WO2022130843A1

WO2022130843A1 - Appearance inspection method, appearance inspection device, and method and device for processing structure

Info

Publication number: WO2022130843A1
Application number: PCT/JP2021/041436
Authority: WO
Inventors: 啓晃笠井; 薫酒井; 真由香大崎; 信昭中須; 淳也小坂
Original assignee: 株式会社日立ハイテクファインシステムズ
Priority date: 2020-12-16
Filing date: 2021-11-10
Publication date: 2022-06-23
Also published as: JP7390278B2; JP2022095139A

Abstract

The purpose of the present invention is to recognize defects such as burrs is occurring on a first side and a second side of a transparent member such as a composite member 10. In order to meet this purpose in this appearance inspection device 20 for inspecting a structure, light reflected in a first direction relative to the transparent member (glass) of the composite member 10 is imaged with a camera 21, or imaging is performed such that the pixel value of reflected light is greater than or equal to a fixed value, and the shape of defects relating to the first side are recognized on the interface of the transparent member. Here, 'surface' means the interface of the transparent member on the side of the camera imaging device.

Description

Visual inspection method, visual inspection equipment, processing method and equipment for structures

The present invention relates to a technique for inspecting the appearance of a structure composed of a plurality of members, particularly a structure including a transparent member and other members. Furthermore, the present invention also relates to a structure processing technique using inspection results. The transparent member may be sufficient to transmit the inspection light used for the inspection in a predetermined amount or more, and includes a so-called translucent member.

Currently, various structures are produced at production sites such as factories, but they are not always produced as designed. That is, the structure may have a defect in the members constituting the structure. For example, protrusions called parting lines and burrs may occur. In addition, there may be missing parts in the design. Therefore, at the production site, the appearance of the structure is inspected. In this appearance inspection, it is general that the inspection is performed by performing image processing (including image recognition) on the image input from the photographing apparatus.

Here, if the structure contains a transparent member, it may be difficult to perform a highly accurate visual inspection. In the inspection of such a structure, it is necessary to consider the presence of a transparent member.

For example, Patent Document 1 discloses "a foreign matter inspection device capable of inspecting foreign matter adhering to the front surface and the back surface of an object with high accuracy". This "foreign matter inspection device is a light projecting unit that projects the inspection light so that the inspection light is obliquely incident on the light-transmitting subject from the light projection position, and the scattering of the foreign matter caused by the inspection light. It includes a light receiving unit that receives light at a light receiving position and a processing unit that processes the light receiving result of the light receiving unit to determine the presence or absence of the foreign matter. "

Japanese Unexamined Patent Publication No. 2016-133357

Here, as a structure, among the transparent members (transparent members), it is necessary to inspect at least two of the interfaces indicating the boundaries of the members for defects including other members. be. FIG. 1 shows a composite member 10 which is an example of this structure. The composite member 10 is formed by integrally molding the glass 101 and the rubber frame 102 provided on the outer peripheral surface of the glass 101. FIG. 1A is a cross-sectional view of the composite member 10, and FIG. 1B shows a top view thereof. Here, as shown in FIG. 1A, defects of

burrs

1021 and 1022 occur on the upper surface and the lower surface of the glass 101 of the rubber frame 102 of the composite member 10, respectively. In such a case, it is necessary to distinguish and recognize the member on the upper surface and the burr on the lower surface in the inspection of the defect that occurs. In the example of FIG. 1, burrs on the upper surface and the lower surface of the glass 101, that is, a defect is recognized, but this also recognizes the rubber frame 102, that is, a member.

However, although Patent Document 1 describes detecting the presence of a foreign substance, it does not consider recognizing a defect related to the shape of the member. Therefore, the present invention relates to an inspection of a structure in which a defect occurs in another member, a member constituting the other member, or another member at the interface of the transparent member, and the following points are the problems. An object of the present invention is to recognize the shape of a defect in shape in a structure having a transparent member and other members.

In order to solve the above problems, in the present invention, in the visual inspection of the structure, the photographing unit performs an imaging with respect to the reflected light in the first direction with respect to the transmissive member, or an imaging in which the pixel value of the reflected light becomes a certain value or more. This is done to recognize the shape of the defect related to the first surface of the interface of the transparent member. Here, the surface means the image pickup unit side of the interface of the transmissive member.

In general production, defects in the narrow sense such as burrs may not occur (within the range of design tolerances). Even in such a case, in the present invention, it is possible to recognize the shape of the member constituting the structure and the portion thereof. Therefore, the shape of the "defect" in the present specification includes members and parts within the design tolerance range.

More specifically, the present invention relates to a visual inspection method using an visual inspection device for inspecting a defect of a structure including a permeable member, wherein the visual inspection device includes a holding portion for holding the structure and the above-mentioned. An imaging unit that photographs the held structure, and a lighting unit that illuminates at least the first surface of the transparent member of the retained structure for imaging by the imaging unit. It has a control unit that controls the holding unit, the photographing unit, and the lighting unit, and the control unit has at least one of the first surfaces in the photographing with the photographing unit with respect to the first surface of the transmissive member. At least one of the lighting unit and the photographing unit is controlled so that the pixel value of the unit becomes equal to or more than a predetermined value, and when the pixel value becomes equal to or more than the predetermined value, at least one of the structures. The imaging unit is controlled so that the unit captures the reflected light in the first direction with respect to the first surface, and the imaging information of at least a part of the imaged structure is used to capture the image of the structure. This is a visual inspection method for recognizing the shape of a defect related to the first surface. In order for the control unit to control the pixel value to be equal to or higher than a predetermined value, (1) the illumination unit irradiates the illumination from a predetermined first direction, and (2) the illumination unit determines. It includes at least one of irradiating the illumination with the above luminance and (3) controlling the gain in the photographing of the photographing unit.

The present invention also includes the above-mentioned visual inspection device and an information processing device for realizing the control unit. Further, a computer program for operating the information processing apparatus and a storage medium for storing the program are also included.

