Vision Inspection Systems and Controllers

Abstract

To eliminate a need for recreating a same inspection configuration separately on a controller side when having switched from smart camera-only operation to controller-implemented operation, thereby reducing a user's burden.SOLUTION: An image inspection system 1 is constituted such that a smart camera 102 and a normal camera 101 are connectable to each other. The smart camera 102 comprises a first environment configuration regarding hardware, and a storage unit 102e that stores a first inspection configuration for captured images. A controller 10 comprises a storage unit 14 that stores a second environment configuration regarding hardware and a first inspection configuration acquired from the smart camera 102, an acquisition unit 11 that acquires the image generated by the smart camera 102 or the normal camera 101, and an inspection unit 12 that executes inspection based on the first inspection configuration upon the image acquired by the acquisition unit 11.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present disclosure relates to an image inspection system and a controller capable of performing various inspections using images of a workpiece. [Background technology]

[0002] 2. Description of the Related Art Conventionally, a controller-type system has been known as an image inspection system, which includes a camera that captures an image of a workpiece and a controller connected to the camera, and in which the controller performs an inspection on the image captured by the camera.

[0003] Apart from this controller type, there are also cases in recent years where smart cameras are used that do not require a controller and can independently perform operations from image capture to image inspection (see, for example, Patent Document 1). [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Laid-Open No. 2013-158889 Summary of the Invention [Problem to be solved by the invention]

[0005] Smart cameras like those described in Patent Document 1 have the advantage of eliminating the need to worry about where to install the controller and simplifying the wiring.

[0006] On the other hand, it is anticipated that there will be situations where it will be necessary to add an inspection line for the same type of workpiece. In this case, image inspection will begin with a smart camera alone, and then operation will be switched to multi-camera operation.

[0007] When starting multi-camera operation, it is possible to increase the number of smart cameras, but because smart cameras are generally expensive, there are times when it is desirable to introduce normal cameras and controllers instead of smart cameras to reduce costs.In such cases, the inspection settings previously used with smart cameras also included hardware environment settings, so they could not be used for inspections using controllers, which are different hardware from smart cameras, and the inspection settings had to be recreated on the controller side, which was time-consuming.

[0008] The present disclosure has been made in consideration of such points, and its purpose is to reduce the burden on users by eliminating the need to recreate the same inspection settings separately on the controller side when switching from operation using only a smart camera to operation using a controller. [Means for solving the problem]

[0009] To achieve the above object, one aspect of the present disclosure can be based on an image inspection system including a normal camera equipped with an image generation function, a smart camera that is a camera equipped with an image inspection function in addition to the image generation function, and a controller to which the normal camera and the smart camera can be connected. The smart camera includes a storage unit that stores first environmental settings related to the smart camera's hardware and first inspection settings for images captured by the smart camera. The controller also includes a storage unit that stores second environmental settings related to the controller's hardware and the first inspection settings acquired from the smart camera, an acquisition unit that acquires images generated by the smart camera or the normal camera, and an inspection unit that performs inspections on the images acquired by the acquisition unit based on the first inspection settings.

[0010] Alternatively, a controller to which the normal camera and the smart camera can be connected may be assumed. In this case, the controller includes a storage unit that stores second environmental settings related to the hardware of the controller and the first inspection settings acquired from the smart camera, an acquisition unit that acquires images generated by the smart camera or the normal camera, and an inspection unit that performs an inspection based on the first inspection settings on the images acquired by the acquisition unit.

[0011] For example, if an inspection line is later added to a site where only smart cameras are in operation, and a normal camera and controller are introduced, the smart camera can also be connected to the controller. The smart camera's memory stores first environmental settings related to the hardware and first inspection settings for images captured by the smart camera. Because the first inspection settings are settings for images, they can also be used for image inspections using a normal camera and controller. Therefore, by storing the first inspection settings in the smart camera's memory in the controller's memory and using them during image inspections using a normal camera and controller, there is no need to recreate the inspection settings on the controller side.

[0012] According to another aspect, a controller may be capable of simultaneously connecting multiple cameras, including the normal camera and / or the smart camera. The image inspection system further includes an input unit that accepts selection of either a camera independent mode in which images acquired from each of the multiple cameras connected to the controller are inspected based on inspection settings set for each camera, or a camera synchronous mode in which images acquired from each of the multiple cameras connected to the controller are inspected based on inspection settings common to the multiple cameras. In this case, the inspection unit accepts the selection of the camera independent mode via the input unit, and can apply the first inspection setting acquired from the smart camera to images acquired from other cameras only when operating in the camera independent mode.

[0013] In other words, the controller can operate in either a camera independent mode, in which images acquired from multiple cameras are inspected based on individual inspection settings, or a camera linked mode, in which images acquired from multiple cameras are inspected based on common inspection settings, enabling a wide range of uses. Only in the camera independent mode can the first inspection setting acquired from a smart camera be used to inspect images acquired from other cameras.

[0014] In another aspect, the first inspection setting includes at least an imaging setting tool, and the imaging setting tool includes a camera designation parameter for designating one of a plurality of cameras connected to the controller. When the input unit accepts selection of the camera linked mode, the input unit accepts designation of a plurality of cameras connected to the controller via the camera designation parameter of the common inspection setting, while when the input unit accepts selection of the camera independent mode, the input unit can accept inspection settings for each camera.

[0015] In other words, in camera linkage mode, it is necessary to specify multiple cameras that are the target of linkage mode within the imaging tool with inspection settings common to each camera, so it is not possible to use inspection settings from smart cameras, which do not require camera specification in the imaging tool.On the other hand, in camera independent mode, inspection settings are assigned to each camera, so there is no need to specify a camera, and inspection settings can be shared between smart cameras.

[0016] According to another aspect, a smart camera may be capable of generating images only without performing the image inspection function upon detecting that it is connected to the controller. In this case, the inspection unit can perform an inspection based on the first inspection setting on images generated by the smart camera connected to the controller. Therefore, by allowing the smart camera to focus on image generation and the controller to focus on inspection, the overall efficiency of the inspection can be improved.

[0017] In another aspect, the storage unit of the controller may store second inspection settings for images captured by a camera connected to the controller. Furthermore, the storage unit of the smart camera may store the second inspection settings acquired from the controller. In this case, the inspection unit can perform an inspection based on the second inspection settings on images generated by the smart camera. This configuration allows the inspection settings on the controller to be reused for the smart camera, making it easy to switch from operation using a controller to operation using the smart camera alone.

[0018] In another aspect, the first environment setting for the smart camera hardware and the second environment setting for the controller hardware may include a correspondence between terminals of each piece of hardware and information about signals received from the outside via the terminals. In other words, which terminal receives instructions for capturing images or lighting depends on the hardware. Therefore, by separating the correspondence from the inspection settings, the inspection settings can be reused.

[0019] According to another aspect, the inspection unit of the controller may further include a plurality of cores for performing an inspection on the image acquired by the acquisition unit. In this case, each core constituting the plurality of cores is assigned to each of a plurality of cameras connected to the controller, and performs an inspection based on the first inspection setting on the image generated by the corresponding camera, thereby suppressing jitter (fluctuations in the signal waveform) between cores.

[0020] In another aspect, the first inspection setting may include first imaging conditions for the smart camera. In this case, if the normal camera cannot reproduce parameter values ​​included in the first imaging conditions, the inspection unit generates second imaging conditions by modifying the parameter values, and performs an inspection on images generated by the normal camera based on an inspection setting in which the first imaging conditions of the first inspection setting are replaced with the second imaging conditions. That is, if the smart camera is equipped with a processing device capable of more advanced processing than the normal camera and parameter values ​​that perform advanced processing are set, it is possible that the normal camera cannot reproduce the parameter values ​​of the imaging conditions set for the smart camera. If reproduction is not possible, according to this aspect, the parameter values ​​can be modified for the normal camera and the inspection setting can be executed. If the adjustment of parameter values ​​does not significantly affect the inspection results, usability can be improved by allowing the inspection setting to be reused.

