than a predetermined value is calculated.
- the inspection unit 858 that compares the original image data XD and the reconstructed image data constitutes a "comparison means".
- step S503 the inspection unit 858 determines whether there is a defective part in the tablet 9. Specifically, the inspection unit 858 determines whether each calculated difference is larger than a predetermined threshold. If the difference is larger than a predetermined threshold, it is determined in step S504 that it is "defective". On the other hand, if the difference is smaller than the predetermined threshold, it is determined to be "good” in step S505.
- step S304 following the quality determination process in step S303 the quality determination process in step S303 is performed based on all the original image data XD related to the image data (region data AD) obtained in step S301. It is determined whether or not the That is, in step S301, a plurality of original image data XD (same number as the number of rows of tablets 9) are obtained from the area data AD, and whether the quality determination process in step S303 based on all of these original image data XD has been performed. It is determined whether or not.
- step S304 YES
- step S305 if the quality determination process based on all the original image data XD is being performed (step S304: YES), the process moves to step S305. On the other hand, if that is not the case (step S304: NO), the process returns to step S302. Therefore, the processes of steps S302 and S303 are repeated until the quality determination process based on all the original image data XD is performed.
- step S305 processing according to the quality determination result is executed. That is, as a result of performing the above-described quality determination process based on all the original image data XD, if all the original image data XD is determined to be "good", the tablet 9 related to the inspection area KA of the imaging target is It is determined that it is a "good product”, and this determination result is stored in the storage section 85d. Further, the display control unit 854 displays information indicating that the product is a “good product” and the image data (wide range data AD) obtained in step S301 on the display units 85b and 85c.
- the display control unit 854 displays information in which information indicating that it is a "defective product" and information for identifying the defective tablet 9x are associated with the image data (area data AD) obtained in step S301. is displayed on the display sections 85b and 85c.
- a mark MK indicating the position is displayed superimposed on the image data (area area data AD) obtained in step S301 (Fig. (see 13). That is, specific marks MK indicating the position of the defective tablet 9x and the position of the defective part in the defective tablet 9x are displayed in a state overlapping with the image data (region data AD).
- the transport control device 7 temporarily stops the transport of the tablet 9.
- a second imaging of the defective tablet 9x is performed, and a quality determination is performed based on the image data obtained by this imaging.
- the product is usually determined to be "defective" as in the first quality determination.
- information for identifying the defective tablet 9x (mark MK) is displayed in association with the image data (area data AD) obtained by the second imaging.
- the image data (area data AD) displayed on the display sections 85b and 85c is acquired while the tablet 9 is temporarily stopped, so it does not reflect the current state of the tablet 9 (position, spacing, orientation, etc.). ) and the state of the tablet 9 shown by the image data displayed on the display sections 85b and 85c almost match.
- the defective tablet 9x is identified by the inspector who performs the visual inspection based on this information, and the defective tablet 9x is removed from the conveyance device 4 using tweezers or the like. At this time, since the current state of the tablet 9 and the state of the tablet 9 indicated by the image data displayed on the display sections 85b and 85c almost match, the inspector can check the display contents on the display sections 85b and 85c. Using this method, defective tablets 9x can be identified very easily.
- the inspector restarts the transportation of the tablet 9.
- simply operating the transport operation section does not restart the transport of the tablets 9, and in order to restart the transport of the tablets 9, the re-judgment operation section 85e must be operated to perform the pass/fail judgment process. It is necessary to operate the conveyance operation section after repeating the process again and satisfying the condition that the product is determined to be "good” by this quality determination process. Note that the determination results and image data regarding this second pass/fail determination process are also displayed on the display sections 85b and 85c.
- the inspection unit 858 determines the quality of the tablet 9 conveyed by the conveyance device 4 (that is, subject to visual inspection). If the inspection unit 858 determines that the tablet is defective, the transportation of the tablet 9 is temporarily stopped. Furthermore, information for identifying the defective tablet 9x is displayed on each display section 85b, 85c.