Furthermore, the present invention also includes a processing method and a processing apparatus using the results of visual inspection. Here, a computer program for executing this processing method and a storage medium for storing this program are also included.

According to the present invention, it is possible to more accurately inspect the appearance of a structure including a permeable member. Issues, configurations and effects other than those described above will be clarified by the following description of the embodiments.

The figure which shows the composite member which is the inspection target in each Example The block diagram which shows the structure of the appearance inspection apparatus in Example 1. A block diagram showing the configuration of the control unit in the first embodiment. Flow chart showing the processing contents in the first embodiment The figure which shows the design information used in Example 1. (A) A diagram showing shooting information in Example 1, and (b) a diagram showing an example of an image when the shooting conditions are not used. (A) A diagram showing shooting information in Example 1, (b) a diagram explaining a process of extracting a cutting line from shooting information, (c) a diagram showing a relationship between brightness and pixels in shooting information, (d) a minimum value. The figure which shows the relationship between the brightness and the pixel in the shooting information when it is difficult to distinguish The block diagram which shows the structure of the manufacturing system in Example 1. The block diagram which shows the structure of the appearance inspection apparatus in Example 2. Flow chart showing the processing contents in the second embodiment The figure which shows the design information used in Example 2. (A) (b) A diagram showing imaging information in Examples 2 and 3, and (c) a diagram illustrating extraction of a cutting line in Examples 2 and 3. A block diagram showing the configuration of the visual inspection apparatus according to the third embodiment.

Hereinafter, each embodiment of the present invention will be described with reference to the drawings, but first, the inspection target common to each embodiment will be described. In each embodiment, the composite member 10 shown in FIG. 1 will be described as an example to be inspected. The composite member 10 is used as a window of a railroad vehicle, and a glass portion and a frame supporting the glass portion are integrally manufactured. The composite member 10 is installed in a railway vehicle as a so-called unit frame. In the present invention, the composite member 10 is not limited to being installed in a railroad vehicle, but can be installed in various means of transportation such as automobiles, aircraft, and ships, and in buildings such as buildings.

As described above, in the composite member 10 to be inspected, the glass 101, which is a transparent member, is surrounded by the rubber frame 102 and held. Then, as shown in FIG. 1A, burrs 1021 and burrs 1022 are generated on the upper surface and the lower surface of the glass 101, respectively, on the rubber frame 102. Examples of these burrs include thorns having a thickness of about 0.1 mm, strips having a thickness of about 0.1 to 0.3 mm, and granules having a thickness of 0.1 mm or less.

Here, when inspecting the appearance of the composite member 10, the image is taken from the upper surface side using a photographing device such as a camera or an photographing unit. For example, the photographing device photographs from the direction of the arrow shown in FIG. 1 (a). In this case, in the rubber frame 102, burrs 1021 are generated on the upper surface side in the drawing. Similarly, burrs 1022 are generated on the lower surface side. At this time, when observing from the direction of the arrow in the figure, since the glass 101 has transparency, both the burrs 1021 and the burrs 1022 can be observed. Therefore, when these are photographed by using a photographing device, not only the burr 1021 but also the burr 1022 is photographed. That is, both the burr 1021 on the side facing the photographing device and the burr 1022 in the direction of passing the transmitted light through the glass 101 are photographed. This means that burrs 1021 and burrs 1022 cannot be distinguished. Therefore, the burr 1021 cannot be extracted or recognized. As a result, the cutting position for removing the burr 1021 cannot be specified.

Therefore, in each embodiment of the present invention, the burr 1021 is recognized, in other words, the process for removing the burr 1022 is executed. As described above, in each embodiment, the recognition process is executed in consideration of the influence of the transparent member such as the glass 101 on the image to be inspected such as the rubber frame 102.

First, Example 1 will be described. FIG. 2 is a block diagram showing the configuration of the visual inspection apparatus 20 in this embodiment. The visual inspection device 20 includes a camera 21 that photographs the composite member 10 that is an inspection object, a coaxial epi-illuminator 22 that illuminates the composite member 10, an XY stage 23 that holds the composite member 10, and each of these configurations, that is, a visual inspection. It has a control unit 24 that controls the device 20.

Hereinafter, each configuration of the visual inspection device 20 will be described. The camera 21 is a means for realizing a photographing device or a photographing unit having a function of photographing the composite member 10. The shooting of the camera 21 is not limited to a still image, and a moving image may be shot.

Further, the coaxial epi-illumination device 22 functions as a lighting unit, and is a means for realizing the lighting unit that illuminates the composite member 10 photographed by the camera 21. The coaxial epi-illuminator 22 reflects the illumination (light) from the light source 221 by a reflector 222 such as a half mirror, and emits illumination parallel to the optical axis of the camera 21. As a result, it becomes possible to irradiate at least a part of the composite member 10 with the illumination from the coaxial epi-illuminator 22. In this embodiment, "coaxial epi-illumination" illumination that irradiates illumination parallel to the optical axis of the camera 21 is used, but other illumination may be used.

Next, the XY stage 23 is a means for realizing the holding portion having a function of holding the composite member 10. The XY stage 23 holds the composite member 10 so that the upper surface, which is the first surface to be the inspection surface of the composite member 10, faces the camera 21 and the coaxial epi-illuminator 22. Here, the direction in which the upper surface faces directly means the direction in which the upper surface, which is the inspection surface, is suitable for inspection.
Therefore, this direction can be appropriately determined according to the shape and characteristics of the inspection target and the inspection surface, and the accuracy of the inspection. Further, in order to turn in this direction, as will be described later, the XY stage 23 may be operated according to the control unit 24, or the inspector may manually operate the XY stage 23 so as to be in that direction. Although not shown, it has a drive mechanism represented by a motor for driving the XY stage 23.

Finally, the control unit 24 controls the operations of the camera 21, the coaxial epi-illuminator 22, and the XY stage 23. That is, the control unit 24 controls the operation of the visual inspection device 20. Therefore, the control unit 24 is connected to each of the camera 21, the coaxial epi-illuminator 22, and the XY stage 23, and sends and receives information for control to and from each other.

