Information processing device, information processing method, and program

Abstract

The present invention provides an information processing device capable of easily identifying the cause of a defect and the like. According to one embodiment of the present invention, an information processing apparatus is provided. The information processing apparatus includes an image acquisition section, a relational model acquisition section, and an output section. The image acquisition unit is configured to acquire a predetermined image from an industrial process. The predetermined image comprises an image which is generated in the industrial process and is related to the defect. The relational model acquisition unit is configured to acquire a relational model. The relational model is a model created by associating analysis data obtained by analyzing a predetermined site in the industrial process with image data of a corresponding image related to the defect. The output unit is configured to output a cause of the defect and / or a countermeasure for the defect on the basis of the predetermined image and the relational model.

Description

Technical Field

[0001] The present invention relates to an information processing device, an information processing method, and a program. Background Art

[0002] Various technologies have been developed to prevent problems in industrial processes. Patent Document 1 discloses a technology for controlling slime in a water system containing a reducing substance by predetermined control.

[0003] Prior art literature

[0004] Patent Literature

[0005] [Patent Document 1]

[0006] Japanese Patent Publication No. 2009-241018 Summary of the Invention

[0007] Problems to be solved by the invention

[0008] Despite the existence of these technologies to prevent defects, unexpected defects can still occur in real industrial processes. Various analyses have been conducted to identify the causes of these defects, but the reality is that these analyses take considerable time. Furthermore, since identifying these defects requires the expertise of analysts, a technology that can easily identify the causes of defects is needed.

[0009] In view of the above circumstances, the present invention provides an information processing apparatus capable of easily identifying the cause of a defect and the like.

[0010] Solutions to Problems

[0011] According to one embodiment of the present invention, an information processing device is provided. The information processing device includes an image acquisition unit, a relational model acquisition unit, and an output unit. The image acquisition unit is configured to acquire predetermined images from an industrial process. The predetermined images include images related to defects generated in the industrial process. The relational model acquisition unit is configured to acquire a relational model. The relational model is a model created by associating analysis data obtained by analyzing predetermined locations in the industrial process with image data of corresponding defect-related images. The output unit is configured to output the cause of the defect and / or a countermeasure for the defect based on the predetermined image and the relational model.

[0012] Specifically, the present invention can be implemented in the following various ways.

[0013] (1) An information processing device, comprising: an image acquisition unit configured to acquire predetermined images from an industrial process, the predetermined images including images related to defects generated in the industrial process; a relational model acquisition unit configured to acquire a relational model, the relational model being a model created by associating analysis data obtained by analyzing a predetermined portion in the industrial process with image data related to an image of a corresponding defect; and an output unit configured to output a cause of the defect and / or a countermeasure for the defect based on the predetermined image and the relational model.

[0014] (2) The information processing device according to (1) above, wherein the output unit is configured to output a cause of the defect and a countermeasure for the defect.

[0015] (3) The information processing device according to (1) or (2) above, wherein the analysis data is data obtained by analyzing the predetermined part by a chemical or biochemical method.

[0016] (4) The information processing device according to (1) or (2) above, wherein the analysis data includes one or more analysis data selected from the group consisting of: optical microscope observation, electron microscope observation, Fourier transform infrared spectroscopy analysis, pyrolysis gas chromatography analysis, ninhydrin reaction test, X-ray fluorescence analysis, X-ray diffraction analysis, bacteria count analysis, fungus count analysis, elemental mapping analysis, iodine starch reaction test, organic solvent-based extraction amount, acid-based extraction amount, and alkali-based extraction amount.

[0017] (5) The information processing device according to any one of (1) to (4) above, wherein the analysis data is data obtained by analyzing the defect in the industrial process as a predetermined portion in the industrial process.

[0018] (6) The information processing device according to any one of (1) to (5) above, wherein the industrial process is a process including a water-based step.

[0019] (7) The information processing device according to any one of (1) to (6) above, wherein the industrial process is a papermaking process.