- the inspector who performs the visual inspection can easily identify the defective tablet 9x from among the plurality of tablets 9 by effectively utilizing the display contents on the display sections 85b and 85c. Further, when a defective tablet 9x is determined, since the tablet 9 is temporarily stopped, the inspector can more accurately remove the identified defective tablet 9x. This can more reliably prevent overlooking the defective tablet 9x, failing to remove the defective tablet 9x, or mistakenly removing a good tablet 9 other than the defective tablet 9x. As a result, defective tablets 9x can be more reliably prevented from flowing downstream, and the reliability of visual inspection can be further improved.
- the reconstruction image obtained by imaging at least the range corresponding to the defective tablet 9x (that is, the range including the location where the defective tablet 9x was located) is added. Only when the determination result based on the determination image data is good, it is possible to restart the conveyance device 4 and restart the conveyance of the tablets 9. Therefore, even if a defective tablet 9x is forgotten or mistakenly taken, it is possible to more reliably prevent the defective tablet 9x from erroneously flowing downstream. As a result, the reliability of visual inspection can be further improved.
- the learning data SD and the original image data XD image data obtained by extracting regions corresponding to one row of multiple tablets 9 adjacent to each other in the transport direction for each row of tablets 9 is used. . Therefore, the amount of data required for learning can be made relatively small, and the cost associated with learning can be reduced. Further, the extraction process of the learning data SD and the original image data XD can be made easier.
- both image data to be compared relate to the same tablet 9. Therefore, since the shape and appearance of tablet 9 are almost the same in both image data to be compared, there is no need to set relatively lenient judgment conditions in order to prevent false detection due to differences in shape and appearance; Judgment conditions can be set. Furthermore, it is possible to match the imaging conditions (for example, the arrangement position and arrangement angle of the tablet 9, the bright/dark state, the viewing angle of the camera, etc.) in both image data to be compared. As a result, the quality of the tablet 9 can be determined with high accuracy.
- the reversing part 42 for reversing the front and back sides of the tablet 9 is provided, visual inspection of the front surface 9a and the back surface 9b of the tablet 9 can be easily performed.
- each of the cameras 83 and 84 is arranged above the conveyance path of the tablet 9, and is configured to image the top surface of the tablet 9, respectively.
- a transparent or semi-transparent transmitting section 46 (the section with a dotted pattern in FIG. 15) is provided in at least a part of the tablet support section 45, and each camera 83 , 84 are arranged to sandwich the tablet support part 45 vertically, and the first camera 83 and the second camera 84 can take images of the top surface of the tablet 9 and the bottom surface of the tablet 9 through the transparent part 46, respectively. good. That is, the first camera 83 captures an image of one of the front surface 9a and the back surface 9b of the tablet 9, and the second camera 84 captures an image of the other of the front surface 9a and the back surface 9b of the tablet 9 through the transparent portion 46. may be configured. Then, the inspection unit 858 may be configured to determine the quality of the tablet 9 based on each image data regarding the front surface 9a and back surface 9b of the tablet 9 obtained by the cameras 83 and 84.
- the second camera 84 is disposed below the tablet support section 45, so it is possible to prevent the second camera 84 from interfering with visual inspection.
- both the front surface 9a and the back surface 9b of the tablet 9 are determined to be good by the quality determination based on the re-determination image data, starting (restarting) of the conveying device 4 is permitted. For this reason, even if a defective tablet 9x is failed to be picked up and the defective tablet 9x is turned over, the transfer device 4 will not be allowed to start (restart). Therefore, it is possible to very effectively prevent defective tablets 9x from flowing downstream.
- the inspection unit 858 is configured to determine the quality of the front surface 9a and back surface 9b of the tablet 9. On the other hand, the inspection unit 858 may also be configured to determine the quality of the side surface 9c of the tablet 9.
- the inspection unit 858 determines the quality of the side surface 9c based on the image data obtained by the side surface camera 86.
- the side camera 86 corresponds to "side image capturing means” and "imaging means”.
- the reversing section 42 can be used to determine the quality of the side surface 9c of the tablet 9. Therefore, the accuracy of determining the quality of the tablet 9 can be further improved. Moreover, it is possible to more reliably prevent defective tablets 9x from flowing downstream, and it is possible to further improve the reliability of visual inspection.
