Visual inspection device

Abstract

The appearance inspection apparatus provided by the present invention comprises: a transport body (1) having a first main surface and a second main surface, wherein the second main surface is opposite to the first main surface; a supply unit (2) supplying an object to be inspected (20) onto the second main surface of the transport body; a drive unit (3) for driving the transport body to transport the object to be inspected; a first imaging unit capable of photographing the object to be inspected on the transport body using imaging light having a wavelength that transmits through the transport body for appearance inspection; an obstacle extraction unit for extracting obstacles contained in the image captured by the first imaging unit based on optical characteristic values ​​of obstacles contained in the transport body; and a rating calculation unit for calculating the rating of the obstacles extracted by the obstacle extraction unit. The rating calculation unit calculates the rating of the obstacles based on the cumulative value of data of multiple obstacles extracted by the obstacle extraction unit.

Description

Technical Field

[0001] This invention relates to an appearance inspection apparatus for inspecting the appearance of an object. Background Technology

[0002] A device is known to inspect the appearance of an object to be inspected by placing it on a transport platform or other transport body and taking pictures of the object with a photographing device.

[0003] As such an appearance inspection device, Patent Document 1 discloses an appearance inspection device that places the object to be inspected on a circular transport platform made of transparent glass, rotates the transport platform to transport the object to be inspected, and takes pictures of the appearance of each surface of the object to be inspected with a camera.

[0004] Patent Document 1: Japanese Patent Application Publication No. 2011-133458

[0005] However, in the visual inspection apparatus described in Patent Document 1, there is a possibility that scratches and dirt on the transport table may be mistaken for scratches and dirt on the object being inspected, thus wrongly determining a qualified object to be unqualified. That is, when photographing the object being inspected on a transparent transport table using a camera positioned below the transport table on the opposite side from the object being inspected, there is a possibility that scratches and dirt on the transport table, which overlaps with the object being inspected, may be mistaken for scratches and dirt on the object being inspected. Therefore, to prevent such misjudgments during visual inspection, it is preferable to clean and replace the transport table according to the level of photographic obstacles such as scratches and dirt on the transport table. However, Patent Document 1 does not disclose a method for calculating the level of photographic obstacles. Summary of the Invention

[0006] The present invention addresses the aforementioned problems and aims to provide an appearance inspection device capable of extracting photographic obstacles such as scratches and dirt from a transported object and calculating their severity.

[0007] The appearance inspection apparatus of the present invention is an appearance inspection apparatus for performing appearance inspection of an object to be inspected, characterized in that it comprises:

[0008] The transporter has a first main surface and a second main surface, wherein the second main surface is opposite to the first main surface;

[0009] The supply department supplies the inspection object to the second main surface of the aforementioned transporter.

[0010] The drive unit moves the object to be inspected on the transport body by driving the transport body.

[0011] The first imaging unit is capable of photographing the object to be inspected on the transport body by means of imaging light having a wavelength that can pass through the transport body in order to perform an appearance inspection of the object to be inspected.

[0012] The obstacle extraction unit extracts the obstacles contained in the image captured by the first imaging unit based on optical characteristic values ​​of the obstacles contained in the transporter; and

[0013] The rating calculation unit calculates the rating of the shooting obstacles extracted by the shooting obstacle extraction unit.

[0014] The aforementioned level calculation unit calculates the level of the photographing obstacle based on the cumulative value of data from multiple photographing obstacles extracted by the aforementioned photographing obstacle extraction unit.

[0015] According to the visual inspection apparatus of the present invention, since it extracts photographic obstacles contained in the captured image and calculates the level of the photographic obstacles based on the cumulative value of data from multiple extracted photographic obstacles, it is possible to appropriately calculate the level of photographic obstacles such as scratches and dirt on the transporter. The level of photographic obstacles can, for example, serve as an indicator for the maintenance of the transporter. Therefore, based on the calculated level of photographic obstacles, it is possible to perform, for example, cleaning or replacing the transporter at appropriate intervals. Attached Figure Description

[0016] Figure 1 This is a schematic diagram illustrating the structure of an appearance inspection device in one embodiment.

[0017] Figure 2 This is a perspective view showing the external shape of a multilayer ceramic capacitor, which is an example of an object under inspection.

[0018] Figure 3 This diagram shows the configuration of a first imaging section for photographing the first main surface of a multilayer ceramic capacitor, an example of an object to be inspected, and a second imaging section for photographing the second main surface.

[0019] Figure 4 This diagram shows the configuration of a third imaging section for photographing the first end face of a multilayer ceramic capacitor, an example of an object to be inspected, and a fourth imaging section for photographing the second end face.

[0020] Figure 5 This diagram shows the configuration of the fifth photographing section for photographing the first side of a multilayer ceramic capacitor, an example of an object to be inspected, and the sixth photographing section for photographing the second side.

[0021] Figure 6 It is a block diagram that schematically represents the functions of the processing unit.

[0022] Figure 7 It is a diagram showing the area in the image captured by the first imaging unit, excluding the area where the object to be inspected is placed.

[0023] Figure 8 This diagram illustrates an example of a method for sorting extracted photographic obstacles into scratches or dirt.

[0024] Figure 9 This is a diagram illustrating an example of a method for sorting scratches and dirt based on coordinate positions on an image.

[0025] Figure 10 (a) is a schematic diagram showing the brightness distribution of the R, G, and B images when the optical axis of the first imaging unit and the optical axis of the first illumination unit are aligned; (b) is a schematic diagram showing the brightness distribution of the R, G, and B images when the optical axis of the green illumination light in the illumination light of the first illumination unit is deviated; and (c) is a schematic diagram showing the brightness distribution of the R, G, and B images when the optical axis of the first imaging unit is deviated.

[0026] Explanation of reference numerals in the attached figures

[0027] 1…Transporter; 2…Supply Unit; 3…Drive Unit; 4…Processing Unit; 4a…Obstacle Extraction Unit; 4b…Sorting Unit; 4c…Grade Calculation Unit; 4d…Pass Rate Determination Unit; 4e…Storage Unit; 5…Alarm Unit; 6…Arrangement Unit; 7…Discharge Unit; 11…First Imaging Unit; 12…Second Imaging Unit; 13…Third Imaging Unit; 14…Fourth Imaging Unit; 15…Fifth Imaging Unit; 16…Sixth Imaging Unit; 20…Inspection Object; 20X…Layered Ceramic Capacitor; 30…Obstacle… Obstacles; 30a…Shooting obstacles sorted as scratches; 30b…Shooting obstacles sorted as dirt; 31…First illumination unit; 32…Second illumination unit; 33…Third illumination unit; 34…Fourth illumination unit; 35…Fifth illumination unit; 36…Sixth illumination unit; 41…First reflector; 42…Second reflector; 43…Third reflector; 50…Image; 51…The area in the image excluding the area containing the object to be inspected; 71…Air ejector; 100…Appearance inspection device. Detailed Implementation

[0028] The following describes embodiments of the present invention and provides a detailed explanation of its features.

[0029] Figure 1This is a schematic diagram illustrating the structure of an appearance inspection device 100 in one embodiment. The appearance inspection device 100 in one embodiment is a device for performing appearance inspection of an object 20, and includes a transport body 1, a supply unit 2, a drive unit 3, a processing unit 4, and a first imaging unit 11. Furthermore, the first imaging unit 11 is positioned opposite the second imaging unit 12 relative to the transport body 1 (see reference 1). Figure 3 The appearance inspection device 100 in this embodiment also includes an alarm unit 5, an arrangement unit 6, an exhaust unit 7, a second camera unit 12, a third camera unit 13, a fourth camera unit 14, a fifth camera unit 15, and a sixth camera unit 16.

