Program product, information processing method, information processing device, and computer readable recording medium

Abstract

The present disclosure provides a program product, an information processing method, an information processing apparatus, and a computer-readable recording medium, which can expect to determine a state related to ejection of a liquid such as a chemical liquid or a cleaning liquid during substrate processing. A computer program according to the present embodiment causes a computer to execute: a process for acquiring a moving image obtained by capturing an image of an ejection unit of a substrate processing apparatus that ejects a liquid onto a substrate to be processed; a learning model input unit that inputs a frame image included in the acquired moving image to a learning model obtained by machine learning so as to receive an image of the ejection unit as an input and output information relating to the ejection state of the liquid ejected from the ejection unit; acquiring information which is output by the learning model and is related to a spraying state; and determining, on the basis of the acquired information, whether or not the discharge of the liquid is correct.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to a program, an information processing method, an information processing apparatus, and a computer-readable recording medium. BACKGROUND

[0002] In Patent Literature 1, there is proposed a substrate processing method including a holding process of holding a substrate by moving the substrate into an inside of a chamber, a supplying process of supplying a fluid to the substrate in the inside of the chamber, a photographing process of sequentially photographing the inside of the chamber by a camera to acquire image data, a condition setting process of determining a monitoring object from among a plurality of monitoring object candidates in the inside of the chamber and changing an image condition based on the monitoring object, and a monitoring process of performing a monitoring process for the monitoring object based on the image data having the image condition corresponding to the monitoring object.

[0003] PRIOR ART DOCUMENTS

[0004] PATENT LITERATURE

[0005] Patent Literature 1: Japanese Patent Application Publication No. 2021-190511 SUMMARY

[0006] PROBLEMS TO BE SOLVED BY THE INVENTION

[0007] The present disclosure provides a computer program, an information processing method, and an information processing apparatus that enable determination of a state related to ejection of a liquid such as a chemical liquid or a cleaning liquid in substrate processing.

[0008] SOLUTION TO PROBLEM

[0009] The computer program according to one embodiment causes a computer to execute the following processes: acquire a moving image obtained by photographing an ejection portion of a substrate processing apparatus that ejects a liquid toward a substrate as a processing target; input a frame image included in the acquired moving image to a learning model that is obtained by machine learning in such a manner that an image of the ejection portion is accepted as input and outputs information related to an ejection state of the liquid ejected from the ejection portion; acquire the information related to the ejection state output by the learning model; and determine whether or not the ejection of the liquid is correct based on the acquired information.

[0010] EFFECT OF THE INVENTION

[0011] According to the present disclosure, it is possible to determine a state related to ejection of a liquid such as a chemical liquid or a cleaning liquid in substrate processing. BRIEF DESCRIPTION OF DRAWINGS

[0012] Figure 1is a schematic view for explaining a configuration example of a substrate processing apparatus according to the present embodiment.

[0013] Figure 2 is a schematic view for explaining an outline of an information processing system according to the present embodiment.

[0014] Figure 3 is a block diagram showing a configuration example of an information processing apparatus according to the present embodiment.

[0015] Figure 4 is a schematic view for explaining a configuration example of a learning model used by an information processing apparatus according to the present embodiment.

[0016] Figure 5 is a schematic view showing an example of a state of ejection.

[0017] Figure 6 is a flowchart showing an example of a procedure of ejection state determination processing performed by an information processing apparatus according to the present embodiment.

[0018] Figure 7 is a flowchart showing an example of a procedure of abnormality determination processing performed by an information processing apparatus according to the present embodiment.

[0019] Figure 8 is a flowchart showing an example of a procedure of abnormality determination processing performed by an information processing apparatus according to the present embodiment.

[0020] Figure 9 is a schematic view showing an example of a display of a warning screen displayed by an information processing apparatus according to the present embodiment.

[0021] Figure 10 is a schematic view showing an example of information display performed by an information processing apparatus according to the present embodiment.

[0022] Figure 11 is a flowchart showing an example of a procedure of abnormality determination processing performed by an information processing apparatus according to Modification 1.

[0023] Figure 12 is a schematic view showing an example of a display of a warning screen displayed by an information processing apparatus according to Modification 2. DETAILED DESCRIPTION

[0024] Hereinafter, specific examples of an information processing system according to the present embodiment will be described with reference to the accompanying drawings. Furthermore, the present disclosure is not limited to these examples, and is shown by the claims, with the intention of including all modifications within the meaning and scope equivalent to the claims.

[0025] <SYSTEM CONFIGURATION>

[0026] Figure 1 is a schematic view for explaining one configuration example of a substrate processing apparatus 1 to which the present embodiment is applied. The substrate processing apparatus 1 to which the present embodiment is applied is an apparatus that performs wet etching, which is a substrate processing that processes a substrate (for example, a wafer on which an oxide film or a nitride film is formed) into a desired shape by supplying a chemical liquid that dissolves the film to the film while rotating the substrate as a processing target. The substrate processing apparatus 1 to which the present embodiment is applied is configured to include a chamber 11, a substrate holding mechanism 12, a spouting portion 13, a recovery cup 14, and the like.

[0027] The chamber 11 is a sealed reaction container, and the substrate holding mechanism 12, the spouting portion 13, the recovery cup 14, and the like are housed in the inside thereof. A FFU (Fan Filter Unit) 15 is provided at the top of the chamber 11. The FFU 15 is used to form a downflow in the chamber 11.

[0028] The substrate holding mechanism 12 has a holding portion 12a, a support portion 12b, and a driving portion 12c. The holding portion 12a is, for example, a disc, and horizontally holds a substrate (wafer) as a processing target on the disc. The support portion 12b is joined to a central portion of a lower surface of the holding portion 12a, and is a cylindrical member that extends in the vertical direction (in the up-down direction in the present embodiment), and horizontally supports the holding portion 12a. In addition, a lower end portion of the support portion 12b is joined to the driving portion 12c, and is rotatably supported by the driving portion 12c. The driving portion 12c has a motor or the like as a prime mover, and rotates the support portion 12b about an axis. Thus, the substrate holding mechanism 12 can rotate the holding portion 12a supported by the support portion 12b by rotating the support portion 12b by the driving portion 12c, and thereby rotate the substrate held by the holding portion 12a. Figure 1

[0029] The spouting portion 13 spouts a liquid such as a chemical liquid or a cleaning liquid to a substrate held by the substrate holding mechanism 12. For example, the chemical liquid uses dilute hydrofluoric acid, and the cleaning liquid uses pure water, but the liquid spouted by the spouting portion 13 is not limited to these. The spouting portion 13 is connected to a liquid supply source 16 provided outside the chamber 11 via a tubular liquid supply path, and spouts a liquid supplied from the supply source 16 to a substrate. In addition, the spouting portion 13 is joined to a driving mechanism not shown, and can move horizontally between a central portion and a peripheral portion of a substrate. By combining the rotation of a substrate by the substrate holding mechanism 12 and the horizontal movement of the spouting portion 13 by the driving mechanism, the substrate processing apparatus 1 can spout a liquid from the spouting portion 13 to an appropriate position of a substrate as a processing target.

