Data processing system, billing determination system, data processing device, program, data processing method

Abstract

Provided are a data processing system, a billing determination system, a data processing device, a program, and a data processing method that enable more accurate identification of the number of sensing units that are connected. The data processing system is provided with: a plurality of sensing units that each generate sensing data; a relay unit for receiving and transmitting a plurality of sensing data items generated by the plurality of sensing units; a processing unit for executing predetermined processing on the plurality of sensing data items transmitted from the relay unit; and a first determination unit for performing a predetermined analysis on information pertaining to the plurality of sensing data items transmitted from the relay unit and determining the number of sensing units included in the plurality of sensing units on the basis of a result of the analysis.

Description

Data processing system, billing judgment system, data processing device, program, and data processing method

[0001] The present invention relates to a data processing system, a billing determination system, a data processing device, a program, and a data processing method.

[0002] Conventionally, there has been a service that provides users with application programs used in data processing systems in the form of containers (see, for example, Patent Literature 1). For example, by building a data processing system that processes sensing data generated by multiple sensing units such as cameras using an application program in a container, it is possible to realize the Internet of Things (IoT). The container provider charges users according to the number of connected sensing units and the amount of container usage.

[0003] Japanese Patent Application Laid-Open No. 2022-142986

[0004] In some data processing systems, a relay unit such as an intermediate server is interposed between multiple sensing units and a processing unit that performs processing related to the container. The relay unit receives sensing data generated by the multiple sensing units and transmits the sensing data to a downstream processing unit. In this configuration, the relay unit may lose identification information indicating the sensing unit that generated the sensing data from the multiple sensing data. This poses a problem in that it is not possible to accurately identify the number of connected sensing units.

[0005] An object of the present invention is to provide a data processing system, a billing determination system, a data processing device, a program, and a data processing method that are capable of more accurately specifying the number of connected sensing units.

[0006] In order to achieve the above object, the invention of the data processing system described in claim 1 is characterized by comprising: a plurality of sensing units that each generate sensing data; a relay unit that receives and transmits the plurality of sensing data generated by the plurality of sensing units; a processing unit that performs a predetermined process on the plurality of sensing data transmitted from the relay unit; and a first judgment unit that performs a predetermined analysis on information regarding the plurality of sensing data transmitted from the relay unit and judges the number of sensing units included in the plurality of sensing units based on the results of the analysis.

[0007] The invention described in claim 2 is characterized in that, in the data processing system described in claim 1, each of the plurality of sensing data includes time series data, and the first determination unit determines the number of sensing units based on the mutual continuity or amount of change of the plurality of sensing data.

[0008] According to a third aspect of the present invention, in the data processing system of the first aspect, the plurality of sensing units are devices of the same type that generate the sensing data of the same type.

[0009] The invention described in claim 4 is characterized in that, in the data processing system described in claim 1, each of the plurality of sensing units is a camera, each of the plurality of sensing data is image data captured by the camera, and the first judgment unit performs image analysis of the image data as the predetermined analysis.

[0010] The invention described in claim 5 is characterized in that, in the data processing system described in claim 1, each of the plurality of sensing units is a microphone, each of the plurality of sensing data is audio data recorded by the microphone, and the first judgment unit performs audio analysis of the audio data as the predetermined analysis.

[0011] The invention described in claim 6 is characterized in that, in the data processing system described in claim 1, the first judgment unit, in the specified analysis, identifies two or more pieces of sensing data that satisfy a specified approximation condition among the plurality of sensing data by performing pattern analysis on the plurality of sensing data, and judges the number of sensing units by a method of determining that the two or more pieces of sensing data that satisfy the approximation condition were generated by one of the sensing units.

[0012] The invention described in claim 7 is characterized in that, in the data processing system described in claim 6, each of the plurality of sensing data includes target data relating to a specified sensing target and background data other than the target data, and the first judgment unit, in the specified analysis, identifies the two or more sensing data among the plurality of sensing data whose background data mutually satisfy the approximation condition.

[0013] The invention described in claim 8 is characterized in that, in the data processing system described in claim 1, it is provided with a notification control unit that performs processing to notify a user operating the processing unit of information regarding the determination result of the number of sensing units by the first determination unit.

[0014] The invention described in claim 9 is characterized in that, in the data processing system described in claim 8, the information regarding the judgment result includes at least one of the number of the sensing units, the response speed of the processing unit depending on the number of the sensing units, the number of faulty sensing units among the plurality of sensing units identified depending on the number of the sensing units, and the amount charged to the user based on the number of the sensing units.

[0015] The invention described in claim 10 is characterized in that, in the data processing system described in claim 1, the predetermined processing executed by the processing unit includes processing of inputting the sensing data into a learning model that has been machine-learned in advance, thereby obtaining an output result from the learning model.

[0016] The invention described in claim 11 is characterized in that, in the data processing system described in claim 1, it comprises: a memory unit that stores contract information regarding a contract with a user who operates the processing unit, which contract specifies the maximum number of sensing units that can be connected to the processing unit via the relay unit; and a second judgment unit that judges whether there is a breach of contract regarding the contract based on the contract information and the judgment result of the first judgment unit regarding the number of sensing units.

[0017] In addition, in order to achieve the above-mentioned object, the invention of the billing judgment system described in claim 12 is characterized by comprising: a data processing system described in any one of claims 1 to 11; and a third judgment unit that determines the amount to be charged to the user operating the processing unit based on the result of the judgment of the number of the sensing units by the first judgment unit.

[0018] In addition, in order to achieve the above object, the invention of a data processing device described in claim 13 is an information processing device connected to a relay unit that receives and transmits multiple sensing data generated by multiple sensing units, characterized in that it comprises: a processing unit that performs a predetermined processing on the multiple sensing data transmitted from the relay unit; and a first judgment unit that performs a predetermined analysis on information related to the multiple sensing data transmitted from the relay unit and judges the number of sensing units included in the multiple sensing units based on the results of the analysis.

[0019] Furthermore, in order to achieve the above-mentioned object, the invention of the program described in claim 14 is characterized in that a computer provided in an information processing device connected to a relay unit that receives and transmits multiple sensing data generated by multiple sensing units functions as: a processing means that executes a predetermined process on the multiple sensing data transmitted from the relay unit; and a first determination means that performs a predetermined analysis on information related to the multiple sensing data transmitted from the relay unit and determines the number of sensing units included in the multiple sensing units based on the results of the analysis.

[0020] Furthermore, in order to achieve the above object, the invention of the data processing method described in claim 15 is a data processing method executed by a computer, characterized in that it includes: a data generation step in which a plurality of sensing units each generate sensing data; a relay step in which a relay unit receives and transmits the plurality of sensing data generated in the data generation step; a processing step in which a predetermined process is performed on the plurality of sensing data transmitted in the relay step; and a first determination step in which a predetermined analysis is performed on information related to the plurality of sensing data transmitted in the relay step, and the number of sensing units included in the plurality of sensing units is determined based on the results of the analysis.