[0020] (8) An information processing device according to any one of (1) to (7) above, wherein the image data associated with the relational model includes any information of the color, size and shape of the defect; and the output unit is configured to output the cause of the defect and / or a countermeasure for the defect based on any information of the color, size and shape of the defect in the predetermined image.

[0021] (9) The information processing device according to any one of (1) to (8) above, wherein the predetermined image is an image obtained by photographing with an RGB (red, green, blue) camera.

[0022] (10) The information processing device according to any one of (1) to (9) above, wherein the output unit is configured to output an operating condition of the industrial process as a countermeasure against the defect.

[0023] (11) The information processing device according to any one of (1) to (10) above, wherein the relational model is a model configured to be updated before the output unit outputs the cause of the defect and / or the countermeasure for the defect.

[0024] (12) An information processing method performed by an information processing device, comprising: an image acquisition step for acquiring a predetermined image from an industrial process, wherein the predetermined image includes an image related to a defect generated in the industrial process; a relational model acquisition step for acquiring a relational model, wherein the relational model is a model created by associating analysis data obtained by analyzing a predetermined part in the industrial process with image data related to an image of a corresponding defect; and an output step for outputting a cause of the defect and / or a countermeasure for the defect based on the predetermined image and the relational model.

[0025] (13) A program that causes a computer to function as the following units: an image acquisition unit configured to acquire predetermined images from an industrial process, the predetermined images including images related to defects generated in the industrial process; a relational model acquisition unit configured to acquire a relational model, the relational model being a model created by associating analysis data obtained by analyzing a predetermined portion in the industrial process with image data related to an image of a corresponding defect; and an output unit configured to output a cause of the defect and / or a countermeasure for the defect based on the predetermined image and the relational model.

[0026] According to these aspects, an information processing device and the like that can easily identify the cause of a defect and the like are provided. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 It is a diagram showing the overall configuration of the information processing system 100 .

[0028] Figure 2 It is a diagram showing the hardware configuration of the information processing device 1 .

[0029] Figure 3 It is a block diagram showing the functions of the information processing device 1 .

[0030] Figure 4 This is an activity diagram showing the flow of information processing using the information processing device 1 and the like.

[0031] Figure 5 11 is a diagram showing an example of output content displayed by the output unit 113 . DETAILED DESCRIPTION

[0032] Hereinafter, embodiments of the present invention will be described. Note that various features described in the following embodiments can be combined with each other.

[0033] In addition, in this embodiment, the program for implementing the software can be provided as a computer-readable non-transitory storage medium (Non-Transitory Computer-Readable Medium), can be downloaded from an external server, and can also be provided by launching the program on an external computer to implement its functions on a client terminal (so-called cloud computing).

[0034] In this embodiment, a "unit" may include, for example, a combination of hardware resources implemented by circuits in a broad sense and information processing software specifically implemented by these hardware resources. Furthermore, this embodiment processes various information, which can be represented by physical values ​​representing voltage and current signals, binary bits consisting of 0s and 1s representing high and low signal values, or quantum superpositions (so-called qubits), and can perform communications and calculations on circuits in a broad sense.

[0035] In a broad sense, a circuit refers to a circuit implemented by appropriately combining at least circuits, circuit types, processors, and memory. This includes application-specific integrated circuits (ASICs), programmable logic devices (PLDs), such as simple programmable logic devices (SPLDs), complex programmable logic devices (CPLDs), and field programmable gate arrays (FPGAs).

[0036] 1. Hardware Structure

[0037] In this section, the hardware configuration of the information processing system 100 of this embodiment will be described. Figure 1 It is a diagram showing the overall configuration of the information processing system 100 .