[0030] ​The recovery cup 14 is configured to surround the holding portion 12a of the substrate holding mechanism 12, and to capture liquid scattered from the substrate due to rotation of the holding portion 12a. A liquid discharge port 14a is provided at the bottom of the recovery cup 14, and liquid captured by the recovery cup 14 is discharged from the liquid discharge port 14a to the outside of the chamber 11. In addition, an exhaust port 14b is provided at the bottom of the recovery cup 14, and gas supplied from the FFU 15 is discharged from the exhaust port 14b to the outside of the chamber 11.

[0031] Further, Figure 1 The substrate processing apparatus 1 shown is a structure provided with one liquid ejection portion 13. The substrate processing apparatus 1 can selectively eject a chemical liquid or a cleaning liquid by switching between the chemical liquid for performing a dissolution process of a substrate and the cleaning liquid for performing a cleaning process of a substrate in the supply source 16. However, the substrate processing apparatus 1 can also be a structure provided with a plurality of liquid ejection portions 13. The substrate processing apparatus 1 can also be provided with a liquid ejection portion 13 for ejecting a chemical liquid and a liquid ejection portion 13 for ejecting a cleaning liquid, for example, separately.

[0032] Figure 2 is a schematic diagram for explaining an outline of an information processing system according to the present embodiment. The information processing system according to the present embodiment is configured to be provided with the substrate processing apparatus 1, the information processing apparatus 3, and the camera 5 described above. The camera 5 has a camera element such as a CCD (Charge Coupled Device) or a CMOS (Complementary Metal Oxide Semiconductor), for example, and can perform so-called moving image capturing by capturing several tens of times per second. The camera 5 is provided in the chamber 11 of the substrate processing apparatus 1, for example, and captures the liquid ejection portion 13 while a substrate processing is performed. The camera 5 transmits data of a moving image obtained by the capturing to the information processing apparatus 3. The data of the moving image is data obtained by concatenating a plurality of still images (frame images) in time series, for example. The camera 5 can be a device provided in the substrate processing apparatus 1, for example, and can also be a device provided separately from the substrate processing apparatus 1, for example.

[0033] The information processing apparatus 3 is an apparatus that performs control and monitoring of the substrate processing performed by the substrate processing apparatus 1 and the like. The information processing apparatus 3 is provided as an apparatus independent of the substrate processing apparatus 1 in the present embodiment, but is not limited thereto and can be an apparatus integral with the substrate processing apparatus 1. The information processing apparatus 3 is connected to the substrate processing apparatus 1 and the camera 5, for example, via a communication cable, and can perform transmission and reception of data with the substrate processing apparatus 1 and the camera 5. The information processing apparatus 3 receives data of a moving image obtained by the camera 5 capturing the ejection section 13, and performs determination of a state related to ejection of liquid by the ejection section 13 based on the received data of the moving image. The information processing apparatus 3 controls the operation of the substrate processing apparatus 1 in accordance with the determined state of ejection of liquid. In addition, the information processing apparatus 3 determines which of normal and abnormal the substrate processing is based on the determined state of ejection of liquid, and notifies the user of the abnormality by output of a message or sound or the like in the case where there is an abnormality.

[0034] Figure 3 Fig. 1 is a block diagram showing one configuration example of the information processing apparatus 3 according to the present embodiment. The information processing apparatus 3 according to the present embodiment can be realized, for example, by installing a prescribed application program or the like in a general-purpose information processing apparatus such as a personal computer or a server computer. The information processing apparatus 3 according to the present embodiment is configured to include a processing section 31, a storage section 32, a communication section 33, a display section 34, an operation section 35, and the like. In the present embodiment, the processing is described as being performed by one information processing apparatus 3, but the processing of the information processing apparatus 3 can be performed by a plurality of apparatuses in a distributed manner.

[0035] The processing section 31 is constituted by an arithmetic processing apparatus such as a CPU (Central Processing Unit), an MPU (Micro-Processing Unit), a GPU (Graphics Processing Unit), or a quantum processor, a ROM (Read Only Memory), and a RAM (Random Access Memory), and the like. The processing section 31 performs various processes such as a process of determining the ejection state of the ejection section 13 of the substrate processing apparatus 1 based on a moving image acquired from the camera 5, a process of controlling the operation of the substrate processing apparatus 1 based on the determined ejection state, and a process of notifying of an abnormality related to the substrate processing based on the determined ejection state, by reading out and executing a program 32a stored in the storage section 32.

[0036] The storage section 32 is configured using a large-capacity storage device such as a hard disk or an SSD (Solid State Drive). The storage section 32 stores various programs executed by the processing section 31 and various data required for the processing of the processing section 31. In the present embodiment, the storage section 32 stores a program 32a executed by the processing section 31. In addition, the storage section 32 is provided with a model information storage section 32b that stores information about a learning model completed by machine learning used by the information processing apparatus 3, and a log information storage section 32c that stores log information about the substrate processing of the substrate processing apparatus 1.

[0037] In the present embodiment, the program (computer program, program product) 32a is provided in the form of being recorded on a recording medium 99 such as a memory card or an optical disk, and the information processing apparatus 3 reads out the program 32a from the recording medium 99 and stores it in the storage section 32. However, the program 32a may, for example, be written in the storage section 32 at the manufacturing stage of the information processing apparatus 3. In addition, for example, the program 32a can be acquired by the information processing apparatus 3 through communication from a remote server device or the like that distributes the program 32a. For example, the program 32a can be read out by a writing device from the recording medium 99 and written in the storage section 32 of the information processing apparatus 3. The program 32a can be provided in the form of being distributed via a network, or in the form of being recorded on the recording medium 99.

[0038] The model information storage section 32b stores information about a learning model obtained by performing machine learning. The information about the learning model may, for example, include information indicating the structure of the learning model and the values of internal parameters determined by machine learning.

[0039] Figure 4 is a schematic diagram for explaining one structure example of a learning model used by the information processing apparatus 3 according to the present embodiment. The learning model according to the present embodiment is a learning model obtained by performing machine learning in advance in a manner of accepting an image (still image) of the ejection section 13 of the substrate processing apparatus 1 as input and outputting information about the ejection state of liquid by the ejection section 13. The learning model according to the present embodiment can employ a structure such as a CNN (Convolutional Neural Network) or a DNN (Deep Neural Network), but is not limited to these structures, and can be a learning model of an arbitrary structure.