[0030] The object of inspection 20 may be, for example, electronic components such as multilayer ceramic capacitors, inductors, thermistors, and module substrates, or semi-finished products in the process of manufacturing electronic components. However, the object of inspection 20 is not limited to electronic components or semi-finished products.

[0031] Figure 2 This is a perspective view showing the external shape of a multilayer ceramic capacitor 20X, which is an example of the object under inspection 20. (See image below.) Figure 2 As shown, the multilayer ceramic capacitor 20X has a ceramic blank 21, a first external electrode 22a and a second external electrode 22b.

[0032] The multilayer ceramic capacitor 20X has a hexahedral shape. That is, the multilayer ceramic capacitor 20X has a first end face 23a and a second end face 23b that are opposite each other, a first main face 24a and a second main face 24b that are opposite each other, and a first side face 25a and a second side face 25b that are opposite each other.

[0033] The transport body 1 has a first main surface 1a and a second main surface 1b opposite to the first main surface 1a (see reference). Figures 3-5 The transporter 1 is not particularly limited in structure or shape as long as it can transport the object to be inspected 20 via the visual inspection device 100. In this embodiment, the transporter 1 is an electrostatic induction adsorption type transporter that uses electrostatic induction to adsorb the object to be inspected 20 onto the second main surface 1b, which serves as the transport surface, for transport. However, the transporter 1 may also have a structure that transports the object to be inspected 20 without adsorbing it.

[0034] In this embodiment, the transport body 1 has a transparent area where at least the object to be inspected 20 is placed, and it has a shape that can rotate about a central axis X1 orthogonal to the first main surface 1a and the second main surface 1b. Specifically, the transport body 1 has a transparent and rotatable frustum-shaped shape, and its diameter is, for example, 100 mm or more and 1000 mm or less.

[0035] In this embodiment, the transport body 1 is made of a transparent insulating material, such as transparent glass materials like borosilicate glass, quartz glass, or float glass, or transparent resin materials like polyethylene terephthalate resin or acrylic resin. However, it is not necessary for the entire transport body 1 to be made of a transparent insulating material; at least the area where the object to be inspected 20 is placed may be made of a transparent insulating material. Furthermore, as will be described later, "transparent" here refers to transmittance that allows the light emitted during inspection of the object to be inspected 20 to pass through, and does not necessarily mean transparent to visible light.

[0036] The transporter 1 can be a single-layer structure or a multi-layer structure. Furthermore, if the transporter 1 is an electrostatic induction adsorption transporter, it may also include a conductive layer. In this case, the conductive layer can be configured as a transparent electrode made of indium tin oxide (ITO), antimony-doped tin oxide (ATO), titanium oxide, graphene, or the like.

[0037] In this embodiment, as described later, the object to be inspected 20, placed on the second main surface 1b of the transporter 1, is photographed from the first main surface 1a side by the first imaging unit 11 for visual inspection. Therefore, the area in the transporter 1 where at least the object to be inspected 20 is placed has a transmittance sufficient to allow the photographic light irradiated during inspection of the object to be inspected 20 to pass through. Higher transmittance is preferable to improve the accuracy of the visual inspection.

[0038] When the transporter 1 is an electrostatic induction adsorption type transporter, the first main surface 1a is positively or negatively charged by a charging device or the like. Here, we will describe a structure where the first main surface 1a is positively charged. In this case, a negative charge appears on the first main surface 1a side and a positive charge appears on the second main surface 1b side inside the transporter 1. In addition, a negative charge appears on the first main surface 24a side of the object to be inspected 20 placed on the second main surface 1b of the transporter 1. As a result, an electrostatic adsorption force is generated between the object to be inspected 20 and the transporter 1, and the object to be inspected 20 is adsorbed onto the second main surface 1b of the transporter 1.

[0039] The supply unit 2 supplies the inspection object 20 onto the second main surface 1b of the transport body 1. The term "second main surface 1b" does not refer to the upper part in the vertical direction, but rather to the surface of the second main surface 1b itself, regardless of direction. The supply unit 2 can be, for example, a vibratory feeder that supplies the inspection object 20 by vibration, a rotary feeder that supplies the inspection object 20 while rotating, a pneumatic feeder that supplies the inspection object 20 pneumatically, a conveyor belt feeder that supplies the inspection object 20 via a conveyor belt, or an individual loading mechanism that supplies the inspection object 20 one by one.

[0040] The supply unit 2 supplies the inspection objects 20 to the transport body 1 at regular time intervals. The supply rate of the supply unit 2 is, for example, 50 to 30,000 per minute. The supply unit 2 can supply the inspection objects 20 in one row or in multiple rows of two or more rows.

[0041] In this embodiment, the object to be inspected 20 supplied to the transporter 1 is attracted to the transporter 1 by electrostatic induction, and as... Figure 1 As shown, they are arranged in a row by the arranging section 6. That is, multiple inspection objects 20 are arranged in a row with their orientations aligned by contacting the arranging section 6. Here, the orientations of the stacked ceramic capacitors 20X, which are inspection objects 20, are aligned, so that the directions of the first external electrode 22a and the second external electrode 22b are aligned with the transport direction.

[0042] As described above, the transport body 1 in this embodiment has a rotatable frustum shape, and the object to be inspected 20 on the second main surface 1b is transported by rotation. Since centrifugal force is applied to the object to be inspected 20, the rotational speed of the transport body 1 cannot be set too high to prevent the object to be inspected 20 from shifting position and falling off the second main surface 1b. However, since the object to be inspected 20 is transported in a state where it is attracted to the second main surface 1b of the transport body 1 by electrostatic induction, the positional shift of the object to be inspected 20 and its fall off the second main surface 1b during transport can be suppressed. Therefore, the rotational speed of the transport body 1 can be made higher. That is, the object to be inspected 20 is preferably transported in a state where it is attracted to the transport body 1.

[0043] Furthermore, in the method of transporting the object 20 to be inspected in a state of electrostatic adsorption, the transport body 1 can be used in a state of being tilted relative to the horizontal plane, rather than in a horizontal state.

[0044] The drive unit 3 moves the object to be inspected 20 on the transport body 1 by driving the transport body 1. The drive unit 3 can be, for example, a DC motor, a servo motor, a stepper motor, a linear motor, an electromagnetic motor, an ultrasonic motor, or compressed air. The drive of the transport body 1 can be continuous or intermittent. When the transport body 1 is driven continuously, the speed is, for example, 100 mm / s or more and 2000 mm / s, and as an example, 800 mm / s. Furthermore, the drive unit 3 can be set to any position.

[0045] The appearance inspection device 100 is configured to perform an appearance inspection of the object 20 being inspected while it is being transported by the transporter 1. The appearance inspection of the object 20 is performed by the first imaging unit 11 to the sixth imaging unit 16 and the processing unit 4.

[0046] The first imaging unit 11 to the sixth imaging unit 16 each capture images of the object 20 to be inspected on the transporter 1. The first imaging unit 11 to the sixth imaging unit 16 are not particularly limited in that they are only required to capture images of the object 20; for example, a CCD camera or a CMOS camera can be used. The shooting speed of the first imaging unit 11 to the sixth imaging unit 16 is, for example, 10 to 400 frames per second, or 150 frames per second as an example. The shooting area is, for example, 1mm × 1mm to 10mm × 10mm, or 5mm × 5mm as an example. When the size of the image along the travel direction of the transporter 1 is set to Wa, the distance the transporter 1 travels when capturing two images consecutively is a distance of 0.5Wa to 5Wa.