[0038] The information processing system 100 of the present embodiment is a system for outputting the cause of a defect generated in an industrial process P or a countermeasure for the defect. The information processing system 100 of the present embodiment includes an information processing device 1 and a camera device 2, which are connected via a communication line. The communication line here includes the Internet or wireless, etc., which is used to mediate the data exchange between devices connected to its own line. On the other hand, the system exemplified by the information processing system 100 includes one or more devices or constituent elements. Therefore, the information processing device 1 itself can be regarded as an example of a system, and the example of the industrial process P including the camera device 2 or as an application object can also be referred to as a system. Below, the various devices that can constitute the information processing system 100 will be described.

[0039] (Information Processing Device 1)

[0040] Figure 2 1 is a diagram showing the hardware configuration of information processing device 1. Information processing device 1 includes a control unit 11, a storage unit 12, an input unit 13, a display unit 14, and a communication unit 15. These components are electrically connected via a communication bus 10. The following describes each component of information processing system 100.

[0041] (Control Unit 11)

[0042] The control unit 11 is, for example, a central processing unit (CPU) (not shown). The control unit 11 implements various functions related to the information processing device 1 by reading predetermined programs stored in the storage unit 12. Specifically, the software information processing stored in the storage unit 12 is specifically implemented by the control unit 11, which is an example of hardware, and can be executed as each functional unit included in the control unit 11. Further details regarding this will be described in the next section. Note that the control unit 11 is not limited to a single one; two or more control units 11 may be provided for each function. Furthermore, combinations of these may also be employed.

[0043] (Storage Unit 12)

[0044] The storage unit 12 is used to store the various information defined in the above description. This can be implemented, for example, as a storage device such as a solid-state drive (SSD) that stores various programs related to the information processing device 1 and executed by the control unit 11, or as a memory device such as random access memory (RAM) that stores temporary information necessary for program operations (parameters, arrays, etc.). The storage unit 12 stores various programs and variables related to the information processing device 1 and executed by the control unit 11.

[0045] (Input unit 13)

[0046] The input unit 13 may be included in the housing of the information processing device 1 or may be externally connected. For example, the input unit 13 may be integrated with the display unit 14 and implemented as a touch panel. If it is a touch panel, the user can input tapping operations, sliding operations, etc. Of course, a switch button, a mouse, a QWERTY keyboard, etc. may also be used instead of a touch panel. In other words, the input unit 13 receives operation input from the user. This input is transmitted as a command signal via the communication bus 10 to the control unit 11, and the control unit 11 can perform predetermined control or calculations as needed.

[0047] (Display unit 14)

[0048] The display unit 14 may be included in the housing of the information processing device 1 or may be externally connected. The display unit 14 displays a graphical user interface (GUI) screen that can be operated by the user. It is preferable to implement a display device such as a CRT display, a liquid crystal display, an organic EL display, or a plasma display, depending on the type of information processing device 1.

[0049] (Ministry of Communications 15)

[0050] The communication unit 15 is configured to be able to transmit various electrical signals from the information processing device 1 to external components. Furthermore, the communication unit 15 is configured to be able to receive various electrical signals transmitted from external components to the information processing device 1. Note that the communication unit 15 may have a network communication function, thereby enabling various information to be transmitted between the information processing device 1 and external devices via a communication line.

[0051] (Camera 2)

[0052] The imaging device 2 captures at least a portion of the industrial process P as an image. The imaging device 2 can be appropriately selected from known devices capable of capturing images of objects, etc. From one perspective, the imaging device 2 may be a digital camera, a camera connected to a smartphone, a camera connected to a computer, or the like. For example, the imaging device 2 may be an RGB (Red, Green, Blue) camera that captures at least a portion of the industrial process P as an image. In this case, the acquired data may be so-called RGB (Red, Green, Blue) data.

[0053] (Industrial Process P)

[0054] The industrial process P to which the information processing system 100 of this embodiment is applied can be appropriately set from among known industrial processes. The industrial process P is generally considered to be a production process implemented in various known factories, such as a papermaking process, a steelmaking process, a power generation process, an oil refining process, a chemical process, a coating process, a semiconductor processing process, and the like. As described later, it is preferred that the industrial process P include a water-based process because predetermined analysis data is used in the information processing system 100 of this embodiment. Among them, it is particularly preferred that the industrial process P be a papermaking process. In the following description, the information processing process will be described by assuming that the industrial process P is a papermaking process using a water system as an example.