Further, the control unit 24 may be configured integrally with the visual inspection device 20 or may be configured in a separate housing. In the former case, the above connection can be realized in the internal communication path. In the latter case, the network enables connection. Further, in the latter case, the control unit 24 can be realized by an information processing device such as a PC, a tablet, or a server. The configuration of the control unit 24 thus realized in a separate housing will be described with reference to FIG. In addition, even when it is configured integrally with the visual inspection device 20, the basic configuration is the same as when it is realized in a separate housing, and the difference will be mentioned in the description of FIG.

FIG. 3 is a block diagram showing the configuration of the control unit 24. The control unit 24 includes a processing unit 241, an input unit 242, a display unit 243, a communication I / F 244, a main storage unit 245, and an auxiliary storage unit 246. These are connected to each other via an internal communication channel such as a bus. Hereinafter, each configuration will be described.

First, the processing unit 241 is realized by an arithmetic unit such as a CPU. That is, the process according to each program stored or expanded in the main storage unit 245 is executed. The details of this process will be described later. In this embodiment, the function of the control unit 24 is realized by using a program, that is, software, but it may be realized by hardware. That is, the processing unit 241 may be configured by a circuit that executes the processing described later.

Next, the input unit 242 can be realized by an input device such as a keyboard or pointing, and accepts the operation of the inspector. This operation includes operations in inspections such as starting and stopping of the camera 21, as well as measures to be taken in the event of a failure. When the visual inspection device 20 is automatically operated, the input unit 242 may be omitted.

Next, the display unit 243 can be realized as a so-called display and outputs various information. Further, the display unit 243 may be integrally molded with the input unit 242 like a touch panel. Further, when the visual inspection device 20 is automatically operated, the display unit 243 may be omitted.

Next, the communication I / F 244 has a function of connecting to various configurations of the visual inspection device 20 via a network. In this embodiment, it is connected to one visual inspection device 20, but it may be connected to a plurality of units. Further, when the control unit 24 is integrally configured with the visual inspection device 20, this configuration may be omitted and the control unit 24 may be directly connected to another configuration.

Next, the main storage unit 245 is realized by a storage medium such as a memory. Then, the main storage unit 245 expands each program for performing the processing in the processing unit 241 to the main storage unit 245.
Therefore, it is usually desirable that each program is stored in another storage unit such as the auxiliary storage unit 246 or a storage medium. Here, the program includes various programs for controlling the visual inspection device 20. These can be exemplified by the following.
Camera control program 2451 that controls the activation of the camera 21
Coaxial epi-illuminator control program 2452 that controls the activation of the coaxial epi-illuminator 22
XY stage control program 2453 that controls the operation of the XY stage 23
Further, the main storage unit 245 also stores, as a program, an image processing program 2454 for executing various image processing on the shooting information shot by the camera 21. Details of these programs will be described later.

Finally, the auxiliary storage unit 246 stores various information. The various information includes the design information 2461 of the composite member 10 and the shooting information 2462 taken by the camera 21. Further, the auxiliary storage unit 246 can be realized by a hard disk drive (HDD), a solid state drive (SSD), various optical disks, or the like. The auxiliary storage unit 246 may also store the above-mentioned various programs. Further, the information stored in the auxiliary storage unit 246 may be stored in an external storage device connected via the communication I / F 244.

Next, the details of various processes using the processing unit 241 will be described with reference to FIG. FIG. 4 is a flowchart showing the processing contents using the processing unit 241 in this embodiment.

First, in step S100, the processing unit 241 activates the camera 21 and the coaxial epi-illuminator 22 according to the camera control program 2451 and the coaxial epi-illuminator control program 2452. The activation condition includes an inspector's operation via the input unit 242 and reception of a detection signal indicating that the composite member 10 is held from the XY stage 23. At this time, the composite member identification information for identifying the composite member 10 held by the XY stage 23 is included in the operation or detection signal for the input unit 242, and the processing unit 241 specifies this.

Next, in step S101, the processing unit 241 specifies the imaging conditions of the composite member 10 to be inspected according to the image processing program 2454. For this purpose, the processing unit 241 reads out the imaging conditions of the corresponding composite member 10 from the design information 2461 of the auxiliary storage unit 246. An example of this design information 2461 is shown in FIG. The design information 2461 includes composite member identification information 2461-1 for identifying the composite member to be inspected, and member 2461-2 for identifying each member constituting the composite member identification information 2461-1. Here, it is desirable to add information indicating whether or not the member is a permeable member that affects the inspection target and whether or not the member is an inspection target.

The design information 2461 also includes a dimension 2461-3 indicating the size and shape of each member and a tolerance 2461-4 for determining whether correction processing is necessary for this dimension. Further, the design information 2461 also includes a shooting condition 2461-5, which is a condition for shooting the composite member for each composite member. It is preferable that the shooting conditions 2461-5 include the illumination brightness thereof and the camera gain in the camera 21. Further, instead of the camera gain, the ISO sensitivity, the aperture, and the shutter speed of the camera 21 for realizing this camera gain may be stored.

In step S101, the record corresponding to the composite member identification information specified in step S100 is specified from the design information 2461. For example, when # 1 is specified in step S100, the first record shown in FIG. 5 is specified. Then, the processing unit 241 specifies the shooting conditions 2461-5 included in the specified record. In the example of FIG. 5, the processing unit 241 specifies “a” as the illumination brightness and “b” as the gamer gain. It should be noted that these shooting conditions indicate conditions in which the pixel value of the reflected light in the transmissive member is equal to or higher than a certain level. This pixel value includes luminance. Further, as an example in which the pixel value is above a certain level, shooting information such as so-called overexposure is saturated.

Next, in step S102, the processing unit 241 outputs a control signal for moving to the XY stage 23 according to the XY stage control program 2453. At this time, the processing unit 241 identifies the position of the composite member 10 photographed by the camera 21, and sends a control signal to the XY stage 23 so that the cutting line for cutting the burr is included in the field of view of the camera 21. Output. This cutting line can be specified by the processing unit 241 using the dimensions of the design information 2461.

