Information processing device, control method for information processing device, and program

Abstract

This system provides a mechanism to improve user experience in image-based inspections. [Solution] An information processing device comprising a display control means for controlling the display of an object that accepts an operation to acquire a first image to be used for machine learning, wherein, when predetermined conditions are met, the object accepts an operation to acquire a second image that is the subject of inspection using a learning model generated by machine learning.

Description

Technical Field

[0008] , , , , , , ,

[0007]

[0001] An information processing apparatus, a control method thereof, and a program, particularly related to machine learning technology.

Background Art

[0002] Conventionally, in inspection by machine learning using images, it is performed in the following flow: 1. "shooting" of an object, 2. "learning" of the captured image, and 3. "inspection" using the learned model.

[0003] Patent Document 1 discloses technologies related to shooting, learning, and inspection.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Disclosure of the Invention

Problems to be Solved by the Invention

[0005] These processes of shooting, learning, and inspection are often performed on a factory line or the like, and an operator who is not familiar with shooting or machine learning will operate the information processing apparatus. <​​​​​​​​​​​​​​

[0009] An information processing device comprising display control means for controlling the display of an object that accepts operations for acquiring a first image to be used for machine learning, characterized in that, when predetermined conditions are met, the object accepts operations for acquiring a second image that is the subject of inspection using a learning model generated by machine learning. [Effects of the Invention]

[0010] According to the present invention, it is possible to provide a mechanism that improves user operability in image-based inspections. [Brief explanation of the drawing]

[0011] [Figure 1] This is a system configuration diagram of information processing system 100. [Figure 2] This is a hardware block diagram of the information processing device 104. [Figure 3] This is an example of a block diagram showing the software configuration. [Figure 4] This is an example of a flowchart for the process of shooting, learning, and testing. [Figure 5] This is an example of the runtime screen 510 and the threshold change dialog 520. [Figure 6] This is an example of the learning model selection screen 610 and the new learning model creation screen 620. [Figure 7] This is an example of the execution button 512 (start inspection). [Figure 8] This is an example of a flowchart for the second embodiment, from shooting to learning to inspection. [Figure 9] This is an example of the execution button 512 (learned execution) of the second embodiment. [Modes for carrying out the invention]

[0012] Embodiments of the present invention will be described in detail below with reference to the drawings.

[0013] FIG. 1 is a system configuration diagram showing an example of the configuration of the information processing system 100 of the present invention.

[0014] FIG. 1 shows a configuration in which the camera 106 and the information processing apparatus 104 are connected via an image transfer cable (USB, Ethernet, Camera Link, etc.).

[0015] Note that, instead of the connection via the image transfer cable, a wireless connection may be used.

[0016] Further, even if such a configuration is not provided, an information processing apparatus equipped with a camera may be used.

[0017] The camera 106 of the present embodiment is a network camera, but is not limited thereto, and may be a digital camera, a wearable camera, an infrared camera, a camera attached to a vehicle, an aircraft, a drone, etc.

[0018] The information processing apparatus 104 acquires an image obtained by photographing a photographing target with the camera 106 via an image transfer cable, and executes a process for determining the completion of each operation.

[0019] Further, the information processing system 100 of the present embodiment may have a configuration in which the camera 106 having the above-described configuration does not exist. In that case, an image is acquired from another terminal on the network or from the external memory 212 of the information processing apparatus 104.

[0020] The user operation terminal 102 is a user operation terminal such as a tablet, a smartphone, a touch-operable display, a wearable terminal, or a notebook PC, and is connected to the information processing apparatus 104 by wire or wirelessly.

[0021] The information processing apparatus 104 performs processing to generate screen information, transmits the screen information to the user operation terminal 102, and the user operation terminal 102 displays the received screen information using a browser, an application, or the like.

[0022] The user terminal 102 receives user input on the screen, buttons, etc., and transmits the received input to the information processing device 104.

[0023] The information processing device 104 may also be an information processing device or virtual machine located in a network environment or cloud environment.

