FIREWORKS DEVELOPMENT SUPPORT DEVICE, TRAINED MODEL GENERATION DEVICE, FIREWORKS DEVELOPMENT SUPPORT METHOD, TRAINED MODEL GENERATION METHOD, PROGRAM, AND RECORDING MEDIUM
The system uses machine learning models to simulate and design firework ball configurations and flowering states, addressing the time-consuming nature of traditional manufacturing and knowledge sharing in firework ball development.
Patent Information
- Application Number
- JP2021030401
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2021-02-26
- Publication Date
- 2024-11-13
- Estimated Expiration
- 2041-02-26
AI Technical Summary
The manufacturing of firework balls is time-consuming and relies heavily on personal experience, making it difficult to develop new designs and share knowledge effectively.
A system utilizing an acquisition unit, generation unit, and output unit to input and output estimated configuration and flowering information using machine learning models, supported by a fireworks development support device and trained model generation device.
Enables efficient simulation and design of firework ball configurations and flowering states without actual manufacturing, supporting knowledge acquisition and development.
Smart Images

Figure 0007582661000001 
Figure 0007582661000002 
Figure 0007582661000003
Abstract
Description
[Technical field]
[0001] The present invention relates to a fireworks development support device, a trained model generation device, a fireworks development support method, a trained model generation method, a program, and a recording medium. [Background technology]
[0002] There are fireworks that have explosives such as stars and explosives packed inside the shell (for example, Patent Document 1, etc.). When the fuse burns, the fire reaches the explosives inside the firework, causing the firework to explode, igniting the stars and scattering them, causing the firework to explode. [Prior art documents] [Patent documents]
[0003] [Patent Document 1] JP 2016-020753 A Summary of the Invention [Problem to be solved by the invention]
[0004] However, manufacturing firework balls and checking the results of launching the manufactured firework balls (whether they bloom in the expected shape) takes time and effort. Meanwhile, knowledge about manufacturing firework balls is passed on from master to apprentice by hearsay. In addition, the development and manufacturing of new firework balls depends on the experience (amount of knowledge, etc.) of the technicians who manufacture firework balls. Therefore, if manufacturing firework balls takes time and effort, not only does it not become easy to develop new firework balls, but it also creates problems such as the time it takes to acquire knowledge about manufacturing firework balls and the difficulty in spreading that knowledge.
[0005] Therefore, an object of the present invention is to provide a fireworks development support device, a trained model generation device, a fireworks development support method, a trained model generation method, a program, and a recording medium that are capable of supporting the development of fireworks balls and the acquisition of knowledge. [Means for solving the problem]
[0006] In order to achieve the above object, the fireworks development support device of the present invention comprises: An acquisition unit, a generation unit, and an output unit, The acquisition unit acquires at least one of planned configuration information indicating a configuration of the planned firework ball and planned blooming information indicating a blooming state of the planned firework ball, When the concept configuration information is acquired, The generation unit inputs the acquired concept configuration information into a flowering information estimation model to generate estimated flowering information corresponding to the concept configuration information, When the above-mentioned concept development information is acquired, The generation unit inputs the acquired conception information into a composition information estimation model to generate estimated composition information of the firework ball corresponding to the conception information; The output unit is a device that outputs at least one of the estimated flowering information and the estimated configuration information.
[0007] The trained model generation device of the present invention comprises: An acquisition unit and a trained model generation unit are included, The acquisition unit acquires teacher data linking actual configuration information and actual flowering information of the fireworks ball, the trained model generation unit generates and outputs estimated flowering information corresponding to concept configuration information when concept configuration information of a planned firework ball is input by machine learning using the teacher data; and A configuration information estimation model that generates and outputs estimated configuration information corresponding to the conception information when conception information of a planned fireworks ball is input; The present invention is an apparatus for generating at least one of the trained models.
[0008] The fireworks development support method of the present invention includes the steps of: The method includes an acquisition step, a generation step, and an output step, The acquiring step acquires at least one of design configuration information indicating a configuration of the designed firework ball and design blooming information indicating a blooming state of the designed firework ball, When the concept configuration information is acquired, The generating step inputs the acquired concept configuration information into a flowering information estimation model to generate estimated flowering information corresponding to the concept configuration information; When the above-mentioned concept development information is acquired, The generating step inputs the acquired conception information into a composition information estimation model to generate estimated composition information of the firework ball corresponding to the conception information; The output step is a method for outputting at least one of the estimated flowering information and the estimated configuration information.