[0047] Here, the object of inspection 20 is the aforementioned stacked ceramic capacitor 20X, and the structure of the CCD camera used in the first imaging unit 11 to the sixth imaging unit 16 will be described. Furthermore, in Figure 1 The second camera section 12 to the sixth camera section 16, which are located above the transport body 1 (on the side of the second main surface 1b of the transport body 1), are shown. The first camera section 11, which is located below the transport body 1 (on the side of the first main surface 1a of the transport body 1), is not shown.

[0048] exist Figure 1 In the image, the first shooting area SA1 is the area for shooting the first side surface 25a and the second side surface 25b of the multilayer ceramic capacitor 20X. The second shooting area SA2 is the area for shooting the first end surface 23a and the second end surface 23b of the multilayer ceramic capacitor 20X. The third shooting area SA3 is the area for shooting the first main surface 24a and the second main surface 24b of the multilayer ceramic capacitor 20X.

[0049] Figure 3 This diagram shows the configuration of a first imaging unit 11 for photographing the first main surface 24a of the multilayer ceramic capacitor 20X and a second imaging unit 12 for photographing the second main surface 24b. The first imaging unit 11 is positioned away from the transport body 1 in a direction orthogonal to the first main surface 1a and is located on the side of the first main surface 1a. The second imaging unit 12 is positioned away from the transport body 1 in a direction orthogonal to the second main surface 1b of the object under inspection and is located on the side of the second main surface 1b.

[0050] In this embodiment, such as Figure 3As shown, the transport body 1 is horizontally arranged, with a first imaging unit 11 located below the transport body 1 and a second imaging unit 12 located above the transport body 1. Therefore, the first imaging unit 11 captures images of the first main surface 24a of the stacked ceramic capacitor 20X mounted on the second main surface 1b of the transport body 1 from the side of the first main surface 1a, i.e., below. Conversely, the second imaging unit 12 captures images of the second main surface 24b of the stacked ceramic capacitor 20X from the side of the second main surface 1b, i.e., above.

[0051] In this embodiment, a first illumination unit 31 is provided to illuminate the first main surface 24a of the multilayer ceramic capacitor 20X, and a second illumination unit 32 is provided to illuminate the second main surface 24b. The first illumination unit 31 and the second illumination unit 32 are, for example, hemispherical dome illuminations, with the aperture for allowing the imaging light to pass through located near the center.

[0052] Figure 4 This diagram shows the configuration of the third imaging unit 13 for photographing the first end face 23a of the multilayer ceramic capacitor 20X, and the fourth imaging unit 14 for photographing the second end face 23b. The third imaging unit 13 and the fourth imaging unit 14 are positioned away from the transport body 1 in a direction orthogonal to the second main face 1b of the object under inspection 20, and are located on the side of the second main face 1b. In this embodiment, as... Figure 4 As shown, since the transport body 1 is arranged horizontally, the third camera unit 13 and the fourth camera unit 14 are arranged above the transport body 1.

[0053] In this embodiment, a third illumination section 33 for illuminating the first end face 23a of the multilayer ceramic capacitor 20X with light, and a fourth illumination section 34 for illuminating the second end face 23b with light are provided. The third illumination section 33 and the fourth illumination section 34 can use the same structure as the first illumination section 31 and the second illumination section 32.

[0054] like Figure 4 As shown, the imaging surface of the third imaging unit 13 faces vertically downward. A first reflector 41 for altering the light path is provided vertically below the imaging surface of the third imaging unit 13. The first reflector 41 is, for example, a mirror or a prism. The third imaging unit 13 images the first end face 23a of the stacked ceramic capacitor 20X by receiving light whose light path has been altered by the first reflector 41.

[0055] like Figure 4 As shown, the imaging surface of the fourth imaging unit 14 faces vertically downward. A second reflector 42 for altering the light path is provided vertically below the imaging surface of the fourth imaging unit 14. The second reflector 42 is, for example, a mirror or a prism. The fourth imaging unit 14 images the second end face 23b of the stacked ceramic capacitor 20X by receiving light whose light path has been altered by the second reflector 42.

[0056] Figure 5 This diagram shows the configuration of the fifth imaging unit 15 for photographing the first side 25a of the multilayer ceramic capacitor 20X and the sixth imaging unit 16 for photographing the second side 25b. The fifth imaging unit 15 is positioned away from the transport body 1 in a direction orthogonal to the second main surface 1b of the object under inspection 20, and is located on the side of the second main surface 1b. In this embodiment, as... Figure 5 As shown, since the transport body 1 is horizontally arranged, the fifth imaging unit 15 is positioned above the transport body 1. Furthermore, the sixth imaging unit 16 is positioned radially outward from the transport body 1, and at approximately the same height as the stacked ceramic capacitor 20X.

[0057] In this embodiment, a fifth illumination unit 35 is provided for illuminating the first side 25a of the stacked ceramic capacitor 20X with light, and a sixth illumination unit 36 ​​is provided for illuminating the second side 25b with light. The fifth illumination unit 35 and the sixth illumination unit 36 ​​can use the same structure as the first illumination unit 31 and the second illumination unit 32.

[0058] like Figure 5 As shown, the imaging surface of the fifth imaging unit 15 faces vertically downward. A third reflector 43 for altering the light path is provided vertically below the imaging surface of the fifth imaging unit 15. The third reflector 43 is, for example, a mirror or a prism. The fifth imaging unit 15 images the first side surface 25a of the stacked ceramic capacitor 20X by receiving light whose light path has been altered by the third reflector 43.

[0059] like Figure 5 As shown, the imaging surface of the sixth imaging unit 16 faces the second side 25b of the stacked ceramic capacitor 20X. That is, the sixth imaging unit 16 directly receives imaging light to image the second side 25b of the stacked ceramic capacitor 20X.

[0060] The appearance inspection of the object 20 is performed by the processing unit 4 based on images captured by the first imaging unit 11 to the sixth imaging unit 16. The processing unit 4 is, for example, a personal computer or a dedicated image processing controller. In the appearance inspection based on the captured images, for example, the external dimensions of the object 20, the dimensions of specific parts, the presence or absence of surface irregularities, the presence or absence of foreign matter, the presence or absence of damage, and the presence or absence of discoloration are investigated.

[0061] Alternatively, the structure may omit the second camera section 12 to the sixth camera section 16 from the first camera section 11 to the sixth camera section 16, or the structure may omit the third camera section 13 to the sixth camera section 16. Furthermore, the structure may include camera sections different from the first camera section 11 to the sixth camera section 16.

[0062] The inspection object 20, which has been photographed for visual inspection, is discharged from the second main surface 1b of the transport body 1 by the discharge unit 7. The length of the transport path from when the inspection object 20 is supplied to the second main surface 1b of the transport body 1 to when it is discharged is, for example, 300 mm or more and 2500 mm or less.

[0063] In this embodiment, the discharge section 7 includes an air ejection section 71. That is, air is ejected towards the object to be inspected 20 via the air ejection section 71, discharging the object to be inspected 20 from the second main surface 1b of the transport body 1. However, the method of discharging the object to be inspected 20 is not limited to using air ejection; for example, it can also be discharged by suction using a suction mechanism (not shown). Alternatively, the object to be inspected 20 can be picked up one by one and discharged. Furthermore, it can be discharged based on the results of visual inspection, allowing qualified and unqualified products to be collected in different containers.