[0021] According to the present invention, the number of connected sensing units can be more accurately identified.

[0022] 1 is a diagram showing a schematic configuration of a data processing system. FIG. 2 is a block diagram showing the main functional configuration of an AI server. FIG. 3 is a block diagram showing the main functional configuration of an intermediate server. FIG. 4 is a block diagram showing the main functional configuration of a management server. FIG. 5 is a diagram showing an example of the contents of container management data. FIG. 6 is a diagram showing an example of the contents of user management data. FIG. 7 is a diagram showing an example of the contents of contract management data. FIG. 8 is a block diagram showing the main functional configuration of a user terminal. FIG. 9 is a diagram explaining a method for identifying an image pattern of image data. FIG. 10 is a diagram showing an example of the contents of image pattern data. FIG. 11 is a diagram showing an example of the contents of analysis result data. FIG. 12 is a diagram explaining another method for identifying an image pattern of image data. A flowchart showing the control procedure of a setup process. A flowchart showing the control procedure of an AI analysis management process. A flowchart showing the control procedure of a connection number analysis process. A flowchart showing the control procedure of a handling process.

[0023] Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

[0024] (Overview of Data Processing System) FIG. 1 is a diagram showing the schematic configuration of a data processing system 1. The data processing system 1 (billing determination system) includes an AI server 10 (data processing device), an intermediate server 20 (relay unit), a display unit 30, multiple cameras 40a-40c (multiple sensing units), a management server 50, and a user terminal 60. The intermediate server 20 is communicatively connected to the AI ​​server 10, the display unit 30, and the cameras 40a-40c. The communication paths between the intermediate server 20 and the AI ​​server 10, the display unit 30, and the cameras 40a-40c may be via a network N. The AI ​​server 10, the management server 50, and the user terminal 60 are communicatively connected to one another via the network N. The network N may be, for example, the Internet, but is not limited to this. Hereinafter, any one of the cameras 40a-40c will be referred to as "camera 40."

[0025] The AI ​​server 10, intermediate server 20, display unit 30, and cameras 40a to 40c are installed in an area where an IoT service using the cameras 40a to 40c is implemented. The area may be, for example, a facility such as a factory or warehouse, but is not limited to these. In this embodiment, an IoT service is described as an example in which the cameras 40a to 40c installed in a factory monitor the actions of workers working in the factory.

[0026] The multiple image data (multiple sensing data) captured and generated by the cameras 40a to 40c are transmitted to the intermediate server 20. In this embodiment, each image data is assumed to be a video. Therefore, each image data includes time-series data. However, each image data may also be a still image. The intermediate server 20 displays the video related to the multiple image data received on the display unit 30. The video displayed on the display unit 30 is monitored by a monitor or the like.

[0027] The intermediate server 20 transmits the received image data to the AI ​​server 10. The AI ​​server 10 includes an AI connection module 101 and an AI container 102. The AI ​​connection module 101 and the AI ​​container 102 include software and / or middleware executed by the CPU 11 (Central Processing Unit) of the AI ​​server 10 shown in FIG. 2 . Hereinafter, for convenience, "the CPU 11 performs operation m by executing the AI ​​connection module 101 (or the AI ​​container 102)" may be referred to as "the AI ​​connection module 101 (or the AI ​​container 102) performs operation m." The AI ​​connection module 101 transfers the received image data to the AI ​​container 102. The AI ​​connection module 101 also analyzes the received image data to determine the number of cameras 40a to 40c. The AI ​​connection module 101 transmits the determination result to the management server 50. The method by which the AI ​​connection module 101 determines the number of cameras will be described in detail later.

[0028] The AI ​​container 102 includes a learning model that has been machine-learned in advance to perform predetermined image recognition processing on image data. The AI ​​server 10 uses the AI ​​container 102 to perform image recognition processing (predetermined processing) on ​​multiple pieces of image data and transmits the processing results to the intermediate server 20. In this embodiment, for example, image recognition processing is performed to extract workers from videos related to the image data and identify the work content of the workers. The intermediate server 20 displays information related to the work content of the workers on the display unit 30 based on the received processing results. However, the content of the processing performed by the AI ​​container 102 is not limited to the above.

[0029] The management server 50 is a server device managed by a service provider that provides the AI ​​connection module 101 and the AI ​​container 102. The user terminal 60 is a terminal device used by a user who operates the AI ​​server 10, the intermediate server 20, the display unit 30, and the cameras 40a to 40c. The user may be, for example, an employee of a company that operates a factory. The management server 50 deploys the AI ​​connection module 101, the AI ​​container 102, and the like to the AI ​​server 10 in response to a request from the user. The management server 50 manages contract information related to a contract with the user for a service that provides the AI ​​container 102. This contract specifies the type of AI connection module 101 used by the user, the number of cameras 40a to 40c the user can use, and the amount charged to the user based on that number.

[0030] (Configuration of Data Processing System) Next, the configuration of each device in the data processing system 1 will be described. Fig. 2 is a block diagram showing the main functional configuration of the AI ​​server 10. The AI ​​server 10 includes a CPU 11, a RAM 42 (Random Access Memory), a storage unit 13, and a communication unit 14. Each unit of the AI ​​server 10 is connected via a communication path such as a bus.

[0031] The CPU 11 is a processor that controls the operation of each unit of the AI ​​server 10 by executing various processes in accordance with the program 131 stored in the storage unit 13. By executing various processes in accordance with the program 131, the CPU 11 functions as a processing unit and a first determination unit. For example, the CPU 11 as a processing unit performs predetermined processing on multiple image data 132 transmitted from the intermediate server 20. This predetermined processing includes inputting the image data 132 into the learning model of the AI ​​container 102 and acquiring output results related to image recognition processing from the AI ​​container 102. Furthermore, the CPU 11 as a first determination unit executes the AI ​​connection module 101 to perform predetermined analysis on the multiple image data 132 and determine the number of cameras 40a to 40c based on the analysis results. In addition to the CPU 11, the AI ​​server 10 may also include an AI accelerator. The AI ​​accelerator is a dedicated processor for high-speed inference processing in the neural network of the learning model of the AI ​​container 102. When an AI accelerator is provided, the CPU 11 and the AI ​​accelerator constitute the processing unit.

[0032] The RAM 12 provides a working memory space for the CPU 11 and stores temporary data.

[0033] The storage unit 13 is a non-transitory recording medium readable by the CPU 11 as a computer. The storage unit 13 is configured with a hard disk drive (HDD) or a solid state drive (SSD), etc. A program 131 is stored in the storage unit 13. The program 131 includes a client application 1311, an AI connection module 101, and an AI container 102. The client application 1311 is an application program for causing the AI ​​server 10 to perform various operations related to IoT services. The client application 1311 is downloaded from the management server 50 and installed in the storage unit 13.