[0024] Furthermore, the user terminal 102 and the information processing device 104 may be integrated into a single unit. The following describes the hardware configuration of an information processing device applicable to the user terminal 102 and information processing device 104 shown in Figure 1, using Figure 2 as an example.

[0025] Figure 2 is a block diagram showing a hardware configuration applicable to the user terminal 102 and information processing device 104 shown in Figure 1.

[0026] In Figure 2, 201 is the CPU, which comprehensively controls each device and controller connected to the system bus 204. The ROM 202 or external memory 212 stores the BIOS (Basic Input / Output System), which is the control program for the CPU 201, the operating system program (hereinafter referred to as the OS), and various programs necessary to implement the functions executed by each PC, as described later.

[0027] 203 is RAM, which functions as the main memory, work area, etc., of the CPU 201. The CPU 201 loads the necessary programs, etc., from ROM 202 or external memory 212 into RAM 203, and then executes the loaded programs to perform various operations.

[0028] 205 is an input controller that controls input from pointing devices such as the keyboard (KB) 210 and a mouse (not shown).

[0029] 206 is a video controller that controls the display on indicators such as display 211.

[0030] 207 is the memory controller, which stores various data on external storage devices (hard disks (HDs)), flexible disks (FDs), or PCMCIA card slots. It controls access to external memory 212, such as CompactFlash® memory, which is connected via an adapter.

[0031] 208 is a communication interface controller that controls the reception of image data from the external PC 213 via the network (TCP / IP). 209 is an image I / F controller that controls the reception of image data from camera 106 via an image transfer cable (USB, Ethernet, Camera Link, etc.).

[0032] The various programs described later for realizing the present invention are stored in RAM 203 and executed by CPU 201.

[0033] Furthermore, the image data used when executing the above program is stored in ROM202, external memory212, external PC213, and camera106 depending on the application, and is stored in RAM203 via various controllers when the program is executed. Figure 3 is an example of a block diagram showing the software configuration of an embodiment of the present invention.

[0034] The user terminal 102 and the information processing device 104 are equipped with the following functional units.

[0035] The display control unit 301 is a functional unit that controls the display of an object that accepts operations for acquiring a first image to be used for machine learning.

[0036] The first display control unit 302 is a functional unit that controls the display of a first object that accepts operations for acquiring a first image to be used for machine learning.

[0037] The second display control unit 303 is a functional unit that controls the display of a second object that accepts an operation to acquire a second image, which is the subject of inspection using a learning model generated by machine learning, when predetermined conditions are met.

[0038] The second display control unit 303 is a functional unit that controls the display of a second object that accepts an operation to acquire a second image, which is the subject of inspection using a learning model generated by machine learning, instead of the first object, when predetermined conditions are met.

[0039] The first display control unit 302 is a functional unit that controls the display of a screen including a first object that accepts operations for acquiring a first image to be used for machine learning.

[0040] The second display control unit 303 is a functional unit that controls the display of a second object at approximately the same position on the screen where the first object was displayed, when predetermined conditions are met, and accepts an operation to acquire a second image that is the subject of inspection using a learning model generated by machine learning.

[0041] The screen display control unit 304 is a functional unit that controls the display of a screen that accepts operations for acquiring a first image to be used for machine learning.

[0042] This concludes the explanation of Figure 3. The process from image capture to learning and testing shown in Figure 4 will be explained.

[0043] The following processes shall be performed by the CPU 201 of the user terminal 102 or the CPU 201 of the information processing device 104.

[0044] <First Embodiment> In S401, the user terminal 102 receives an operation from the user to launch an application that performs shooting, learning, and inspection processing, and sends a request to launch the application to the information processing device 104.

[0045] In this embodiment, the application is a web application, but it may also be an application that does not use internet technology, or a dedicated application may be installed on the user terminal 102 and communicate with the information processing device 104.

[0046] In S402, the information processing device 104 receives an application launch request from the user terminal 102.