Further, in this step, the processing unit 241 outputs a control signal to the camera 21 that moves the position of the camera so that the cutting line is included in the field of view of the camera 21 as described above according to the camera control program. May be good. In this way, the cutting line may be controlled to be included in the field of view of the camera 21 by the relative positional relationship between the camera 21 and the XY stage (or the composite member 10 held by the camera 21). The cutting line is a kind of processing position for processing a member according to a design.

Next, in step S103, the processing unit 241 outputs a control signal for shooting according to the shooting conditions specified in step S101 to the camera 21 according to the camera control program 2451. Further, the processing unit 241 outputs a control signal for lighting according to the shooting conditions specified in step S101 to the coaxial epi-illuminator 22 according to the coaxial epi-illuminator control program 2452.

As a result of these, the coaxial epi-illuminator 22 irradiates the illumination under the illumination conditions included in the photographing conditions specified in step S101. Here, the coaxial epi-illuminator 22 generates reflected light in the first direction from the composite member 10 or the glass 101. It is desirable that this reflected light is substantially specular reflected light. The substantially specularly reflected light means the reflected light including the specularly reflected light and falling within a certain range from the reflection angle of the specularly reflected light. Further, the reflected light in the first direction can be defined as having a pixel value of a certain value or more in the shooting information. An example of this is saturation including so-called overexposure. In this embodiment, the reflected light in the first direction is a substantially specular reflected light, and will be described as causing saturation.

Further, the camera 21 performs shooting including substantially specular reflected light with the output control signal, that is, the camera gain specified in step S101. That is, the camera 21 captures an image in which the pixel value of a part of the glass 101 is saturated.

As described above, since the pixel value may be equal to or higher than a certain value, the processing unit 241 may be controlled as follows in step S103. (1) As described above, the coaxial epi-illuminator 22 irradiates the illumination from a predetermined first direction, (2) the coaxial epi-illuminator 22 irradiates the illumination with a predetermined brightness or more, (3). At least one of controlling the gain in the shooting of the camera 21 is included. That is, it suffices if any one or a combination of these can realize shooting in which the pixel value is equal to or higher than a certain value.

Here, apart from the explanation of the flowchart, the significance of photographing the first reflected light will be explained in this embodiment. First, FIG. 6A shows an example schematically showing the shooting information 2462 acquired in step S013. As shown in FIG. 6A, the glass portion 101-1 of the photographing information 2462 is saturated. The rubber frame portion 102-1 is present on the right side of FIG. 6A. The rubber frame portion 102-1 shows a burr portion 1021-1 indicating the burr 1021 generated on the upper surface side of the glass 101.

On the other hand, an image when the shooting conditions of step S101 are not used is shown in FIG. 6 (b). In the image shown in FIG. 6B, the glass portion 101-1 is not saturated. Therefore, in this image, the burr portion 1022-1 showing the burr 1022 generated on the lower surface side of the glass 101 is also included in the shooting information. As described above, when the shooting conditions specified in step S101 are not used, the burr portion 1022-1 on the lower surface side is also included in the shooting information 2462, which makes it difficult to specify the burr portion 1021-1. Therefore, when removing burrs, it becomes difficult to identify an accurate cutting line.

Note that FIGS. 6 (a) and 6 (b) schematically show shooting information and images, respectively. Therefore, in the photographing information 2462 shown in FIG. 6A, a negligible burr portion 1022-1 may be included.

Further, in this embodiment, in step S101, the shooting conditions 2461-5 included in the design information 2461 are used. However, the camera 21 may take a picture on a trial basis, and the processing unit 241 may determine whether the glass portion 101-1 is saturated according to the image processing program 2454. In this case, the shooting of the camera 21 and the determination by the processing unit 241 are repeatedly performed. Further, the inspector may use the image displayed on the display unit 243 to determine whether or not the image is saturated.

This completes the explanation of the significance of using the reflected light in the first direction, and returns to the explanation of the flowchart of FIG.

In step S103, the shooting information, which is an image shot by the camera 21, is output to the control unit 24. In response to this, the processing unit 241 of the control unit 24 stores this shooting information in the auxiliary storage unit 246 as shooting information 2462. The shooting information 2462 may be the image itself taken by the camera 21, and it is preferable that the shooting information 2462 includes information for identifying the individual of the composite member 10 to be shot from this image.

Next, in step S104, the processing unit 241 extracts a cutting line from the shooting information 2462 according to the image processing program 2454. This extraction will be described with reference to FIG. FIG. 7 (a) is the same diagram as in FIG. 6 (a), and is a diagram schematically showing shooting information 2462. On the other hand, in FIG. 7B, a cutting line is extracted using the information showing the relationship between the pixel and the luminance shown in FIG. 7C. For example, the processing unit 241 has a pixel value from the rubber frame portion 102-1 (right side in the drawing) shown in FIG. 7 (c), that is, a pixel whose brightness first shows a minimum value, and a glass portion 101-1 and rubber. The cutting line 103 is extracted by sequentially specifying along the boundary region of the frame portion 102-1.

Note that the processing unit 241 of this embodiment can execute the extraction of the cutting line 103 not only by using the minimum value of the luminance value but also by using the characteristics of the luminance value. For example, the processing unit 241 of the present embodiment may detect and realize a portion where the rate of change in the slope of the luminance is larger than a predetermined value. As shown in FIG. 7 (d), it may be difficult to determine the minimum value. In the composite member 10, depending on the shape of the rubber frame 102, a minimum value may be formed, or a minimum value may not be obtained and the change in inclination may be large. Therefore, the processing unit 241 may extract the cutting line 103 by utilizing both the minimum value and the change in the slope.

Further, when the cutting line is extracted, the processing unit 241 extracts the burr end face on the left side from the cutting line 103 (see FIG. 7 (c)). In this case, the processing unit 241 sets the difference between the burr end face position and the cutting line 103 position as the burr size. In some cases, the cutting line 103 cannot be extracted, that is, there is no need for cutting without burrs. Similarly in this case, the processing unit 241 sets the difference between the burr end face position and the cutting line 103 position as the burr size (see FIG. 7 (d)).