[0021] According to another aspect, when the inspection unit determines that the normal camera does not have a function for executing the first imaging condition, the inspection unit may invalidate the first imaging condition in the first inspection setting and output error information related to the invalidation to the outside. According to this configuration, when the normal camera does not have a function for reproducing the imaging condition set for the smart camera, the controller can invalidate the imaging condition. When the imaging condition is invalidated, the controller may output error information to the outside.

[0022] In another aspect, the storage unit of the smart camera can further store a first image capture setting for the smart camera, and the controller can further include an input unit that accepts input of a second image capture setting for the normal camera, and the inspection unit can perform an inspection based on the first inspection setting on an image generated by the normal camera based on the second image capture setting.

[0023] With this configuration, the inspection settings can be separated not only from the environment settings but also from the imaging settings. By inputting only the second imaging settings and adding them separately, the inspection settings acquired from the smart camera can be reused even for a normal camera that is different from the smart camera that acquired the inspection settings. [Effects of the Invention]

[0024] As explained above, when switching from operation using only a smart camera to operation using a controller, there is no need to recreate the same inspection settings separately on the controller side, thereby reducing the burden on the user. [Brief explanation of the drawings]

[0025] [Figure 1] 1 is a schematic diagram illustrating a configuration of an image inspection system according to an embodiment of the present invention. [Figure 2] FIG. 1 is a block diagram of an image inspection system. [Figure 3] FIG. 10 is a diagram illustrating an overview of a camera linkage mode. [Figure 4] FIG. 10 is a diagram illustrating an overview of a camera independent mode. [Figure 5] FIG. 10 is a diagram illustrating an example of connecting an existing smart camera to an additional controller. [Figure 6] 10 is a flowchart showing an example of a procedure for transferring a first inspection setting of a smart camera to a controller via a PC. [Figure 7] 10 is a flowchart showing an example of a procedure for directly transferring a first inspection setting of a smart camera to a controller. [Figure 8] FIG. 1 is a diagram illustrating an example of a system in which multiple smart cameras are replaced with an equal number of normal cameras. [Figure 9] 10 is a flowchart showing an example of a procedure for transferring first inspection settings of multiple smart cameras to a controller via a PC. [Figure 10] FIG. 10 is a diagram showing an example of constructing an image inspection system using a smart camera and a normal camera in combination. [Figure 11] 10 is a flowchart showing an example of a procedure for transferring a first inspection setting set in the smart camera to the controller. [Figure 12] FIG. 1 is a diagram illustrating an example of operation of an image inspection system. [Figure 13] 10 is a flowchart showing an example of a procedure for selecting a mode of the image inspection system. [Figure 14] 10A to 10C are diagrams illustrating four usage patterns of the second mode of the image inspection system. [Figure 15] FIG. 1 is a diagram illustrating an example of operation of an image inspection system. [Figure 16] 10 is a flowchart showing an example of a setting procedure based on the type recognition of the connected camera. [Figure 17] FIG. 10 is a diagram showing a display example of an image generation process that can be used with the smart camera. DETAILED DESCRIPTION OF THE INVENTION

[0026] Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. Note that the following description of the preferred embodiments is merely exemplary in nature and is not intended to limit the present invention, its applications, or its uses.

[0027] FIG. 1 is a schematic diagram showing the configuration of an image inspection system 1 according to an embodiment of the present invention, and FIG. 2 is a block diagram of the image inspection system 1. The image inspection system 1 includes a normal camera 101 equipped with an image generation function, a smart camera 102 that is a camera equipped with an image inspection function in addition to the image generation function, and a controller 10 to which the normal camera 101 and the smart camera 102 can be connected. The image generation function is a function that captures an image of a workpiece (also referred to as an inspection target) and generates an inspection target image. The image inspection function is a function that can perform one or more types of inspection processing on the inspection target image, and is not included in the normal camera 101. In other words, a normal camera refers to a camera that does not include an image inspection function. The image inspection function is included not only in the smart camera 102 but also in the controller 10.

[0028] The image inspection system 1 is a system that inspects a workpiece imaged by a normal camera 101 or a smart camera 102. The normal camera 101 and the smart camera 102 are installed, for example, on a line along which a plurality of workpieces are transported in sequence, and are capable of sequentially capturing images of the workpieces as they are transported.

[0029] Since the normal camera 101 is not equipped with an image inspection function, inspection of the inspection target image generated by the normal camera 101 is performed by the image inspection function of the controller 10. On the other hand, since the smart camera 102 is equipped with an image inspection function, inspection of the inspection target image generated by the smart camera 102 can be performed by the smart camera 102. Note that, as will be described later, all or part of the inspection of the inspection target image generated by the smart camera 102 may be performed by the image inspection function of the controller 10.

[0030] 1 and 2 show an operation mode in which one normal camera 101 and one smart camera 102 are connected to the controller 10 at the same time, but the number of cameras connected is not limited to this, and an operation mode in which multiple normal cameras 101 and multiple smart cameras 102 are connected to the controller 10 at the same time can also be adopted.

[0031] It is also possible to adopt an operational configuration in which one or more normal cameras 101 are connected to the controller 10 and the smart camera 102 is not connected to the controller 10, or to adopt an operational configuration in which one or more smart cameras 102 are connected to the controller 10 and the normal camera 101 is not connected to the controller 10.

[0032] The smart camera 102 can inspect the image of the inspection target by itself, and can be used without being connected to the controller 10. For example, if there is one line on which workpieces are transported and only one location on the workpiece needs to be inspected, there will be no particular problem if one smart camera 102 is introduced to the site and operated without introducing a controller 10.

[0033] However, with the development of image inspection technology using AI (artificial intelligence), the load of inference and learning may cause a situation where the processing power of the smart camera 102 alone is insufficient. In other words, after starting image inspection operation using the smart camera 102 alone, there may be a need to switch to inspection operation using the controller 10 in order to increase processing power.

[0034] Furthermore, after installing the smart camera 102, if you want to add more lines or inspect multiple locations on a workpiece simultaneously, you will need to install multiple additional cameras to increase processing capacity. Installing multiple smart cameras 102 is an option, but the cost of installing multiple smart cameras 102 is high because they are equipped with an image inspection function. Therefore, installing multiple cheaper normal cameras 101 to increase processing capacity is an option. Because the normal camera 101 does not have an image inspection function, it must be installed together with the controller 10. However, since multiple normal cameras 101 can be connected to the controller 10 and images generated by each normal camera 101 can be inspected at high speed, fewer controllers 10 can be installed, and the total installation cost can be reduced compared to installing multiple smart cameras 102.

[0035] A problem that arises when a normal camera 101 and controller 10 are introduced later is that when an inspection using the newly introduced normal camera 101 is to be performed on the controller 10, the inspection settings must be newly created on the controller 10 side, which is time-consuming and laborious. The image inspection system 1 according to this embodiment is equipped with a function that can solve this problem. The configuration of each camera 101, 102 and the configuration of the controller 10 will be specifically described below.

[0036] (Normal camera configuration) As shown in FIG. 2, the normal camera 101 includes an imaging unit 101a and an illumination unit 101b. The imaging unit 101a is configured with an image sensor such as a CCD (charge-coupled device) or a CMOS (complementary metal-oxide semiconductor), but any image sensor capable of capturing an image of a workpiece will suffice. The illumination unit 101b is a unit that illuminates the workpiece when capturing an image with the imaging unit 101a, and includes a light-emitting element such as a light-emitting diode. Also, as shown in FIG. 1, the normal camera 101 is provided with an optical system 101d that includes a lens into which light reflected from the workpiece is incident.

[0037] The imaging unit 101a and the illumination unit 101b are controlled by a camera-side control unit 101c built into the normal camera 101. For example, when a trigger signal is input from the outside, the camera-side control unit 101c controls the imaging unit 101a and the illumination unit 101b according to a preset imaging setting. The illumination unit 101b illuminates the workpiece at a predetermined timing, and the imaging unit 101a captures the illuminated workpiece for a predetermined exposure time. The light intensity of the illumination unit 101b and the gain of the imaging unit 101a are specified by the imaging setting. The camera-side control unit 101c may be built into a device other than the normal camera 101, such as the controller 10. The illumination unit 101b may be separate from the normal camera 101 and may be configured to illuminate the workpiece from a location different from the imaging unit 101a. In this case, the illumination unit 101b can be controlled by the controller 10.