- a cover may be provided to cover all or part of the upstream conveyance section 41, the reversing section 42, and the downstream conveyance section 43.
- a camera imaging means
- the quality of the tablet 9 may be determined based on image data obtained by the camera.
- the state of the cover may be used as a condition for resuming transportation of the tablet 9 or as a condition for performing imaging by the cameras 83 and 84 to obtain image data for re-judgment.
- the conditions for resuming transportation or performing imaging may be that the cover is properly installed or that the opening provided in the cover is closed.
- each of the cameras 83 and 84 is configured to image one tablet 9 at least twice, but the number of times one tablet 9 is imaged may be changed as appropriate.
- the transportation of the tablet 9 is temporarily stopped at the timing when the second imaging of the tablet 9 is performed.
- the timing of the temporary stop may be changed as appropriate.
- the transport of the tablets 9 may be temporarily stopped immediately after a defective determination is made.
- the actual position of the defective tablet 9x is used for the pass/fail judgment.
- the position of the defective tablet 9x may be slightly different from the position of the defective tablet 9x in the image data (area data AD) used in the above.
- the average transport amount of the tablets 9 from the execution of imaging to the completion of the quality determination is obtained and estimated in advance, and as shown in FIG.
- the mark MK1 indicating the downstream position by the average transport amount may be displayed in association with the image data (region data AD) related to the defective tablet 9x. Even with this configuration, defective tablets 9x can be identified relatively easily.
- a mark indicating a relatively wide range along the conveyance direction of the tablet 9 may be used as the mark MK1.
- a mark MK2 indicating a column including a defective tablet 9x may be displayed in association with the image data (area data AD) related to the defective tablet 9x. Even with this configuration, defective tablets 9x can be identified relatively easily.
- a mark MK3 for indicating the position of the defective tablet 9x may be displayed in a frame WK corresponding to the inspection area KA (the imaging range of the cameras 83, 84, 86).
- a frame line indicating the area to be inspected KA is attached to the top surface of the transport device 4 (upstream transport section 41, downstream transport section 43, etc.). This makes it easier to identify defective tablets 9x based on the displayed content.
- auxiliary lines ML or the like corresponding to the conveyance grooves 41a, 42a, and 43a may be provided within the frame WK.
- a disc-shaped flat tablet having a circular shape in plan view is exemplified as the tablet 9, but the type, shape, etc. of the tablet are not limited to the above embodiment.
- tablets include not only medicines but also tablets used for consumption.
- tablets include not only plain tablets, but also sugar-coated tablets, film-coated tablets, orally disintegrating tablets, enteric-coated tablets, gelatin-coated tablets, and various other capsules such as hard capsules and soft capsules. This also includes tablets.
- the shape of the tablet 9 is not limited to a circular shape in a plan view, but may be a polygonal shape in a plan view, an elliptical shape in a plan view, an ellipse shape in a plan view, or the like.
- the tablets 9 are supplied from the visual inspection auxiliary device 1 (chute 5) to the counter 6a and the bottling device 6b, but the supply target of the tablets 9 may be determined as appropriate. May be changed. Therefore, for example, a blister sheet (for example, a PTP sheet) is manufactured in which the tablet 9 is accommodated in a pocket formed in a container film, and a cover film is attached to the container film so as to close the pocket.
- the tablets 9 may be supplied to a blister packaging machine (for example, a PTP packaging machine) for packaging the tablets. Further, the blister packaging machine may be configured to include the visual inspection auxiliary device 1.
- the configuration of the AI model 100 (neural network 90) as the "identification means" and its learning method are not limited to the above embodiment.
- a configuration may be adopted in which processing such as normalization is performed on various data as necessary.
- the structure of the neural network 90 is not limited to that shown in FIG. 9, and may have a configuration in which a pooling layer is provided after the convolutional layer 93, for example.
- the number of layers of the neural network 90, the number of nodes in each layer, the connection structure of each node, etc. may be different.
- the AI model 100 is a generative model having the structure of a convolutional autoencoder (CAE), but is not limited to this, for example, a variational autoencoder (VAE) It may also be a generative model with the structure of a different type of autoencoder, such as Variational Autoencoder).