[0009] The trained model generation method of the present invention includes the steps of: The method includes an acquisition step and a trained model generation step, The acquiring step acquires teacher data linking actual configuration information and actual blooming information of the fireworks ball, The trained model generating step generates and outputs estimated flowering information corresponding to concept configuration information of a designed firework ball by machine learning using the training data, when concept configuration information of the designed firework ball is input. A configuration information estimation model that generates and outputs estimated configuration information corresponding to the conception information when conception information of a planned fireworks ball is input; A method for generating at least one of the trained models. Effect of the Invention
[0010] According to the present invention, it is possible to support the development of fireworks and the acquisition of knowledge about them. [Brief description of the drawings]
[0011] [Figure 1] FIG. 1 is a block diagram showing an example of a configuration of a fireworks development support device according to the first embodiment. [Diagram 2] FIG. 2 is a block diagram showing an example of a hardware configuration of the fireworks development support device according to the first embodiment. [Diagram 3] FIG. 3 is a flowchart showing an example of processing in the fireworks development assistance device of the first embodiment. [Figure 4] FIG. 4(A) is a block diagram showing an example of the configuration of a trained model generation device of embodiment 2, and FIG. 4(B) is a block diagram showing an example of the hardware configuration of the trained model generation device of embodiment 2. [Diagram 5] FIG. 5 is a flowchart showing an example of processing in the trained model generation device of the second embodiment. [Figure 6] FIG. 6 is a schematic diagram showing an example of the configuration information and the flowering information. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0012] In the fireworks development support device of the present invention, for example, The flowering information estimation model and the configuration information estimation model may be the same trained model.
[0013] In the fireworks development support device of the present invention, for example, The generating unit may input the acquired concept configuration information to the flowering information estimation model to generate the estimated flowering information and first evaluation information for the estimated flowering information.
[0014] The fireworks development support device of the present invention includes, for example, Further, a learning processing unit is included, The learning processing unit may construct at least one of the flowering information estimation model and the composition information estimation model by machine learning using teacher data that links actual composition information and actual flowering information of fireworks balls.
[0015] In the fireworks development support device of the present invention, for example, The acquisition unit further acquires second evaluation information for the flowering information, The learning processing unit may construct the flowering information estimation model by machine learning using teacher data in which the flowering information and the second evaluation information are linked together.
[0016] The fireworks development support device of the present invention includes, for example, Further, an evaluation information generating unit is included, the evaluation information generation unit inputs flowering information into a fireworks evaluation model to generate second evaluation information corresponding to the flowering information; The acquisition unit acquires the generated second evaluation information, The learning processing unit may construct the flowering information estimation model by machine learning using teacher data in which the flowering information and the second evaluation information are linked together.
[0017] In the trained model generation device of the present invention, for example, The acquisition unit acquires teacher data linking the blooming information of the fireworks ball with the second evaluation information, The trained model generation unit may be configured to generate a flowering information estimation model that generates and outputs the estimated flowering information and first evaluation information corresponding to the estimated flowering information when the concept configuration information is input through machine learning using the teacher data.
[0018] In the fireworks development support method of the present invention, for example, The flowering information estimation model and the configuration information estimation model may be the same trained model.
[0019] In the fireworks development support method of the present invention, for example, The generating step may be such that the acquired concept configuration information is input to the flowering information estimation model to generate the estimated flowering information and first evaluation information for the estimated flowering information.
[0020] The fireworks development support method of the present invention includes, for example, Further, a learning process is included, The learning processing step may be a step of constructing at least one of the flowering information estimation model and the composition information estimation model by machine learning using teacher data that links actual composition information and actual flowering information of fireworks balls.
[0021] In the fireworks development support method of the present invention, for example, The acquiring step further includes acquiring second evaluation information for the flowering information, The learning process may be implemented in such a manner that the flowering information estimation model is constructed by machine learning using teacher data in which the flowering information and the second evaluation information are linked together.