[0040] In this embodiment, the liquid ejection state by the ejection unit 13 is classified into four states: "with liquid column", "liquid column broken and falling", "with droplets", and "no liquid". The learning model outputs four values ​​corresponding to these four states, and the information processing device 3 can determine the ejection state at that time point as the state corresponding to the largest value among the four values ​​output by the learning model. Furthermore, the ejection state is not limited to the above four states; states other than the above four states may also be used, and three or fewer states or five or more states may also be used.

[0041] To perform machine learning for generating a learning model, images of liquid being ejected from the ejector section 13 of the substrate processing device 1 captured by camera 5 are collected in advance. Designers and others then label the ejection state of the ejector section 13 in the collected images according to which of the four states described above it corresponds (an annotation task). Based on the learning data obtained by establishing a correspondence between the images of the ejector section 13 and the labels of the ejection states, the information processing device 3 can generate a learning model through supervised machine learning. The information processing device 3 stores information related to the learning model generated by machine learning in the model information storage unit 32b.

[0042] Furthermore, in this embodiment, the information processing device 3 is configured to store information related to the learning model and perform processing based on the learning model, but this is not a limitation. The information related to the learning model may also be stored in a device different from the information processing device 3, where the processing based on the learning model may be performed, and the information processing device 3 may obtain the processing result from that device. Additionally, the machine learning processing of the learning model may be performed by the information processing device 3 or by a device different from the information processing device 3.

[0043] The log information storage unit 32c stores various types of information obtained during substrate processing performed by the substrate processing apparatus 1, such as date and time and identification information of the substrate processing. The information stored in the log information storage unit 32c may include, for example, moving images captured by the camera 5, frame images extracted from the moving images, the ejection state determined by the learning model based on the frame images, whether there are any abnormalities in the substrate processing based on the ejection state, or control content related to the substrate processing based on the ejection state.

[0044] The communication unit 33 transmits and receives data between itself and the board processing apparatus 1 and the camera 5, for example, via a wired or wireless network N. In this embodiment, the communication unit 33 receives motion image data transmitted from the camera 5 and provides it to the processing unit 31. In addition, the communication unit 33 is given control information by the processing unit 31, including settings and instructions for actions related to board processing, and sends the given control information to the board processing apparatus 1, thereby controlling the operation of the board processing apparatus 1.

[0045] The display unit 34 is constructed using a liquid crystal display or the like, and displays various images and characters based on the processing of the processing unit 31. For example, the display unit 34 displays images (moving or still images) captured by the camera 5, information related to the ejection state determined by the learning model, or notifications related to abnormalities in the substrate processing, and other such information.

[0046] The operation unit 35 accepts user operations and notifies the processing unit 31 of the accepted operations. For example, the operation unit 35 accepts user operations via input devices such as mechanical buttons or a touch panel provided on the surface of the display unit 34. Alternatively, the operation unit 35 may also be an input device such as a mouse and keyboard, which can be detachable from the information processing device 3.

[0047] Furthermore, the storage unit 32 may also be an external storage device connected to the information processing device 3. Additionally, the information processing device 3 may be configured as a multi-computer system including multiple computers, or it may be a virtual machine constructed virtually through software. Furthermore, the information processing device 3 is not limited to the above-described structure; for example, it may include a reading unit for reading information stored on a removable storage medium, and it may not include, for example, a display unit 34 and an operation unit 35.

[0048] Furthermore, in the information processing apparatus 3 according to this embodiment, by reading and executing the program 32a stored in the storage unit 32, the image acquisition unit 31a, the ejection state determination unit 31b, the abnormality determination unit 31c, the display processing unit 31d, and the control processing unit 31e, which are software functional units, are implemented in the processing unit 31. In addition, in this figure, the functional units of the processing unit 31 are shown as functional units that perform processing related to the ejection state of the ejection unit 13 of the substrate processing apparatus 1, and the illustrations of functional units related to other processing are omitted.

[0049] The image acquisition unit 31a performs the following processing: by communicating with the camera 5 via the communication unit 33, it acquires image data of the ejection section 13 of the substrate processing apparatus 1 captured by the camera 5. In this embodiment, the camera 5 is a camera that captures moving images by taking approximately several dozen shots per second. The image data acquired by the image acquisition unit 31a can be in the format of moving images or in the format of still images (frame images) included in the moving images. The image acquisition unit 31a repeatedly acquires images from the camera 5 and stores information such as the acquisition date and time of the images acquired from the camera 5, as well as identification information of the substrate being processed, in the log information storage unit 32c. Furthermore, by repeatedly acquiring images from the camera 5 during substrate processing, the image acquisition unit 31a can obtain still images in a time sequence obtained by capturing images of the ejection section 13.

[0050] The ejection state determination unit 31b performs the following processing: based on the image (frame image included in the motion image) acquired by the image acquisition unit 31a, it determines the ejection state of the ejection section 13 of the substrate processing apparatus 1. As described above, in this embodiment, the ejection state determination unit 31b uses a learning model stored in the model information storage unit 32b to determine the ejection state. The ejection state determination unit 31b inputs the image of the ejection section 13 acquired by the image acquisition unit 31a to the learning model to obtain four values ​​output by the learning model. The ejection state determination unit 31b compares the four values ​​obtained from the learning model to determine whether the ejection state of the ejection section 13 is "with liquid column", "liquid column broken and falling", "with liquid droplets", or "no liquid". The ejection state determination unit 31b stores the information related to the determined ejection state in the log information storage unit 32c in correspondence with the original image. In addition, the ejection state determination unit 31b inputs multiple images acquired by the image acquisition unit 31a from the camera 5 in a time sequence to the learning model to determine the ejection state, thereby obtaining the ejection state determination result in the time sequence.

[0051] The anomaly determination unit 31c determines whether the substrate processing performed by the substrate processing apparatus 1 is normal or abnormal based on the determination result of the ejection state determined by the ejection state determination unit 31b. In this embodiment, the anomaly determination unit 31c determines the moment when the ejection state changes based on multiple ejection states in a time sequence determined by the ejection state determination unit 31b from multiple images acquired by the image acquisition unit 31a in a time sequence, and determines whether the moment is appropriate based on whether the time is normal or abnormal.

[0052] In this embodiment, the information processing device 3 controls the substrate processing performed by the substrate processing device 1. For example, the information processing device 3 sends a command to the substrate processing device 1 to open or close a valve provided in the liquid supply path from the liquid supply source 16 to the ejection section 13, thereby controlling the start and stop of liquid ejection from the ejection section 13. The anomaly determination unit 31c calculates the time from when the command to close the valve is given to the substrate processing device 1 until the liquid column ejected from the ejection section 13 breaks, based on multiple determination results in the time sequence of the ejection state determination unit 31b. In this case, the anomaly determination unit 31c, for example, determines the moment when the ejection state changes from "with liquid column" to "liquid column breaks and is falling," and calculates the time from when the command to close the valve is given until that moment. If the calculated time exceeds a predetermined threshold (e.g., 0.6 seconds to 0.8 seconds), the anomaly determination unit 31c determines that an anomaly has occurred in the liquid ejection during substrate processing.