[0064] Figure 6 This is a block diagram schematically illustrating the functions of processing unit 4. Figure 6 The alarm unit 5, described later, is also shown together with the processing unit 4. The processing unit 4 functions as an obstacle extraction unit 4a and a grade calculation unit 4c. The obstacle extraction unit 4a extracts the obstacles contained in the image captured by the first imaging unit 11 based on the optical characteristic values ​​of the obstacles contained in the transporter 1. The grade calculation unit 4c calculates the grade of the obstacles extracted by the obstacle extraction unit 4a. The processing unit 4 may also function as a sorting unit 4b and a pass rate determination unit 4d, described later. The processing unit 4 may also have a storage unit 4e that stores the images captured by the first imaging unit 11. However, the obstacle extraction unit 4a and the grade calculation unit 4c may be provided separately from the processing unit 4. In addition, the sorting unit 4b and the pass rate determination unit 4d may be provided separately from the processing unit 4.

[0065] The aforementioned "imaging obstacle included in the transport body 1" refers to an object that obstructs the imaging process used for visual inspection of the object 20, such as scratches or dirt on the transport body 1. Furthermore, the so-called "optical characteristic value related to the imaging obstacle included in the transport body 1" is, for example, the brightness of the imaging obstacle in the captured image. Using the obtained brightness information, it is possible to use the area of ​​brightness, the coordinates of brightness, maximum brightness, minimum brightness, differential value (the change in brightness per unit distance), color information, the difference in brightness with the average brightness of the background, and the difference in brightness with any specified brightness, etc.

[0066] In this embodiment, the obstacle extraction unit 4a extracts obstacles based on the optical characteristic values ​​contained in the image captured by the first imaging unit 11 and stored in the storage unit 4e. That is, the first imaging unit 11 is used not only for visual inspection of the object 20 but also for extracting obstacles. Therefore, since there is no need to provide new equipment for extracting obstacles, the size of the visual inspection device 100 can be suppressed.

[0067] Furthermore, since the first imaging unit 11 captures the inspection object 20 on the transporter 1 from the first main surface 1a side of the transporter 1, it can obtain an image focused on the first main surface 1a as an image for extracting images of obstacles. That is, since the second imaging unit 12 captures images with the second main surface 24b of the inspection object 20 on the transporter 1 focused, there are cases where the focus is not accurately aligned with the second main surface 1b of the transporter 1, depending on the height of the inspection object 20. However, since the first imaging unit 11 captures images with the focus aligned with the first main surface 24a of the inspection object 20 on the transporter 1, it can obtain an image that is also focused on the first main surface 1a of the transporter 1 at approximately the same distance from the first imaging unit 11.

[0068] The following is a detailed explanation of the method for extracting obstacles during photography.

[0069] In order to capture obstacles, the first imaging unit 11 captures a wider area than the area where the object to be inspected 20 is placed on the transporter 1. For example... Figure 7 As shown, the obstacle extraction unit 4a extracts the obstacle within a range 51 from the image 50 captured by the first imaging unit 11, excluding the area where the object to be inspected 20 is placed. However, even if the object to be inspected 20 is not captured in the image 50 captured by the first imaging unit 11, the obstacle can be extracted within the entire range of the captured image 50.

[0070] That is, if the first imaging unit 11 only captures the area on the transporter 1 where the object to be inspected 20 is placed, it is difficult to determine whether the obstacle in the image captured from the first main surface 1a side of the transporter 1 originates from the transporter 1 or from the object to be inspected 20 if the object to be inspected 20 is placed on the transporter 1. However, since the first imaging unit 11 captures a wider area than the area on the transporter 1 where the object to be inspected 20 is placed, it is possible to capture obstacles that exist within the range 51 after removing the area where the object to be inspected 20 is placed, even when the object to be inspected 20 is placed on the transporter 1. Furthermore, since the processing unit 4 extracts the obstacles within the range 51 after removing the area where the object to be inspected 20 is placed from the image captured by the first imaging unit 11, it is able to extract the obstacles contained in the transporter 1 with high precision.

[0071] In this embodiment, the average brightness of the area 51 after removing the area where the object to be inspected 20 is placed from the image 50 captured by the first imaging unit 11 and stored in the storage unit 4e is calculated, and areas whose brightness is higher than the calculated average brightness by a predetermined value and whose area is higher than a predetermined area are extracted as imaging obstacles. The predetermined value is, for example, 10, and the predetermined area is, for example, 10 μm. 2 Here, the difference between the brightness of the obstacle being photographed and the average brightness of the range 51 excluding the range containing the object under inspection 20 is called the brightness difference of the obstacle being photographed.

[0072] The sorting unit 4b sorts the shooting obstacles extracted by the shooting obstacle extraction unit 4a into scratches on the transport body 1 and dirt adhering to the transport body 1. For example, such as Figure 8 As shown, the sorting unit 4b sorts the elongated photographic obstacles 30a extracted by the photographic obstacle extraction unit 4a as scratches. That is, the elongated photographic obstacles 30a are considered as scratches generated during the cleaning of the surface of the transport body 1 and are distinguished from dirt. Specifically, when the direction in which the elongated shape extends is taken as the first direction and the direction orthogonal to the first direction is taken as the second direction, photographic obstacles 30a whose ratio L1 / L2 of the size L1 in the first direction to the size L2 in the second direction is 3 or more are sorted as scratches generated on the transport body 1. In addition, the sorting unit 4b sorts the photographic obstacles 30b that were not sorted as scratches as dirt adhering to the transport body 1.

[0073] Furthermore, the method for sorting extracted obstacles into scratches and dirt is not limited to the methods described above. For example, obstacles that remain even after wiping the surface of the transporter 1 can be sorted into scratches. Alternatively, multiple colors of illumination can be used for imaging, and the captured images, such as R-images, G-images, and B-images, can be differentiated by color to sort scratches and dirt. That is, as with R-images, G-images, and B-images, if it is known in advance that scratches or dirt are easily detected in a specific color image, scratches and dirt can be sorted based on that specific color image.

[0074] Additionally, scratches and dirt can be sorted based on their coordinate positions on the image. For example, ... Figure 9 As shown, if there is a tendency for scratches to occur at the edge of the object 20 along the transport direction at the location where the object 20 is placed, then the photographing obstacle 30a at that coordinate position is classified as a scratch. In this case, the photographing obstacle 30a located at the edge of the object 20 along the transport direction, and which is elongated in shape, can also be classified as a scratch.

[0075] Furthermore, when taking pictures to extract obstacles, a dedicated lighting unit, different from the first lighting unit 31 used for visual inspection, can be used. That is, if it is known in advance that scratches or dirt are easy to detect in images taken using lighting of a specific color, scratches and dirt can be sorted with high precision by using dedicated lighting for taking pictures.

[0076] Alternatively, after cleaning the surface of the transport body 1, a dedicated inspection for sorting out scratches can be performed without transporting the object 20 to be inspected. After cleaning the surface of the transport body 1, since the transport body 1 is free of dirt or has almost no dirt, the shooting obstacles extracted by the shooting obstacle extraction unit 4a can be sorted into scratches with high precision.

[0077] In this way, the sorting unit 4b sorts the shooting obstacles extracted by the shooting obstacle extraction unit 4a into scratches on the transport body 1 and dirt attached to the transport body 1, so that the transport body 1 can be cleaned and replaced appropriately. That is, if the transport body 1 contains scratches, the transport body 1 needs to be replaced, but if it does not contain scratches, the transport body 1 does not need to be replaced, and cleaning is sufficient.

[0078] As described above, the grade calculation unit 4c calculates the grade of the photographed obstacle based on the cumulative value of data from multiple photographed obstacles extracted by the photographed obstacle extraction unit 4a. The grade of the photographed obstacle refers to the degree of photographing the obstacle, such as the quantity and size of the obstacle. In this embodiment, the grade calculation unit 4c calculates the grade of the photographed obstacle separately for scratches and dirt sorted by the sorting unit 4b.