[0047] In S403, the information processing device 104 acquires information related to the application. Information related to the application includes, but is not limited to, the values ​​of the items on the runtime screen 510 (Figure 5), specifically the selected learning model name, threshold, and inspection results 515 (total number of inspections, number of normal items, number of abnormal items), and refers to information used for processing and displaying the application.

[0048] In S404, the information processing device 104 proceeds to S405 if no learning model has been selected, and to S439 if one has been selected.

[0049] In S405, the information processing device 104 creates runtime screen information by setting the text of the execution button 512 to be displayed on the runtime screen 510 to "Start shooting".

[0050] In other words, this step is an example of a process that controls the display of an object that accepts operations for acquiring the first image to be used for machine learning.

[0051] In S406, the information processing device 104 transmits the created runtime screen information to the user operation terminal 102.

[0052] In S407, the user terminal 102 receives runtime screen information.

[0053] In S408, the user terminal 102 uses the received runtime screen information to display the runtime screen 510 in the browser.

[0054] In other words, this step is an example of a process that controls the display of an object that accepts operations for acquiring the first image to be used for machine learning.

[0055] In other words, this step is an example of a process that controls the display of a screen containing a first object that accepts operations for acquiring a first image to be used for machine learning.

[0056] In other words, this step is an example of a process that controls the display of a screen that accepts operations for acquiring the first image to be used for machine learning.

[0057] In S409, the user terminal 102 accepts the press of the execution button 512, which displays the words "Start shooting".

[0058] In S410, the user terminal 102 transmits a shooting start instruction to the information processing device 104.

[0059] In S411, the information processing device 104 receives an instruction to start shooting.

[0060] In S412, the information processing device 104 acquires a photograph of the item to be studied and stores it in the RAM 203 or external memory 212. The photograph may be acquired as the camera 106 takes a photograph, or it may be acquired from the camera 106 or another information processing device that has been photographed in advance.

[0061] In S413, if the information processing device 104 receives an instruction to end the shooting process, it proceeds to S414; otherwise, it returns to S412.

[0062] In S414, the information processing device 104 terminates the image capture process. Specifically, it sends an instruction to the camera 106 to terminate the image capture process, or terminates the acquisition of images from other information processing devices.

[0063] In S415, the information processing device 104 obtains a list of captured images stored in S412.

[0064] In S416, the information processing device 104 creates screen information that includes a list of captured images.

[0065] In S417, the information processing device 104 transmits the screen information created in S416 to the user operation terminal 102.

[0066] In S418, if the user terminal 102 receives a press of the stop button 513 (Figure 5), it proceeds to S419; otherwise, it repeats S418. Alternatively, in S418, the captured image acquired by the information processing device 104 in S412 may be displayed on the screen of the user terminal 102. This allows the user to check what kind of image is currently being acquired (captured).

[0067] In S419, the user terminal 102 transmits a recording completion instruction to the information processing device 104.

[0068] In S420, the user terminal 102 receives screen information that includes a list of captured images.

[0069] In S421, the user terminal 102 displays a list of captured images using the received screen information. Specifically, it displays a screen similar to the learning image list 622 on the new learning model creation screen 620 (Figure 6). This screen is used to create a new learning model.

[0070] In S422, the user terminal 102 accepts the selection of images to be trained. Specifically, it accepts the user's selection of images to be used for machine learning from the training image list 622, and displays a selected mark superimposed on the image or near the image.

[0071] In S423, the user terminal 102 accepts the press of the learning execution button 623 (Figure 6).

[0072] In S424, the user terminal 102 sends a learning execution instruction to the information processing device 104. At this time, it also sends the identification information of the learning target image selected in S422.

[0073] In S425, the information processing device 104 receives a learning execution instruction. At this time, it also receives identification information of the selected image to be learned.

[0074] In S426, the information processing device 104 acquires the training target image based on the identification information of the training target image received in S425 and stores it in the RAM 203 or external memory 212.

[0075] In S427, the information processing device 104 performs machine learning using the acquired training target images.