[0034] The AI ​​container 102 includes the above-described learning model. The AI ​​container 102 is an application environment built on the host OS of the AI ​​server 10 using various data and middleware for utilizing this learning model. Specifically, the AI ​​container 102 is managed by a container engine (not shown) running on the host OS. Therefore, when using the AI ​​container 102, there is no need to install a guest OS on the host OS. Therefore, compared to conventional virtualization technology using a guest OS, hardware resources such as the CPU 11 can be used more efficiently. Furthermore, the AI ​​container 102 has the advantage of being able to be deployed on any device on which the container engine runs. The AI ​​connection module 101 may also be in the form of a container.

[0035] In addition to the program 131, the storage unit 13 stores image data 132, image pattern data 133, analysis result data 134, etc. The image data 132 is data of moving images generated by the cameras 40a to 40c. The storage unit 13 stores a plurality of image data 132. The contents of the image pattern data 133 and the analysis result data 134 will be described later.

[0036] The communication unit 14 performs communication operations in accordance with a predetermined communication standard. Through this communication operation, the communication unit 14 transmits and receives data to and from the intermediate server 20, the management server 50, and the user terminal 60.

[0037] 3 is a block diagram showing the main functional configuration of the intermediate server 20. The intermediate server 20 includes a CPU 21, a RAM 22, a storage unit 23, and a communication unit 24. The components of the intermediate server 20 are connected to each other via a communication path such as a bus.

[0038] The CPU 21 is a processor that controls the operation of each part of the intermediate server 20 by executing various processes in accordance with the program 231 stored in the storage unit 23. For example, the CPU 21 transmits control signals to the cameras 40a to 40c to cause them to take pictures and receives image data 132. The CPU 21 transmits the image data 132 and a control signal to the display unit 30 to display an image related to the image data 132. The CPU 21 transmits the received image data 132 to the AI ​​server 10.

[0039] The RAM 22 provides a working memory space for the CPU 21 and stores temporary data.

[0040] The storage unit 23 is a non-transitory recording medium that can be read by the CPU 21 as a computer. The storage unit 23 is configured with an HDD, an SSD, etc. The storage unit 23 stores a program 231 and various data required for executing the program 231.

[0041] The communication unit 24 performs communication operations in accordance with a predetermined communication standard, and through this communication operation, the communication unit 24 transmits and receives data between the AI ​​server 10, the display unit 30, and the cameras 40a to 40c.

[0042] 4 is a block diagram showing the main functional configuration of the management server 50. The management server 50 includes a CPU 51, a RAM 52, a storage unit 53, and a communication unit 54. The components of the management server 50 are connected to each other via a communication path such as a bus.

[0043] The CPU 51 is a processor that controls the operation of each unit of the management server 50 by executing various processes in accordance with the program 531 stored in the storage unit 53. By executing various processes in accordance with the program 531, the CPU 51 functions as a notification control unit, a second determination unit, and a third determination unit. For example, the CPU 51 as a notification control unit performs processing to notify the user operating the AI ​​server 10 of information regarding the determination result of the number of cameras 40a to 40c received from the AI ​​server 10. The CPU 51 as a second determination unit determines whether there is a breach of contract based on the contract management data 534 (contract information) and the determination result of the number of cameras 40a to 40c. The CPU 51 as a third determination unit determines the amount to be charged to the user based on the determination result of the number of cameras 40a to 40c.

[0044] The RAM 52 provides a working memory space for the CPU 51 and stores temporary data.

[0045] The storage unit 53 is a non-transitory recording medium that can be read by the CPU 51 as a computer. The storage unit 53 is configured with an HDD, an SSD, etc. The storage unit 53 stores a program 231, container management data 532, user management data 533, contract management data 534, etc.

[0046] FIG. 5 is a diagram showing an example of the contents of the container management data 532. The container management data 532 includes information on multiple different types of AI containers that can be provided by the management server 50. FIG. 5 illustrates information on three AI containers: "Container A," "Container B," and "Container C." Of these, "Container A" is assumed to be the AI ​​container 102 deployed on the AI ​​server 10 of this embodiment. The container management data 532 includes information on the "recommended number of connections" and "connection device" for each AI container.

[0047] The "recommended number of connections" indicates the maximum number of cameras 40 that can be connected to an AI container to achieve the intended function. In other words, if more cameras 40 than the "recommended number of connections" are connected to an AI container, the performance of the AI ​​container, such as its response speed, will be lower than the intended level. Here, the number of connected cameras 40 refers to the number of cameras 40 that directly or indirectly supply image data 132 to the AI ​​container. The "connected device" indicates the type of camera 40 that can be connected to the AI ​​container. In the example shown in FIG. 5, "camera X" can be connected to "container A" and "container B." "camera Y" can be connected to "container C." "Camera X" and "camera Y" differ from each other, for example, in terms of image resolution, etc.

[0048] FIG. 6 is a diagram showing an example of the contents of the user management data 533. The user management data 533 includes information about multiple users who receive the AI ​​container. In FIG. 6, information about "User U1" and "User U2" is shown as an example. Of these, "User U1" is assumed to be the user who receives the AI ​​container 102 of this embodiment. The user management data 533 includes information about the "server used," "container used," "number of connections," "connected device," and "contract" for each user.

[0049] "Server in use" refers to the AI ​​server being used (operated) by the user. "AI server S1" corresponding to user U1 in Figure 6 corresponds to the AI ​​server 10 in this embodiment. "Container in use" refers to the type of AI container being provided to the user. "Container A" corresponding to user U1 in Figure 6 corresponds to the AI ​​container 102 in this embodiment. "Number of connections" refers to the number of cameras 40 connected to the AI ​​container being used (operated) by the user. Information on this number of connections is sent from the AI ​​server 10 to the management server 50. "Connected device" refers to the type of camera 40 connected to the AI ​​container being used by the user. "Contract" refers to a contract regarding the provision of AI container services, concluded with the user as one of the parties. The contents of the contract are registered in the contract management data 534.

[0050] FIG. 7 is a diagram showing an example of the contents of the contract management data 534. The contract management data 534 includes information regarding the contents of the contract concluded with each user. In FIG. 7, information on "Contract Z1" and "Contract Z2" is shown as examples. Of these, "Contract Z1" is assumed to be the contract concluded with the user who is provided with the AI ​​container 102 of this embodiment. The contract management data 534 includes information on "container used," "maximum number of connections," and "charge amount" for each contract.