Therefore, in this embodiment, it is desirable that the information indicating the relationship between the brightness and the pixels is stored in the auxiliary storage unit 246 as the shooting information 2462 or in association with the shooting information 2462. In this case, the processing unit 241 stores information indicating the cutting line so that it is possible to determine which inner side of the rubber frame 102 is the cutting line.

Next, in step S105, the processing unit 241 determines the presence or absence of burrs based on the extracted cutting line 103 according to the image processing program 2454. This determination is executed as follows. If the cutting line 103 is not extracted, the processing unit 241 determines that there is no burr. Further, when the cutting line 103 is extracted, the processing unit 241 uses the tolerance 2461-4 of the design information 2461. That is, the processing unit 241 determines whether or not it is within the range of the tolerance 2461-4 from the cutting line. Using this result, the processing unit 241 determines that there is no burr if it is within the range.

As a result of this determination, in this step, it becomes possible to determine whether the rubber frame 102 needs to be cut. It is desirable that the processing unit 241 associates this determination result with the cutting line extracted in step S104 and stores it in the auxiliary storage unit 246.

Then, in step S106, the processing unit 241 determines, according to the image processing program 2454, whether or not the entire circumference of the upper surface side of the composite member 10, that is, each inner side of the rubber frame 102 on the upper surface side has been photographed. For this reason, it is desirable that the processing unit 241 use the design information 2461. As a result, if it is not completed (NO), the process returns to step S101.
When finished (YES), the process of this flowchart ends.

Although the above flowchart shows the inspection of the upper surface side of the composite member 10, the lower surface side can be similarly performed. That is, the lower surface side can be inspected by performing the same processing as in FIG. 4 on the composite member 10 installed on the XY stage 23 at the top and bottom (front and back) in the opposite direction to the above.

It is also possible to omit the use of the imaging conditions in step S101 for the inspection of the lower surface side. That is, if the burr 1021 is cut as described later, the burr 1021 on the upper surface that is not the inspection target is not photographed. Therefore, in the inspection of the lower surface, it is possible to omit step S101.

This is the end of the description of the process in the visual inspection of Example 1. Here, in the first embodiment, processing for burrs such as cutting of burrs described below may be further performed.

In this embodiment, the cutting device for cutting burrs may be integrally configured with the visual inspection device 20, or may be configured with a separate housing. However, it is desirable to be connected to the control unit 24 which may be used in any case. Further, it may be realized as the manufacturing system 200 shown in the block diagram of FIG.
The manufacturing system 200 of FIG. 8 has an inspection station 201 having the same function as the visual inspection device 20, and a processing station 202 which is a kind of cutting device. That is, the camera 21 and the coaxial epi-illuminator 22 are installed in the inspection station 201. Further, a laser oscillator 25 for cutting burrs is installed in the processing station 202. In this embodiment, the laser oscillator 25 is used as the cutting device, but the present invention is not limited to this. Further, in this embodiment, cutting is taken as an example of processing, but other processing such as deletion is included. That is, a processing device other than the cutting device may be used.

Further, the control unit 24 is connected to each configuration of the inspection station 201 and the processing station 202, and outputs a control signal to control them. The processing unit 241 of the control unit 24 may be provided at each of the inspection station 201 and the processing station 202.

Next, the processing in the manufacturing system 200 of FIG. 8 will be briefly described. After the processing in the flowchart of FIG. 4 is completed, the processing unit 241 outputs a control signal for moving the composite member 10 to the processing station 202 side to the XY stage 23-a according to the XY stage control program 2453. As a result, as shown in FIG. 8, the XY stage 23-a moves the composite member 10 to the processing station 202 side by using its driving function. In addition, in order to move in this way, the calibration result of the coordinates of the inspection station 201 and the processing station 202 is stored in the auxiliary storage unit 246.

Then, the processing unit 241 outputs a control signal to the laser oscillator 25 according to a cutting processing control program (not shown). This control signal is a signal instructing to cut the inner side of the rubber frame 102 determined to have burrs (YES) in step S105 according to the cutting line extracted in step S104.

For this purpose, the processing unit 241 uses information indicating the presence or absence of cutting lines and burrs stored in the auxiliary storage unit 246. As a result, the laser oscillator 25 outputs a laser along the cutting line extracted in step S104, and cuts the burr 1021 of the rubber frame 102 of the composite member 10.

In this embodiment, the burr is cut after taking a picture of each inner side (entire circumference) of the rubber frame 102. However, this may be cut off each time a photograph is taken.

This is the end of the description of Example 1, and the following will explain Example 2.

In the second embodiment, the illuminator 26 included in the illuminating unit is added as compared with the visual inspection device 20 of the first embodiment. FIG. 9 is a block diagram showing the configuration of the visual inspection apparatus 20 in the second embodiment. As described above, the illuminator 26 is added as compared with the first embodiment, and the other configurations are the same as those of the first embodiment. Further, the illuminator 26 is connected to the control unit 24 and is installed so that the illuminator is illuminated at a position and angle different from that of the coaxial epi-illumination illuminator 22. This Example 2 is suitable when it is difficult to extract a cutting line from the photographing information when the illumination of the coaxial epi-illuminator 22 is used.

For example, when the shooting information as shown in FIG. 12A using the illumination of the coaxial epi-illuminator 22 is acquired, the difference between the pixel values of the burr portion 1021-1 and the rubber frame portion 102-1 is small and the cutting line is extracted. May be difficult. In such a case, it is preferable to use this embodiment.

For this purpose, in this embodiment, an appearance inspection is performed using two shooting information using two lightings. In this embodiment, two lights and two shooting information are used, but three or more lights and shooting information may be used. Hereinafter, the details of the process for this inspection will be described below with reference to the flowchart of FIG.