[0053] Furthermore, the anomaly determination unit 31c calculates the time from the end of liquid ejection by the ejection unit 13 to the final droplet falling. In this case, the anomaly determination unit 31c, for example, determines the first moment when the ejection state changes from "with liquid column" to "liquid column broken and falling" and the second moment when it changes from "liquid column broken and falling" to "no liquid," and calculates the time from the determined first moment to the second moment. However, if, after changing from "liquid column broken and falling" to "no liquid," it further changes to the state of "with droplets" within a predetermined time, the anomaly determination unit 31c sets the moment when the droplet falls and changes back to the state of "no liquid" as the second moment. If the calculated time exceeds a predetermined threshold (e.g., a few seconds to tens of seconds), the anomaly determination unit 31c determines that an anomaly has occurred in the liquid ejection during substrate processing.

[0054] Alternatively, the anomaly determination unit 31c can also determine an anomaly using methods other than these. For example, the anomaly determination unit 31c can determine an anomaly only if a "droplet-containing" ejection state occurs. Alternatively, for example, the anomaly determination unit 31c can calculate the size of the droplets captured in the frame image that is determined to be in a "droplet-containing" ejection state, and determine an anomaly if the calculated droplet size exceeds a threshold. Alternatively, for example, the anomaly determination unit 31c can determine an anomaly if the number of times droplets are generated exceeds a threshold.

[0055] The display processing unit 31d performs the processing of displaying various characters and images on the display unit 34. In this embodiment, when the anomaly determination unit 31c determines that an anomaly has occurred, the display processing unit 31d displays a warning screen for the user on the display unit 34 to notify the user of the occurrence of the anomaly. The display processing unit 31d may display, for example, an image (moving image or frame image) that is the main factor in determining the occurrence of the anomaly, information related to the anomaly that has occurred, and identification information of the substrate that is the object of processing on the warning screen. In addition, the display processing unit 31d may also display various other types of information on the display unit 34.

[0056] The control processing unit 31e performs control related to the ejection of liquid during substrate processing in the substrate processing apparatus 1 based on the ejection state determination result made by the ejection state determination unit 31b and the abnormality determination result made by the abnormality determination unit 31c. In this embodiment, a valve is provided in the liquid supply path from the liquid supply source 16 to the ejection unit 13 in the substrate processing apparatus 1, and the control processing unit 31e can control the ejection of liquid by the ejection unit 13 by giving the substrate processing apparatus 1 an opening and closing command to the valve. The valve control can be, for example, simply opening and closing control, or it can be, for example, adjusting the opening and closing amount or opening and closing speed. In addition, in the case where the structure can selectively eject medicine or cleaning liquid from one ejection unit 13, the control processing unit 31e can also perform switching control of which type of liquid is ejected.

[0057] For example, if the anomaly determination unit 31c determines that an anomaly exists, the control processing unit 31e stops the liquid ejection from the ejection unit 13 by controlling the shut-off valve. In this case, the control processing unit 31e may also eject the cleaning fluid from the ejection unit 13. Furthermore, if, for example, the time from controlling the shut-off valve to the breakage of the liquid column ejected from the ejection unit 13 exceeds a threshold and is excessively long, the control processing unit 31e can control the speed of the shut-off valve to accelerate.

[0058] Alternatively, the ejection state of the ejection unit 13 can be determined not during actual substrate processing, but during the preparation phase such as startup of the substrate processing apparatus 1. In this case, the control processing unit 31e can, for example, control the valve opening by gradually opening / closing it while adjusting the valve opening to explore the droplet generation rate. Furthermore, the control processing unit 31e can, for example, control the rate of droplet generation or the rate at which the time until the liquid column breaks exceeds a threshold by repeatedly opening / closing the valve while adjusting the opening and closing speed. Based on the information obtained through such exploration, it is expected that the user will determine appropriate settings for the substrate processing apparatus 1 to perform substrate processing.

[0059] <Ejection Status Determination and Abnormality Determination>

[0060] The information processing device 3 in this embodiment determines the ejection state of liquids such as medicine or cleaning solution ejected from the ejection unit 13 based on motion images obtained by the camera 5 capturing images of the ejection unit 13 of the substrate processing device 1. In this embodiment, the ejection state is categorized into four types: "with liquid column", "liquid column broken and falling", "with liquid droplets", and "no liquid". Figure 5 This is a schematic diagram illustrating an example of the ejection state. In Figure 5 The image shown is an example of an image of the ejector section 13 captured by camera 5. The upper image corresponds to the ejection state with a "liquid column", the middle image corresponds to the ejection state with "liquid column broken and falling", and the lower image corresponds to the ejection state with "liquid droplets". In addition, the illustration of the ejection state without liquid is omitted.

[0061] In this example, the ejector section 13 is cylindrical and positioned above the substrate to be processed at predetermined intervals. Liquids such as medicine or cleaning solution are ejected from the opening at the lower end of the ejector section 13, and the ejected liquid falls onto the upper surface of the substrate to be processed. The shooting range of the camera 5 is defined to include the area from the lower end of the ejector section 13 to the upper surface of the substrate. Furthermore, the ejector section 13 can be moved horizontally via a drive mechanism (not shown), and the camera 5 moves with the ejector section 13 or captures images of the entire area within the movement range of the ejector section 13. Therefore, the ejector section 13 can be positioned at any location to capture images of the ejected liquid.

[0062] The "liquid column" ejection state is a state in which liquid is continuously ejected from the ejection section 13, and the columnar liquid (liquid column) connects the lower end of the ejection section 13 to the upper surface of the substrate. The "liquid column broken and falling" ejection state is a state in which the ejection of liquid from the ejection section 13 has just stopped, and there is space between the lower end of the ejection section 13 and the upper end of the liquid column, with the liquid column standing upright on the upper surface of the substrate. The "droplet" ejection state is a state in which there is no liquid column between the lower end of the ejection section 13 and the upper surface of the substrate, but there are one or more spherical liquid (droplets). The "no liquid" ejection state is a state in which there is neither a liquid column nor a droplet between the lower end of the ejection section 13 and the upper surface of the substrate.

[0063] The information processing apparatus 3 involved in this embodiment acquires motion images captured by the camera 5 and extracts frame images contained in the motion images for input to... Figure 4The learning model shown is used to acquire information related to the ejection state output by the learning model. Based on the information acquired from the learning model, the information processing device 3 determines whether the ejection state of the ejection section 13 captured in the frame image is "with liquid column", "liquid column broken and falling", "with droplets", or "no liquid".