[0051] The "used container" indicates the type of AI container used (operated) by the user. The "maximum number of connections" indicates the maximum number of cameras 40 that can be connected to the AI ​​container and its processing unit. This maximum number of connections is the same as the "recommended number of connections" of the corresponding AI container in the container management data 532. However, the maximum number of connections may be adjusted according to the specifications of the AI ​​server 10 and / or the specifications of the cameras 40 operated by the contracting user. In the example of FIG. 7 , contract Z1 specifies the maximum number of connections as "10." Therefore, the user of this embodiment can connect up to 10 cameras 40 to the AI ​​container 102 via the intermediate server 20. The "charge amount" indicates the amount charged to the user, determined according to the type of "used container" and the "maximum number of connections," etc.

[0052] 4, the communication unit 54 performs communication operations in accordance with a predetermined communication standard. Through this communication operation, the communication unit 54 transmits and receives data between the AI ​​server 10 and the user terminal 60.

[0053] 8 is a block diagram showing the main functional configuration of the user terminal 60. The user terminal 60 includes a CPU 61, a RAM 62, a storage unit 63, a display unit 64, an operation unit 65, and a communication unit 66. The components of the user terminal 60 are connected via a communication path such as a bus. The user terminal 60 may be, for example, a tablet terminal, a notebook PC, or a smartphone.

[0054] The CPU 61 is a processor that controls the operation of each part of the user terminal 60 by executing various processes in accordance with a program 631 stored in the storage unit 63 .

[0055] The RAM 62 provides a working memory space for the CPU 61 and stores temporary data.

[0056] The storage unit 63 is a non-transitory recording medium that can be read by the CPU 61 as a computer. The storage unit 63 is configured with a flash memory, an HDD, an SSD, etc. The storage unit 63 stores a program 631 and various data necessary for executing the program 631.

[0057] The display unit 64 includes a display device such as a liquid crystal display, etc. The display unit 64 displays various information display screens, operation screens, etc. in accordance with control signals and image data input from the CPU 61.

[0058] The operation unit 65 includes operation buttons operated by the user, and a touch panel superimposed on the screen of the display unit 64. When an operation is performed on the operation buttons or the touch panel, the operation unit 65 outputs an operation signal corresponding to the operation to the CPU 61.

[0059] The communication unit 66 performs communication operations in accordance with a predetermined communication standard. Through this communication operation, the communication unit 66 transmits and receives data between the AI ​​server 10 and the management server 50.

[0060] The display unit 30 shown in FIG. 1 displays images captured by the cameras 40a-40c based on image data 132 transmitted from the intermediate server 20. The display unit 30 may include monitors in a number corresponding to the number of cameras 40a-40c. The display unit 30 may be capable of simultaneously displaying multiple images of image data 132 captured by the multiple cameras 40a-40c on these multiple monitors. The display unit 30 may receive information related to the analysis results by the AI ​​container 102 via the intermediate server 20 and display the information together with the image related to the image data 132. For example, the display unit 30 may display the work content being performed by the worker together with the image related to the image data 132.

[0061] The cameras 40a to 40c operate in a local environment where the AI ​​server 10 and intermediate server 20 are installed. The cameras 40a to 40c are each installed at a predetermined position in the factory at a predetermined angle. The cameras 40a to 40c capture video at a predetermined frame rate, generate image data 132, and send it to the intermediate server 20. For example, the cameras 40a to 40c generate image data 132 and send it to the intermediate server 20 each time they capture video for a predetermined period of time.

[0062] (Operation of Data Processing System) Next, the operation of the data processing system 1 will be described in detail. As described above, in the data processing system 1 of this embodiment, the image data 132 generated by the cameras 40a to 40c is transmitted to the AI ​​server 10 via the intermediate server 20. In this configuration, the image data 132 transmitted from the intermediate server 20 to the AI ​​server 10 may not include identification information for identifying which camera 40 captured the image. For example, even if the image data 132 generated by the camera 40 includes metadata including the identification information of the camera 40, the metadata may be deleted by the intermediate server 20. In this case, the AI ​​server 10 cannot identify which camera 40 captured the image from the image data 132 received from the intermediate server 20. Therefore, even if an over-connection state occurs in which more cameras 40 than the number specified in the contract are connected to the AI ​​container 102, the management server 50, which acquires information from the AI ​​server 10, cannot determine that the over-connection state exists.

[0063] Therefore, in the data processing system 1 of this embodiment, the AI ​​connection module 101 installed in the AI ​​server 10 analyzes multiple pieces of image data 132 to determine the number of connected cameras 40. As the above analysis, the AI ​​connection module 101 performs image analysis of the image data 132. Specifically, the AI ​​connection module 101 performs pattern analysis on the multiple pieces of image data 132 to identify two or more pieces of image data 132 that satisfy predetermined approximation conditions among the multiple pieces of image data 132. The AI ​​connection module 101 determines the number of cameras 40 by a method of determining that two or more pieces of image data 132 that satisfy the approximation conditions were generated by one camera 40.

[0064] 9 is a diagram illustrating a method for image analysis of image data. Here, it is assumed that image patterns P1 and P2 shown on the left side of FIG. 9 are known as image patterns that can be captured by the camera 40 of the data processing system 1. Image patterns P1 and P2 correspond to background images captured at different locations within the factory. These image patterns P1 and P2 are background images extracted from image data 132 that has been received from the intermediate server 20 and analyzed.

[0065] The center of FIG. 9 shows the most recent image data 132 received from the intermediate server 20. The file name of this image data 132 is "hij.jpg." This image data 132 includes target data 1321 corresponding to a worker as a predetermined sensing target, and background data 1322 (background image) other than the target data 1321. The AI ​​connection module 101 identifies the area of ​​the target data 1321 in the image data 132 using various known image recognition processes and deletes the target data 1321 from the image data 132. The image recognition process for identifying the area of ​​the target data 1321 may be, for example, a process for identifying a moving portion of a video as the area of ​​the target data 1321. Alternatively, the image recognition process may be a process for inputting the image data 132 into a learning model that has been machine-trained to identify areas in the image data 132 where people appear, and identifying the area of ​​the target data 1321 based on the output from the learning model.

[0066] The AI ​​connection module 101 performs pattern analysis on background data 1322, which is obtained by removing the target data 1321 from the image data 132. Alternatively, instead of the background data 1322, complementary background data may be used, which is obtained by complementing the area from which the target data 1321 has been removed so that the area is continuous with the background data 1322. The AI ​​connection module 101 compares the background data 1322 with image pattern P1 and image pattern P2. If the background data 1322 and the image pattern P1 or image pattern P2 satisfy a predetermined approximation condition, the AI ​​connection module 101 determines that the image data 132 corresponds to one of the image pattern P1 and the image pattern P2 that satisfies the approximation condition. The approximation condition may be satisfied if, in a predetermined image pattern matching process, the features of a portion of the background data 1322 that accounts for a predetermined percentage or more (e.g., 80% or more) are common to the features of the image pattern P1 or the image pattern P2. In the example shown in Figure 9, the background data 1322 of the image data 132 satisfies the approximation condition with the image pattern P2. Therefore, the AI ​​connection module 101 determines that the image data 132 corresponds to the image pattern P2. The determination result is shown on the right side of Figure 9. As described above, the image patterns P1 and P2 are extracted from other image data 132 that have already been analyzed. Therefore, the above determination operation corresponds to identifying two or more image data 132 among the multiple image data 132 whose background data 1322 satisfy the approximation condition.