[0038] Furthermore, the camera-side control unit 101c generates an inspection object image by performing various processes, such as preprocessing, on the image acquired by the imaging unit 101a. Therefore, the image generation function of the normal camera 101 is configured by the imaging unit 101a and the camera-side control unit 101c. The inspection object image generated by the image generation function of the normal camera 101 is output to the controller 10 via a signal line. Note that the part that controls the imaging unit 101a and the part that performs various processes on the image acquired by the imaging unit 101a may be configured separately. Alternatively, the image acquired by the imaging unit 101a may be output directly to the controller 10. In this case, the image generation function of the normal camera 101 is configured by the imaging unit 101a.

[0039] (Smart camera configuration) The smart camera 102 includes an imaging unit 102a, an illumination unit 102b, a camera control unit 102c, and an optical system 102d. The imaging unit 102a and illumination unit 102b of the smart camera 102 are configured similarly to the imaging unit 101a and illumination unit 101b of the normal camera 101. Therefore, the image generation function of the smart camera 102 is configured by the imaging unit 102a and the camera control unit 102c. The camera control unit 102c and optical system 102d of the smart camera 102 are also configured similarly to the camera control unit 101c and optical system 101d of the normal camera 101.

[0040] The image generation function of the smart camera 102 includes a function that processes acquired images into images suitable for image inspection, thereby improving the performance of image inspection and reducing the processing load. For example, the image generation function of the smart camera 102 also includes a function that processes images with edge emphasis using a Sobel filter, and processes images with defect areas of the inspection target emphasized based on information learned in advance.

[0041] The image generation function of the smart camera 102 also includes image synthesis processing. For example, it is possible to acquire multiple images captured under different imaging conditions and synthesize them to create an image suitable for image inspection. Specifically, it has the function of acquiring multiple images with different exposure times and synthesizing them to generate a high dynamic range image, and the function of synthesizing multiple images acquired under different lighting conditions (light intensity, light wavelength, lighting direction, etc.) to generate an image that highlights defective parts of the inspection object. Defective parts are, for example, scratches.

[0042] The image generation function of the smart camera 102 also includes ROI processing, which cuts out a part of the acquired image. For example, this includes processing to cut out a part of the acquired image based on pre-set pixel position information, and processing to search for the position of the inspection object within the acquired image and cut out the part of the image that shows the inspection object.

[0043] The smart camera 102 also includes a storage unit 102e and an inspection unit 102f. The storage unit 102e stores first environmental settings related to the hardware of the smart camera 102 and first inspection settings for images captured by the smart camera 102. The first environmental settings serve to absorb differences between the hardware of the smart camera 102 and the first inspection settings, and the first environmental settings differ depending on the hardware.

[0044] The first environmental settings include settings for the input / output terminals of the smart camera 102 hardware, network settings, industrial Ethernet settings such as data size, the camera name, account settings, etc. The network settings also include IP address settings and FTP server settings as the output destination of the output settings. Examples of industrial Ethernet include EtherNet / IP, PROFINET, and EtherCAT. The account settings also include a password for administrator privileges and whether or not each operation can be performed with operator privileges (whether or not inspection settings can be switched, whether or not inspection settings can be saved, etc.). The first environmental settings also include the correspondence between the input / output terminals of the smart camera 102 hardware and information about signals received from outside via those terminals.

[0045] On the other hand, the first inspection setting does not depend on the hardware of the smart camera 102 and is composed only of settings related to the inspection. Therefore, the first inspection setting may be the same even if the hardware is different. This first inspection setting includes an imaging setting tool related to the imaging setting of the camera, a positioning setting tool related to the positioning setting for positioning the inspection area relative to the inspection object image captured by the camera, an inspection setting tool related to the inspection setting for setting the inspection content for the positioned inspection area, an output tool related to the output setting for setting the output of the inspection results, and an operation screen setting. The positioning setting tool is executed after the imaging setting tool, the inspection setting tool is executed after the positioning setting tool, and the output tool is executed after the inspection setting tool.

[0046] The imaging setting tool includes first imaging conditions of the smart camera 102. Furthermore, the imaging setting tool also includes first imaging settings related to imaging settings of the smart camera 102, and the first imaging settings are stored in the storage unit 102e of the smart camera 102.

[0047] When multiple cameras are connected to the controller 10, the imaging setting tool may include a camera designation parameter for designating one of the multiple cameras connected to the controller 10. The camera designation parameter makes it possible to designate a specific camera.

[0048] When the smart camera 102 acquires a second inspection setting from the controller 10 (described later), the memory unit 102e stores the second inspection setting acquired from the controller 10. The second inspection setting is an inspection setting for an image captured by a camera connected to the controller 10, and is stored in the memory unit 14 of the controller 10.

[0049] The inspection unit 102f of the smart camera 102 may be configured as hardware or a combination of hardware and software. The inspection unit 102f acquires the first inspection setting stored in the storage unit 102e and performs an inspection based on the first inspection setting. Specifically, the inspection unit 102f performs a positioning process for an inspection area relative to the inspection target image generated by the smart camera 102, then performs an inspection process for the positioned inspection area using preset inspection content, and performs an output process to output the inspection results to an external device. The positioning process by the inspection unit 102f uses a preset positioning setting. For example, if only a portion of a workpiece is to be inspected, the positioning setting includes information specifying the relative position and size of the inspection target area on the workpiece. The positioning setting also includes a process for extracting the inspection target area, rotating the inspection target area so that it has a desired orientation, and enlarging or reducing the inspection target area so that it has a desired size.

[0050] The inspection process by the inspection unit 102f uses pre-set inspection settings. The inspection settings include the inspection details for the positioned inspection area. Specific examples include the presence or absence of parts assembled on the workpiece, the presence or absence of scratches, whether the dimensions are within the standard, and the presence or absence of printing.

[0051] The output process by the inspection unit 102f uses preset output settings, which include settings related to the output of the inspection results, such as the data to be output (whether or not to output the inspection target image, etc.), the output destination of the inspection results, the output timing, etc.

[0052] The smart camera 102 is configured to automatically detect whether or not it is connected to the controller 10. That is, for example, when the smart camera 102 is connected to the controller 10 via an industrial Ethernet, the smart camera 102 determines whether or not the controller 10 is connected to the industrial Ethernet of the smart camera 102, and when it detects that it is connected to the controller 10 based on the determination result, it is configured to be able to perform only image generation without performing the image inspection function. In this case, the smart camera 102 operates like the normal camera 101. Note that the smart camera 102 may also operate like the normal camera 101 by user setting.

[0053] (Controller and peripheral device configuration) A display unit 30 and an operation unit 40 are connected to the controller 10. The display unit 30 and the operation unit 40 constitute the image inspection system 1. The display unit 30 is configured with, for example, a liquid crystal display device or an organic EL display device, and is controlled by the controller 10. The display unit 30 displays, for example, a display image generated by the controller 10, an inspection object image generated by the normal camera 101 and the smart camera 102, inspection results, etc.

[0054] The operation unit 40 is composed of operation devices and the like that allow the user to perform various input operations. When the operation unit 40 is operated, the operation content is detected by the input unit 13 of the controller 10. The operation unit 40 includes, for example, a keyboard 40a, a mouse 40b, a touch panel 40c, and the like. The touch panel 40c is configured to be able to detect touch operations by the user. The touch panel 40c and the display unit 30 may be integrated together, in which case, for example, a user interface displayed on the display unit 30 may be directly operated using the touch panel 40c.

[0055] The image inspection system 1 may include a personal computer (hereinafter referred to as PC) 35, which is shown only in Fig. 1. When the image inspection system 1 includes the PC 35, the display unit 30 and the operation unit 40 may be connected to the main body of the PC 35. In this case, the display unit 30 is controlled via the main body of the PC 35, and the operation state of the operation unit 40 is acquired by the controller 10 via the main body of the PC 35.