- CAE convolutional autoencoder
- VAE variational autoencoder
- VAE Variational Autoencoder
- the neural network 90 is configured to learn using the error backpropagation method, but the configuration is not limited to this, and a configuration may be adopted in which learning is performed using various other learning algorithms.
- the neural network 90 may be configured by a circuit dedicated to AI processing, such as a so-called AI chip.
- the AI model 100 may be configured by storing only learning information such as parameters in the storage unit 85d, reading this information by the AI processing dedicated circuit, and setting it in the neural network 90.
- control system 85 includes a learning section 857 and is configured to perform learning of the neural network 90 within the control system 85, but the present invention is not limited to this.
- the learning unit 857 is omitted, the learning of the neural network 90 is performed outside the control system 85, and the AI model 100 (trained neural network 90) that has been trained externally is stored in the storage unit 85d. It may also be a configuration.
- the learning data SD may be generated outside the control system 85.
- the configuration of the conveyance device 4 is not limited to that mentioned in the above embodiment, and for example, the conveyance device may be configured by a conveyor or the like.
- Decoder section decoding section
- 100...AI model identification means
- 853...camera control unit imaging control unit
- 854...display control unit display control unit
- 858...inspection unit determination unit, reconstructed image data acquisition unit, comparison unit
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Pathology (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- Immunology (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Quality & Reliability (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Multimedia (AREA)
- Medicinal Chemistry (AREA)
- Signal Processing (AREA)
- Pharmacology & Pharmacy (AREA)
- Animal Behavior & Ethology (AREA)
- Public Health (AREA)
- Veterinary Medicine (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
- Medical Preparation Storing Or Oral Administration Devices (AREA)
- Image Analysis (AREA)
Abstract
Description
前記搬送手段により搬送される錠剤を撮像する撮像手段と、
前記撮像手段により得られた画像データに基づき、錠剤の良否を判定する判定手段と、
錠剤の目視検査を行う検査者から視認可能な位置に配置される表示手段と、
前記表示手段において、少なくとも前記判定手段により不良判定がなされた錠剤を特定するための情報を表示させることが可能な表示制御手段と、
前記判定手段により不良判定がなされた場合に、前記搬送手段による錠剤の搬送を一時的に停止させる搬送制御手段とを備えることを特徴とする目視検査補助装置。
前記撮像制御手段は、錠剤の一時停止中に、少なくとも不良判定された錠剤に対応する範囲を撮像して、再判定用の前記画像データである再判定用画像データを得るように前記撮像手段を制御するとともに、
前記判定手段は、前記再判定用画像データに基づき、錠剤の良否を判定し、