[0055] 2. Functional structure

[0056] In this section, the functional structure of this embodiment will be described. Figure 3 1 is a block diagram showing the functions of the information processing device 1. As described above, information processing by software (stored in the storage unit 12) is embodied by hardware (control unit 11) and can be executed as each functional unit included in the control unit 11.

[0057] Specifically, the information processing device 1 (control unit 11) may include an image acquisition unit 111, a relational model acquisition unit 112, an output unit 113, a relational model creation unit 114, and a storage management unit 115 as functional units. Note that these functional units may be added or omitted as appropriate depending on the intended use of the information processing device 1.

[0058] (Image Acquisition Unit 111)

[0059] The image acquisition unit 111 is configured to execute an image acquisition step. During the image acquisition step, the image acquisition unit 111 acquires predetermined images from the industrial process P. Here, the predetermined images include images related to defects generated in the industrial process P. During this acquisition process, the image acquisition unit 111 obtains various information from the imaging device 2 capable of capturing at least a portion of the industrial process P, for example, via the communication unit 15.

[0060] (Relational Model Acquisition Unit 112)

[0061] The relational model acquisition unit 112 is configured to execute a relational model acquisition step. In this step, the relational model acquisition unit 112 acquires a relational model. Here, the relational model is created by correlating analysis data obtained from analyzing predetermined locations in the industrial process P with corresponding image data of defect-related images. This model will be described in more detail in subsequent sections.

[0062] (Output unit 113)

[0063] The output section 113 is configured to execute an output step. In the output step, the output section 113 generates various output objects. Typically, the output objects are configured to be recognizable by a user or the like. In this case, the output section 113 is configured to create display information and control the display information so as to be visible to a user or the like. The display information can be visual information itself, such as a screen, an image, an icon, text, or the like generated in a manner recognizable by a user, or can be rendering information for displaying visual information (e.g., a screen, an image, an icon, text, or the like) on various devices or terminals. Further, the output section 113 can be configured to output a signal to drive a printing device or to drive a predetermined device. In the information processing system 100 of the present embodiment, the output section 113 is configured to output a cause of a defect and / or a countermeasure against the defect based on a predetermined image and a relational model. The specific manner of the output content will be described later.

[0064] (Relational model creation section 114)

[0065] The relational model creation section 114 is configured to execute a relational model creation step. In the relational model creation step, the relational model creation section 114 creates or updates a relational model used for the above output step or the like.

[0066] (Storage management section 115)

[0067] The storage management section 115 is configured to execute a storage management step. In the storage management step, the storage management section 115 is configured to manage various information to be stored in association with the information processing system 100 of the present embodiment. Typically, the storage management section 115 is configured to store information processed by the information processing device 1 or the like in a storage area. Examples of the storage area include the storage section 12 of the information processing device 1 or a storage section of various devices or terminals, but the storage area does not necessarily have to be inside the information processing system 100, and the storage management section 115 can manage various information stored in an external storage section or the like.

[0068] 3. Information processing details

[0069] In the third section, an information processing method executed by the information processing device 1 or the like will be described with reference to an activity diagram or the like. Figure 4 is an activity diagram indicating an information processing flow using the information processing device 1 or the like.

[0070] First, in the information processing method of the present embodiment, the image acquisition section 111 acquires a predetermined image from the industrial process P (step S1).

[0071] Step S1 can typically be accomplished by using the camera device 2 to capture a predetermined portion of the industrial process P, with the image acquisition unit 111 acquiring image data via a communication link. It should be noted that the object captured by the camera device 2 can be any object capable of being captured, whether organic or inorganic. For example, the object can be predetermined equipment present in the industrial process P (including various equipment such as reactors, storage tanks, and transportation equipment), raw materials used in the industrial process P, or products or intermediate products manufactured in the industrial process P.