[0056] Furthermore, when the PC 35 is included, some or all of the functions of the controller 10 may be executable by the PC 35. That is, since some or all of the functions of the controller 10 can be configured by a CPU (Central Processing Unit), ROM, RAM, etc. built into the PC 35, the functions of the controller 10 can also be executed by devices other than the controller 10.

[0057] The controller 10 includes an acquisition unit 11 that acquires images captured by a smart camera 102 or a normal camera 101 connected to the controller 10, an inspection unit 12, an input unit 13, a storage unit 14, and a setting unit 15. The storage unit 14 is configured, for example, with a solid state drive or a hard disk drive. The storage unit 14 may be provided in the PC 35, and in this case, various data can be stored in the same way as when the storage unit 14 is provided in the controller 10. The storage unit 14 may also be configured as an external storage device connected to the controller 10 or the PC 35 via a network.

[0058] The acquisition unit 11, the inspection unit 12, the input unit 13, and the setting unit 15 may be configured as hardware or a combination of hardware and software. For example, the controller 10 has a built-in CPU. This CPU is connected to a ROM, a RAM, etc., processes given signals and data, performs various calculations, and outputs the calculation results. The acquisition unit 11, the inspection unit 12, the input unit 13, and the setting unit 15 may be configured by a CPU, a ROM, a RAM, etc. capable of performing such operations. Furthermore, the acquisition unit 11, the inspection unit 12, the input unit 13, and the setting unit 15 may each be configured as an independent arithmetic processing device.

[0059] The acquisition unit 11 acquires not only the inspection target image generated by the smart camera 102 but also the first inspection setting set in the smart camera 102. The first inspection setting acquired by the acquisition unit 11 from the smart camera 102 is stored in the storage unit 14. The storage unit 14 also stores a second inspection setting for the image captured by the normal camera 101 or the smart camera 102 connected to the controller 10.

[0060] Furthermore, the storage unit 14 also stores second environmental settings related to the hardware of the controller 10. The second environmental settings have the role of absorbing differences in hardware and include items similar to the first environmental settings related to the hardware of the smart camera 102. That is, the second environmental settings include settings for input / output terminals of the hardware of the controller 10, network settings, industrial Ethernet settings such as data size, camera name, account settings, etc. The second environmental settings also include the correspondence between the input / output terminals of the hardware of the controller 10 and information related to signals received from outside via those terminals.

[0061] The inspection unit 12 of the controller 10 is a part that performs an inspection based on a first inspection setting on an image acquired by the acquisition unit 11. When the acquisition unit 11 acquires an image to be inspected from the normal camera 101, the inspection unit 12 performs an inspection based on the first inspection setting on the image to be inspected acquired from the normal camera 101. The inspection unit 12 of the controller 10 is configured to be able to perform at least the same processing as the inspection unit 102f of the smart camera 102. In this embodiment, the inspection unit 12 of the controller 10 has higher processing performance than the inspection unit 102f of the smart camera 102, and can complete complex processing at high speed. The inspection unit 12 of the controller 10 can also perform inference processing using AI, for example, at high speed.

[0062] Furthermore, when the acquisition unit 11 acquires an inspection target image from the smart camera 102, the inspection unit 12 can also perform an inspection based on the first inspection setting on the inspection target image acquired from the smart camera 102. In this embodiment, as described above, when it is detected that the smart camera 102 is connected to the controller 10, the inspection unit 12 of the controller 10 is configured to be able to perform only the generation of the inspection target image without performing the image inspection function. Therefore, the inspection unit 12 of the controller 10 performs an inspection based on the first inspection setting on the image generated by the smart camera 102 connected to the controller 10. Furthermore, when the storage unit 14 stores a second inspection setting, the inspection unit 12 performs an inspection based on the second inspection setting on the image generated by the smart camera 102. This makes it possible to use the second inspection setting on the controller 10 side in the smart camera 102, thereby easily switching from operation using the controller 10 to operation using the smart camera 102 alone.

[0063] In this embodiment, the inspection unit 12 of the controller 10 includes multiple cores. When multiple cameras (normal camera 101 and / or smart camera 102) are connected to the controller 10, each core constituting the multiple cores is assigned to a corresponding camera of the multiple cameras connected to the controller 10, and performs an inspection based on a first inspection setting on images generated by the corresponding camera. FIG. 2 illustrates an example in which the inspection unit 12 includes a first core 12a and a second core 12b. In this example, assuming that a first camera and a second camera are connected to the controller 10, the first core 12a is assigned to the first camera and performs an inspection based on the first inspection setting on images generated by the corresponding first camera, and the second core 12b is assigned to the second camera and performs an inspection based on the first inspection setting on images generated by the corresponding second camera. The number of cores may be three or more, and the number of cameras connected to the controller 10 may also be three or more. In this way, by assigning each core of the multi-core to each camera, jitter (fluctuations in the signal waveform) between cores can be suppressed.

[0064] The smart camera 102 is often a higher performance camera than the normal camera 101, and there are cases where an imaging condition that can be executed by the smart camera 102 cannot be executed by the normal camera 101. In response to this, when the normal camera 101 cannot reproduce a parameter value included in the first imaging condition of the smart camera 102, the inspection unit 12 of the controller 10 generates a second imaging condition in which the parameter value is corrected, and performs an inspection on the image generated by the normal camera 101 based on the inspection setting in which the first imaging condition of the first inspection setting is replaced with the second imaging condition.

[0065] That is, the inspection unit 12 reads parameter values ​​executable by the normal camera 101 and compares the parameter values ​​included in the first imaging condition of the smart camera 102 with the parameter values ​​executable by the normal camera 101. Then, the inspection unit 12 determines whether the normal camera 101 can reproduce the parameter values ​​included in the first imaging condition of the smart camera 102. If it is determined that the normal camera 101 can reproduce the parameter values ​​included in the first imaging condition of the smart camera 102, the inspection unit 12 performs inspection on the image generated by the normal camera 101 under the first imaging condition of the smart camera 102. On the other hand, if it is determined that the normal camera 101 cannot reproduce the parameter values ​​included in the first imaging condition of the smart camera 102, the inspection unit 12 performs inspection on the image generated by the normal camera 101 under the second imaging condition in which the parameter values ​​have been corrected to be compatible with the normal camera 101.

[0066] For example, the smart camera 102 may have a zoom function, but the normal camera 101 may not. In such a case, among the parameter values ​​of the smart camera 102, the parameter values ​​that can be reproduced by the normal camera 101 include the exposure time, gain, ROI, etc.

[0067] On the other hand, among the parameter values ​​of the smart camera 102 with a zoom function, parameter values ​​that cannot be reproduced by the normal camera 101 include the zoom lens position, focus lens position, downscaling coefficient, etc. Downscaling is a process of reducing the pixel resolution of a target image. For example, by performing downscaling on a captured image corresponding to an output area, which is all or part of the area of ​​an image sensor, it is possible to generate an inspection target image with a smaller number of pixels than the captured image.

[0068] Furthermore, the first imaging condition of the smart camera 102 may include a function that can be executed only by the smart camera 102, i.e., a function that cannot be executed by the normal camera 101. In this way, if the normal camera 101 does not have the function to execute the first imaging condition of the smart camera 102, the inspection unit 12 of the controller 10 disables the first imaging condition in the first inspection setting. An example of a function that cannot be executed by the normal camera 101 is advanced image processing.

[0069] When the inspection unit 12 of the controller 10 invalidates the first imaging condition, the invalidation may be output to the outside as error information. For example, the user can be notified that the first imaging condition has been invalidated by displaying the error information on the user interface of the display unit 30.

[0070] The input unit 13 is configured to be able to accept input of second imaging settings related to imaging by the normal camera 101. A user can input the second imaging settings using the operation unit 40 while viewing the user interface displayed on the display unit 30. The user's input operation is accepted by the input unit 13 and stored in the storage unit 14 as the second imaging settings. In this case, the inspection unit 12 of the controller 10 performs an inspection based on the first inspection settings on an image generated by the normal camera 101 based on the second imaging settings. In other words, the inspection settings can be separated not only from the environment settings but also from the imaging settings. Therefore, the first inspection settings acquired from the smart camera 102 can be applied to a normal camera different from the smart camera 102 that acquired the inspection settings by separately adding only the imaging settings as the second imaging settings.