前記搬送制御手段は、一時停止中の錠剤の搬送が再開されるように前記搬送手段を再起動させることが可能であり、
前記搬送制御手段による前記搬送手段の再起動は、前記再判定用画像データに基づく前記判定手段による判定結果が良であることを条件として行われるように構成されていることを特徴とする手段1に記載の目視検査補助装置。
前記搬送手段は、少なくとも一部に透明又は半透明な透過部が設けられるとともに、搬送される錠剤を支持する錠剤支持部を有し、
前記撮像手段は、前記錠剤支持部を上下に挟む位置に配置される第一撮像手段と第二撮像手段とを備え、
前記第一撮像手段によって錠剤の表面及び裏面のうちの一方が撮像され、前記第二撮像手段によって前記透過部を通して該錠剤の表面及び裏面のうちの他方が撮像されるように構成されており、
前記判定手段は、前記第一撮像手段及び前記第二撮像手段により得られた、錠剤の表面及び裏面に係る画像データに基づき、錠剤の良否を判定するように構成されていることを特徴とする手段1乃至3のいずれかに記載の目視検査補助装置。
前記判定手段は、
入力される画像データから特徴量を抽出する符号化部、及び、該特徴量から画像データを再構成する復号化部を有するニューラルネットワークに対し、良品の錠剤に係る画像データに基づく学習データのみを学習させて生成した識別手段と、
前記撮像手段により得られた画像データに基づく元画像データを前記識別手段へ入力して再構成された画像データを再構成画像データとして取得可能な再構成画像データ取得手段と、
前記元画像データ及び前記再構成画像データを比較可能な比較手段とを備え、
前記比較手段による比較結果に基づき、錠剤の良否を判定可能に構成されており、
前記学習データは、良品の錠剤に係る画像データから1列分の複数個の錠剤に対応する領域を錠剤の列ごとに抽出して得た画像データであり、
前記元画像データは、前記撮像手段により得られた画像データから1列分の複数個の錠剤に対応する領域を錠剤の列ごとに抽出して得た画像データであることを特徴とする手段1に記載の目視検査補助装置。
前記搬送手段は、搬送される錠剤の表裏を反転させる反転部を有し、
前記撮像手段は、前記反転部に位置する錠剤の側面を撮像する側面撮像手段を備え、
前記判定手段は、前記側面撮像手段により得られた画像データに基づき、錠剤の側面に関する良否を判定可能に構成されていることを特徴とする請求項1に記載の目視検査補助装置。
Claims (6)
- 目視検査の対象となる錠剤を搬送する搬送手段を備えた目視検査補助装置であって、
前記搬送手段により搬送される錠剤を撮像する撮像手段と、
前記撮像手段により得られた画像データに基づき、錠剤の良否を判定する判定手段と、
錠剤の目視検査を行う検査者から視認可能な位置に配置される表示手段と、
前記表示手段において、少なくとも前記判定手段により不良判定がなされた錠剤を特定するための情報を表示させることが可能な表示制御手段と、
前記判定手段により不良判定がなされた場合に、前記搬送手段による錠剤の搬送を一時的に停止させる搬送制御手段とを備えることを特徴とする目視検査補助装置。 - 前記表示制御手段は、前記表示手段において、少なくとも前記判定手段により不良判定がなされた錠剤を特定するための情報を該錠剤に係る前記画像データに対応付けて表示させることが可能に構成されていることを特徴とする請求項1に記載の目視検査補助装置。
- 前記撮像手段を制御する撮像制御手段を有し、
前記撮像制御手段は、錠剤の一時停止中に、少なくとも不良判定された錠剤に対応する範囲を撮像して、再判定用の前記画像データである再判定用画像データを得るように前記撮像手段を制御するとともに、
前記判定手段は、前記再判定用画像データに基づき、錠剤の良否を判定し、
前記搬送制御手段は、一時停止中の錠剤の搬送が再開されるように前記搬送手段を再起動させることが可能であり、
前記搬送制御手段による前記搬送手段の再起動は、前記再判定用画像データに基づく前記判定手段による判定結果が良であることを条件として行われるように構成されていることを特徴とする請求項1に記載の目視検査補助装置。 - 錠剤は、表面及び裏面を有するものであり、
前記搬送手段は、少なくとも一部に透明又は半透明な透過部が設けられるとともに、搬送される錠剤を支持する錠剤支持部を有し、
前記撮像手段は、前記錠剤支持部を上下に挟む位置に配置される第一撮像手段と第二撮像手段とを備え、
前記第一撮像手段によって錠剤の表面及び裏面のうちの一方が撮像され、前記第二撮像手段によって前記透過部を通して該錠剤の表面及び裏面のうちの他方が撮像されるように構成されており、
前記判定手段は、前記第一撮像手段及び前記第二撮像手段により得られた、錠剤の表面及び裏面に係る画像データに基づき、錠剤の良否を判定するように構成されていることを特徴とする請求項1乃至3のいずれか1項に記載の目視検査補助装置。 - 前記搬送手段は、錠剤を複数列に並んだ状態で搬送するものであり、
前記判定手段は、
入力される画像データから特徴量を抽出する符号化部、及び、該特徴量から画像データを再構成する復号化部を有するニューラルネットワークに対し、良品の錠剤に係る画像データに基づく学習データのみを学習させて生成した識別手段と、
前記撮像手段により得られた画像データに基づく元画像データを前記識別手段へ入力して再構成された画像データを再構成画像データとして取得可能な再構成画像データ取得手段と、
前記元画像データ及び前記再構成画像データを比較可能な比較手段とを備え、
前記比較手段による比較結果に基づき、錠剤の良否を判定可能に構成されており、
前記学習データは、良品の錠剤に係る画像データから1列分の複数個の錠剤に対応する領域を錠剤の列ごとに抽出して得た画像データであり、
前記元画像データは、前記撮像手段により得られた画像データから1列分の複数個の錠剤に対応する領域を錠剤の列ごとに抽出して得た画像データであることを特徴とする請求項1に記載の目視検査補助装置。 - 錠剤は、表面及び裏面と、該表面及び該裏面間に位置する側面とを有するものであり、
前記搬送手段は、搬送される錠剤の表裏を反転させる反転部を有し、