[0064] Furthermore, the moving images captured by camera 5 contain, for example, dozens of frames per second. The information processing device 3 repeatedly determines the ejection state using a learning model for multiple frames within the moving images in a time-series order. Thus, the information processing device 3 can, for example, obtain ejection state determination results dozens of times per second. Based on the determination results of multiple ejection states in the time series, the information processing device 3 can, for example, determine the moment when the ejection state changes from "with liquid column" to "liquid column broken and falling," and the moment when it changes from "liquid column broken and falling" to "no liquid." Additionally, the information processing device 3 can calculate the time during which one ejection state is maintained based on the moment the ejection state changes. For example, the information processing device 3 can calculate the time during which the ejection state "liquid column broken and falling" is maintained based on the number of frames in the moving images existing between the moment the ejection state changes from "with liquid column" to "liquid column broken and falling" and the moment it changes from "liquid column broken and falling" to "no liquid."

[0065] Furthermore, the information processing device 3 according to this embodiment controls the start and stop of liquid ejection from the ejection section 13 of the substrate processing device 1 by issuing opening and closing commands to the valve provided in the liquid supply path from the liquid supply source 16 to the ejection section 13. For example, the information processing device 3 can calculate the elapsed time between the time when the command to close the valve is given to the substrate processing device 1 and the time determined according to the image captured by the camera 5 as described above.

[0066] The information processing device 3 determines whether there is an abnormality related to the liquid ejection during substrate processing performed by the substrate processing device 1 based on whether the time calculated from the motion image captured by the camera 5 exceeds a threshold. In this embodiment, the information processing device 3 determines the presence or absence of an abnormality by judging whether the time from issuing the command to close the valve until the liquid column breaks exceeds a threshold, and whether the time from the break of the liquid column until the end of liquid ejection exceeds a threshold.

[0067] Figure 6This is a flowchart illustrating an example of the ejection state determination process performed by the information processing apparatus 3 according to this embodiment. The ejection state determination unit 31b of the processing unit 31 of the information processing apparatus 3 according to this embodiment acquires the earliest frame image in the time series from a plurality of frame images included in the motion image captured by the camera 5 that has not been determined for ejection state (step S1). The ejection state determination unit 31b inputs the frame image acquired in step S1 into the machine learning completed learning model pre-stored in the model information storage unit 32b. Figure 4 (Step S2). The ejection state determination unit 31b acquires the ejection state determination result output by the learning model based on the image input in step S2 (Step S3).

[0068] The ejection state determination unit 31b stores the ejection state obtained in step S3, along with various other information such as the frame image obtained in step S1, the date and time of obtaining the frame image, or the identification information of the substrate being processed, in the log information storage unit 32c (step S4). The ejection state determination unit 31b determines whether the substrate processing performed by the substrate processing apparatus 1 has ended (step S5). If the substrate processing has not ended (S5: "No"), the ejection state determination unit 31b returns the processing to step S1 and performs the same processing on the next frame image according to the time sequence. If the substrate processing has ended (S5: "Yes"), the ejection state determination unit 31b ends the ejection state determination process.

[0069] Figure 7 and Figure 8 This is a flowchart illustrating an example of the anomaly detection process performed by the information processing apparatus 3 according to this embodiment. Furthermore, the process shown in this flowchart begins at the state where the substrate processing apparatus 1 is spraying a liquid such as medicine or cleaning solution from the ejection section 13. Additionally, the information processing apparatus 3 controls the substrate processing of the substrate processing apparatus 1 according to a predetermined process; for example, after spraying a liquid for a time or amount set as a process, it stops the spraying of liquid by issuing a stop spraying command to the substrate processing apparatus 1. The anomaly detection unit 31c of the processing unit 31 of the information processing apparatus 3 according to this embodiment determines whether the time to stop the spraying of liquid by the ejection section 13 of the substrate processing apparatus 1 has been reached (step S11). If the stop time has not been reached (S11: "No"), the anomaly detection unit 31c enters standby mode until the time to stop the spraying of liquid is reached.

[0070] When the moment for stopping liquid ejection is reached (S11: "Yes"), the control processing unit 31e of the processing unit 31 performs valve closing control by issuing a command to the substrate processing device 1 to close the valve provided in the liquid supply path from the liquid supply source 16 to the ejection unit 13 (step S12). Additionally, the anomaly determination unit 31c determines the liquid ejection state from the ejection unit 13 based on the motion image captured by the camera 5 (step S13). Furthermore, in step S13, the following steps are performed... Figure 6 The flowchart shown illustrates the processing. Based on the determination result of step S13, the anomaly determination unit 31c determines whether the ejection state of the ejection unit 13 is "liquid column broken and falling" (step S14). If the ejection state is not "liquid column broken and falling" (S14: "No"), the anomaly determination unit 31c returns the processing to step S13 and repeats the ejection state determination.

[0071] When the ejection state is "liquid column broken and falling" (S14: "Yes"), the anomaly determination unit 31c calculates the time from the moment the command to close the valve was given in step S12 to the moment it was determined in step S14 that the ejection state is "liquid column broken and falling" (liquid breakage time) (step S15). The anomaly determination unit 31c determines whether the liquid breakage time calculated in step S15 exceeds a predetermined threshold (step S16). If the liquid breakage time exceeds the threshold (S16: "Yes"), the display processing unit 31d of the processing unit 31 notifies the user of the anomaly by displaying a warning screen on the display unit 34 (step S17) and ends the processing.

[0072] If the liquid breakage time does not exceed the threshold (S16: "No"), the anomaly determination unit 31c determines the ejection state of the liquid ejected from the ejector 13 based on the motion image captured by the camera 5 (step S18). Based on the determination result of step S18, the anomaly determination unit 31c determines whether the ejection state of the ejector 13 is "no liquid" (step S19). If the ejection state is not "no liquid" (S19: "No"), the anomaly determination unit 31c returns the process to step S18 and repeats the ejection state determination. If the ejection state is "no liquid" (S19: "Yes"), the anomaly determination unit 31c calculates the time (end time) from the moment in step S14 when the ejection state was determined to be "liquid column broken and falling" to the moment in step S19 when the ejection state was determined to be "no liquid" (step S20).

[0073] Next, the anomaly determination unit 31c determines the ejection state of the liquid ejected from the ejector 13 based on the motion image captured by the camera 5 (step S21). Based on the determination result of step S21, the anomaly determination unit 31c determines whether the ejection state of the ejector 13 is "with droplets" (step S22). If the ejection state is "with droplets" (S22: "Yes"), the anomaly determination unit 31c returns the process to step S18 and repeats the ejection state determination. If the ejection state is not "with droplets" (S22: "No"), the anomaly determination unit 31c determines whether a predetermined time has elapsed since the moment in step S19 when the ejection state was determined to be "without droplets" (step S23). If the predetermined time has not elapsed (S23: "No"), the anomaly determination unit 31c returns the process to step S21 and repeats the ejection state determination.