[0071] The inspection unit 12 of the controller 10 can also be configured to determine a final result based on the result of the inspection process performed on the inspection target image acquired from the normal camera 101 and the result of the inspection process performed by the smart camera 102 received from the smart camera 102. In this case, if either the result of the inspection process performed on the inspection target image acquired from the normal camera 101 or the result of the inspection process performed by the smart camera 102 is bad, the final result may be determined to be bad, or the final result may not be determined to be good unless both the result of the inspection process performed on the inspection target image acquired from the normal camera 101 and the result of the inspection process performed by the smart camera 102 are good.

[0072] (Camera independent mode / Camera linked mode) The controller 10 is configured to be switchable between a camera independent mode and a camera linked mode. The camera independent mode is a mode in which an inspection of images acquired from each of a plurality of cameras connected to the controller 10 is performed based on the inspection settings set for each camera. In the camera independent mode, each camera connected to the controller operates independently like a smart camera. Separate inspection settings can be set for each camera, and the settings may be different or the same for each camera.

[0073] On the other hand, the camera linkage mode is a mode in which the images acquired from the multiple cameras connected to the controller 10 are inspected based on inspection settings that are common to the multiple cameras. In the camera linkage mode, all the cameras connected to the controller 10 operate in coordination, and the inspection settings are assigned to the controller 10. For example, this mode can be applied to cases in which a single workpiece is photographed from different angles and methods, and the multiple images are inspected based on a single inspection setting.

[0074] The user can select either the camera independent mode or the camera linked mode by operating the operation unit 40. This operation is a mode selection operation. The input unit 13 is configured to be able to accept the selection of either mode.

[0075] When the input unit 13 receives the selection of the camera linked mode, the mode of the controller 10 becomes the camera linked mode, whereas when the input unit 13 receives the selection of the camera independent mode, the mode of the controller 10 becomes the camera independent mode. This allows for a wide range of usage.

[0076] When the controller 10 is in camera linkage mode, the smart camera 102 and the controller 10 cannot use the same inspection settings. This is because the inspection settings in camera linkage mode are used for multiple cameras, and it is necessary to specify the camera in the imaging setting tool. On the other hand, when performing an inspection in camera independent mode or using the smart camera 102, there is only one imaging unit, so it is not necessary to specify the camera in the imaging setting tool.

[0077] In response to this, the inspection unit 12 of the controller 10 is configured so that the input unit 13 accepts the selection of the camera independent mode, and the first inspection setting acquired from the smart camera 102 can be applied to images acquired from other cameras only when operating in the camera independent mode.

[0078] Furthermore, when the input unit 13 receives a selection of the camera linkage mode, it accepts the specification of multiple cameras connected to the controller 10 via camera specification parameters of a common inspection setting. On the other hand, when the input unit 13 receives a selection of the camera independent mode, it is configured to be able to accept inspection settings for each camera. That is, in the camera linkage mode, multiple cameras that are the targets of the camera linkage mode are specified in the imaging setting tool included in the inspection setting common to each camera. Because this camera specification is required, the inspection setting of the smart camera 102, which does not require camera specification in the imaging setting tool, cannot be used. However, in the camera independent mode, inspection settings are assigned to each camera, so camera specification is not required and inspection settings can be shared between the smart camera 102 and the camera.

[0079] For example, as shown in the overview of the camera linkage mode in Fig. 3, when normal cameras A and B are connected to the controller 10, both the inspection settings and the environment settings are common. On the other hand, Fig. 4 shows a case where, for example, there are multiple lines on the site that perform the same inspection, and the controller 10 is in the camera independent mode.

[0080] 4, there is a controller 10 to which a smart camera 102 and two normal cameras, camera A and camera B, are connected, and inspection settings and environment settings are stored in the smart camera 102. The inspection settings of the smart camera 102 include information on whether imaging is performed in response to a trigger signal or periodically without a trigger signal, as well as information on whether to illuminate the workpiece when imaging. The "Terminal 1" in the terminal settings included in the environmental settings of the smart camera 102 is a trigger signal as an imaging execution signal, and the "Terminal 2" is a lighting signal as an illumination execution signal. This relationship is a correspondence between the terminals of the hardware and information about signals received from the outside via the terminals.

[0081] The inspection settings of the controller 10 include information on whether imaging is performed in response to a trigger signal or periodically without a trigger signal, as well as information on whether the workpiece is illuminated when imaging is performed. "Terminal 1" of the terminal settings included in the environment settings of the controller 10 is a trigger signal that serves as an imaging execution signal for camera A, "Terminal 2" is an illumination signal that serves as an illumination execution signal for camera A, "Terminal 3" is a trigger signal that serves as an imaging execution signal for camera B, and "Terminal 4" is an illumination signal that serves as an illumination execution signal for camera B. In this way, the environment settings of the controller 10 include physical information that indicates which terminal receives the imaging execution signal or the illumination signal. The inspection settings include information on whether imaging is performed in response to a trigger signal or periodically without a trigger signal, as well as information on whether the workpiece is illuminated when imaging is performed.

[0082] (Specific example 1) For example, as shown in the upper part of Fig. 5, inspections were previously performed using only the smart camera 102, but when an item to be inspected was added or an inspection tool was added, the processing time became longer and the inspection takt time could not be met. On the other hand, the controller 10 has a higher processing capacity than the smart camera 102. Therefore, as shown in the lower part of Fig. 5, we decided to address this issue by adding a controller 10 in addition to the existing smart camera 102.

[0083] In this example, an existing smart camera 102 is connected to the controller 10, and the acquisition unit 11 of the controller 10 acquires the first inspection setting used by the smart camera 102 and stores it in the memory unit 14. The inspection unit 12 of the controller 10 performs an inspection based on the first inspection setting on the image generated by the smart camera 102. The smart camera 102 connected to the controller 10 does not use the image inspection function, so it operates as a camera similar to the normal camera 101.

[0084] There are at least two ways to acquire the first inspection settings used in the smart camera 102 in the controller 10. One is to transfer the first inspection settings of the smart camera 102 to the controller 10 via the PC 35, and the other is to transfer the first inspection settings of the smart camera 102 directly to the controller 10.

[0085] FIG. 6 is a flowchart showing an example of a procedure for transferring the first inspection setting of the smart camera 102 to the controller 10 via the PC 35. After starting, in step SA1, the PC 35 is connected to the smart camera 102. In step SA2, the first inspection setting of the smart camera 102 is transferred to the PC 35 and temporarily saved in the PC 35. After step SA2, in step SA3, the smart camera 102 is connected to the controller 10. In step SA4, the PC 35 is connected to the controller 10. The order of steps SA3 and SA4 may be reversed, or they may be performed simultaneously. After step SA4, in step SA5, the first inspection setting saved in the PC 35 is transferred to the controller 10.

[0086] 7 is a flowchart showing an example of a procedure for transferring the first inspection setting of the smart camera 102 directly to the controller 10 without going through the PC 35. In step SB1 after starting, the smart camera 102 is connected to the controller 10. Then, the process proceeds to step SB2, where the first inspection setting of the smart camera 102 is transferred to the controller 10.

[0087] (Specific example 2) As shown on the left side of Figure 8, assume that an image inspection system is configured using three smart cameras 102 on a certain line. If an image inspection system capable of performing an inspection equivalent to this is needed on another line, and it can be constructed inexpensively, the expensive smart cameras 102 can be replaced with normal cameras 101 and an additional controller 10 can be installed. This reduces the total cost. Generally, there is a cost advantage in replacing two or more smart cameras 102 with a combination of the same number of normal cameras 101 and controllers 10.

[0088] In this case, the existing smart cameras 102 are connected to the controller 10, and the acquisition unit 11 of the controller 10 acquires the first inspection settings used by each smart camera 102 and stores them in the memory unit 14. The first inspection settings of the three smart cameras 102 may differ from one another, in which case the acquisition unit 11 of the controller 10 acquires the three first inspection settings and stores them in the memory unit 14. The inspection unit 12 of the controller 10 assigns a corresponding first inspection setting to the image generated by each normal camera 101, and performs an inspection based on the assigned first inspection setting.