[0067] If the background data 1322 of the image data 132 and the image pattern P1 and image pattern P2 do not satisfy the approximation conditions, the AI ​​connection module 101 registers the background data 1322 as a new image pattern. Furthermore, the AI ​​connection module 101 determines that the image data 132 is an image taken by a third camera 40 different from the camera 40 corresponding to the image pattern P1 and the image pattern P2.

[0068] The image pattern of background data 1322 extracted from image data 132 is registered in image pattern data 133. Fig. 10 is a diagram showing an example of the contents of image pattern data 133. In the example shown in Fig. 10, pattern data D1 to D3 are registered for three image patterns P1 to P3, respectively. Each of pattern data D1 to D3 is background data 1322 extracted from one of the image data 132, or complementary background data obtained by complementing the background data 1322.

[0069] The image pattern determination results for each piece of image data 132 received from the intermediate server 20 are registered in the analysis result data 134. Fig. 11 is a diagram showing an example of the content of the analysis result data 134. In the example shown in Fig. 11, the analysis results for four pieces of image data 132 with file names "abcd.jpg", "efgh.jpg", "hij.jpg", and "klmn.jpg" are registered.

[0070] For the first image data 132 with the file name "abcd.jpg", its background data 1322 is registered as the corresponding image pattern P1 as is. This is because there is no image pattern to compare it with at this point. The background data 1322 of this first image data 132 is also registered as the pattern data D1 of the image pattern P1 in the image pattern data 133 of FIG. 10.

[0071] It is assumed that the background data 1322 of the second image data 132, whose file name is "efgh.jpg," is determined not to satisfy the approximation condition with respect to image pattern P1. As a result, the background data 1322 of the second image data 132 is registered as is as the corresponding image pattern P2. Furthermore, the background data 1322 of this second image data 132 is registered as pattern data D2 of image pattern P2 in the image pattern data 133 of FIG. 10 .

[0072] As for the third image data 132 having the file name "hij.jpg", as explained with reference to Fig. 9, it is determined that the background data 1322 satisfies the approximation condition with respect to image pattern P2. As a result, the third image data 132 is recorded as corresponding to image pattern P2. From this, it is determined that the second image data 132 and the third image data 132 were photographed by the same camera 40.

[0073] It is assumed that the background data 1322 of the fourth image data 132, whose file name is "klmn.jpg," is determined not to satisfy the similarity conditions with either image pattern P1 or P2. As a result, the background data 1322 of the fourth image data 132 is registered as is as the corresponding image pattern P3. Furthermore, the background data 1322 of this fourth image data 132 is registered as the pattern data D3 of image pattern P3 in the image pattern data 133 of FIG. 10.

[0074] This process is repeated, and the number of types of corresponding image patterns that are ultimately registered in the analysis result data 134 is determined to be the number of cameras 40 connected to the intermediate server 20 .

[0075] Note that the image patterns may be generated by taking pictures of the inside of the factory in advance, instead of the background data 1322 extracted from the image data 132. In this case, it is determined whether each image data 132 satisfies the similarity condition with all the image patterns. Furthermore, if the orientation of the camera 40 changes or the camera 40 moves, the corresponding image pattern may be determined by cutting out some still images from a continuous video and comparing them with the image patterns.

[0076] 12 is a diagram illustrating another method for identifying image patterns in image data 132. The identification method in FIG. 12 can be used in cases where video image data 132 generated by cameras 40a to 40c is input to the AI ​​server 10 as a single continuous piece of data. In this case, the AI ​​connection module 101 divides the image data 132 into multiple pieces of image data 132 at predetermined time intervals. The AI ​​connection module 101 then determines the number of types of image patterns, i.e., the number of cameras 40, based on the mutual continuity or amount of change in the background data 1322 of the multiple pieces of image data 132.

[0077] Specifically, as shown in FIG. 12 , it is assumed that the image data 132 is divided into "opq.jpg," "rst.jpg," and "uvw.jpg" from the beginning. The AI ​​connection module 101 evaluates the continuity or amount of change between the background data 1322 of the first "opq.jpg" and the second "rst.jpg" (e.g., the first frame image data). For example, the background data 1322 of the two divided image data 132 may be divided into minute regions, and pattern matching processing may be performed on corresponding minute regions, and the number of minute regions with common features may be used as an evaluation value for continuity. Alternatively, the number of minute regions with no common features may be used as an evaluation value for change. The first image data 132 and the second image data 132 in FIG. 12 are assumed to have an evaluation value for continuity of the background data 1322 that is lower than a first reference value, or an evaluation value for change that is greater than a second reference value. In this case, the AI ​​connection module 101 determines that the first image data 132 and the second image data 132 have different image patterns, i.e., were taken by different cameras 40. Also, the second image data 132 and the third image data 132 in FIG. 12 have a continuity of background data 1322 that is higher than a first reference value, or the amount of change is smaller than a second reference value. In this case, the AI ​​connection module 101 determines that the second image data 132 and the third image data 132 have the same image pattern, i.e., were taken by the same camera 40.

[0078] If it is determined that the image data 132 was taken by a different camera 40 than the immediately preceding image data 132, the AI ​​connection module 101 evaluates the continuity or amount of change between any other image data 132 that has already been determined and the background data 1322. If there is image data 132 of background data 1322 whose continuity is higher than a first reference value or whose amount of change is smaller than a second reference value, the AI ​​connection module 101 determines that these image data 132 were taken by the same camera 40. This method also makes it possible to determine the number of connected cameras 40 based on the number of ultimately registered image patterns. The continuity of the background data 1322 being higher than the first reference value or the amount of change being smaller than the second reference value corresponds to satisfying the above-mentioned approximation condition.

[0079] Next, a description will be given of the control procedures for the setup process and the AI ​​analysis management process that are executed in the data processing system 1 to realize the above operations.

[0080] Fig. 13 is a flowchart showing the control procedure of the setup process. Fig. 13 shows the processes executed by the CPU 11 of the AI ​​server 10, the CPU 51 of the management server 50, and the CPU 61 of the user terminal 60. In Fig. 13, the process flow is depicted with solid lines, and data transmitted and received between devices is depicted with dashed lines.

[0081] When the setup process is started, the CPU 61 of the user terminal 60 sends an account creation request to the management server 50 in response to an input operation by the user (step S101). Upon receiving the account creation request, the CPU 51 of the management server 50 creates an account for the user and creates a record for the user in the user management data 533 (step S102).