[0074] If the predetermined time has elapsed (S23: "Yes"), the anomaly determination unit 31c determines whether the end time calculated in step S20 exceeds a predetermined threshold (step S24). If the end time exceeds the threshold (S24: "Yes"), the display processing unit 31d notifies the user of the anomaly by displaying a warning screen on the display unit 34 (step S25) and ends the processing. If the end time does not exceed the threshold (S24: "No"), the anomaly determination unit 31c does not perform an anomaly notification based on the display of the warning screen and ends the processing.

[0075] Display Processing

[0076] Figure 9 This is a schematic diagram showing an example of a warning screen displayed by the information processing apparatus 3 according to this embodiment. For example, the information processing apparatus 3 according to this embodiment displays the warning screen on the display unit 34 in steps S17 and S25 of the flowchart described above. Figure 9 The warning screen shown is an example of a warning screen displayed in step S25 when it is determined that droplets have been generated and the end time has exceeded a threshold. In this warning screen, the information processing device 3 obtains identification information such as the batch number and slot number attached to the substrate to be processed, and displays a warning message such as "Droplets generated at slot Y in batch X!" at the top of the screen. In addition, the identification information of the substrate to be processed can be input by the user into the information processing device 3 or the substrate processing device 1 before the substrate processing begins.

[0077] Additionally, the information processing device 3 displays one of the frame images determined to be "with droplets" or a motion image containing multiple frame images determined to be "with droplets" below the warning message on the warning screen. Furthermore, the information processing device 3 displays the information "Droplet Falling Time: xx seconds" based on the end time calculated in step S20 of the flowchart described above. Moreover, the information processing device 3 can also count the number of droplets produced to display the information "Number of Droplets: y", and can also display the information "Droplet Volume: zzmL" based on the size of the produced droplets. Furthermore, in Figure 6 to Figure 8 The flowchart omits calculations for the number of droplets and their capacity.

[0078] Figure 10 This is a schematic diagram illustrating an example of information display performed by the information processing apparatus 3 according to this embodiment. The information processing apparatus 3 according to this embodiment can also display information related to droplet generation separately from (or together with) the warning screen described above. For example, the information processing apparatus 3 stores the droplet fall time (end time of step S20) when droplets are generated in the log information storage unit 32c. The information processing apparatus 3 reads multiple past droplet fall time values ​​stored in the log information storage unit 32c and displays, for example, a histogram with the droplet fall time as the horizontal axis and the number of data points as the vertical axis on the display unit 34.

[0079] The information processing device 3, for example, highlights the matching parts by using color differentiation to indicate which part of the histogram the latest droplet fall time is included in. Furthermore, the information processing device 3 overlays the histogram of droplet fall times with a distribution curve assuming the droplet fall time distribution follows a normal distribution. Additionally, the information processing device 3 calculates a predetermined confidence interval for the distribution of droplet fall times and displays the range of droplet fall times corresponding to the confidence interval on the histogram. Moreover, the information processing device 3 can, for example, determine the threshold used for judgment in step S24 of the flowchart above based on the upper limit of the confidence interval for the distribution of past droplet fall times. The information processing device 3 can also periodically perform the process of calculating the predetermined confidence interval based on information from multiple past droplet fall times stored in the log information storage unit 32c, and periodically update the threshold used for judgment.

[0080] also, Figure 9 and Figure 10 The displayed screen is an example, and is not limited to this; the information processing device 3 can notify the substrate of any abnormality in the display format. Furthermore, the information processing device 3 does not only display this information when an abnormality occurs, but can also display this information when no abnormality occurs.

[0081] <Variation Example>

[0082] (Variation Example 1)

[0083] In the information processing system described in Modification 1, in addition to a camera 5 for capturing images of the ejection section 13, a second camera for capturing images of the surface of the substrate to be processed is also provided in the chamber 11 of the substrate processing apparatus 1. The information processing apparatus 3 acquires data of moving images captured by the second camera and determines the dryness or wetness of the substrate surface (whether it is dry or wet) based on the frame images contained in the acquired moving images.

[0084] To determine the dryness or wetness of a substrate surface, a learning model generated beforehand through machine learning can be used, for example. This learning model can be generated, for example, through supervised machine learning using training data (labeled data), which is obtained by mapping images of the substrate surface to indicators indicating whether the surface is dry or wet. However, the dryness or wetness of the substrate surface can also be determined without using a learning model, for example, by comparing pixel values ​​of images of the substrate surface with a threshold.

[0085] The information processing apparatus 3 involved in Modification Example 1 simultaneously determines the ejection state based on frames in a moving image obtained by photographing the ejection section 13, and determines the wet / dry condition based on frames in a moving image obtained by photographing the substrate surface. For example, if it is determined that the ejection state is "with droplets" and the substrate surface is dry, the information processing apparatus 3 can determine that an abnormality has occurred in the ejection of liquid during substrate processing and notify the user.

[0086] Furthermore, the above-mentioned abnormality determination conditions are just one example and are not limited to this. The information processing device 3 can also perform arbitrary notifications and controls based on any combination of the determination results of the liquid ejection state and the determination results of the dryness and wetness of the substrate surface. Alternatively, instead of providing a separate camera for photographing the substrate surface in addition to the camera 5 that photographs the ejection section 13, the camera 5 can simultaneously photograph both the ejection section 13 and the substrate surface.

[0087] (Variation Example 2)

[0088] In the above-described embodiments, the presence or absence of an abnormality is determined when the substrate processing apparatus 1 stops ejecting liquid from the ejection section 13, but this is not limited to this. The information processing apparatus 3 according to Modification 2 determines the presence or absence of an abnormality when the substrate processing apparatus 1 begins ejecting liquid from the ejection section 13. For example, the information processing apparatus 3 according to Modification 2 determines that an abnormality has occurred if, after a command to open the valve to eject liquid from the ejection section 13 is given, and during the period until a columnar liquid (liquid column) is ejected from the ejection section 13, droplets are ejected from the ejection section 13, and a warning screen is displayed or control processing corresponding to the abnormality is performed.

[0089] Figure 11 This is a flowchart illustrating an example of the anomaly detection process performed by the information processing apparatus 3 according to Modification 1. In Modification 2, the control processing unit 31e of the processing unit 31 of the information processing apparatus 3 performs control (step S41) to open the valve provided in the liquid supply path from the liquid supply source 16 to the ejection unit 13 by communicating with the substrate processing apparatus 1 through the communication unit 33.