[0072] In the information processing method of this embodiment, the images acquired in step S1 include images related to defects generated in the industrial process P.

[0073] A "defect" refers to a state in which industrial process P is not operating as designed (in other words, an abnormal state) or its products. The information processing method of this embodiment can be applied to situations where an abnormality exists in industrial process P, or where the product image differs from that of a normal industrial process P. Typical examples include equipment defects or malfunctions, unusual colors or patterns of raw materials or products (intermediate products), or unusual dimensions of raw materials or products (intermediate products). In the information processing method of this embodiment, the imaging device 2 captures images containing such defects.

[0074] On the other hand, in the information processing method of this embodiment, the relational model acquisition unit 112 acquires the relational model (step S2). The order of executing steps S1 and S2 is arbitrary: step S1 may be executed before step S2, step S2 may be executed before step S1, or both steps may be executed simultaneously.

[0075] In step S2, a relational model is created by correlating analysis data obtained by analyzing a predetermined portion of the industrial process P with corresponding image data of defect-related images. Here, the relational model is a model of the relationship between the predetermined analysis data and the corresponding image data of the defect-related images. This model can be, for example, a function or a lookup table that indicates the relationship between the predetermined analysis data and the corresponding image data of the defect-related images, or it can be a learned model that has learned the relationship between the predetermined analysis data and the corresponding image data of the defect-related images.

[0076] Regarding this relational model, the relationship between the analysis data and the corresponding image data of the defect-related image can be analyzed based on known analysis methods. Typically, regression analysis (linear models, generalized linear models, generalized linear mixed models, ridge regression, Lasso regression, elastic net, support vector regression, projection pursuit regression, etc.), time series analysis (VAR models, SVAR models, ARIMAX models, SARIMAX models, state-space models, etc.), decision trees (decision trees, regression trees, regression trees, random forests, XGBoost, etc.), neural networks (simple parsers, multi-layer parsers, DNN (Deep Neural Network), CNN (Convolution Neural Network), RNN (Recurrent Neural Network), LSTM, etc.), Bayesian methods (simple Bayesian, etc.), clustering (k-means, k-means++, etc.), and integrated learning (Boosting, AdaBoost, etc.) can be used to analyze and obtain the desired prediction model.

[0077] In this embodiment, a neural network is preferably used in the above-mentioned analysis method. Furthermore, the software or program used to create and apply the relational model in this embodiment can be selected from a variety of software or programs that can use neural networks. For example, frameworks such as Keras, TensorFlow, and PyTorch, as well as software such as Teachable Machine (Google) or DataRobot (DataRobot) can also be used.

[0078] While the relational model creation example is performed by relational model creation unit 114, this is not limiting. The information processing method of this embodiment can also be executed by creating such a relational model outside of information processing system 100 and installing it in information processing apparatus 1, etc. Furthermore, the relational model can be configured to be updated before output unit 113 outputs the cause of a defect and / or a countermeasure therefor (before executing the output step). In this case, relational model creation unit 114 can be configured to sequentially update the relational model based on the accumulation of information on countermeasures for the defect, etc.

[0079] The analytical data associated with the relational model does not necessarily have to be data obtained by analyzing the defect itself. For example, in the case of evaluating product defects, the analytical data associated with the relational model does not necessarily have to be data related to the product itself; it can also be data obtained by analyzing the manufacturing equipment or raw materials used. Alternatively, the analytical data is preferably data obtained by analyzing defects in industrial process P as predetermined locations within the process. This latter approach makes it easier to assess defects with higher accuracy.