[0022] The fireworks development support method of the present invention includes, for example, Further, the method includes an evaluation information generating step, the evaluation information generating step inputs flowering information into a fireworks evaluation model to generate second evaluation information corresponding to the flowering information; The acquiring step acquires the generated second evaluation information, The learning process may be implemented in such a manner that the flowering information estimation model is constructed by machine learning using teacher data in which the flowering information and the second evaluation information are linked together.
[0023] In the trained model generation method of the present invention, for example, The acquiring step acquires teacher data linking the blooming information of the fireworks ball with the second evaluation information, The trained model generation process may be configured to generate a flowering information estimation model by machine learning using the training data, which generates and outputs the estimated flowering information and first evaluation information corresponding to the estimated flowering information when the concept configuration information is input.
[0024] The program of the present invention is a program for causing a computer to execute each step of the method of the present invention as a procedure.
[0025] The recording medium of the present invention is a computer-readable recording medium having the program of the present invention recorded thereon.
[0026] In the present invention, the term "firework ball" refers to a firework ball for launching fireworks, and is also called a firework ball. The use of the firework ball is not particularly limited, and may be for signals, for decoration, for testing, etc. Furthermore, the firework ball may be, for example, spherical or cylindrical.
[0027] Next, an embodiment of the present invention will be described with reference to the drawings. The present invention is not limited to the following embodiment. In each of the drawings, the same parts are given the same reference numerals. In addition, the explanations of each embodiment can be mutually incorporated unless otherwise specified, and the configurations of each embodiment can be combined unless otherwise specified.
[0028] [Embodiment 1] Fig. 1 is a block diagram showing an example of the configuration of a fireworks ball manufacturing support device 10 of this embodiment. As shown in Fig. 1, this device 10 includes an acquisition unit 11, a generation unit 12, and an output unit 13. This device 10 may further include, as optional components, an evaluation information generation unit 14, a learning processing unit 15, etc. This device 10 may also include, for example, a fireworks information database 1, an evaluation information database 2, etc.
[0029] The device 10 may be, for example, a single device including each of the above-mentioned units, or a device to which each of the above-mentioned units can be connected via a communication line network. The device 10 may be connected to an external device described later via the communication line network. The communication line network is not particularly limited, and a known network may be used, for example, wired or wireless. Examples of the communication line network include an Internet line, a World Wide Web (WWW), a telephone line, a Local Area Network (LAN), a Storage Area Network (SAN), a Delay Tolerant Networking (DTN), a Low Power Wide Area (LPWA), and a Local 5G (L5G). Examples of wireless communication include Wi-Fi (registered trademark), Bluetooth (registered trademark), a Local 5G, and a LPWA. Examples of the wireless communication may include a form in which each device directly communicates with each other (Ad Hoc communication), an infrastructure (infrastructure communication), and an indirect communication via an access point. The device 10 may be incorporated into a server as a system, for example. The device 10 may be, for example, a personal computer (PC, for example, desktop type or notebook type) on which the program of the present invention is installed, a smartphone, a tablet terminal, etc. Furthermore, the device 10 may be in the form of cloud computing or edge computing, for example, in which at least one of the units is on a server and the other units are on a terminal.
[0030] 2 illustrates a block diagram of the hardware configuration of the device 10. The device 10 includes, for example, a central processing unit (CPU, GPU, etc.) 101, a memory 102, a bus 103, a storage device 104, an input device 105, a display device 106, and a communication device 107. The components of the device 10 are connected to each other via the bus 103 and their respective interfaces (I / F).
[0031] The central processing unit 101 is responsible for the overall control of the present device 10. In the present device 10, the central processing unit 101 executes, for example, the program of the present invention and other programs, and also reads and writes various information. Specifically, for example, the central processing unit 101 functions as an acquisition unit 11, a generation unit 12, and an output unit 13. The central processing unit 101 also functions, for example, as an evaluation information generation unit 14, a learning processing unit 15, etc.
[0032] The bus 103 can also be connected to, for example, an external device. Examples of the external device include an external storage device (external database, etc.), a printer, an external input device, an external display device, an external imaging device, etc. The present device 10 can be connected to an external network (the above-mentioned communication line network) by, for example, a communication device 107 connected to the bus 103, and can also be connected to other devices via the external network.