[0079] Here, multiple obstacles are extracted from multiple images captured by the first imaging unit 11, and the level of the obstacles is calculated. The number of images used to calculate the level of the obstacles is preferably large, for example, between 100 and 10,000 images; 1,000 images are an example. This number is, for example, the number of images captured by the first imaging unit 11 during one rotation of the transporter 1 in the rotational direction. However, when calculating the level of the obstacles, it is not necessary to use all the images captured by the first imaging unit 11 during one rotation of the transporter 1 in the rotational direction; the images can be selected by extracting them at regular intervals. By calculating the level of the obstacles based on the cumulative value of the data of multiple obstacles extracted from multiple images captured during one rotation of the transporter 1, the level of the obstacles can be calculated with high accuracy.

[0080] In this embodiment, the level calculation unit 4c calculates the level of the obstacle based on the brightness and area of ​​the obstacle included in the image captured by the first imaging unit 11. Specifically, the level calculation unit 4c calculates the level of the obstacle based on the cumulative value of the product of the brightness difference and the area of ​​each extracted obstacle. For example, 100 obstacles are extracted from 1000 images, and the brightness difference of each obstacle is 10 and the area is 1000 μm. 2 In this case, the cumulative value of the product of the brightness difference and area of ​​each photographed obstacle is 10 × 1000 × 100, which equals 1,000,000. The more obstacles are photographed, and the wider the area of ​​the obstacles, the larger this cumulative value becomes. When using 1,000 images, the above cumulative value is, for example, more than 1,000 but less than 1 million.

[0081] Regarding the cumulative value of the product of the brightness difference and area of ​​each of the aforementioned obstacles, since its value varies depending on the size of the captured image, it is preferable to divide the product of the brightness difference and area of ​​each obstacle by the size of the captured image; more specifically, to divide by a normalized area of ​​the region from which the obstacles are extracted. By performing normalization, the cumulative value of the obstacles per unit area can be extracted regardless of the image size.

[0082] The level calculation unit 4c calculates the level of the obstacle based on its brightness and area, enabling a higher accuracy in determining the level of the obstacle.

[0083] For example, when a predetermined number of objects 20 undergoing visual inspection are referred to as a batch, the cumulative value can be calculated using a predetermined number of images from the start of visual inspection of a batch, or by tracing back from the end of visual inspection of a batch to calculate the cumulative value using a predetermined number of images. Furthermore, since the rotation of the transporter 1 is controlled by a rotation angle centered on the central axis X1, the cumulative value can be calculated based on the rotation angle using images taken during one revolution of the transporter 1. Alternatively, multiple images can be acquired and combined into a single image based on the maximum brightness of the same pixel in each image to obtain the cumulative value, or the cumulative value can be calculated by adding the brightness of each pixel in multiple images and dividing by the number of images.

[0084] Next, the rating calculation unit 4c calculates the average value of the accumulated value distributed among each image based on the number of images used in calculating the rating of the captured obstacle, and uses the calculated average value as the rating of the captured obstacle. The more scratches, dirt, and other captured obstacles there are, and the larger the size of the captured obstacle, the higher the rating value of the captured obstacle. In the example above, since 1000 images were used, the rating of the captured obstacle is 1000000 / 1000 = 1000. Therefore, the rating of the captured obstacle as a whole of the transport body 1 can be numerically quantified.

[0085] As described above, in this embodiment, the rating calculation unit 4c calculates the rating of the shooting obstacle for each scratch and dirt sorted by the sorting unit 4b. However, it is also possible to calculate the rating of the shooting obstacle including scratches and dirt, rather than for each scratch and dirt.

[0086] Next, the pass rate determination unit 4d calculates the pass rate of the inspected object 20 that passed the visual inspection. As an example, the pass rate of the inspected object 20 can be calculated by (1 - number of defects / number of inspections) × 100. Here, the number of defects refers to the number of inspected objects 20 that are determined to be "defective" based on at least one of the following surfaces: the first end face 23a, the second end face 23b, the first main face 24a, the second main face 24b, the first side face 25a, and the second side face 25b, in visual inspections based on images captured by the first imaging unit 11 to the sixth imaging unit 16. However, the pass rate of any surface can also be calculated based on images obtained from any surface of the inspected object 20, including the first end face 23a, the second end face 23b, the first main face 24a, the second main face 24b, the first side face 25a, and the second side face 25b.

[0087] In this embodiment, the pass rate determination unit 4d determines the pass rate of the inspection object 20 based on the images of the inspection object 20 captured by the first imaging unit 11 and the second imaging unit 12 stored in the storage unit 4e. Specifically, the pass rate determination unit 4d calculates the pass rate of the first main surface 24a of the inspection object 20 (hereinafter referred to as the lower surface pass rate) based on the image of the inspection object 20 captured by the first imaging unit 11, and calculates the pass rate of the second main surface 24b of the inspection object 20 (hereinafter referred to as the upper surface pass rate) based on the image of the inspection object 20 captured by the second imaging unit 12. The lower surface pass rate and the upper surface pass rate are both examples of the pass rate of the inspection object 20. In the presence of imaging obstacles, the lower surface pass rate tends to be lower than the upper surface pass rate due to the influence of the imaging obstacles. Furthermore, the criteria used here to determine the pass rate of the upper surface and the pass rate of the lower surface for sorting into pass and fail does not necessarily have to be the same as the criteria used in the visual inspection of the object 20. Different criteria used to obtain information related to the photographing of obstacles can also be used.

[0088] The pass rate for the lower surface is calculated by (1 - number of lower surface defects / number of lower surface inspections) × 100. The number of lower surface inspections is the number of visual inspections performed on the first main surface 24a of the inspected object 20, and the number of lower surface defects is the number of items judged to be defective during the visual inspection of the first main surface 24a. Similarly, the pass rate for the upper surface is calculated by (1 - number of upper surface defects / number of upper surface inspections) × 100. The number of upper surface inspections is the number of visual inspections performed on the second main surface 24b of the inspected object 20, and the number of upper surface defects is the number of items judged to be defective during the visual inspection of the second main surface 24b. For example, if visual inspections are performed on 1000 inspected objects 20, and based on images captured by the first imaging unit 11, the lower surface pass rate is 98% if 20 inspected objects 20 have their first main surface 24a judged to be defective. Furthermore, in cases where the second main surface 24b of 10 out of 1000 inspection objects 20 is determined to be unqualified based on the image captured by the second imaging unit 12, the pass rate of the upper surface is 99%.

[0089] The alarm unit 5 issues an alarm based on the level of the obstruction being photographed, calculated by the level calculation unit 4c. The level of the obstruction used to determine whether the alarm unit 5 should issue an alarm can be the level of damage to the obstruction, the level of dirt, or the level of an obstruction containing both damage and dirt. By having the alarm unit 5 issue an alarm based on the level of the obstruction calculated by the level calculation unit 4c, for example, the cleaning and replacement of the transporter 1 can be performed at appropriate time intervals.

[0090] In this embodiment, the alarm unit 5 issues an alarm based on the pass rate determined by the pass rate determination unit 4d and the level of the obstacle to be photographed calculated by the level calculation unit 4c. By having the alarm unit 5 issue an alarm based on the pass rate determined by the pass rate determination unit 4d and the level of the obstacle to be photographed calculated by the level calculation unit 4c, alarms can be issued at more appropriate timing.