[0080] The analytical data of this embodiment can be based on any known analytical method. For example, the analytical data according to this embodiment can be data obtained by analyzing a predetermined portion using a chemical or biochemical method. From another perspective, the analytical data of this embodiment can include one or more analytical data selected from the group consisting of: optical microscopy observation, electron microscopy observation, Fourier transform infrared spectroscopy analysis, pyrolysis gas chromatography analysis, ninhydrin reaction test, X-ray fluorescence analysis, X-ray diffraction analysis, bacterial count analysis, fungal count analysis, elemental mapping analysis, iodine starch reaction test, organic solvent-based extraction amount, acid-based extraction amount, and alkali-based extraction amount.

[0081] That is, in this embodiment, the component characteristics associated with the defect can be estimated based on this type of analysis data. Therefore, the cause of the defect or a countermeasure can be accurately output from the image information acquired by the image acquisition unit 111. Note that in the case of the aforementioned papermaking production process (including water systems), the use of this type of analysis data helps improve the accuracy of the output results.

[0082] On the other hand, the image data associated with the relational model may include any of the color, size, and shape of the defect. The output unit 113 is preferably configured to output the cause of the defect and / or a countermeasure for the defect based on any of the color, size, and shape of the defect in the predetermined image.

[0083] As described above, in information processing system 100 of this embodiment, an RGB camera can be used as imaging device 2, and the color, size, and shape of a defect can be associated with the RGB data of the image information captured by the RGB camera. In other words, the combination of RGB data for each pixel can serve as a basis for outputting the cause, etc., in the output step described later.

[0084] However, the information processing method of this embodiment outputs the cause of the defect and / or a countermeasure against the defect based on the image acquired by the image acquisition unit 111 and the relational model acquired by the relational model acquisition unit 112 (step S3 ).

[0085] This step can be achieved by the function of the output unit 113. Although the output unit 113 may output only one of the cause of the defect and the countermeasure for the defect, it is preferable that the output unit 113 outputs both the cause of the defect and the countermeasure for the defect.

[0086] The following will refer to Figure 5 An example of content output by the output unit 113 will be described. Figure 5 11 is a diagram showing an example of output content displayed by the output unit 113 .

[0087] Figure 5 The display screen D in FIG. 1 is visual information output by the output unit 113 of the information processing device 1 and is typically displayed as a screen on the display unit 14 of the information processing device 1. The display screen D displays an image related to a defect acquired by the image acquisition unit 111, as well as the type of defect (fault type), cause, and countermeasures associated with the image related to the defect.

[0088] "Defect Type" refers to the broad classification of defects (faults), and more detailed causes are shown in the "Cause" item. Figure 5 Among the output contents shown, the operating conditions of the industrial process P are output as the countermeasures against the defects (refer to Figure 5 in the “Countermeasures” section).

[0089] Note that if the industrial process P is a paper production process, the operating conditions as countermeasures are listed below.

[0090] (Countermeasure A) Add water treatment chemicals, change the type of chemicals, adjust the addition amount, change the addition location, or suspend addition.

[0091] Examples of the water treatment chemicals described herein include: slime control agents, corrosion inhibitors, pitch control agents, defoaming agents, scale inhibitors, process cleaning agents (acidic agents, alkaline agents, chelating agents, etc.), retention and drainage aids (including organic and inorganic types), flocculants (including organic and inorganic types), coagulants (including organic and inorganic types), charge regulators, deaerators, cleaning aids, felt conditioners, paper machine roll stain inhibitors, paper machine clothing stain inhibitors, wrinkle agents, evaporation aids, deinking agents or bleaching agents (surfactants, sodium hydroxide, sodium hypochlorite, chlorine dioxide, ozone, hydrogen peroxide, chelating agents, sodium silicate, etc.), beating aids, release agents, etc.

[0092] (Countermeasure B) Add functional chemicals that affect water treatment, change the type of chemicals, adjust the addition amount, change the addition location, or suspend addition.

[0093] Examples of the functional chemicals described herein include: paper strengthening agents (including dry strength agents and wet strength agents), sizing agents, dyes, adhesives, fillers (such as calcium carbonate and kaolin), and the like.