[0076] In S428, the information processing device 104 generates a learning model by performing machine learning.

[0077] In S429, the information processing device 104 proceeds to S439 if a learning model has been selected, and to S430 if it has not been selected.

[0078] In S430, the information processing device 104 obtains a list of learning models. Specifically, it obtains a list of names of learning models stored in RAM 203 or external memory 212.

[0079] In S431, the information processing device 104 generates screen information for the learning model list using the acquired learning model list information.

[0080] In S432, the information processing device 104 transmits the screen information of the generated learning model list to the user operation terminal 102.

[0081] In S433, the user terminal 102 receives screen information of the learning model list.

[0082] In S434, the user terminal 102 displays the learning model list screen. Specifically, it displays the model list on the left side of the learning model selection screen 610 (Figure 6). This screen is used to select the learning model to be used for inspecting the items.

[0083] In S435, the user terminal 102 accepts the selection of a learning model to be used for inspecting the item. The screen information for the learning image list on the right side of the learning model selection screen 610 may be received when the learning model list screen information is received in S433, or the screen information for the learning image list corresponding to the selected learning model may be received from the information processing device 104 when the selection of a learning model is accepted in S435.

[0084] In S436, the user terminal 102 transmits identification information of the selected learning model to the information processing device 104.

[0085] In S437, the information processing device 104 receives identification information of the selected learning model.

[0086] In S438, the information processing device 104 stores the identification information of the selected learning model in the RAM 203 or external memory 212.

[0087] In S439, the information processing device 104 creates runtime screen information by setting the text of the execution button 512 to "Start Inspection" on the runtime screen 510.

[0088] In other words, this step is an example of a process that controls the display of a second object that accepts an operation to acquire a second image to be examined using a machine learning-generated learning model, provided that predetermined conditions are met.

[0089] In this embodiment, the wording of the same execution button 512 is changed from "Start Shooting" to "Start Inspection," but this is not limited to the same button. Different buttons (a shooting start button and an inspection start button) may be swapped when certain conditions are met, or one of them may be activated when certain conditions are met. The screens displaying the shooting start button and the inspection start button may be separate screens, but the shooting start button and the inspection start button may be displayed in approximately the same position. Alternatively, the screens displaying the shooting start button and the inspection start button may be the same screen, and when certain conditions are met, the screen may automatically scroll or move the buttons to display the shooting start button and the inspection start button in approximately the same position. In this context, "displaying in approximately the same position" means that the buttons on the screen can be in exactly the same position, or they can be displayed in approximately the same position to prevent the user from accidentally pressing other buttons. In other words, rather than being strictly within a range of millimeters or centimeters, it means displaying them in a position that prevents the user from accidentally pressing other buttons, and is a measure to make operation easier for the user.

[0090] In other words, this step is an example of a process that controls the display of a second object that accepts an operation to acquire a second image to be examined using a machine learning-generated learning model, instead of the first object, when certain conditions are met.

[0091] In other words, this step illustrates an example of a process that controls the display of a second object at approximately the same position on the screen where the first object was displayed, when predetermined conditions are met. This second object accepts an operation to acquire a second image, which is the subject of inspection using a machine learning-generated learning model.

[0092] In other words, the runtime screen 510 is an example of a screen that accepts an operation to acquire a second image to be inspected using a learning model generated by machine learning, when predetermined conditions are met.

[0093] In this embodiment, the condition for changing the text of the execution button 512 to "Start Inspection" is set in S404 and S429 to "A learning model has been selected." However, the condition is not limited to this, and may be at least one of the following: the imaging device is inaccessible, the imaging device is not connected, a first image exists, the first image is accessible, the first image has been acquired, an operation to acquire the first image has been received, a screen for acquiring the first image has been displayed, machine learning is complete, a learning model has been generated, a learning model exists, the learning model is accessible, the learning model is loaded into memory, the learning model is available for inspection, a learning model to be used for inspection has been selected, the learning model to be used for inspection has been determined, and the system is in a state where inspection can be started.