[0091] Specifically, alarm unit 5 issues an alarm when the level of the obstacle being photographed is above a first threshold, and the difference between the lower surface defect rate and the upper surface defect rate is above a second threshold. The first threshold is, for example, 1000, and the second threshold is, for example, 3%. In this case, alarm unit 5 issues an alarm when the level of the obstacle being photographed is above 1000, and the difference between the lower surface defect rate and the upper surface defect rate is above 3%. As described above, when an obstacle is present, the lower surface pass rate becomes lower than the upper surface pass rate due to the influence of the obstacle. Therefore, by issuing an alarm based on the difference between the lower and upper surface defect rates and the level of the obstacle, an alarm can be issued at a more appropriate timing.

[0092] There are no particular restrictions on the method by which the alarm unit 5 issues an alarm; for example, it can be done by displaying an alarm on a monitor or by issuing an alarm sound.

[0093] However, the conditions under which the alarm unit 5 issues an alarm are not limited to the conditions described above. For example, the alarm unit 5 may also issue an alarm when the level of the obstacle to be photographed, calculated by the level calculation unit 4c, is above the first threshold, and the pass rate of the inspected object 20 is below the third threshold. As described above, the pass rate of the inspected object 20 can be calculated by (1 - number of defects / number of inspections) × 100.

[0094] In addition, the alarm unit 5 can also issue an alarm when the level of the obstacle to be photographed calculated by the level calculation unit 4c is above the first threshold and the pass rate of the lower surface of the object to be inspected 20 is below the fourth threshold.

[0095] Furthermore, instead of calculating the average value of each image based on the product of the brightness difference and area of ​​the obstacle, the method of determining the obstacle's level can be modified by collecting the product of the brightness difference and area of ​​the obstacles from all images, and calculating the average, median, maximum, and quartile values ​​to determine the obstacle's level. In this case, indicators such as the pass rate of the inspected object 20 over a certain period can also be appropriately combined to determine whether to issue an alarm.

[0096] If alarm unit 5 sounds an alarm, it is preferable to clean and replace transport body 1. That is, if the level of obstacles to imaging, such as scratches or dirt, on transport body 1 increases, there is a possibility that a high-precision visual inspection cannot be performed. Therefore, if an alarm is sounded, transport body 1 is cleaned and replaced. By cleaning and replacing transport body 1, a high-precision visual inspection of the subsequent inspection object 20 can be performed.

[0097] According to the appearance inspection device 100 in this embodiment, the grade calculation unit 4c calculates the grade of the photographing obstacle based on the cumulative value of data of multiple photographing obstacles extracted by the photographing obstacle extraction unit 4a. Therefore, it is possible to appropriately calculate the grade of photographing obstacles such as scratches and dirt on the transport body 1. Thus, based on the calculated grade of the photographing obstacle, the transport body 1 can be cleaned and replaced, for example, at appropriate time intervals.

[0098] Furthermore, in conventional devices where the transporter cannot be cleaned and replaced at appropriate intervals, when an object to be inspected is supplied to a transport table with scratches and dirt, the object's posture may become disordered due to the scratches and dirt, potentially leading to misjudgment of the object's appearance. However, in the appearance inspection device 100 of this embodiment, as described above, since the transporter 1 can be cleaned and replaced at appropriate intervals based on the level of the obstacle being photographed, the disorder of the object's posture caused by scratches and dirt can be suppressed, enabling high-precision appearance inspection of the object 20. Additionally, the alarm unit 5 can also issue an alarm based on the level of the obstacle being photographed calculated by the level calculation unit 4c and the number / proportion of objects 20 that were not properly inspected due to improper posture. Moreover, the so-called objects 20 that were not properly inspected due to improper posture are those objects 20 that were not properly inspected based on image processing due to improper posture. For example, if the length of the object under inspection 20 detected by the sensor is shorter than the reference length due to an incorrect posture, the object under inspection 20 is judged to have an incorrect posture.

[0099] Furthermore, in the visual inspection device 100 of this embodiment, at least the area where the inspection object 20 is placed on the transport body 1 is transparent, and it has a shape capable of rotating about a central axis X1. The drive unit 3 is configured to transport the inspection object 20 in the rotational direction by driving the transport body 1 in the rotational direction centered on the central axis X1. Therefore, since the position of the transport body 1 is easy to manage, the position of the extracted imaging obstacle is easily determined. For example, if the drive unit 3 is a servo motor, the position of the transport body 1 can be managed by an encoder.

[0100] Furthermore, when the transporter 1 has a rotatable shape, the transport path tends to be shorter compared to other shapes such as straight lines, thus increasing the impact of a photographing obstacle on the visual inspection. Therefore, the visual inspection device 100 in this embodiment is more effective at extracting photographing obstacles and calculating the level of the extracted obstacles.

[0101] In addition, the information obtained by the appearance inspection device 100 in this embodiment (batch name, variety name, inspection time, equipment number, light intensity setting value, inspection item setting value, etc.), as well as the pass rate and statistical value of inspection results of each appearance inspection device 100 / shooting unit / inspection item, and the output value of all inspection items of the inspection object 20, can be collected to a server, database, etc., and the level of the obstacle to be photographed can be calculated by referring to or comparing the data of each appearance inspection device 100 among the multiple appearance inspection devices 100.

[0102] In addition to the extraction of photographic obstacles included in the transport body 1 described above, the appearance inspection device 100 in this embodiment can also be used for the purposes described in (i) to (iv) below.

[0103] (i) Detection of foreign objects attached to the lens of the imaging part

[0104] The appearance inspection device 100 in this embodiment can detect foreign objects attached to the lenses of the first imaging unit 11 to the sixth imaging unit 16.

[0105] As described above, the obstacle extraction unit 4a extracts obstacles based on the optical characteristic values ​​contained in the image captured by the first imaging unit 11. In the case of detecting foreign objects attached to the lens, as described above, the obstacle can be extracted from the captured image within the area excluding the area where the object to be inspected 20 is placed, or it can be extracted within the area including the area where the object to be inspected 20 is placed. Here, after extracting the obstacle, the area, brightness, and coordinates of the extracted obstacle are recorded.

[0106] When an obstacle with similar area and brightness exists at approximately the same coordinate in multiple images captured by the first imaging unit 11, the processing unit 4 determines that the obstacle is a foreign object attached to the lens of the first imaging unit 11. That is, when a foreign object is attached to the lens of the first imaging unit 11, an obstacle with similar area and brightness is reflected at approximately the same coordinate in multiple captured images. Therefore, when an obstacle with similar area and brightness exists at approximately the same coordinate in multiple images, it can be determined that the obstacle is a foreign object attached to the lens of the first imaging unit 11. Furthermore, since the amount of illumination may change during each capture, or obstacles may be captured in conjunction with other obstacles, it is not simply a matter of "an obstacle with the same area and brightness at the same coordinate," but rather "an obstacle with similar area and brightness at approximately the same coordinate."

[0107] By making the same judgment based on images captured by imaging units other than the first imaging unit 11, namely the second imaging unit 12, the third imaging unit 13, the fourth imaging unit 14, the fifth imaging unit 15, and the sixth imaging unit 16, foreign objects attached to each imaging unit can be detected.

[0108] If a foreign object is detected attached to the lens of the imaging unit using the above method, an alarm can be issued by the alarm unit 5. By issuing an alarm using the alarm unit 5, the lens can be cleaned or replaced, and cleaning or replacement of the lens can be performed at an appropriate time. In addition, by detecting foreign objects attached to the lens of the imaging unit, misjudgments of the appearance inspection of the object 20 caused by foreign objects attached to the lens of the imaging unit can be suppressed. Furthermore, it is possible to distinguish whether the obstacle in the image is caused by the lens of the imaging unit or by the transporter 1.