前記撮像手段は、前記反転部に位置する錠剤の側面を撮像する側面撮像手段を備え、
前記判定手段は、前記側面撮像手段により得られた画像データに基づき、錠剤の側面に関する良否を判定可能に構成されていることを特徴とする請求項1に記載の目視検査補助装置。
Priority Applications (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202280089109.XA CN118613714A (zh) | 2022-05-17 | 2022-11-24 | 目视检查辅助装置 |
| KR1020247022127A KR20240116523A (ko) | 2022-05-17 | 2022-11-24 | 육안 검사 보조 장치 |
| US18/888,881 US20250012735A1 (en) | 2022-05-17 | 2024-09-18 | Visual inspection assistance device |
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP2022-080921 | 2022-05-17 | ||
| JP2022080921A JP7330328B1 (ja) | 2022-05-17 | 2022-05-17 | 目視検査補助装置 |
Related Child Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| US18/888,881 Continuation US20250012735A1 (en) | 2022-05-17 | 2024-09-18 | Visual inspection assistance device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2023223579A1 true WO2023223579A1 (ja) | 2023-11-23 |
Family
ID=87577113
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/JP2022/043352 Ceased WO2023223579A1 (ja) | 2022-05-17 | 2022-11-24 | 目視検査補助装置 |
Country Status (5)
| Country | Link |
|---|---|
| US (1) | US20250012735A1 (ja) |
| JP (1) | JP7330328B1 (ja) |
| KR (1) | KR20240116523A (ja) |
| CN (1) | CN118613714A (ja) |
| WO (1) | WO2023223579A1 (ja) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP7642117B1 (ja) | 2024-02-26 | 2025-03-07 | Ckd株式会社 | 検査装置及びブリスタ包装機 |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR102742281B1 (ko) * | 2024-07-31 | 2024-12-16 | 주식회사 스마트링크 | 정제 알약 검사장치를 포함하는 알약 포장 검사시스템 |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0826229A (ja) * | 1994-07-19 | 1996-01-30 | Shionogi & Co Ltd | カプセル剤のバンドシール検出方法及びバンドシール検出装置 |
| JP2001095898A (ja) * | 1999-10-01 | 2001-04-10 | Hayashi Yakuhin Kikai Kk | 錠剤目視検査装置 |
| JP2006220448A (ja) * | 2005-02-08 | 2006-08-24 | Ckd Corp | 検査装置及びptp包装機 |
| JP2019039925A (ja) * | 2018-09-25 | 2019-03-14 | Ckd株式会社 | 外観検査装置及びブリスター包装機 |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4747313B2 (ja) | 2006-12-28 | 2011-08-17 | 公立大学法人高知工科大学 | 薄板状部品の検査方法と検査装置 |
| JP7312560B2 (ja) | 2019-01-31 | 2023-07-21 | 株式会社Screenホールディングス | 情報処理装置、情報処理方法、情報処理プログラム、学習方法および学習済モデル |
-
2022
- 2022-05-17 JP JP2022080921A patent/JP7330328B1/ja active Active
- 2022-11-24 KR KR1020247022127A patent/KR20240116523A/ko active Pending
- 2022-11-24 WO PCT/JP2022/043352 patent/WO2023223579A1/ja not_active Ceased
- 2022-11-24 CN CN202280089109.XA patent/CN118613714A/zh active Pending
-
2024
- 2024-09-18 US US18/888,881 patent/US20250012735A1/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0826229A (ja) * | 1994-07-19 | 1996-01-30 | Shionogi & Co Ltd | カプセル剤のバンドシール検出方法及びバンドシール検出装置 |