[0094] As a result, even workers unfamiliar with photography or machine learning, workers wearing gloves, and workers performing tasks while paying attention to their surroundings can easily initiate "start photography" and "start inspection," thereby improving user operability in image-based inspections.

[0095] In S440, the information processing device 104 transmits the created runtime screen information to the user operation terminal 102.

[0096] In S441, the user terminal 102 receives runtime screen information.

[0097] In S442, the user terminal 102 uses the received runtime screen information to display the runtime screen 510 in the browser.

[0098] In this embodiment, the wording of the same execution button 512 is changed from "Start Shooting" to "Start Inspection," but this is not limited to the same button. Different buttons (a shooting start button and an inspection start button) may be swapped when certain conditions are met, or one of them may be activated when certain conditions are met. The screens displaying the shooting start button and the inspection start button may be separate screens, but the shooting start button and the inspection start button may be displayed in approximately the same position. Alternatively, the screens displaying the shooting start button and the inspection start button may be the same screen, and when certain conditions are met, the screen may automatically scroll or move the buttons to display the shooting start button and the inspection start button in approximately the same position. In this context, "displaying in approximately the same position" means that the buttons on the screen can be in exactly the same position, or they can be displayed in approximately the same position to prevent the user from accidentally pressing other buttons. In other words, rather than being strictly within a range of millimeters or centimeters, it means displaying them in a position that prevents the user from accidentally pressing other buttons, and is a measure to make operation easier for the user.

[0099] In other words, this step is an example of a process that controls the display of a second object that accepts an operation to acquire a second image to be examined using a machine learning-generated learning model, provided that predetermined conditions are met.

[0100] In other words, this step is an example of a process that controls the display of a second object that accepts an operation to acquire a second image to be examined using a machine learning-generated learning model, instead of the first object, when certain conditions are met.

[0101] In other words, this step illustrates an example of a process that controls the display of a second object at approximately the same position on the screen where the first object was displayed, when predetermined conditions are met. This second object accepts an operation to acquire a second image, which is the subject of inspection using a machine learning-generated learning model.

[0102] In S443, the user terminal 102 accepts the press of the execution button 512 (Figure 7) which displays the words "Start Inspection".

[0103] In other words, the runtime screen 510 is an example of a screen that accepts an operation to acquire a second image to be inspected using a learning model generated by machine learning, when predetermined conditions are met.

[0104] As a result, even workers unfamiliar with photography or machine learning, workers wearing gloves, and workers performing tasks while paying attention to their surroundings can easily initiate "start photography" and "start inspection," thereby improving user operability in image-based inspections.

[0105] In S444, the user terminal 102 sends an instruction to start the inspection to the information processing device 104. At the same time, it also sends the threshold values ​​shown in Figure 5 to the information processing device 104.

[0106] In S445, the information processing device 104 receives an instruction to start the inspection. At that time, it also receives the threshold value.

[0107] In S446, the information processing device 104 acquires a photograph of the item to be inspected and stores it in the RAM 203 or external memory 212. The photograph may be acquired as the camera 106 takes a photograph, or it may be acquired from the camera 106 or another information processing device that has taken a photograph in advance.

[0108] In S447, the information processing device 104 performs an inspection of the captured images acquired in S446 using the selected learning model and threshold. It counts up each item of the inspection results.

[0109] In S448, the information processing device 104 proceeds to S450 if the inspection reveals no abnormalities in the captured image, and to S449 if abnormalities are found.

[0110] In S449, the information processing device 104 creates an image in which marks indicating abnormalities are superimposed on the abnormal areas of the captured image. In this embodiment, an image in which marks indicating abnormalities are superimposed is created, but this method is not limited to this method. Alternatively, the coordinates of the abnormal areas may be obtained and these coordinates may be used to identify and display the abnormal areas on the captured image.

[0111] In S450, the information processing device 104 creates screen information that includes the inspection results and captured images.