[0082] In response to the user's input operation, the CPU 61 of the user terminal 60 sends a license purchase request and an AI server registration request to the management server 50 (step S103). The license purchase request is an application to purchase a license required to connect and use the camera 40 to the AI ​​container 102. The AI ​​server registration request is an application to register the AI ​​server 10 that deploys the AI ​​container 102. Upon receiving the license purchase request and the AI ​​server registration request, the CPU 51 of the management server 50 registers information about the license and the intermediate server 20 in the user management data 533 and the contract management data 534 (step S104). Thereafter, the user remits the fee amount specified in the contract to the service provider using a predetermined method.

[0083] The CPU 51 of the management server 50 transmits the client application 1311 and the AI ​​connection module 101 to the AI ​​server 10 (step S105). The CPU 11 of the AI ​​server 10 installs and starts the received client application 1311 and the AI ​​connection module 101 (step S106). The CPU 51 of the management server 50 transmits the licensed AI container 102 to the AI ​​server 10 (step S107). The CPU 11 of the AI ​​server 10 installs the received AI container 102 (step S108).

[0084] When step S108 is completed, each device completes the setup process. After this, the intermediate server 20 sets the AI ​​server 10 as the destination of image data 132 from each connected camera 40, although this is omitted in FIG.

[0085] When the setup process is completed, the AI ​​analysis management process is executed. Figure 14 is a flowchart showing the control procedure for the AI ​​analysis management process. Figure 14 shows the processes executed by the camera 40, the CPU 21 of the intermediate server 20, the CPU 11 of the AI ​​server 10, and the CPU 51 of the management server 50. In Figure 14, the process flow is depicted with solid lines, and data transmitted and received between devices is depicted with dashed lines.

[0086] When the AI ​​analysis management process is started, each camera 40 captures video to generate image data 132 and transmits the image data 132 to the intermediate server 20 at a predetermined timing (step S201: data generation step). The CPU 21 of the intermediate server 20 displays an image of the received image data 132 on the display unit 30 (step S202). The CPU 21 also transmits the received image data 132 to the AI ​​server 10 (step S203: relay step). Although omitted in FIG. 14 , steps S201 to S203 are repeatedly executed during the operating hours of the factory.

[0087] The CPU 11 of the AI ​​server 10 inputs the image data 132 into the AI ​​container 102 and executes image recognition processing (step S204: processing step). That is, the CPU 11 obtains the output of the image recognition processing using the learning model of the AI ​​container 102. The CPU 11 also transmits the result of the image recognition processing to the intermediate server 20 (step S205).

[0088] The CPU 11 of the AI ​​server 10 determines whether it is time to analyze the number of connections of the cameras 40 (step S206). The analysis timing may be set, for example, once a day outside of factory operating hours. This makes it possible to analyze the number of connections of the cameras 40 for each day. If it is determined that it is not time to analyze the number of connections ("NO" in step S206), the CPU 11 returns the process to step S204. If it is determined that it is time to analyze the number of connections ("YES" in step S206), the CPU 11 executes connection number analysis processing (step S207: first determination step).

[0089] 15 is a flowchart showing the control procedure for the connection number analysis process. When the connection number analysis process is called, the CPU 11 of the AI ​​server 10 selects one unanalyzed image data 132 (step S301). The CPU 11 extracts background data 1322 of the selected image data 132 using the method described above (step S302).

[0090] The CPU 11 determines whether one or more image patterns are registered in the image pattern data 133 (step S303). If it determines that one or more image patterns are registered ("YES" in step S303), the CPU 11 determines whether the extracted background data 1322 satisfies the approximation condition with any of the image patterns (step S304). If it determines that the approximation condition with any of the image patterns is satisfied ("YES" in step S304), the CPU 11 associates the selected image data 132 with the image pattern that satisfies the approximation condition and registers them in the analysis result data 134 (step S305).

[0091] On the other hand, if the process branches to "NO" in step S303 or S304, the CPU 11 registers the extracted background data 1322 as pattern data of a new image pattern in the image pattern data 133 (step S306). The CPU 11 associates the selected image data 132 with the registered new image pattern and registers them in the analysis result data 134 (step S307).

[0092] Upon completion of step S305 or S307, the CPU 11 determines whether or not there is unanalyzed image data 132 (step S308). If it is determined that there is unanalyzed image data 132 ("YES" in step S308), the CPU 11 selects the next unanalyzed image data 132 (step S309) and returns the process to step S302. If it is determined that all image data 132 have been analyzed ("NO" in step S308), the CPU 11 proceeds to step S310. In step S310, the CPU 11 determines that the number of types of image patterns associated with the image data 132 in the analysis result data 134 is the number of connections of the cameras 40, and stores the determination result in the RAM 12 or the storage unit 13. Upon completion of step S310, the CPU 11 terminates the connection number analysis process and returns the process to the AI ​​analysis management process of FIG. 14.

[0093] 14 is completed (step S207), the CPU 11 transmits information about the number of connections of the camera 40 to the management server 50 (step S208). The CPU 51 of the management server 50 registers the received information about the number of connections in the user management data 533 (step S209). Thereafter, the CPU 51 executes a handling process for handling the number of connections (step S210).

[0094] 16 is a flowchart showing the control procedure for the handling process. When the handling process is called, the CPU 51 generates determination result information to be notified to the user and adds information on the number of connected cameras 40 to the determination result information (step S401). The CPU 51 references the container management data 532 and determines whether the number of connected cameras 40 is greater than the recommended number of connected cameras (step S402). If it determines that the number of connected cameras 40 is greater than the recommended number of connected cameras (step S402: YES), the CPU 51 adds a warning to the determination result information indicating that the number of connected cameras 40 exceeds the recommended number of connected cameras and information on the response speed corresponding to the number of connected cameras (step S403). The determination result information may also include a warning indicating that the AI ​​container 102 may not operate normally.

[0095] When step S403 is completed, or when it is determined in step S402 that the number of connections is equal to or less than the recommended number of connections ("NO" in step S402), the CPU 51 executes step S404. In step S404, the CPU 11 references the contract management data 534 and determines whether the number of connections is greater than the maximum number of connections stipulated in the contract. When it is determined that the number of connections is greater than the maximum number of connections ("YES" in step S404), the CPU 51 calculates the amount to be charged to the user based on the contents of the contract (step S405). The CPU 51 adds information indicating that the contract is violated and the calculated amount to be charged to the determination result information (step S406).

[0096] When step S406 is completed, or when it is determined in step S404 that the number of connections is equal to or less than the maximum number of connections stipulated in the contract ("NO" in step S404), the CPU 51 executes step S407. In step S407, the CPU 51 determines whether the number of connections has decreased since the previous notification process. When it is determined that the number of connections has decreased ("YES" in step S407), the CPU 51 determines that the decrease in the number of connections is due to a malfunction of a camera 40, and adds information on the number of malfunctioning cameras 40 to the determination result information (step S408).