[0033] An example of the memory 102 is a main memory (primary storage device). When the central processing unit 101 performs processing, the memory 102 reads various operation programs, such as the program of the present invention, stored in the storage device 104 described below, and the central processing unit 101 receives data from the memory 102 and executes the program. The main memory is, for example, a RAM (random access memory). The memory 102 may also be, for example, a ROM (read only memory).
[0034] The storage device 104 is also called an auxiliary storage device, for example, in contrast to the main memory (primary storage device). As described above, the storage device 104 stores an operation program including the program of the present invention. The storage device 104 may be, for example, a combination of a recording medium and a drive for reading and writing from and to the recording medium. The recording medium is not particularly limited, and may be, for example, an internal or external type, such as a hard disk (HD), CD-ROM, CD-R, CD-RW, MO, DVD, flash memory, or memory card. The storage device 104 may be, for example, a hard disk drive (HDD) in which a recording medium and a drive are integrated, or a solid state drive (SSD).
[0035] In the present device 10, the memory 102 and the storage device 104 can also store various information such as log information, information acquired from an external database (not shown) or an external device, information generated by the present device 10, and information used when the present device 10 executes processing. Note that at least a portion of the information may be stored, for example, in an external server other than the memory 102 and the storage device 104, or may be stored in a distributed manner in multiple terminals using block chain technology or the like.
[0036] The device 10 may further include, for example, an input device 105 and a display device 106. The input device 105 is, for example, a touch panel, a keyboard, a mouse, etc. The display device 106 is, for example, an LED display, a liquid crystal display, etc.
[0037] Next, an example of the firework ball manufacturing support method of this embodiment will be described with reference to the flowchart of Fig. 3. The firework ball manufacturing support method of this embodiment is carried out as follows, for example, using the firework ball manufacturing support device 10 of Fig. 1. Note that the firework ball manufacturing support method of this embodiment is not limited to the use of the firework ball manufacturing support device 10 of Fig. 1. Also, the steps shown in parentheses in Fig. 3 are optional steps.
[0038] First, the acquisition unit 11 acquires at least one of plan configuration information indicating the configuration of the planned firework ball and plan blooming information indicating the blooming state of the planned firework ball (S11a, acquisition step). Here, the configuration of the firework ball includes elements such as information on the internal structure of the firework ball, such as the arrangement of each explosive (star, explosive, etc.); information on the explosive, such as the type and amount of each explosive; and information on the shape of the firework ball, such as spherical, cylindrical, etc. Also, the blooming state of the firework ball means a state in which the firework ball rises to a predetermined height, the fuse is ignited, the fire reaches the explosive by the fuse, the firework ball explodes, and the stars are scattered while burning. The plan configuration information and the plan blooming information may include image information, such as a moving image or a still image. The acquisition unit 11 may acquire the concept construction information and the concept development information from an external server or an external database via the communication line network, or may acquire the concept construction information and the concept development information by reading the concept construction information and the concept development information stored in the memory 102 and the storage device 104. Specifically, the concept construction information and the concept development information can be generated using a technology such as 3D CAD.
[0039] Next, when the concept configuration information is acquired, the generation unit 12 inputs the acquired concept configuration information into a flowering information estimation model to generate estimated flowering information corresponding to the concept configuration information (S12a, generation step). The flowering information estimation model is a model that has been trained to generate estimated flowering information corresponding to the concept configuration information from the input concept configuration information of the firework ball. The flowering information estimation model may be constructed by, for example, the learning processing unit 15 described later, or may be stored in advance in the memory 102 and the storage device 104, or may be acquired by the acquisition unit 11 from an external trained model generation device (for example, the trained model generation device 20 described later) via a communication line network. The estimated flowering information is information indicating the flowering state of the firework ball estimated from the configuration of the firework ball in the concept configuration information. The estimated flowering information may include image information such as a video or a still image.