[0112] In S451, the information processing device 104 transmits screen information that includes the inspection results and the captured image. The captured image to be transmitted may be an image with an abnormality mark superimposed on it, or the captured image may be transmitted together with the coordinates of the abnormal location.

[0113] In S452, the user terminal 102 receives screen information that includes the inspection results and captured images.

[0114] In S453, the user terminal 102 displays a screen containing the inspection results and captured images using screen information that includes the inspection results and captured images. The captured images to be displayed may be images with marks indicating abnormalities superimposed on them, or the coordinates of the abnormal areas may be used to identify and display the abnormal areas on the captured images.

[0115] In S454, if the user terminal 102 receives a press of the stop button 513 (Figure 5), it proceeds to S455; otherwise, it returns to S452. The stop button 513 is both a button to stop shooting and a button to stop inspection. Therefore, even workers unfamiliar with shooting or machine learning, workers wearing gloves, or workers performing tasks while paying attention to their surroundings can easily perform the "stop shooting" and "stop inspection" commands, thereby improving user operability in image-based inspections.

[0116] In S455, the user terminal 102 transmits an inspection completion instruction to the information processing device 104.

[0117] In S456, if the information processing device 104 does not receive an instruction to end the inspection, it returns to S446; if it does receive an instruction to end the inspection, it terminates the process shown in Figure 4. When terminating, the information acquired or selected during the shooting-learning-inspection process (captured image, created learning model, identification information of the image used for learning, selected learning model, threshold, inspection results, etc.) is stored in the external memory 212 so that it can be acquired as application setting information in S403 when the application is launched again or when the shooting-learning-inspection process is executed again.

[0118] As a result, even workers unfamiliar with photography or machine learning, workers wearing gloves, and workers performing tasks while paying attention to their surroundings can easily perform the instruction operations for photography, learning, and inspection, thereby improving user operability in image-based inspection.

[0119] This concludes the explanation of the process shown in Figure 4. <Second Embodiment> The first embodiment was designed to allow users to easily initiate "start shooting" and "start inspection" commands. However, the shooting-learning-inspection process shown in Figure 4 also includes a command operation called "execute learning."

[0120] Therefore, the second embodiment is a mechanism that allows users to easily perform the "execute learning" and "start testing" commands.

[0121] Figure 8 illustrates the shooting, learning, and inspection process of the second embodiment.

[0122] Of the processes in Figure 8 that are the same as those in Figure 4 (S401-S456), we will omit the explanation and explain the newly added process (S801).

[0123] In S801, the information processing device 104 creates runtime screen information by setting the text of the execution button 512 (Figure 9) displayed on the runtime screen 510 to "Learning Execution".

[0124] In other words, this step, or S421, is a step that illustrates an example of a process that controls the display of an object that accepts operations for performing machine learning.

[0125] In other words, this step, or S421, is a step that illustrates an example of a process that controls the display of a first object that accepts operations for performing machine learning.

[0126] In other words, this step, or S421, is a step that illustrates an example of a process that controls the display of a screen containing a first object that accepts operations for performing machine learning.

[0127] In other words, this step, or S421, is a step that illustrates an example of a process that controls the display of a screen that accepts operations for performing machine learning.

[0128] As a result, even workers unfamiliar with photography or machine learning, workers wearing gloves, and workers performing tasks while paying attention to their surroundings can easily perform the "run learning" and "start inspection" commands, thereby improving user operability in image-based inspections. As described above, it goes without saying that the object of the present invention can also be achieved by supplying a recording medium containing a program that realizes the functions of the embodiments described above to a system or device, and by having the computer (or CPU or MPU) of that system or device read and execute the program stored on the recording medium.

[0129] In this case, the program read from the recording medium itself realizes the novel function of the present invention, and the recording medium on which that program is recorded constitutes the present invention.

[0130] For recording media used to supply programs, examples include flexible disks, hard disks, optical disks, magneto-optical disks, CD-ROMs, CD-Rs, DVD-ROMs, magnetic tapes, non-volatile memory cards, ROMs, EEPROMs, silicon disks, and the like.