[0097] When step S408 is completed, or when it is determined in step S407 that the number of connections has not decreased ("NO" in step S407), the CPU 51 transmits determination result information to the user terminal 60 (step S409). The determination result information may be transmitted in the form of an email, a message on a messaging app, or the like. When step S409 is completed, the CPU 51 terminates the handling process and returns the process to the AI ​​analysis management process of FIG. 14. Returning to FIG. 14, when the notification process (step S210) is completed, each device terminates the AI ​​analysis management process.

[0098] (Variation 1) Next, Variation 1 of the above embodiment will be described. Differences from the above embodiment will be described below, and commonalities with the above embodiment will not be described. The AI ​​connection module 101 may be provided outside the AI ​​server 10. For example, the AI ​​connection module 101 may be provided in the management server 50. In this case, an image transfer module is provided in the AI ​​server 10 instead of the AI ​​connection module 101. This image transfer module inputs image data 132 received by the AI ​​server 10 into the AI ​​container 102 and transmits it to the management server 50. The AI ​​connection module 101 of the management server 50 determines the number of connected cameras 40 based on the received image data 132, as in the above embodiment. Therefore, in this case, the CPU 51 of the management server 50 that executes the AI ​​connection module 101 corresponds to the first determination unit.

[0099] (Variation 2) Next, Variation 2 of the above embodiment will be described. Differences from the above embodiment will be described below, and commonalities with the above embodiment will be omitted. Variation 2 may be combined with Variation 1. In the above embodiment, a camera 40 was used as an example of the sensing unit, but the sensing unit may be any device capable of acquiring environmental data as sensing data. For example, a microphone that records sound and generates voice data may be used as the sensing unit. In this case, the AI ​​connection module 101 performs voice analysis on multiple voice data generated by multiple microphones to determine the number of connected microphones. The voice data also includes target data related to the sensing target and background data other than the target data. The target data is, for example, the voice of a person talking, and the background data is, for example, environmental sounds. The AI ​​connection module 101 determines whether the background data of multiple voice data satisfy the approximation condition. The AI ​​connection module 101 determines that two or more voice data whose background data satisfy the approximation condition were generated by the same microphone.

[0100] (Effects) As described above, the data processing system 1 according to this embodiment includes a plurality of cameras 40 serving as a plurality of sensing units, an intermediate server 20 serving as a relay unit, and an AI server 10 having a CPU 11 serving as a processing unit and a first determination unit. Each of the plurality of cameras 40 generates image data 132 as sensing data. The intermediate server 20 receives and transmits the plurality of image data 132 generated by the plurality of cameras 40. The CPU 11 serving as a processing unit performs predetermined processing on the plurality of image data 132 transmitted from the intermediate server 20. The CPU 11 serving as a first determination unit performs predetermined analysis on information related to the plurality of image data 132 transmitted from the intermediate server 20 and determines the number of cameras 40 included in the plurality of cameras 40 based on the results of the analysis. When the intermediate server 20 is involved, identification information of the camera 40 that captured the image may be deleted from the image data 132. In contrast, the data processing system 1 according to this embodiment can more accurately identify the number of connected cameras 40 by analyzing the image data 132. Identifying the number of cameras 40 can be used, for example, for billing purposes or to warn of performance degradation due to excessive connections.

[0101] Each of the plurality of image data 132 includes time-series data. In this case, the CPU 11 as the first determination unit may determine the number of cameras 40 based on the mutual continuity or amount of change of the plurality of image data 132. This makes it possible to determine the number of cameras 40 by a simple process of evaluating the continuity or amount of change of the image data 132. Furthermore, even when the plurality of image data 132 captured by the plurality of cameras 40 is input in a connected state, it is possible to determine the number of cameras 40 that generated the plurality of image data 132 included in the input data.

[0102] The multiple cameras 40 serving as multiple sensing units are devices of the same type that generate image data 132 as the same type of sensing data. This makes it possible to identify the number of sensing units by analyzing and comparing the same type of sensing data.

[0103] As the above analysis, the CPU 11 as the first determination unit performs image analysis of the image data 132. This makes it possible to more accurately identify the number of cameras 40 from the contents of the image data 132.

[0104] Each of the plurality of sensing units may be a microphone. Each of the plurality of sensing data may be audio data recorded by a microphone. The CPU 11 as the first determination unit may perform audio analysis of the audio data as the above analysis. This makes it possible to more accurately identify the number of microphones from the content of the audio data.

[0105] In the analysis, the CPU 11 as the first determination unit performs pattern analysis on the plurality of image data 132 to identify two or more image data 132 that satisfy a predetermined approximation condition from among the plurality of image data 132. The CPU 11 determines the number of cameras 40 by a method of determining that two or more image data 132 that satisfy the approximation condition were generated by one camera 40. This makes it possible to more accurately identify the number of cameras 40 from the contents of the image data 132.

[0106] Each of the plurality of image data 132 includes target data 1321 relating to a predetermined sensing target and background data 1322 other than the target data 1321. In the analysis, the CPU 11 as a first determination unit identifies, among the plurality of image data 132, image data 132 whose background data 1322 satisfy the approximation condition as the two or more image data 132. In this way, the number of cameras 40 can be more accurately identified by utilizing the approximation of the background data 1322 of image data 132 captured by the same camera 40.

[0107] The data processing system 1 includes a CPU 51 of the management server 50 as a notification control unit. The CPU 51 performs processing to notify the user of determination result information regarding the determination result of the number of cameras 40. This makes it possible to prompt the user to take necessary measures regarding the number of connected cameras 40.

[0108] The determination result information includes at least one of the number of cameras 40, the response speed according to the number of cameras 40, the number of malfunctioning cameras 40 identified according to the number of cameras 40, and the amount charged to the user based on the number of cameras 40. This makes it possible to provide the user with various information related to the number of connected cameras 40 and prompt the user to take necessary measures.

[0109] The predetermined processing executed by the CPU 11 as a processing unit includes processing for acquiring an output result from a learning model that has been machine-learned in advance by inputting image data 132 to the learning model of the AI ​​container 102. This makes it possible to prevent problems from occurring due to inputting image data 132 from an excessive number of cameras 40 to the learning model of the AI ​​container 102.

[0110] The data processing system 1 includes a storage unit 53 of the management server 50 and a CPU 51 of the management server 50 serving as a second determination unit. The CPU 53 stores contract management data 534 relating to a contract with a user that specifies the maximum number of cameras 40 that can be connected to the AI ​​container 102 and the CPU 11 via the intermediate server 20. The CPU 51 serving as the second determination unit determines whether or not there has been a breach of the contract based on the contract management data 534 and the determination result regarding the number of cameras 40. This makes it possible to prompt the user to take necessary measures regarding the breach of contract.