| JP2001095898A (ja) * | 1999-10-01 | 2001-04-10 | Hayashi Yakuhin Kikai Kk | 錠剤目視検査装置 |
| JP2006220448A (ja) * | 2005-02-08 | 2006-08-24 | Ckd Corp | 検査装置及びptp包装機 |
| JP2019039925A (ja) * | 2018-09-25 | 2019-03-14 | Ckd株式会社 | 外観検査装置及びブリスター包装機 |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP7642117B1 (ja) | 2024-02-26 | 2025-03-07 | Ckd株式会社 | 検査装置及びブリスタ包装機 |
| WO2025182156A1 (ja) * | 2024-02-26 | 2025-09-04 | Ckd株式会社 | 検査装置及びブリスタ包装機 |
| JP2025129540A (ja) * | 2024-02-26 | 2025-09-05 | Ckd株式会社 | 検査装置及びブリスタ包装機 |
Also Published As
| Publication number | Publication date |
|---|---|
| JP7330328B1 (ja) | 2023-08-21 |
| JP2023169667A (ja) | 2023-11-30 |
| US20250012735A1 (en) | 2025-01-09 |
| KR20240116523A (ko) | 2024-07-29 |
| CN118613714A (zh) | 2024-09-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP7405915B2 (ja) | 製品を分析する方法及び装置、訓練方法、システム、コンピュータプログラム、並びにコンピュータ読み取り可能記憶媒体 | |
| WO2023223579A1 (ja) | 目視検査補助装置 | |
| WO2004036198A1 (ja) | ガラス壜の検査装置における基準画像の作成方法及び装置 | |
| JP3332543B2 (ja) | 錠剤検査システム | |
| WO2004036197A1 (ja) | ガラス壜の検査装置 | |
| JP6749655B1 (ja) | 検査装置、異常検出方法、コンピュータプログラム、学習モデルの生成方法、及び学習モデル | |
| EP3399302A1 (en) | Egg surface inspection apparatus | |
| JP6696323B2 (ja) | パターン検査装置およびパターン検査方法 | |
| WO2014103617A1 (ja) | 位置合せ装置、欠陥検査装置、位置合せ方法、及び制御プログラム | |
| TW201937161A (zh) | 用於檢測玻璃片的設備及方法 | |
| JP5609433B2 (ja) | 円筒形状容器の検査方法 | |
| CN109297973B (zh) | 缺陷检查系统及缺陷检查方法 | |
| JP3974015B2 (ja) | 物品検査システム | |
| US10852244B2 (en) | Image processing apparatus, image processing method, and recording medium | |
| JP2006214814A (ja) | 検査装置及びptp包装機 | |
| JP7786050B2 (ja) | 検査装置、検査方法、ガラス板の製造方法、および検査プログラム | |
| JP2019066198A (ja) | 容器口部検査装置及び容器口部検査方法 | |
| JPH06258226A (ja) | 錠剤の外観検査方法 | |
| JP7309017B1 (ja) | 錠剤検査装置及び錠剤検査方法 | |
| JP7642117B1 (ja) | 検査装置及びブリスタ包装機 | |
| JP7632622B2 (ja) | 物体判別装置 | |
| US20250209597A1 (en) | Substrate inspecting apparatus and method of inspecting substrate | |
| JP2025084454A (ja) | 検査システム | |
| JP2010151463A (ja) | オフライン演算機能を備えた検査機 | |
| JP2025023415A (ja) | 検査装置、検査方法及びプログラム |
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: 22942780 Country of ref document: EP Kind code of ref document: A1 |
|
| ENP | Entry into the national phase |
Ref document number: 20247022127 Country of ref document: KR Kind code of ref document: A |
|
| WWE | Wipo information: entry into national phase |
Ref document number: 202280089109.X Country of ref document: CN |
|
| WWE | Wipo information: entry into national phase |
Ref document number: 202447087656 Country of ref document: IN |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 22942780 Country of ref document: EP Kind code of ref document: A1 |