[0090] Next, the anomaly determination unit 31c of the processing unit 31 determines the ejection state of the liquid from the ejection unit 13 based on the moving image captured by the camera 5 (step S42). Based on the determination result of step S42, the anomaly determination unit 31c determines whether the ejection state of the ejection unit 13 is "with droplets" (step S43). If the ejection state is "with droplets" (S43: "Yes"), the anomaly determination unit 31c notifies the user of the anomaly by displaying a warning screen on the display unit 34 (step S44) and ends the processing.

[0091] If the ejection state is not "with droplets" (S43: "No"), the anomaly determination unit 31c determines whether the ejection state of the ejection unit 13 is "with a liquid column" (step S45) based on the determination result of step S42. If the ejection state is not "with a liquid column" (S45: "No"), the anomaly determination unit 31c returns the process to step S42 and repeats the ejection state determination. If the ejection state is "with a liquid column" (S45: "Yes"), the anomaly determination unit 31c ends the anomaly determination process at the start of ejection.

[0092] Figure 12 This is a schematic diagram illustrating a display example of a warning screen displayed by the information processing apparatus 3 according to Modification 2. For example, in step S44 of the flowchart described above, the information processing apparatus 3 in Modification 2 displays the warning screen on the display unit 34. Figure 12In the warning screen shown, the information processing device 3 acquires identification information such as batch number and slot number attached to the substrate being processed, and displays a warning message at the top of the screen stating, "Droplets are generated at slot Y in batch X at the start of ejection!" Additionally, the information processing device 3 displays one of the frame images indicating "droplets are present," or a motion image containing multiple frame images indicating "droplets are present," below the warning message on the warning screen. Furthermore, the information processing device 3 can also count the number of droplets generated to display "Number of droplets: y," or display "Droplet volume: zzmL" based on the size of the generated droplets.

[0093] Summary

[0094] In the information processing system of this embodiment with the above structure, the information processing device 3 acquires a motion image obtained by capturing a photograph of the ejection section 13 of the substrate processing device 1, which ejects liquid towards the substrate being processed. The information processing device 3 inputs frame images contained in the acquired motion image to a learning model, which is a learning model obtained by performing machine learning in advance by taking the image of the ejection section 13 as input and outputting information related to the ejection state of the liquid ejected from the ejection section 13. The information processing device 3 acquires the information output by the learning model and determines whether the ejection of liquid in the substrate processing is correct (normal / abnormal) based on the acquired information. Therefore, it is expected that the information processing system of this embodiment can automatically determine the correctness of liquid ejection based on the motion image captured by the camera 5.

[0095] Furthermore, in the information processing system of this embodiment, the ejection state determined by the information processing device 3 includes: a first state "with liquid column", which is a state in which liquid is ejected in a columnar shape from the ejection section 13 onto the substrate being processed; a second state "liquid column breaks and falls", which is a state in which the liquid ejected from the ejection section 13 breaks and the columnar liquid falls onto the substrate; a third state "with droplets", which is a state in which droplets fall onto the substrate from the ejection section 13; and a fourth state "no liquid", which is a state in which no liquid is ejected from the ejection section 13. Therefore, it is expected that the information processing system of this embodiment can use a learning model obtained by learning these ejection states to make highly accurate determinations related to liquid ejection.

[0096] Furthermore, in the information processing system of this embodiment, the information processing device 3 calculates the time from the point when the control to stop the ejection of liquid from the ejection section 13 is performed until the point when the ejection state of the liquid changes from "with liquid column" to "liquid column broken and falling" based on information obtained from the learning model, and determines normal / abnormal conditions related to the ejection of liquid in substrate processing based on the calculated time. For example, the information processing device 3 stops the ejection of liquid by controlling the closure of a valve provided in the flow path from the liquid supply source 16 to the ejection section 13. Therefore, it is expected that the information processing system of this embodiment can determine an abnormality, for example, if a predetermined time has elapsed since the ejection of liquid was stopped but the ejection has not stopped.

[0097] Furthermore, in the information processing system of this embodiment, the information processing device 3 determines whether the liquid ejection related to the substrate processing is normal or abnormal based on whether the information obtained from the learning model indicates "droplets are present." Therefore, it is expected that the information processing system of this embodiment can determine the correctness of the substrate processing based on whether droplets have been generated. Additionally, the information processing device 3 may also calculate the size of the droplets and determine normal / abnormal status based on the calculated size.

[0098] Furthermore, in the information processing system of this embodiment, the information processing device 3 calculates the time from when the information obtained from the learning model changes from "liquid column breaks and falls" to "no liquid," and determines normal / abnormal conditions related to the ejection of liquid from the substrate processing based on the calculated time. Therefore, it is expected that the information processing system of this embodiment can accurately determine abnormalities based on the time until the liquid ejection stops completely.

[0099] Furthermore, in the information processing system of this embodiment, the information processing device 3 controls the opening and closing of valves provided in the flow path from the liquid supply source 16 to the ejection section 13 based on information obtained from the learning model, thereby controlling the ejection of liquid from the ejection section 13. For example, the information processing device 3 can control the ejection of liquid from the ejection section 13 to stop it if an abnormality is determined based on information obtained from the learning model. Therefore, it is expected that the information processing system of this embodiment can achieve high-precision control over the ejection of liquid during substrate processing.

[0100] Furthermore, in the information processing system of this embodiment, if the information processing device 3 determines that the ejection of droplets during substrate processing is incorrect, it notifies the user of the abnormality, for example, by displaying a warning screen on the display unit 34. At this time, the information processing device 3 may, for example, notify the user of the presence or absence of droplets, the number of droplets, the amount of droplets, or the droplet's fall time. Additionally, it may also notify the user of identification information for the substrate being processed. Therefore, it is expected that the information processing system of this embodiment will notify the user of the presence or absence of an abnormality and related information regarding the ejection of liquid during substrate processing, thereby prompting a response to the abnormality.

[0101] Furthermore, in the information processing system according to this embodiment, the information processing device 3 stores at least the frame image of when it is determined that the liquid is spraying incorrectly, and the information obtained from the learning model, in the log information storage unit 32c. Alternatively, the information processing device 3 may store the information of when it is determined that the liquid is spraying correctly in the log information storage unit 32c, or it may choose not to store this information. Therefore, it is expected that the information processing system according to this embodiment can be used by users to verify the cause of an anomaly based on the information stored in the log information storage unit 32c.

[0102] Furthermore, in the information processing system of this embodiment, the information processing device 3 acquires a moving image obtained by photographing a substrate that is the object of processing. Based on the frame images contained in the acquired moving image, the surface state of the substrate is determined. Based on information related to the ejection state obtained from a learning model and the determined surface state of the substrate, a normal / abnormal determination related to the ejection of liquid processed on the substrate is made. The surface state of the substrate includes, for example, a dry surface state and a wet surface state. Therefore, it is expected that the information processing system of this embodiment can accurately determine normal / abnormal conditions by considering the surface state of the substrate that is the object of processing.