[0094] (C) Use of water treatment equipment, replacement of equipment type, change of operation mode, adjustment of operation load, or suspension of operation

[0095] Examples of the water treatment equipment described herein include pressurized flotation equipment, coagulation sedimentation devices, strainers, filters, and washing equipment such as dewaterers, and foreign matter removal devices such as screens, cleaners, and floaters. In addition, the change of the operation mode includes a change in the amount of make-up water or the type of make-up water of the system, a change in the amount of process water discharged from the system, and the like.

[0096] (D) Use of production equipment that affects water treatment, replacement of equipment type, change of operation mode, adjustment of operation load, or suspension of operation

[0097] Examples of the production equipment described herein include pulp liberators, dispersers, beaters, paper machines, steam equipment, bleaching equipment, and the like. In addition, the change of the operation mode includes a change in the type or mixing ratio of raw materials, a production order of production items, a timing of regular maintenance, a timing of internal cleaning of the system, and the like.

[0098] Of course, the content of the countermeasures is not limited to the above-described examples, and can be appropriately set according to the type and scale of the industrial process P and the like.

[0099] In addition, the content output by the output section 113 is not limited to such a display screen, and the cause or the countermeasures can be output in the form of a printed matter, for example. In addition, as part of the countermeasures, the output section 113 can output the operation conditions of the industrial process P as a signal. In this case, part of the equipment provided in the industrial process P can be configured to automatically operate according to the content of the signal.

[0100] The content output by the output section 113 can be stored in a predetermined storage area by the function of the storage management section 115. In addition, the relational model can be updated according to the output content stored in this manner. Such updating of the relational model can be achieved by the function of the relational model creation section 114.

[0101] As described above, the information processing method executed by the information processing apparatus 1 according to the present embodiment can easily determine the cause of defects and the like.

[0102] 4. Modification

[0103] In the fourth section, a modification of the information processing method executed by the above-described information processing apparatus 1 and the like will be described.

[0104] Although the above-described embodiment has been described as the structure of the information processing apparatus 1, a program can be provided to cause a computer to function as each section of the information processing apparatus.

[0105] The above embodiment illustrates an information processing method using a relational model, but the information associated when creating a relational model is not limited thereto. Specifically, the relational model used in this embodiment can be associated with various other conditions, such as weather conditions, conditions related to the region, and conditions related to the age of the device.

[0106] In the above embodiment, although the information processing device 1 performs various storage and control functions, two or more external devices may be used instead of the information processing device 1. That is, defect-related images, causes, countermeasures, and other information may be distributed and stored in two or more external devices using blockchain technology or the like.

[0107] (Example)

[0108] Hereinafter, the present invention will be described in further detail with reference to Examples and Comparative Examples. Note that the present invention is not limited to the following Examples.

[0109] (Creation and validation of relational models)

[0110] Various analyses (chemical analysis, etc.) were performed on 23 defect samples from cardboard production equipment to identify the main components of the defects and to infer their causes and water treatment methods as countermeasures. Furthermore, image data of the defects was added to create a database containing 23 samples. Using this database, a relational model was created using teacher learning to infer the causes suggested by the analytical data from the defect image data. The relational model was created using a deep neural network model created using Teachable Machine (Google).

[0111] Next, to verify the accuracy of the relational model, 15 defective samples were prepared in addition to the 23 samples described above and analyzed in the same manner. Applying the relational model to the image data of these 15 samples and estimating the components and causes of the defects showed a high degree of agreement with the actual analysis results. This demonstrates that the information processing device, etc., of the present invention can easily determine the cause of a defect.

[0112] Finally, although various embodiments of the present invention have been described, these are presented only as examples and are not intended to limit the scope of the invention. Other novel embodiments may be implemented in various forms, and various omissions, substitutions, or modifications may be made within the scope of the invention without departing from the spirit of the invention. These embodiments and their variations are intended to be within the scope and spirit of the invention, as well as the invention described in the claims and their equivalents.