[0089] 9 is a flowchart showing an example of the procedure for transferring the first inspection settings of multiple smart cameras 102 to the controller 10 via the PC 35. After starting, in step SC1, the PC 35 is connected to the smart cameras 102. In step SC2, the first inspection settings of the smart cameras 102 are transferred to the PC 35 and temporarily saved in the PC 35. In step SC3, steps SC1 and SC2 are repeated the number of times equal to the number of smart cameras 102. This allows the first inspection settings of all of the multiple smart cameras 102 to be saved in the PC 35.

[0090] Thereafter, the process proceeds to step SC4, where the PC 35 is connected to the controller 10. In step SC5, the first inspection setting transferred to the PC 35 in step SC2 is transferred from the PC 35 to the controller 10. In step SC6, step SC5 is repeated the number of times equal to the number of normal cameras 101.

[0091] (Specific example 3) FIG. 10 shows an example of building an image inspection system using a smart camera 102 and a normal camera 101 together. As a premise, the maximum number of cameras that can be connected to the controller 10 is four. There are five lines on the site along which workpieces are transported, and a camera needs to be installed on each of lines 1 to 5. When using the controller 10 and normal camera 101 together, one camera is missing, as the maximum number of cameras that can be connected to the controller 10 is four. One option is to introduce a second controller 10 and normal camera 101, but the combined cost of the controller 10 and normal camera 101 would be higher than the cost of one smart camera 102. Therefore, one smart camera 102 is used to enable inspection of lines 1 to 5.

[0092] In this case, the acquisition unit 11 of the controller 10 acquires the first inspection setting of the smart camera 102 and stores it in the storage unit 14. The inspection unit 12 of the controller 10 assigns the corresponding first inspection setting to the image generated by each normal camera 101, and performs an inspection based on the assigned first inspection setting.

[0093] Specifically, as shown in the flowchart of FIG. 11, in step SD1 after starting, the first inspection setting is made in the smart camera 102. In step SD2, the PC 35 is connected to the smart camera 102. In step SD3, the first inspection setting of the smart camera 102 is transferred to the PC 35 and saved in the PC 35. In step SD4, the PC 35 is connected to the controller 10. In step SD5, the first inspection setting saved in the PC 35 in step SD3 is transferred from the PC 35 to the controller 10. Although not shown, if the inspection setting has been made in the controller 10 first, the inspection setting of the controller 10 may be transferred to the smart camera 102 and the inspection may be performed by the inspection process of the smart camera 102.

[0094] (Operation example) 12 is a diagram showing an example of operation of the image inspection system 1. The image inspection system 1 includes a normal camera 101 connected to a controller 10 and a smart camera 102 connected to a PLC (programmable logic controller) 9. An inspection start button 9a is connected to the PLC 9.

[0095] A first part W1 and a second part W2 are assembled to the workpiece W. The workpiece W assembled with the first part W1 and the second part W2 is placed on a mounting table (not shown) and inspected. The normal camera 101 is a camera for detecting the presence or absence of the first part W1, and the smart camera 102 is a camera for detecting the presence or absence of the second part W2.

[0096] When the user operates the inspection start button 9a, a trigger signal is output from the PLC 9 to the controller 10 and the smart camera 102. Receiving the trigger signal, the normal camera 101 and the smart camera 102 perform imaging processing to generate an inspection object image. The inspection object image generated by the normal camera 101 is inspected by the inspection unit 12 of the controller 10, and the inspection result is output from the controller 10 to the PLC 9. On the other hand, the inspection object image generated by the smart camera 102 is inspected by the inspection unit 102f of the smart camera 102, and the inspection result is output from the smart camera 102 to the PLC 9.

[0097] The PLC 9 determines whether the workpiece W is a good or defective product by combining the inspection results output from the controller 10 and the smart camera 102. The determination result is displayed on a display unit (not shown). If the workpiece W is defective, the defective part is clearly indicated on the image.

[0098] (Smart camera internal inspection mode / controller internal inspection mode) The controller 10 is configured to be switchable between a smart camera internal inspection mode (first mode) and a controller internal inspection mode (second mode). The smart camera internal inspection mode is a mode in which the smart camera 102 connected to the controller 10 independently performs the process of generating an inspection target image and the inspection process for the inspection target image, and the inspection process for the inspection target image is performed by the inspection unit 102f of the smart camera 102.

[0099] On the other hand, the in-controller inspection mode is a mode in which the smart camera 102 connected to the controller 10 performs the process of generating an image to be inspected and the process of transferring the image to the controller 10, and the controller 10 performs the inspection process on the transferred image to be inspected, and the inspection process on the image to be inspected is performed by the inspection unit 12 of the controller 10.

[0100] In the in-controller inspection mode, the inspection unit 12 of the controller 10 can also acquire, from the smart camera 102, an inspection target image generated by applying image processing to an image acquired by the smart camera 102. In this case, the inspection unit 12 of the controller 10 executes the inspection process on the inspection target image after the image processing by the smart camera 102.

[0101] Furthermore, in the in-controller inspection mode, the inspection unit 12 of the controller 10 can also acquire, from the smart camera 102, an inspection target image (composite image) generated by the smart camera 102 synthesizing multiple images acquired by the smart camera 102. In this case, the inspection unit 12 of the controller 10 executes inspection processing on the composite image synthesized by the smart camera 102.

[0102] Furthermore, in the in-controller inspection mode, the inspection unit 12 of the controller 10 can also acquire three-dimensional shape data generated by the smart camera 102 based on a plurality of captured images from the smart camera 102. The three-dimensional shape data is data that constitutes the image to be inspected, and in this case, the inspection unit 12 of the controller 10 executes inspection processing on the three-dimensional shape data generated by the smart camera 102.

[0103] The user can operate the operation unit 40 to set either the smart camera internal inspection mode or the controller internal inspection mode. This operation is a mode setting operation. The controller 10 may also perform a setting process to set either the smart camera internal inspection mode or the controller internal inspection mode, in which case it is a mode setting process. The mode setting operation and the mode setting process are accepted by the setting unit 15.

[0104] When the setting unit 15 accepts the setting of the smart camera internal inspection mode, the mode of the controller 10 becomes the smart camera internal inspection mode, whereas when the setting unit 15 accepts the selection of the controller internal inspection mode, the mode of the controller 10 becomes the controller internal inspection mode, and the inspection unit 12 of the controller 10 performs inspection processing on the image to be inspected.

[0105] When the intra-controller inspection mode is set, the setting unit 15 is configured to further accept selection of a first mode in which inspection processing is not performed on an inspection target image generated by one of the smart cameras 102 connected to the controller 10, but all of the inspection processing is performed on the transferred inspection target image. When the setting unit 15 accepts selection of the first mode, the inspection unit 12 of the controller 10 operates to execute all of the inspection processing. That is, there are cases in which multiple inspection processing is set to be performed on an inspection target image, which may impose a heavy processing load on the smart camera 102. In this case, by having the inspection unit 12 of the controller 10, which has a higher processing capacity than the smart camera 102, execute all of the inspection processing, the smart camera 102 can be dedicated to image generation, thereby improving inspection efficiency.

[0106] This is not a limitation, and when multiple smart cameras 102 are connected to the controller 10, the following is also possible. For example, when the in-controller inspection mode is set, the setting unit 15 may be configured to further accept selection of a second mode in which, without performing inspection processing on the inspection target images generated by the multiple smart cameras 102 connected to the controller 10, all of the inspection processes are performed on the multiple inspection target images transferred, and an overall judgment is made based on the results of each inspection process on the multiple inspection target images. When the setting unit 15 accepts the selection of the second mode, the inspection unit 12 of the controller 10 performs all of the inspection processes on the multiple inspection target images on which inspection processing has not been performed. Then, the inspection unit 12 acquires multiple inspection results and makes an overall judgment based on the results of each inspection process. For example, if even one of the multiple inspection results is bad, the overall judgment is made bad, or if all of the multiple inspection results are bad, the overall judgment is not made good.