[0103] Furthermore, in the information processing system of this embodiment, the information processing device 3 determines whether the liquid ejection from the substrate processing is normal or abnormal based on whether a "droplet" state occurs during the period from when the liquid ejection from the ejection section 13 is started and until the liquid ejection state becomes "liquid column". Therefore, it is expected that the information processing system of this embodiment will not only determine normal / abnormal conditions when the liquid ejection from the ejection section 13 stops, but also when the liquid ejection begins.

[0104] It should be considered that all points in the embodiments disclosed herein are illustrative rather than restrictive. The scope of this disclosure is shown not by the foregoing meaning but by the claims, and is intended to include all modifications within the meaning and scope equivalent to the claims.

[0105] The items described in each embodiment can be combined with each other. Furthermore, the independent and dependent claims described in the claims can be combined with each other in all possible combinations, regardless of their referencing form. Moreover, the claims can be described in a form that refers to two or more other claims (multiple claim form), but are not limited to this. It is also possible to use a form that describes multiple claims that refer to at least one other multiple claim (multiple-referencing-multiple-claims).

[0106] Explanation of reference numerals in the attached figures

[0107] 1: Substrate processing device; 3: Information processing device (computer); 5: Camera; 11: Chamber; 12: Substrate holding mechanism; 12a: Holding part; 12b: Support part; 12c: Drive part; 13: Ejection part; 14: Recovery cup; 14a: Drain port; 14b: Exhaust port; 15: FFU; 16: Supply source; 31: Processing unit; 31a: Image acquisition unit; 31b: Ejection status determination unit; 31c: Abnormal determination unit; 31d: Display processing unit; 31e: Control processing unit; 32: Storage unit; 32a: Program (computer program); 32b: Model information storage unit; 32c: Log information storage unit; 33: Communication unit; 34: Display unit; 35: Operation unit; N: Network.

Claims

1. A program product comprising a computer program that causes a computer to perform the following processes: A motion image is obtained by taking a picture of the ejection section of a substrate processing apparatus that ejects liquid toward a substrate to be processed. The learning model is obtained by inputting the frame images contained in the acquired motion images into the learning model, which is a learning model obtained by machine learning in a way that takes the image of the ejection section as input and outputs information related to the ejection state of the liquid ejected from the ejection section. Obtain information related to the ejection state output by the learning model; as well as Based on the information obtained, the correctness of the information related to the ejection of liquid is determined.

2. The program product according to claim 1, wherein, The ejection state includes: The first state is a state in which liquid is ejected in a columnar shape from the ejection section onto the substrate to be processed. The second state is a state in which the liquid ejected from the ejection part breaks and the columnar liquid falls down onto the substrate. The third state is the state in which droplets fall vertically onto the substrate from the ejection portion; and The fourth state is a state in which no liquid is ejected from the ejector.

3. The program product according to claim 2, wherein, The calculation covers the time from the point when the ejection of liquid by the ejector was stopped to the point when the ejection state of the liquid changed from the first state to the second state based on the information obtained from the learning model. The accuracy of the judgment related to the ejection of liquid is determined based on the calculated time.

4. The program product according to claim 3, wherein, Control the opening and closing of the valve located in the flow path of the liquid. As a stop control, the valve is closed.

5. The program product according to claim 2, wherein, The correctness of the liquid ejection is determined based on whether the ejection state of the liquid is the third state, according to the information obtained from the learning model.

6. The program product according to claim 5, wherein, Calculate the droplet size when the liquid is ejected in the third state. The accuracy of the determination of whether or not the liquid ejection is correct is based on the calculated size.

7. The program product according to claim 2, wherein, Calculate the time taken for the liquid ejection state to change from the second state to the fourth state, based on the information obtained from the learning model. The accuracy of the judgment related to the ejection of liquid is determined based on the calculated time.

8. The program product according to claim 1, wherein, The opening and closing of valves located in the flow path of the liquid are controlled based on the information obtained from the learning model.

9. The program product according to claim 1, wherein, Notification will be issued if it is determined that the liquid is being sprayed incorrectly.

10. The program product according to claim 9, wherein, Notify the presence or absence of droplets, the number of droplets, the amount of droplets, or the time of droplet fall.

11. The program product according to claim 9, wherein, The notification provides identification information about the substrate being processed.

12. The program product according to claim 1, wherein, The frame image when it is determined that the liquid is spraying incorrectly, and the information obtained from the learning model are stored in the storage unit.

13. The program product according to claim 1, wherein, Acquire moving images obtained by photographing the substrate to be processed. The surface state of the substrate is determined based on the frame images contained in the acquired motion images. Based on the information related to the ejection state output by the learning model and the determined surface state of the substrate, the correctness of the ejection of liquid is determined.

14. The program product according to claim 13, wherein, The surface state includes the dry state of the substrate surface and the wet state of the substrate surface.

15. The program product according to claim 2, wherein, The correctness of the liquid ejection is determined based on whether the third state occurs during the period from the start of control over the ejection of the liquid by the ejection unit until the ejection state of the liquid becomes the first state.

16. An information processing method, wherein, An information processing device acquires a moving image by taking a picture of the ejection unit of a substrate processing device that ejects liquid toward a substrate that is being processed. The information processing device inputs frame images contained in the acquired motion images into the learning model. The learning model is a learning model obtained by machine learning in which the image of the ejection section is taken as input and the output is information related to the ejection state of the liquid ejected from the ejection section. The information processing device acquires the ejection state-related information output by the learning model. The information processing device determines the correctness of the information related to the ejection of liquid based on the information obtained.

17. An information processing apparatus, wherein, Equipped with a processing department, The processing unit acquires a moving image by taking a picture of the ejection unit of the substrate processing apparatus that ejects liquid toward the substrate to be processed. The processing unit inputs the frame images contained in the acquired motion images into the learning model. The learning model is a learning model obtained by machine learning in which the image of the ejection section is processed as input and the information related to the ejection state of the liquid ejected from the ejection section is output. The processing unit acquires information related to the ejection state output by the learning model. The processing unit determines the correctness of the liquid ejection based on the acquired information.

18. A computer-readable recording medium storing a computer program that causes a computer to perform the following processes: A motion image is obtained by taking a picture of the ejection section of a substrate processing apparatus that ejects liquid toward a substrate to be processed. The learning model is obtained by inputting the frame images contained in the acquired motion images into the learning model, which is a learning model obtained by machine learning in a way that takes the image of the ejection section as input and outputs information related to the ejection state of the liquid ejected from the ejection section. Obtain information related to the ejection state output by the learning model; as well as Based on the information obtained, the correctness of the information related to the ejection of liquid is determined.

Images