[0040] On the other hand, when the conception information is acquired, the generation unit 12 inputs the acquired conception information into a configuration information estimation model to generate estimated configuration information of a firework ball corresponding to the conception information (S12b, generation step). The configuration information estimation model is a model that has been trained to generate estimated configuration information of a firework ball corresponding to the conception information from the input conception information. The configuration information estimation model may be constructed by, for example, the learning processing unit 15 described later, or may be stored in advance in the memory 102 and the storage device 104, or may be acquired by the acquisition unit 11 from an external trained model generation device (for example, the trained model generation device 20 described later) via a communication line network. The estimated configuration information is information indicating the configuration of a firework ball estimated from the state of the firework ball in the conception information. The estimated configuration information may include image information such as a video or a still image.
[0041] The generating unit 12 may include, for example, an estimated flowering information generating unit 121 that executes the step (S12a) and an estimated configuration information generating unit 122 that executes the step (S12b).
[0042] The flowering information estimation model includes, for example, an input layer that inputs concept configuration information of a firework ball, an output layer that outputs the estimated flowering information, and at least one intermediate layer provided between the input layer and the output layer. The configuration information estimation model includes, for example, an input layer that inputs concept flowering information, an output layer that outputs the estimated configuration information, and at least one intermediate layer provided between the input layer and the output layer. The flowering information estimation model and the configuration information estimation model may be program modules that are part of artificial intelligence software. Examples of the multi-layered network include neural networks. Examples of the neural network include, for example, convolution neural networks (CNNs), but are not limited to CNNs, and may be trained models constructed with other learning algorithms such as neural networks other than CNNs, SVMs (Support Vector Machines), Bayesian networks, and regression trees. In this way, the flowering information estimation model and the configuration information estimation model may be, for example, independent trained models. On the other hand, the flowering information estimation model and the configuration information estimation model may be, for example, the same trained model. The same trained model can also be said to be a trained model with bidirectionality. The trained model with bidirectionality includes, for example, an input layer that inputs at least one of the concept configuration information and concept blooming information of the fireworks ball, an output layer that outputs at least one of the estimated blooming information and the estimated configuration information, and at least one intermediate layer provided between the input layer and the output layer. For example, the above-mentioned descriptions regarding the blooming information estimation model and the configuration information estimation model can be used for the trained model with bidirectionality.
[0043] Then, the output unit 13 outputs at least one of the estimated flowering information and the estimated configuration information (hereinafter also referred to as the estimated flowering information, etc.) (S13), and the process ends (END). The output unit 13 may output the estimated flowering information, etc. to an external device via a communication network, for example, or may output the information to the display device 106. The estimated flowering information, etc. may be stored in the memory 102 and the storage device 104.
[0044] For example, in the step (S12a), the generating unit 12 may input the acquired concept configuration information to the flowering information estimation model to generate the estimated flowering information and first evaluation information for the estimated flowering information. The first evaluation information is, for example, information on an evaluation of the flowering state of the firework ball indicated by the estimated flowering information. Specifically, the evaluation is, for example, an evaluation of the appearance of the flowered firework, such as the shape, spread, shoulder tension, disappearance, and color, but is not limited thereto. The first evaluation information may be expressed, for example, by a score for each arbitrary item. The items include, for example, shape, color, change, novelty, and the like. That is, the flowering information estimation model may be, for example, a model trained to generate estimated flowering information corresponding to the concept configuration information and first evaluation information for the estimated flowering information from the concept configuration information of the input firework ball. The flowering information estimation model in this case can be described above, except that, for example, instead of an output layer that outputs the estimated flowering information, an output layer that outputs the estimated flowering information and the first evaluation information for the estimated flowering information is included.
[0045] As described above, the device 10 may further include, for example, a learning processing unit 15. The learning processing unit 15 constructs at least one of the blooming information estimation model and the configuration information estimation model by machine learning using teacher data in which real configuration information and real blooming information of a firework ball are linked (S15a, learning processing step). The real configuration information is information about the configuration of a real firework ball, and is, for example, information in which information about gunpowder is linked to a structural diagram obtained by performing a 3D scan or the like on a firework ball. The configuration of the firework ball is the same as described above. The real blooming information is information about the blooming state of a real firework ball, and is, for example, information obtained by analyzing a video or still image of a firework. The blooming state is the same as described above. The machine learning is not particularly limited, and is, for example, machine learning using a CNN, SVM, Bayesian network, regression tree, or the like. The step (S15a) may be performed before the step (S12a) and the step (S12b), and is not limited to being performed after the step (S11a) as shown in FIG. 3 and a step (S11b) described below.