[0107] Furthermore, when the in-controller inspection mode is set, the setting unit 15 may be configured to further accept the selection of a third mode in which part of the inspection process is executed on an inspection target image generated by one of the smart cameras 102 connected to the controller 10, and the remaining part of the inspection process is executed on the transferred inspection target image. When the setting unit 15 accepts the selection of the third mode, the inspection unit 12 of the controller 10 executes the remaining part of the inspection process that has not been executed.

[0108] Furthermore, when the in-controller inspection mode is set, the setting unit 15 may be configured to further accept selection of a fourth mode in which a part of each inspection process is executed on a plurality of inspection target images generated by a plurality of smart cameras 102 connected to the controller 10, the remaining processes of each inspection process are executed on the plurality of inspection target images transferred, and an overall judgment is made based on the results of each inspection process on the plurality of inspection target images. When the setting unit 15 accepts selection of the fourth mode, the inspection unit 12 of the controller 10 executes the remaining inspection processes on the plurality of inspection target images.

[0109] (Mode Selection) As described above, the image inspection system 1 of this embodiment allows the smart camera 102 and normal camera 101 to be freely connected and used. The procedure for selecting a mode in the image inspection system 1 will be described with reference to the flowchart shown in FIG. 13. In step SE1 after the start of the flowchart shown in FIG. 13, the time required for image inspection processing is calculated. In step SE2, it is determined whether the takt time is met in the first mode, in which the smart camera performs the inspection processing alone. If the takt time is not met in the first mode, the process proceeds to step SE3, in which the second mode, in which the smart camera and controller work together to perform the inspection processing, is selected. If the takt time is met in the first mode, the process proceeds to step SE4, in which the first mode is selected.

[0110] Step SE5 is a change in the inspection content. If the inspection content is changed, for example, because the workpiece has been changed, the process proceeds to step SE6, where the processing content is changed and the time required for processing is calculated. Step SE7 determines whether the takt time is met in the first mode. If the takt time is not met in the first mode, the process proceeds to step SE8, where the mode is changed to the second mode. If the takt time is met in the first mode, the process proceeds to step SE9, where the first mode is continued.

[0111] There may be cases where the system switches to the second mode due to insufficient processing performance while operating in the first mode. In this case, the parameters set in the first mode can be easily reused in the second mode. One method for doing this is for the controller 10 to acquire the parameters required for the image inspection from the camera and then execute the image inspection.

[0112] Next, usage forms 1, 2, 3, and 4 of the second mode in which the smart camera and the controller work together to execute the inspection process will be described with reference to Fig. 14. Fig. 14A in Fig. 14 shows usage form 1, in which an image is acquired by the smart camera 102, an inspection target image is generated, and then the inspection target image is acquired by the controller 10, and the inspection unit 12 of the controller 10 executes the inspection process.

[0113] FIG. 14B in FIG. 14 shows usage mode 2, in which smart cameras A and B each acquire an image, and smart cameras A and B each generate an image of the object to be inspected. After that, controller 10 acquires the multiple images of the object to be inspected, and inspection unit 12 of controller 10 performs inspection processing on the multiple images of the object to be inspected, and integrates the results (determines the final result).

[0114] FIG. 14C in FIG. 14 shows usage mode 3, in which an image is acquired by the smart camera 102, the smart camera 102 generates an image to be inspected, and the inspection unit 102f executes part of the inspection process (the part handled by the camera), and then the image to be inspected is acquired by the controller 10, and the inspection unit 12 of the controller 10 executes the remaining inspection process (the part handled by the controller) on the image to be inspected.

[0115] FIG. 14D in FIG. 14 shows usage mode 4, in which smart cameras A and B each acquire an image, smart cameras A and B each generate an image to be inspected, and inspection unit 102f executes part of the inspection process (the part handled by the camera), and then controller 10 acquires multiple images to be inspected, and inspection unit 12 of controller 10 executes the remaining inspection process (the part handled by the controller) for each image to be inspected.

[0116] (Operation example) FIG. 15 shows two examples of operation of the image inspection system 1. FIG. 15A of FIG. 15 shows an example of performing marking inspection. When performing marking inspection, first, the smart camera 102 captures images of the workpiece multiple times to obtain multiple images. The obtained images are then combined to generate a shape image. The smart camera 102 transfers the generated shape image to the controller 10. The controller 10 identifies a character area based on the shape image transferred from the smart camera 102, and performs OCR processing (character reading processing) on ​​the identified character area. The controller 10 outputs the obtained reading result to the outside.

[0117] Two smart cameras are used when performing marking inspection and scratch inspection. First, smart camera A takes multiple images of the workpiece to acquire multiple images. The acquired images are then combined to generate a shape image. Smart camera B also takes multiple images of the workpiece to acquire multiple images, and then combines the acquired images to generate a shape image. Smart camera A and smart camera B each transfer the generated shape images to controller 10. Controller 10 identifies character areas based on the shape images transferred from smart camera A and performs OCR processing on the identified character areas. Controller 10 outputs the obtained reading results to an external device. Controller 10 also determines whether or not there are scratches based on the shape images transferred from smart camera B. If controller 10 determines that the character reading results are the expected value and that there are no scratches, it determines the product as "good."

[0118] FIG. 15B in FIG. 15 shows an example in which the smart camera 102 performs the process up to the identification of the character area. The smart camera 102 captures multiple images of the workpiece by capturing multiple images. The captured images are then combined to generate a shape image. The smart camera 102 identifies the character area based on the generated shape image and transfers only the identified character area from the shape image to the controller 10. This improves the transfer speed. The controller 10 then performs OCR processing on the character area identified by the smart camera 102. The controller 10 outputs the obtained reading results to an external device. Note that the smart camera 102 may transfer the shape image to the controller 10 in parallel with the process of identifying the character area based on the shape image, and then transfer position information of the identified character area to the controller 10. The controller can then perform OCR processing on the character area based on the position information for the transferred shape image. Since the process of transferring position information generally takes less time than the process of transferring an image, this case also shortens the processing time.

[0119] When performing marking inspection and assembly inspection of parts with different colors, a smart camera 102, a normal camera 101, and a controller 10 are used. The smart camera 102 captures images of the workpiece multiple times to acquire multiple images. The acquired images are combined to generate a shape image. The smart camera 102 transfers the generated shape image to the controller 10. The controller 10 identifies a character area based on the shape image transferred from the smart camera 102 and performs OCR processing and calculates the ease of identification for the identified character area. The controller 10 identifies the type of workpiece from the OCR processing results and determines the color of the part that should be located at a predetermined position in the inspection target image generated by capturing an image of the workpiece with the normal camera 101. The normal camera 101 generates an inspection target image and transfers it to the controller 10. The controller 10 determines whether a part of the specified color is present at a predetermined position in the inspection target image transferred from the normal camera 101. The controller 10 determines the product as "good" if the ease of identification in the OCR processing is above a threshold and a part of the specified color is present at the inspection position in the inspection target image. Here, an example has been shown in which the controller 10 performs OCR processing and calculates the ease of identification on the shape image generated and transferred by the smart camera 102, but the smart camera 102 may also perform OCR processing and calculate the ease of identification and transfer the results of the OCR processing and the ease of identification to the controller 10. In this case, the inspection unit 12 of the controller 10 can determine the final result based on the result of the inspection processing performed on the inspection target image acquired from the normal camera and the result of the inspection processing received from the smart camera.

[0120] (Setting procedure based on the type of connected camera) The controller 10 is configured to be able to recognize the type of the connected camera, whether it is a smart camera 102 or a normal camera 101. The smart camera 102 and the normal camera 101 output signals or information that enable type recognition to the controller 10. If the camera model identified from the information output from the smart camera 102 and the normal camera 101 is included in the models pre-registered in the controller 10, the camera type is recognized. Based on the camera model, the number of pixels, whether color imaging is possible, functions, etc. can be identified.