[0131] Furthermore, it goes without saying that the functions of the aforementioned embodiments are realized not only by the computer executing the program it has read, but also by the operating system (OS) running on the computer performing some or all of the actual processing based on the instructions of that program, thereby realizing the functions of the aforementioned embodiments.

[0132] Furthermore, it goes without saying that this also includes cases where, after a program read from a recording medium is written to the memory of a function expansion board inserted into a computer or a function expansion unit connected to a computer, the CPU or other components of the function expansion board or function expansion unit perform some or all of the actual processing based on the instructions of the program code, and the functions of the aforementioned embodiments are realized through that processing.

[0133] Furthermore, the present invention may be applied to a system consisting of multiple devices or to a device consisting of a single device. It goes without saying that the present invention can also be applied when the results are achieved by supplying a program to a system or device. In this case, by reading a recording medium containing a program for achieving the present invention into the system or device, the system or device can enjoy the effects of the present invention.

[0134] The above program may consist of object code, program code executed by an interpreter, script data supplied to the OS (operating system), and the like.

[0135] Furthermore, by downloading and reading the program for achieving the present invention from a server, database, etc. on a network using a communication program, the system or device can enjoy the effects of the present invention. It should be noted that all configurations combining the above-described embodiments and their modified forms are also included in the present invention. [Explanation of symbols]

[0136] 100 Information Processing Systems 102 User Operation Terminal 104 Information Processing Device 106 Camera

Claims

1. It includes a display control means for controlling the display of an object that accepts operations for acquiring a first image to be used for machine learning, When certain conditions are met, the object accepts an operation to acquire a second image to be inspected using the learning model generated by machine learning. An information processing device characterized by the following.

2. A first display control means that controls the display of a first object that accepts operations for acquiring a first image to be used for machine learning, A second display control means controls the display of a second object that accepts an operation to acquire a second image, which is the subject of inspection using the learning model generated by machine learning, when predetermined conditions are met. An information processing device characterized by comprising:

3. A first display control means that controls the display of a first object that accepts operations for acquiring a first image to be used for machine learning, A second display control means controls the display of a second object that accepts an operation to acquire a second image, which is the subject of inspection using the machine learning model generated by the machine learning method, in place of the first object, when predetermined conditions are met. An information processing device characterized by comprising:

4. A first display control means that controls the display of a screen containing a first object that accepts operations for acquiring a first image to be used for machine learning, When predetermined conditions are met, a second display control means controls the display of a second object at approximately the same position on the screen where the first object was displayed, which accepts an operation to acquire a second image that is the subject of inspection using the learning model generated by machine learning. An information processing device characterized by comprising:

5. It includes a screen display control means that controls the display of a screen that accepts operations for acquiring a first image to be used for machine learning, When predetermined conditions are met, the screen accepts an operation to acquire a second image to be examined using the learning model generated by the machine learning. An information processing device characterized by the following.

6. The conditions described above are met if at least one of the following is satisfied: the imaging device is inaccessible; the imaging device is not connected; the first image exists; the first image is accessible; the first image has been acquired; an operation to acquire the first image has been received; a screen for acquiring the first image has been displayed; the machine learning is completed; the learning model has been generated; the learning model exists; the learning model is accessible; the learning model is loaded into memory; the learning model is available for the inspection; the learning model to be used for the inspection has been selected; the learning model to be used for the inspection has been determined; and the inspection is in a state where it can be started. An information processing apparatus according to any one of claims 1 to 5, characterized by the above.

7. It includes a display control step that controls the display of an object that accepts operations for acquiring a first image to be used for machine learning, When certain conditions are met, the object accepts an operation to acquire a second image to be inspected using the learning model generated by machine learning. A control method for an information processing device characterized by the following.

8. At least one computer, A program for causing each means of the information processing apparatus described in any one of claims 1 to 5 to function.

Images