[0113] Description of Reference Numerals

[0114] 1: Information processing device,

[0115] 2: Camera device,

[0116] 3: Countermeasure type,

[0117] 4: Countermeasure type,

[0118] 10: Communication bus,

[0119] 11: Control Department,

[0120] 12: Storage Department,

[0121] 13: Input part,

[0122] 14: Display unit,

[0123] 15: Ministry of Communications,

[0124] 100: Information processing systems,

[0125] 111: Image acquisition unit,

[0126] 112: Relational model acquisition department,

[0127] 113: Output unit,

[0128] 114: Relational Model Creation Department,

[0129] 115: Storage Management Department,

[0130] D: Display screen,

[0131] P: Industrial process.

Claims

1. An information processing device, comprising: an image acquisition unit configured to acquire predetermined images from an industrial process, wherein the predetermined images include images related to defects generated in the industrial process; a relational model acquisition unit configured to acquire a relational model created by associating analysis data obtained by analyzing a predetermined portion in the industrial process with image data related to an image of a corresponding defect; and An output unit is configured to output a cause of the defect and / or a countermeasure for the defect based on the predetermined image and the relational model.

2. The information processing device according to claim 1, wherein The output unit is configured to output a cause of the defect and a countermeasure for the defect.

3. The information processing device according to claim 1 or 2, wherein: The analysis data is data obtained by analyzing the predetermined part by a chemical or biochemical method.

4. The information processing device according to claim 1 or 2, wherein: The analytical data comprises one or more analytical data selected from the group consisting of: optical microscopy observation, electron microscopy observation, Fourier transform infrared spectroscopy analysis, pyrolysis gas chromatography analysis, ninhydrin reaction test, X-ray fluorescence analysis, X-ray diffraction analysis, bacterial count analysis, fungal count analysis, elemental mapping analysis, iodine starch reaction test, organic solvent-based extraction amount, acid-based extraction amount, and alkali-based extraction amount.

5. The information processing device according to any one of claims 1 to 4, wherein: The analysis data is data obtained by analyzing the defect in the industrial process as a predetermined location in the industrial process.

6. The information processing device according to any one of claims 1 to 5, wherein: The industrial process is a process including a water-based step.

7. The information processing device according to any one of claims 1 to 6, wherein: The industrial process is a papermaking process.

8. The information processing device according to any one of claims 1 to 7, wherein: The image data associated with the relational model includes any one of information on color, size, and shape of the defect; and The output unit is configured to output a cause of the defect and / or a countermeasure for the defect based on any one of information on color, size, and shape of the defect in the predetermined image.

9. The information processing device according to any one of claims 1 to 8, wherein: The predetermined image is an image captured by an RGB (red, green, blue) camera.

10. The information processing apparatus according to any one of claims 1 to 9, wherein: The output unit is configured to output an operating condition of the industrial process as a countermeasure against the defect.

11. The information processing apparatus according to any one of claims 1 to 10, wherein: The relational model is a model configured to be updated before the output unit outputs the cause of the defect and / or the countermeasure for the defect.

12. An information processing method performed by an information processing device, comprising: An image acquisition step for acquiring predetermined images from an industrial process, wherein the predetermined images include images related to defects generated in the industrial process; a relational model acquisition step for acquiring a relational model, wherein the relational model is a model created by associating analysis data obtained by analyzing a predetermined portion in the industrial process with image data related to an image of a corresponding defect; as well as An output step for outputting a cause of the defect and / or a countermeasure for the defect based on the predetermined image and the relational model.

13. A program that causes a computer to function as: an image acquisition unit configured to acquire predetermined images from an industrial process, wherein the predetermined images include images related to defects generated in the industrial process; a relational model acquisition unit configured to acquire a relational model created by associating analysis data obtained by analyzing a predetermined portion in the industrial process with image data related to an image of a corresponding defect; and An output unit is configured to output a cause of the defect and / or a countermeasure for the defect based on the predetermined image and the relational model.

Images