[0046] The acquisition unit 11 may further acquire second evaluation information related to the flowering information (S11b, acquisition step). The flowering information includes, for example, actual flowering information, estimated flowering information, and planned flowering information. The description of the first evaluation information described above can be used for the second evaluation information. The second evaluation information includes, for example, the first evaluation information for estimated flowering information generated in the past, a user's evaluation of the flowering information, a preset appreciation index, and a degree of novelty (the degree to which a similar structure does not exist in the past). The second evaluation information may be, for example, second evaluation information generated by the evaluation information generation unit 14 described later. The acquisition unit 11 may acquire the second evaluation information from an external server or an external database via the communication line network, or may acquire the second evaluation information by reading out the second evaluation information stored in the memory 102 and the storage device 104. The step (S11b) may be executed in parallel with the step (S11a) as shown in FIG. 3, or may be executed in order. The order is not particularly limited.
[0047] Then, the learning processing unit 15 may construct the flowering information estimation model by machine learning using teacher data linking the flowering information and the second evaluation information in the step (S15a), for example. In other words, when the acquisition unit 11 acquires the first evaluation information as the second evaluation information, it can be said that the learning processing unit 15 automatically reconstructs the flowering information estimation model, for example.
[0048] As described above, the device 10 may further include, for example, an evaluation information generating unit 14. The evaluation information generating unit 14, for example, inputs flowering information into a fireworks evaluation model to generate second evaluation information corresponding to the flowering information (S14, evaluation information generating step). The fireworks evaluation model is a model that has been trained to generate second evaluation information corresponding to the flowering information from the input flowering information. The above description can be used for the flowering information and the second evaluation information. The step (S14) may be executed before the step (S11b) as shown in FIG. 3, for example, but if the flowering information is the estimated flowering information, it may be executed after the step (S12a) and before the step (S13). Also, if the flowering information is the planned flowering information, it may be executed after the step (S11a) and before the step (S11b). Then, the acquisition unit 11 acquires the second evaluation information generated by the evaluation information generating unit 14 in the step (S11b). As described above, in the step (S15a), the learning processing unit 15 constructs the flowering information estimation model by machine learning using teacher data in which the flowering information and the second evaluation information are linked together.
[0049] The fireworks evaluation model includes, for example, an input layer that inputs the flowering information, an output layer that outputs the second evaluation information, and at least one intermediate layer provided between the input layer and the output layer. For example, the descriptions regarding the flowering information estimation model and the configuration information estimation model can be used for the fireworks evaluation model.
[0050] The fireworks evaluation model may be constructed, for example, prior to the step (S14), by machine learning performed by the learning processing unit 15 using training data that links the flowering information and the second evaluation information (S15b).
[0051] The device 10 may include, for example, a fireworks information database 1. The fireworks information database 1 is a database that stores configuration information and flowering information. The configuration information is the planned configuration information, the estimated configuration information, the actual configuration information, etc. The flowering information is the planned flowering information, the estimated configuration information, the actual configuration information, etc. An example of the configuration information and the flowering information is shown in FIG. 6. Note that in FIG. 6, the configuration information and the flowering information are shown as plan views, but they may also be three-dimensional views or may include multiple plan views.
[0052] The present device 10 may include, for example, a review information database 2. The review information database 2 is a database that stores the review information.
[0053] According to this embodiment, for example, by generating estimated flowering information of a firework ball, it is possible to simulate the result of launching a planned firework ball. Also, according to this embodiment, for example, by generating estimated configuration information of a firework ball, it is possible to design the configuration of a firework ball from the flowering state of the planned firework ball. In this way, according to this embodiment, it is possible to support the acquisition of knowledge regarding the development and manufacturing of firework balls without actually manufacturing or launching firework balls. Also, by generating the estimated flowering information using the evaluation information, it is possible to support, for example, the development and acquisition of knowledge of firework balls with higher evaluations.
[0054] [Embodiment 2] Fig. 4(A) is a block diagram showing an example of the configuration of the trained model generation device 20 of this embodiment. As shown in Fig. 4(A), this device 20 includes an acquisition unit 21 and a trained model generation unit 22. Each trained model generated by this device 20 may be used, for example, in the fireworks development assistance device 10 described in the first embodiment.