First, step S100 is the same process as in the first embodiment. Next, step S201 is executed, and this process is basically the same as step S101 of the first embodiment. In step S201, in addition to step S101, the posture and angle of the illuminator 26 are specified.
Specifically, the processing unit 241 specifies the "illumination posture / position" of the illuminator 26 as an imaging condition of the composite member 10 to be inspected according to the image processing program 2454. This indicates the irradiation angle of the illuminator 26 with respect to the composite member 10. This irradiation angle is different from that of the coaxial epi-illumination device 22, and is obtained as shown below. Therefore, in the photographing using the illuminating device 26, the reflected light in the second direction, which is a direction different from the first direction, is photographed.

The processing unit 241 uses the design information 2461a shown in FIG. 11 instead of the design information 2461. That is, the processing unit 241 specifies the illumination posture / position of the shooting conditions 2461-5a of the design information 2461a. The lighting posture / position is recorded as being optimized according to the shape of the edge portion of the rubber frame 102 of the composite member 10. That is, as the illumination posture / position, the optimum value of the irradiation angle at which the pixel value (for example, brightness) near the cutting line becomes larger than a predetermined condition when the illuminator 26 is illuminated is used. This optimum value can be obtained from the shape of the rubber frame 102. Therefore, in this step, the processing unit 241 may obtain the shape of the rubber frame 102 included in the design information 2461 to the design information 2461a.

Next, step S102 is executed in the same manner as in the first embodiment. That is, the XY stage 23 moves the composite member 10 to the photographing position.

Next, in step S203, the same image acquisition as in step S103 is performed. However, the lighting has changed from the coaxial epi-illuminator 22 to the illuminator 26. That is, in step S203, the processing unit 241 outputs a control signal for shooting according to the shooting conditions specified in step S201 to the camera 21 according to the camera control program 2451. Further, the processing unit 241 outputs to the illuminator 26 a control signal for illuminating according to the shooting conditions specified in step S201 according to the coaxial epi-illuminator control program 2452 or the illumination control program (not shown).

As a result, the processing unit 241 acquires the shooting information taken by the camera 21 and stores the reflected light in the second direction in the lighting of the illuminator 26 in the auxiliary storage unit 246. As shown in FIG. 12 (b), this shooting information has the following characteristics as compared with the shooting information in the coaxial epi-illuminator 22 of FIG. 12 (a). (1) Pixels of both the burr portion 1021-1 and the burr portion 1022-1 on the upper surface side and the lower surface side are included, and (2) the burr portion 1021-1, the burr portion 1022-1 and the rubber frame portion 102-1. The difference in values is large, and it is easier to extract the cutting line.

Next, in step S204, the processing unit 241 performs the same processing as in step S104 of the first embodiment according to the image processing program 2454. That is, the processing unit 241 extracts the cutting line from the shooting information acquired in step S203. The extraction of the cutting line is performed by using the characteristics of the pixel value as in the first embodiment. Since this extraction is the same as in Example 1, the details thereof will be omitted.

Next, in step S205, the processing unit 241 determines whether the burr portion 1021-1 on the upper surface side can be extracted from the shooting information acquired in step S203 according to the image processing program 2454. That is, it is determined whether or not the burr portion 1021-1 and the burr portion 1022-1 in FIG. 12B are present on the upper surface side. As a criterion for this determination, for example, whether or not the pixel value of the glass portion 101-1 in FIG. 12B is a certain value or more (for example, whether it is saturated) can be used.

Note that the process of step S205 may be omitted, and a configuration in which the process proceeds from step S204 to step S103 may be adopted.

Next, the same process as in step S103 of the first embodiment is executed. That is, the processing unit 241 acquires the shooting information using the coaxial epi-illuminator 22.

Next, in step S206, the processing unit 241 draws a cutting line extracted in step S204 on the composite member 10 according to the image processing program 2454, and displays this on the display unit 243. For this purpose, the processing unit 241 synthesizes the photographing information acquired in step S103 and the cutting line extracted in S204. As a result, the processing unit 241 acquires the information in which (1) the burr portion 1021-1 on the upper surface side is specified and (2) the cutting line 103 is extracted as shown in FIG. 12 (c).

Next, the same steps S105 and S106 as in the first embodiment are executed. Further, in the second embodiment as well, the burr can be cut in the same manner as in the first embodiment.

In this embodiment, the shooting conditions (illumination brightness, illumination posture / angle) of the coaxial epi-illuminator 22 and the illuminator 26 were obtained in step S201, respectively. However, in step S201, the shooting conditions of the illuminator 26 may be obtained, and the shooting conditions of the coaxial epi-illuminator 22 may be obtained immediately before step S103.

Further, in this embodiment, the coaxial epi-illuminator 22 is photographed (step S103) after the image of the illuminator 26 is photographed (step S203), but the order may be reversed. In this case, in step S205, it is determined whether or not the cutting line can be extracted as follows.

The processing unit 241 determines whether or not the cutting line could be extracted in step S203 according to the image processing program 2454. This judgment can be made based on whether the number or length of pixels extracted as the minimum value is a predetermined value or more. Further, the processing unit 241 may display the pixel showing the minimum value on the photographing information on the display unit 243, and the inspector may make the determination. As a result, if it is determined that the cutting line can be extracted (YES), the process proceeds to step S106. If it is determined that the extraction could not be performed (NO), the process proceeds to step S103.

Further, in the case of this processing order, the shooting condition of the coaxial epi-illuminator 22 may be obtained in step S201, and the shooting condition of the illuminator 26 may be obtained immediately before the step S203 which is turned later.

Further, in this embodiment, the coaxial epi-illuminator 22 and the illuminator 26 and two illuminators are physically used, but it may be configured to acquire two shooting information with one illuminator. In this case, the lighting is moved under the control of the control unit 24 to irradiate the lighting. Further, when the coaxial epi-illuminator 22 is used as the illumination, the processing unit 241 can acquire two shooting information by changing the angle of the reflector 222 according to the coaxial epi-illuminator control program 2452. In this case, as described above, it may be configured to acquire three or more shooting information.