[0111] The data processing system 1 as a billing determination system includes a CPU 51 of a management server 50 as a third determination unit. The CPU 51 determines the amount to be charged to the user based on the determination result of the number of cameras 40. This makes it possible to notify the user of the appropriate amount to be charged according to the number of connected cameras 40 and prompt the user to take the necessary measures.

[0112] The AI ​​server 10 as a data processing device according to this embodiment includes a CPU 11 as a processing unit (processing means) and a first determination unit (first determination means). The program 131 according to this embodiment causes the CPU 11 of the AI ​​server 10 to function as the processing means and the first determination means. This makes it possible to more accurately identify the number of connected cameras 40.

[0113] The data processing method according to this embodiment includes a generation step, a relay step, a processing step, and a first determination step. In the generation step, the multiple cameras 40 each generate image data 132. In the relay step, the intermediate server 20 receives and transmits the multiple image data 132 generated in the data generation step. In the processing step, a predetermined process is performed on the multiple image data 132 transmitted in the relay step. In the first determination step, a predetermined analysis is performed on information related to the multiple image data 132 transmitted in the relay step, and the number of cameras 40 included in the multiple cameras 40 is determined based on the results of the analysis. This makes it possible to more accurately identify the number of connected cameras 40.

[0114] (Others) The present invention is not limited to the above-described embodiment, and various modifications are possible. For example, the functions of the user terminal 60 may be integrated into the AI ​​server 10 or the intermediate server 20.

[0115] The method of determining which camera 40 captured the image data 132 is not limited to the method of comparing the background data 1322. For example, different targets may be placed in the capture range of each camera 40, and the camera 40 that captured each image data 132 may be identified from the targets captured in each of the multiple image data 132. In this case, analyzing whether or not a target is captured in the image data 132 corresponds to "pattern analysis."

[0116] Instead of the AI ​​container 102, a container that executes an application program that does not use a learning model may be used.

[0117] Although several embodiments of the present invention have been described, the scope of the present invention is not limited to the above-described embodiments, but includes the scope of the invention described in the claims and its equivalents.

[0118] The present invention can be used in a data processing system, a billing determination system, a data processing device, a program, and a data processing method.

[0119] 1 Data processing system 10 AI server (data processing device) 11 CPU (processing unit, first determination unit) 101 AI connection module 102 AI container 132 Image data (sensing data) 1321 Target data 1322 Background data 20 Intermediate server (relay unit) 40, 40a to 40c Camera (sensing unit) 50 Management server 51 CPU (second determination unit, third determination unit, communication control unit) 60 User terminal

Claims

1. A data processing system comprising: a plurality of sensing units each generating sensing data; a relay unit receiving and transmitting the plurality of sensing data generated by the plurality of sensing units; a processing unit executing a predetermined processing on the plurality of sensing data transmitted from the relay unit; and a first judgment unit performing a predetermined analysis on information relating to the plurality of sensing data transmitted from the relay unit and determining the number of sensing units included in the plurality of sensing units based on the results of the analysis.

2. The data processing system according to claim 1, characterized in that each of the plurality of sensing data includes time series data, and the first judgment unit judges the number of sensing units based on the mutual continuity or amount of change of the plurality of sensing data.

3. The data processing system according to claim 1, wherein said plurality of sensing units are devices of the same type that generate said sensing data of the same type.

4. The data processing system according to claim 1, characterized in that each of the plurality of sensing units is a camera, each of the plurality of sensing data is image data captured by the camera, and the first judgment unit performs image analysis of the image data as the predetermined analysis.

5. The data processing system according to claim 1, characterized in that each of the plurality of sensing units is a microphone, each of the plurality of sensing data is voice data recorded by the microphone, and the first judgment unit performs voice analysis of the voice data as the predetermined analysis.

6. The data processing system of claim 1, characterized in that the first judgment unit, in the specified analysis, identifies two or more pieces of sensing data that satisfy a specified approximation condition among the multiple sensing data by performing a pattern analysis on the multiple sensing data, and determines that the two or more pieces of sensing data that satisfy the approximation condition were generated by one of the sensing units.

7. The data processing system according to claim 6, characterized in that each of the plurality of sensing data includes target data relating to a specified sensing target and background data other than the target data, and the first judgment unit, in the specified analysis, identifies two or more sensing data among the plurality of sensing data whose background data mutually satisfy the approximation condition.

8. A data processing system as described in claim 1, further comprising a notification control unit that performs processing to notify a user operating the processing unit of information regarding the result of the determination of the number of sensing units by the first determination unit.

9. The data processing system of claim 8, wherein the information relating to the judgment result includes at least one of the number of the sensing units, the response speed of the processing unit according to the number of the sensing units, the number of faulty sensing units among the plurality of sensing units identified according to the number of the sensing units, and the amount charged to the user based on the number of the sensing units.

10. The data processing system according to claim 1, characterized in that the specified processing executed by the processing unit includes a process of inputting the sensing data into a learning model that has been machine-learned in advance, thereby obtaining an output result from the learning model.

11. The data processing system according to claim 1, further comprising: a memory unit that stores contract information relating to a contract with a user who operates the processing unit, the contract specifying the maximum number of the sensing units that can be connected to the processing unit via the relay unit; and a second judgment unit that judges whether or not there has been a breach of the contract based on the contract information and the judgment result of the first judgment unit regarding the number of the sensing units.

12. A billing determination system comprising: a data processing system according to any one of claims 1 to 11; and a third determination unit which determines an amount to be charged to a user who operates the processing unit based on the result of determination of the number of the sensing units by the first determination unit.

13. An information processing device connected to a relay unit that receives and transmits multiple sensing data generated by multiple sensing units, comprising: a processing unit that executes a predetermined processing on the multiple sensing data transmitted from the relay unit; and a first judgment unit that performs a predetermined analysis on information related to the multiple sensing data transmitted from the relay unit and judges the number of sensing units included in the multiple sensing units based on the results of the analysis.

14. A program that causes a computer provided in an information processing device connected to a relay unit that receives and transmits multiple sensing data generated by multiple sensing units to function as: a processing means that executes a predetermined process on the multiple sensing data transmitted from the relay unit; and a first determination means that performs a predetermined analysis on information related to the multiple sensing data transmitted from the relay unit and determines the number of sensing units included in the multiple sensing units based on the results of the analysis.

15. A data processing method executed by a computer, comprising: a data generation step in which a plurality of sensing units each generate sensing data; a relay step in which a relay unit receives and transmits the plurality of sensing data generated in the data generation step; a processing step in which a predetermined process is performed on the plurality of sensing data transmitted in the relay step; and a first judgment step in which a predetermined analysis is performed on information relating to the plurality of sensing data transmitted in the relay step, and the number of sensing units included in the plurality of sensing units is judged based on the results of the analysis.

Images