[0121] The setting unit 15 is configured to change the processes that can be set for the camera depending on the recognized type. In the case of the smart camera 102, the processes that can be set for the smart camera 102 are set, while in the case of the normal camera 101, the processes that can be set for the normal camera 101 are set.

[0122] The setting procedure based on the type recognition of the connected camera will be specifically described below with reference to the flowchart shown in FIG. 16. After starting, in step SF1, the type of the camera connected to the controller 10 is recognized using the method described above. In step SF2, it is determined whether or not a smart camera 102 is connected to the controller 10. If a smart camera 102 is connected to the controller 10, the process proceeds to step SF3. In step SF3, the controller 10 displays on the display unit 30 image generation processes that can be used by the smart camera 102 connected to the controller 10. FIG. 17 shows an example of a display of available image generation processes. In this display example, a plurality of image generation processes are displayed in a list format, and the user can select any one of them. In step SF3, the image generation process selected by the user is identified and temporarily stored.

[0123] On the other hand, if the normal camera 101 is connected to the controller 10, the process proceeds directly to step SF4. In step SF4, an image inspection process that can be executed by either the controller 10 or the smart camera 102 is set.

[0124] Thereafter, the process proceeds to step SF5, where it is determined whether or not to automatically set the allocation of image inspection processing. When the image inspection processing includes multiple processes, the allocation of image inspection processing means separating the processes executed by the smart camera 102 from the processes executed by the controller 10. This determination is made based on the input operation of the user. If the user inputs that the allocation of image inspection processing is to be automatically set, the process proceeds to step SF6, whereas if the user inputs that the allocation of image inspection processing is not to be automatically set, the process proceeds to step SF7.

[0125] In step SF6, the controller 10 automatically determines the processing to be executed by the smart camera 102 and the processing to be executed by the controller 10. For example, the first half of the image inspection processing is to be executed by the smart camera 102, and the second half of the image inspection processing is to be executed by the controller 10.

[0126] In step SF7, the user designates the process to be executed by the smart camera 102 and the process to be executed by the controller 10. This completes the setting based on the type recognition of the connected camera.

[0127] (Effects of the embodiment) As described above, for example, if an inspection line is later added and a normal camera 101 and a controller 10 are introduced at a site where only a smart camera 102 was in operation, the smart camera 102 can also be connected to the controller 10. Because the first inspection setting of the smart camera 102 is a setting for an image, it can also be used in image inspection using the normal camera 101 and the controller 10. Therefore, by storing the first inspection setting of the smart camera 102 in the storage unit 14 of the controller 10 and using it during image inspection using the normal camera 101 and the controller 10, it is no longer necessary to recreate the inspection setting on the controller 10 side, thereby reducing the burden on the user.

[0128] Furthermore, the image inspection system 1 allows selection of a mode in which the smart camera 102 performs everything from generation processing of the image to be inspected by itself to inspection processing, or a mode in which the smart camera 102 performs generation processing of the image to be inspected and transfer processing to the controller 10, with the controller 10 performing the inspection processing. In other words, if the smart camera 102 is capable of performing up to the inspection processing, the smart camera 102 can complete the inspection processing by itself, and if the processing capacity of the smart camera 102 by itself is insufficient, the controller 10 can perform the inspection processing, so the smart camera 102 and the controller 10 can work together as needed.

[0129] The above-described embodiments are merely examples in all respects and should not be construed as limiting. Furthermore, all modifications and variations within the scope of the claims are within the scope of the present invention. [Industrial Applicability]

[0130] As described above, the present invention can be used as an image inspection system that can perform various inspections using images of a workpiece, for example. [Explanation of symbols]

[0131] 1. Image inspection system 10 Controller 11 Acquisition Department 12 Inspection Department 12a 1st Core 12b Second Core 13 Input section 14 Storage section 15 Setting section 101 Normal Camera 102 Smart Camera 102e Storage section 102f Inspection Department

Claims

1. An image inspection system including a normal camera equipped with an image generation function, a smart camera which is a camera equipped with an image inspection function in addition to the image generation function, and a controller to which the normal camera and the smart camera can be connected, the smart camera includes a storage unit that stores first environmental settings related to hardware of the smart camera and first inspection settings for images captured by the smart camera; The controller includes a storage unit that stores second environmental settings related to the hardware of the controller and the first inspection settings acquired from the smart camera, an acquisition unit that acquires images generated by the smart camera or the normal camera, and an inspection unit that performs an inspection based on the first inspection settings on the images acquired by the acquisition unit. An image inspection system comprising:

2. A controller connectable to a normal camera equipped with an image generation function and a smart camera equipped with an image inspection function in addition to the image generation function and including a storage unit that stores first environmental settings related to hardware and first inspection settings for captured images, a storage unit that stores second environmental settings related to the hardware of the controller and the first inspection settings acquired from the smart camera; an acquisition unit that acquires an image generated by the smart camera or the normal camera; an inspection unit that performs an inspection based on the first inspection setting on the image acquired by the acquisition unit; A controller comprising:

3. 2. The image inspection system according to claim 1, the controller can simultaneously connect to a plurality of cameras including the normal camera and / or the smart camera; an input unit that accepts selection of either a camera independent mode in which an inspection of an image acquired from each of a plurality of cameras connected to the controller is performed based on an inspection setting set for each of the cameras, or a camera linked mode in which an inspection of an image acquired from each of the cameras connected to the controller is performed based on an inspection setting common to the plurality of cameras; An image inspection system, wherein the inspection unit is configured such that the input unit accepts selection of the camera independent mode, and the first inspection setting acquired from the smart camera can be applied to images acquired from other cameras only when operating in the camera independent mode.

4. 4. The image inspection system according to claim 3, the first inspection setting includes at least an imaging setting tool, and the imaging setting tool includes a camera designation parameter for designating one of a plurality of cameras connected to the controller; When the input unit receives a selection of the camera linkage mode, it accepts the specification of multiple cameras connected to the controller via the camera specification parameters of the common inspection settings, while when the input unit receives a selection of the camera independent mode, it accepts inspection settings for each camera.

5. 2. The image inspection system according to claim 1, When the smart camera detects that it is connected to the controller, the smart camera can perform only image generation without performing the image inspection function; An image inspection system, wherein the inspection unit is capable of performing an inspection based on the first inspection setting on an image generated by the smart camera connected to the controller.

6. 2. The image inspection system according to claim 1, the storage unit of the controller stores a second inspection setting for an image captured by a camera connected to the controller; the storage unit of the smart camera stores the second inspection setting acquired from the controller; The inspection unit performs an inspection based on the second inspection setting on the image generated by the smart camera.

7. 2. The image inspection system according to claim 1, An image inspection system, wherein the first environmental setting regarding the hardware of the smart camera and the second environmental setting regarding the hardware of the controller include a correspondence between terminals of each hardware and information regarding signals received from the outside via the terminals.

8. 2. The image inspection system according to claim 1, the inspection unit of the controller further includes a plurality of cores for performing inspection on the image acquired by the acquisition unit; An image inspection system in which each core constituting the plurality of cores is assigned to each of a plurality of cameras connected to the controller, and performs inspection based on the first inspection setting on images generated by the corresponding camera.

9. 2. The image inspection system according to claim 1, the first inspection setting includes a first imaging condition of the smart camera; An image inspection system in which, if the normal camera cannot reproduce a parameter value included in the first imaging condition, the inspection unit of the controller generates a second imaging condition in which the parameter value is modified, and performs an inspection on the image generated by the normal camera based on an inspection setting in which the first imaging condition of the first inspection setting is replaced with the second imaging condition.

10. 2. The image inspection system according to claim 1, the first inspection setting includes a first imaging condition of the smart camera; An image inspection system in which the inspection unit of the controller disables the first imaging condition among the first inspection settings if the normal camera does not have the function to execute the first imaging condition.

11. 2. The image inspection system according to claim 1, the storage unit of the smart camera further stores a first image capture setting related to image capture by the smart camera; the controller further includes an input unit that accepts input of a second image capture setting related to image capture by the normal camera; An image inspection system, wherein the inspection unit performs an inspection based on the first inspection setting on an image generated by the normal camera based on the second imaging setting.

Images