Further, in Example 2, the same burr cutting process as in Example 1 may be performed. That is, the second embodiment can be applied to the manufacturing system 200 shown in FIG. This is the end of the description of the second embodiment.

In Example 2, two lights were used to acquire two shooting information in order to cope with the case where it is difficult to extract the cutting line. In this embodiment, two cameras are used in order to solve the same problems as in the second embodiment. FIG. 13 is a block diagram showing the configuration of the visual inspection apparatus 20 in this embodiment. A camera 27, which is a photographing unit, is added as compared with the first embodiment shown in FIG. The camera 27 is connected to the control unit 24, and the processing unit 241 controls shooting according to the camera control program 2451.

Hereinafter, the processing of this embodiment is basically the same as that of the chart of Example 2. However, since step S201, step S203, and step S103 are different, the third embodiment will be described below with a focus on these.

First, in step S201, the design information 2461a shown in FIG. 11 was used in Example 2, but in this embodiment, the design information 2461 shown in FIG. 5 is used as in Example 1.

Then, in this embodiment, after step S102, step S203 is executed as follows. In step S203, the processing unit 241 outputs a control signal for shooting according to the shooting conditions specified in step S201 to the camera 21 according to the camera control program 2451. Further, the processing unit 241 outputs a control signal for lighting according to the shooting conditions specified in step S201 to the coaxial epi-illuminator 22 according to the coaxial epi-illuminator control program 2452.

As a result, in this step, the processing unit 241 acquires shooting information from the camera 21. The shooting information acquired in this way can be expressed as shown in FIG. 12A as in the second embodiment.

Then, after step S205, step S103 is performed. In step S103, the processing unit 241 outputs a control signal for shooting according to the shooting conditions specified in step S201 to the camera 27 according to the camera control program 2451. Further, similarly to step S103, the processing unit 241 outputs a control signal for performing illumination according to the shooting conditions specified in step S201 to the coaxial epi-illuminator 22 according to the coaxial epi-illuminator control program 2452.

As a result, in this step, the processing unit 241 acquires shooting information from the camera 27. The shooting information acquired in this way can be expressed as shown in FIG. 12 (b) as in the second embodiment.

Then, in this embodiment, the processes after step S206 are executed in the same manner as in the second embodiment.
Further, also in the third embodiment, the order of the steps S203 and the step S103 may be changed as in the second embodiment. In this case, it is desirable to change the process of step S205 as in the embodiment. Further, in the third embodiment, since the two cameras 21 and the camera 27 are used, it is possible to perform step S203 and step S103 together. In this case, the processing unit 241 outputs a control signal for photographing to the camera 21 and the camera 27 in parallel according to the camera control program 2451.

Further, in this embodiment, one camera (camera 21 or camera 27) may be moved to acquire two or more shooting information. Further, the camera 21 and the camera 27 have different shooting angles, that is, different fields of view. Therefore, it is desirable that the processing unit 241 performs image conversion so as to match the shooting information of the camera 21 with the shooting information of the camera 27 according to the image processing program 2454.

It should be noted that the burr may be cut in the third embodiment as in the first and second embodiments. That is, the third embodiment can be applied to the manufacturing system 200 shown in FIG. This is the end of the description of the third embodiment.

As described above, in each embodiment, the image is taken so that the pixel value of the reflected light or the reflected light in the first direction with respect to the transmissive member becomes a certain value or more. Then, in the second embodiment, the shape relating to the defect that the first surface (eg, the upper surface on the imaging device side) of the transmissive member constituting the structure is generated by further photographing the reflected light in the second direction. Recognize. In each embodiment, the burr of the member (rubber frame 102) constituting the structure is exemplified as a defect, but the object in which the defect occurs is not limited to this. For example, recognizing an object that does not form a structure, such as adhesion of impurities, is also included in each embodiment.

Further, a structure in which different members or different parts of the same member are arranged at a plurality of interfaces of the transparent member can also be inspected.

In addition, although burrs have been described as an example of defects in this embodiment, the structure should be inspected by recognizing the shape of defects such as excess / deficiency with respect to dimensions such as parting line, chipping, and thickness, in addition to burrs. Can be applied to.

Although glass 101 was used as the transparent member, it is also possible to use another transparent member such as an acrylic plate. Further, in this embodiment, the rubber frame 102 is used as a member in which a defect occurs, but it can also be applied to other members and materials. As the other member, a transparent member such as plastic or glass may be used.

Furthermore, the inspection of each embodiment can be applied to a structure in which a defect occurs regarding the interface of the transparent member constituting the structure. Here, what causes a defect related to the interface means that when the structure is observed from a certain direction, a defect observed through the transmissive member and a defect observed without the defect occur. Therefore, the interface associated with the defect is not limited to the upper surface and the lower surface. That is, the expression such as front surface and back surface, front surface and side surface, etc. does not matter. Therefore, each interface can be expressed as a first surface, a second surface, and so on. Further, the shape of the interface includes not only a flat surface but also a curved surface.

Furthermore, in each embodiment, there is an effect that the shape of the defect on the upper surface (first surface) can be recognized. Further, as an additional effect, it may be possible to distinguish between defects on the upper surface and the lower surface.