[0109] (ii) Detect the relative positional offset between the imaging unit and the illumination unit

[0110] The visual inspection device 100 in this embodiment can detect the relative positional offset between the optical axis of the imaging unit and the optical axis of the illumination unit.

[0111] When the first illumination unit 31 illuminates with multiple colors of light, it is necessary to align the optical axis of the first imaging unit 11 with the optical axis of the first illumination unit 31. However, due to reasons such as adjustment errors, there may be inconsistencies. When the optical axes of the first imaging unit 11 and the first illumination unit 31 are aligned, the brightness distribution of the captured image becomes symmetrical with respect to the field of view in all directions. However, when the optical axes are not aligned, the brightness distribution of the image becomes asymmetrical.

[0112] Figure 10(a) is a diagram schematically showing the brightness distribution of the R image, G image, and B image when the optical axis of the first imaging unit 11 is aligned with the optical axis of the first illumination unit 31. Figure 10 (b) is a diagram schematically showing the brightness distribution of the R, G, and B images when the optical axis of the green illumination light in the illumination light of the first illumination unit 31 is deviated. Figure 10 (c) is a schematic diagram showing the brightness distribution of the R, G, and B images when the optical axis of the first imaging unit 11 is deviated. Furthermore, in Figure 10 In, it is shown that relative to Figure 3 The positional relationship shown is such that the positional relationship between the first imaging unit 11 and the object being inspected 20 is flipped upside down.

[0113] like Figure 10 As shown in (a), when the optical axis of the first imaging unit 11 is aligned with the optical axis of the first illumination unit 31, the brightness distribution of the R image, G image, and B image is symmetrical from left to right. In addition, although the illustration is omitted, the brightness distribution of the R image, G image, and B image is symmetrical from top to bottom.

[0114] In contrast, when the optical axis of the green illumination light in the illumination light of the first illumination unit 31 deviates, such as Figure 10 As shown in (b), the brightness distribution of the G image becomes asymmetrical. Furthermore, when the optical axis of the first imaging unit 11 is deviated, as... Figure 10 As shown in (c), the brightness distribution of the R, G, and B images all becomes asymmetrical.

[0115] In this case, when the optical axis of only one color of illumination light out of a variety of colors is deviated, the brightness distribution of only that color image becomes asymmetrical. Conversely, when the optical axis of the first imaging unit 11 is deviated, the brightness distribution of all color images becomes asymmetrical. Therefore, it is possible to distinguish between cases where the optical axis of a specific color of illumination light is deviated and cases where the optical axis of the first imaging unit 11 is deviated, based on the brightness distribution of the color images.

[0116] The relative positional offset of the optical axis described above can also be detected between other imaging units and other illumination units. That is, the positional offset between the optical axis of the second imaging unit 12 and the optical axis of the second illumination unit 32, the positional offset between the optical axis of the third imaging unit 13 and the optical axis of the third illumination unit 33, the positional offset between the optical axis of the fourth imaging unit 14 and the optical axis of the fourth illumination unit 34, the positional offset between the optical axis of the fifth imaging unit 15 and the optical axis of the fifth illumination unit 35, and the positional offset between the optical axis of the sixth imaging unit 16 and the optical axis of the sixth illumination unit 36 ​​can be detected by the same method.

[0117] (iii) Management of the amount of light emitted by the lighting department

[0118] The appearance inspection device 100 in this embodiment can manage the amount of light from the first illumination unit 31 to the sixth illumination unit 36. For example, the average brightness in the background area of ​​the image captured by the first imaging unit 11 can be calculated, and the amount of light from the first illumination unit 31 can be managed based on the calculated average brightness. The background area of ​​the image is the area of ​​the entire image excluding the area where the object to be inspected 20 is captured.

[0119] Specifically, the average brightness of the background area of ​​the captured images is calculated and recorded. The average brightness of the background area is observed at regular intervals, for each fixed number of images, at each regular time period, or in each batch to confirm whether the newly calculated average brightness of the background area falls within a specified range. When the amount of illumination light from the first illumination unit 31 is appropriate, the average brightness of the background area of ​​the image falls within the specified range; however, when the amount of light is insufficient or excessive, the average brightness of the background area of ​​the image does not fall within the specified range. In this case, it is determined that the amount of illumination light from the first illumination unit 31 is inappropriate. This determination can be made by the processing unit 4. When it is determined that the amount of illumination light from the first illumination unit 31 is inappropriate, for example, an alarm can be issued by the alarm unit 5, or the amount of light from the first illumination unit 31 can be controlled to achieve an appropriate amount of light.

[0120] The above-mentioned management of the amount of illumination light can also be performed on each of the second illumination section 32 to the sixth illumination section 36.

[0121] Furthermore, when the transport body 1 contains an obstacle to be photographed, the obstacle to be photographed in the image captured by the first photographing unit 11 also moves when the transport body 1 is rotated. Therefore, due to the obstacle to be photographed, there may be a situation where the average brightness in the background area of ​​the image changes. However, by understanding the correlation between the level of the obstacle to be photographed and the average brightness in the background area, it is possible to distinguish between changes in the average brightness of the background area caused by changes in the amount of illumination light and changes in the average brightness of the background area caused by the obstacle to be photographed.

[0122] (iv) Detection of lighting degradation

[0123] The appearance inspection device 100 in this embodiment can detect the deterioration of the first illumination unit 31 to the sixth illumination unit 36.

[0124] When the first illumination unit 31 includes multiple LEDs, the degree of degradation varies depending on the LED. Therefore, by creating a thermal image of the brightness distribution in the background area of ​​the captured image, and determining that the LED corresponding to a certain area is degraded when the brightness in that area decreases, this determination can be made by the processing unit 4. If LED degradation is determined, for example, an alarm can be issued by the alarm unit 5 to urge the replacement of the LED. Thus, the detection and replacement of degraded LEDs can be easily performed.

[0125] Furthermore, since the brightness of the same area decreases each time under specific LED degradation conditions, recording the coordinates of the location where the brightness has decreased can suppress misjudgments in visual inspection caused by LED degradation.

[0126] The aforementioned detection of lighting degradation can also be performed on the second lighting section 32 to the sixth lighting section 36.

[0127] This invention is not limited to the embodiments described above, and various applications and modifications can be applied within the scope of this invention. For example, in the embodiments described above, the obstacle extraction unit 4a is described as a structure that extracts obstacles based on optical characteristic values ​​contained in the image captured by the first imaging unit 11, but it is not limited to the optical characteristic values ​​being contained in the image captured by the first imaging unit 11. For example, an optical sensor that emits light toward the transport body 1 may also be provided, and obstacles may be extracted based on the optical characteristic values ​​contained in the light emitted from the optical sensor and reflected by the transport body 1.

[0128] In the above embodiment, the transport body 1 has been described as having a transparent area where at least the object to be inspected 20 is placed. However, the area where the object to be inspected 20 is placed may not be transparent. In this case, the first imaging unit 11 can photograph the object to be inspected 20 on the transport body 1 using imaging light, such as infrared light, which has a wavelength that can pass through the transport body 1.

[0129] In one embodiment described above, the transporter 1 is described as a transparent structure with a rotatable circular platform shape, but it may also have a bendable sheet shape or a continuous annular shape.

[0130] The appearance inspection device in this application is as follows.

[0131] <1>. A visual inspection device for performing visual inspection of an object, characterized in that it comprises:

[0132] The transporter has a first main surface and a second main surface, wherein the second main surface is opposite to the first main surface;

[0133] The supply department supplies the inspection object to the second main surface of the aforementioned transporter.