10 ... Composite member, 101 ... Glass, 102 ... Rubber frame, 1021 ... Bali, 1022 ... Bali, 20 ... Visual inspection device, 21 ... Camera, 22 ... Coaxial epi-illuminator, 221 ... Light source, 222 ... Reflector, 23 ... XY stage, 24 ... control unit, 241 ... processing unit, 242 ... input unit, 243 ... display unit, 244 ... communication I / F, 245 ... main memory unit, 2451 ... camera control program, 2452 ... coaxial epi-illuminator control program , 2453 ... XY stage control program, 2454 ... image processing program, 246 ... auxiliary storage, 2461 ... design information, 2462 ... shooting information

Claims

In the visual inspection method using the visual inspection device for inspecting the defect of the structure including the permeable member,
The visual inspection device is directed toward at least the first surface of the transparent member among the holding portion for holding the structure, the photographing portion for photographing the held structure, and the held structure. It has a lighting unit that irradiates lighting for photography in the imaging unit, and a control unit that controls the holding unit, the imaging unit, and the illumination unit.
The control unit
At least one of the photographing unit and the lighting unit so that the pixel value of at least a part of the first surface is equal to or higher than a predetermined value in the photographing by the photographing unit with respect to the first surface of the transmissive member. Control and
When the pixel value becomes equal to or higher than the predetermined value, the photographing unit is controlled so that at least a part of the structure is photographed with the reflected light in the first direction with respect to the first surface.
A visual inspection method comprising recognizing the shape of a defect relating to the first surface of the structure by using the photographed information of at least a part of the photographed structure.
In the visual inspection method according to claim 1,
The control unit
In the shooting by the shooting unit, the lighting unit is controlled so as to irradiate the first illumination so that the pixel value of at least a part of the first surface becomes equal to or higher than a predetermined value.
When the first illumination is irradiated, at least a part of the structure is controlled to capture the reflected light in the first direction generated by the first illumination. Visual inspection method.
In the visual inspection method according to claim 2,
The control unit
The photographing unit is controlled so as to capture the reflected light in the second direction, which is a direction different from the reflected light in the first direction.
A visual inspection method comprising recognizing the shape of the defect by using the photographing information and the second photographing information obtained by photographing the reflected light in the second direction.
In the visual inspection method according to claim 3,
A visual inspection method comprising controlling the imaging unit so that the control unit captures the reflected light in the second direction when the control unit cannot recognize the shape of the defect of the structure.
In the visual inspection method according to claim 2,
The holding portion holds the structure so that the first surface of the transmissive member faces the photographing portion and the illuminating portion.
The visual inspection method, wherein the reflected light in the first direction is a substantially specular reflected light on the first surface of the transmissive member.
In the visual inspection method according to claim 2,
A visual inspection method, wherein the control unit recognizes the shape of a defect related to a second surface existing in a transmission direction of transmitted light from the first surface.
In the visual inspection method according to claim 6,
The control unit
When the lighting unit is controlled so as to irradiate at least a part of the second surface existing in the transmission direction of the transmitted light from the first surface with the second illumination, and the second illumination is irradiated. The imaging unit is controlled so that at least a part of the structure captures the reflected light in the third direction with respect to the second illumination.
A visual inspection method comprising recognizing the shape of a defect relating to the second surface of the structure by using the photographing information for the reflected light in the third direction.
In the visual inspection method according to claim 7,
A visual inspection characterized in that the control unit controls the illumination unit so as to irradiate the second illumination so that the pixel value of at least a part of the second surface becomes equal to or higher than a predetermined value. Method.
The control unit in the visual inspection method according to claim 1 outputs a control signal for processing the defective shape to the processing apparatus for processing the structure.
A processing method for a structure, wherein the processing apparatus processes the structure according to the control signal.
In a visual inspection device that inspects defects in structures including permeable members,
A holding portion that holds the structure and
An imaging unit that photographs the held structure, and
Among the held structures, a lighting unit that irradiates at least the first surface of the transmissive member with illumination for photographing by the photographing unit, and a lighting unit.
It has a control unit that controls the holding unit, the photographing unit, and the lighting unit.
The control unit
At least one of the photographing unit and the lighting unit so that the pixel value of at least a part of the first surface is equal to or higher than a predetermined value in the photographing by the photographing unit with respect to the first surface of the transmissive member. Control and
When the pixel value becomes equal to or higher than the predetermined value, the photographing unit is controlled so that at least a part of the structure is photographed with the reflected light in the first direction with respect to the first surface.
A visual inspection apparatus comprising recognizing the shape of a defect relating to the first surface of the structure by using the photographed information of at least a part of the photographed structure.
In the visual inspection apparatus according to claim 10,
The control unit
In the shooting by the shooting unit, the lighting unit is controlled so as to irradiate the first illumination so that the pixel value of at least a part of the first surface becomes equal to or higher than a predetermined value.
When the first illumination is irradiated, at least a part of the structure is controlled to capture the reflected light in the first direction generated by the first illumination. Visual inspection equipment.
In the visual inspection apparatus according to claim 11,
The control unit
The photographing unit is controlled so as to capture the reflected light in the second direction, which is a direction different from the reflected light in the first direction.
A visual inspection apparatus characterized by recognizing the shape of the defect by using the photographing information and the second photographing information obtained by photographing the reflected light in the second direction.
In the visual inspection apparatus according to claim 12,
The control unit is a visual inspection device that controls the imaging unit so as to capture the reflected light in the second direction when the shape of the defect of the structure cannot be recognized.
In the visual inspection apparatus according to claim 11,
The holding portion holds the structure so that the first surface of the transmissive member faces the photographing portion and the illuminating portion.
The visual inspection apparatus, characterized in that the reflected light in the first direction is a substantially specular reflected light on the first surface of the transmissive member.
In the visual inspection apparatus according to claim 11,
The control unit is a visual inspection apparatus characterized in that it recognizes the shape of a defect related to a second surface existing in a transmission direction of transmitted light from the first surface.
In the visual inspection apparatus according to claim 15,
The control unit
When the lighting unit is controlled so as to irradiate at least a part of the second surface existing in the transmission direction of the transmitted light from the first surface with the second illumination, and the second illumination is irradiated. The imaging unit is controlled so that at least a part of the structure captures the reflected light in the third direction with respect to the second illumination.
A visual inspection apparatus comprising recognizing the shape of a defect relating to the second surface of the structure by using the photographing information for the reflected light in the third direction.
In the visual inspection apparatus according to claim 16,
The visual inspection unit is characterized in that the control unit controls the illumination unit so as to irradiate the second illumination so that the pixel value of at least a part of the second surface becomes equal to or higher than a predetermined value. Device.
The control unit of the visual inspection device according to claim 10 outputs a control signal for processing the defective shape to the processing device for processing the structure.
A processing device for a structure, wherein the processing device processes the structure according to the control signal.