[0134] The drive unit moves the object to be inspected on the transport body by driving the transport body.

[0135] The first imaging unit is capable of photographing the object to be inspected on the transport body by means of imaging light having a wavelength that can pass through the transport body in order to perform an appearance inspection of the object to be inspected.

[0136] The obstacle extraction unit extracts the obstacles contained in the image captured by the first imaging unit based on optical characteristic values ​​of the obstacles contained in the transporter; and

[0137] The rating calculation unit calculates the rating of the shooting obstacles extracted by the shooting obstacle extraction unit.

[0138] The aforementioned level calculation unit calculates the level of the photographing obstacle based on the cumulative value of data from multiple photographing obstacles extracted by the aforementioned photographing obstacle extraction unit.

[0139] <2>. The appearance inspection device according to <1> is characterized in that the obstacle extraction unit extracts the obstacle based on the optical characteristic values ​​contained in the image captured by the first imaging unit.

[0140] <3>. The appearance inspection device according to <1> or <2> is characterized in that the above-mentioned optical characteristic value is brightness.

[0141] <4>. The appearance inspection device according to any one of <1> to <3>, characterized in that the first imaging unit takes a picture of the object to be inspected on the transport body from the first main surface side.

[0142] <5>. The appearance inspection device according to <4> is characterized in that the first imaging unit captures a wider range than the range on which the object to be inspected is placed on the transport body.

[0143] <6>. The appearance inspection device according to <5> is characterized in that the obstacle extraction unit extracts the obstacle from the image captured by the first imaging unit within a range after removing the area containing the object to be inspected.

[0144] <7>. The appearance inspection apparatus according to any one of <1> to <6>, characterized in that, for the transport body, at least the area on which the object to be inspected is placed is transparent, and has a shape capable of rotating about a central axis orthogonal to the first main surface and the second main surface.

[0145] The drive unit drives the transport body in a rotational direction centered on the central axis, thereby transporting the inspection object on the transport body in the rotational direction.

[0146] <8>. The appearance inspection device according to <7> is characterized in that the grade calculation unit calculates the grade of the photographing obstacle based on the cumulative value of data of multiple photographing obstacles extracted by the photographing obstacle extraction unit from multiple images captured by the first photographing unit during one revolution of the transport body in the rotation direction.

[0147] <9>. The appearance inspection apparatus according to any one of <1> to <8>, characterized in that the grade calculation unit calculates the grade of the photographed obstacle based on the brightness of the photographed obstacle and the area of ​​the photographed obstacle contained in the image captured by the first photographing unit.

[0148] <10>. The appearance inspection device according to any one of <1> to <9> is characterized in that it further comprises an alarm unit, which issues an alarm based on the level of the photographing obstacle calculated by the level calculation unit.

[0149] <11>. The appearance inspection device according to any one of <1> to <10> is characterized in that it further comprises a sorting unit, which sorts the photographing obstacles extracted by the photographing obstacle extraction unit into scratches generated on the transport body and dirt adhering to the transport body.

[0150] <12>. The appearance inspection device according to any one of <1> to <9> is characterized in that it further comprises a second imaging unit, which takes a picture of the object to be inspected on the transporter from the second main surface side.

[0151] <13>. The appearance inspection device according to <12> is characterized in that it further comprises a pass rate determination unit, which determines the pass rate of the inspected object.

[0152] <14>. The appearance inspection device according to <13> is characterized in that the pass rate determination unit determines the pass rate of the inspection object based on the image of the inspection object captured by the first imaging unit and the image of the inspection object captured by the second imaging unit.

[0153] <15>. The appearance inspection device according to <13> or <14> is characterized in that it further comprises an alarm unit that issues an alarm based on the level of the photographing obstacle calculated by the level calculation unit.

[0154] The alarm unit issues the alarm based on the pass rate determined by the pass rate determination unit and the level of the obstacle to be photographed calculated by the level calculation unit.

Claims

1. A visual inspection device, used for visually inspecting an object, characterized in that, have: The transporter has a first main surface and a second main surface, wherein the second main surface is opposite to the first main surface; The supply department supplies the inspection object to the second main surface of the aforementioned transporter. The drive unit moves the object to be inspected on the transport body by driving the transport body. The first imaging unit is capable of photographing the object to be inspected on the transport body by means of imaging light having a wavelength that can pass through the transport body in order to perform an appearance inspection on the object to be inspected. The obstacle extraction unit extracts the obstacles in the image captured by the first imaging unit based on optical characteristic values ​​related to the obstacles contained in the transporter. The rating calculation unit calculates the rating of the shooting obstacles extracted by the shooting obstacle extraction unit. The aforementioned level calculation unit calculates the level of the photographing obstacle based on the cumulative value of data from multiple photographing obstacles extracted by the aforementioned photographing obstacle extraction unit.

2. The appearance inspection device according to claim 1, characterized in that, The obstacle extraction unit extracts the obstacles based on the optical characteristic values ​​contained in the image captured by the first imaging unit.

3. The appearance inspection device according to claim 2, characterized in that, The optical characteristic values ​​mentioned above refer to brightness.

4. The appearance inspection device according to claim 1, characterized in that, The first imaging unit photographs the object to be inspected on the transporter from the first main surface side.

5. The appearance inspection device according to claim 4, characterized in that, The first imaging unit captures a wider area than the area on which the object to be inspected is placed on the transporter.

6. The appearance inspection device according to claim 5, characterized in that, The aforementioned obstacle extraction unit extracts the aforementioned obstacle from the image captured by the first imaging unit within the range after removing the area containing the object to be inspected.

7. The appearance inspection device according to claim 1, characterized in that, For the aforementioned transporter, at least the area where the object to be inspected is placed is transparent, and it has a shape capable of rotating about a central axis orthogonal to the first and second main surfaces. The drive unit drives the transport body in a rotational direction centered on the central axis, thereby transporting the inspection object on the transport body in the rotational direction.

8. The appearance inspection device according to claim 7, characterized in that, The rating calculation unit calculates the rating of the photographed obstacle based on the cumulative value of data of multiple photographed obstacles extracted by the photographed obstacle extraction unit from multiple images captured by the first imaging unit during one revolution of the transporter in the rotation direction.

9. The appearance inspection device according to claim 1, characterized in that, The rating calculation unit calculates the rating of the obstacle based on the brightness and area of ​​the obstacle contained in the image captured by the first imaging unit.

10. The appearance inspection device according to claim 1, characterized in that, It also includes an alarm unit that issues an alarm based on the level of the obstacle to be photographed, calculated by the level calculation unit.

11. The appearance inspection device according to claim 1, characterized in that, It also includes a sorting section, which sorts the shooting obstacles extracted by the shooting obstacle extraction section into scratches generated on the transport body and dirt attached to the transport body.

12. The appearance inspection device according to claim 1, characterized in that, It also includes a second camera unit, which takes a picture of the object to be inspected on the transporter from the second main surface side.

13. The appearance inspection device according to claim 12, characterized in that, It also has a pass rate determination department, which determines the pass rate of the inspected object.

14. The appearance inspection device according to claim 13, characterized in that, The pass rate determination unit determines the pass rate of the inspected object based on the image of the inspected object captured by the first imaging unit and the image of the inspected object captured by the second imaging unit.

15. The appearance inspection device according to claim 13 or 14, characterized in that, It also includes an alarm unit that issues an alarm based on the level of the obstacle being photographed, calculated by the level calculation unit. The alarm unit issues the alarm based on the pass rate determined by the pass rate determination unit and the level of the obstacle to be photographed calculated by the level calculation unit.

Images