[0055] The device 20 may be, for example, a single device including the above-mentioned units, or a device to which the above-mentioned units can be connected via the communication line network. The device 20 may be connected to the above-mentioned external device via the communication line network. The device 20 may be, for example, incorporated in a server as a system. The device 20 may be, for example, a personal computer (PC, for example, desktop type or notebook type) in which the program of the present invention is installed, a smartphone, a tablet terminal, or the like. Furthermore, the device 20 may be in the form of cloud computing or edge computing, for example, in which at least one of the above-mentioned units is on a server and the other units are on a terminal.
[0056] 4(B) illustrates an example block diagram of the hardware configuration of trained model generation device 20. Trained model generation device 20 includes, for example, a central processing unit (CPU, GPU, etc.) 101, a memory 102, a bus 103, a storage device 104, an input device 105, a display device 106, a communication device 107, etc. Unless otherwise specified, the descriptions of the respective parts of the hardware configuration of fireworks development support device 10 shown in FIG. 3 can be used for the respective parts of the hardware configuration of trained model generation device 20.
[0057] The central processing unit 101 functions as, for example, an acquisition unit 21 and a trained model generation unit 22.
[0058] The memory 102 and the storage device 104 may store the trained model generated by the device 20.
[0059] Next, an example of a method for producing a trained model according to this embodiment will be described with reference to the flowchart of Fig. 5. The method for producing a trained model according to this embodiment is carried out as follows, for example, using the trained model generation device 20 of Fig. 4. Note that the method for producing a trained model according to this embodiment is not limited to the use of the trained model generation device 20 of Fig. 4.
[0060] First, the acquisition unit 21 acquires teacher data linking the actual configuration information and actual flowering information of the fireworks ball (S21, acquisition step). Next, the trained model generation unit 22 generates at least one trained model by machine learning using the teacher data: a flowering information estimation model that generates and outputs estimated flowering information corresponding to the concept configuration information when concept configuration information of a planned fireworks ball is input, and a configuration information estimation model that generates and outputs estimated configuration information corresponding to the concept flowering information when concept flowering information of a planned fireworks ball is input (S22, trained model generation step), and then ends (END).
[0061] In other words, the present device 20 can be said to construct (generate) at least one of a neural network that receives the concept configuration information as an input and outputs the estimated flowering information, and a neural network that receives the concept configuration information as an input and outputs the estimated configuration information, by performing machine learning with a machine. The neural network is, for example, the same as described above. The neural network may include, for example, an input layer that receives the concept configuration information or the composition flowering information, an output layer that outputs the estimated flowering information or the estimated configuration information, and at least one intermediate layer provided between the input layer and the output layer. The flowering information estimation model and the composition information estimation model may be, for example, independent trained models. On the other hand, the flowering information estimation model and the composition information estimation model may be, for example, the same trained model. The same trained model can be said to be a trained model with bidirectionality. For the trained model with bidirectionality, for example, the description regarding each of the above-mentioned neural networks can be used.
[0062] The input layer includes, for example, a plurality of neurons that accept the input of the concept configuration information or the composition flowering information, and passes the input concept configuration information or the composition flowering information to the intermediate layer. The intermediate layer includes a plurality of neurons that extract features of the concept configuration information or the composition flowering information, and passes the extracted features to the output layer. The output layer includes one or more neurons that output the estimated flowering information or the estimated composition information, and generates the estimated flowering information or the estimated composition information based on the features output from the intermediate layer. The features are not particularly limited, and may be, for example, parameters such as the number of stars, the number of offspring generation, the number of grandchild generation, the light duration, and the tail survival time.
[0063] Further, the acquisition unit 21 may acquire, for example, teacher data linking the flowering information and the second evaluation information. The trained model generation unit 22 may generate a flowering information estimation model that generates and outputs the estimated flowering information and the second evaluation information corresponding to the estimated flowering information when the concept configuration information is input, for example, by machine learning using the teacher data. In other words, the present device 20 can be said to construct a neural network that inputs the concept configuration information and outputs the estimated flowering information and the second evaluation information, instead of a neural network that inputs the concept configuration information and outputs the estimated flowering information, as an input. The neural network is the same as described above, except that the output layer outputs the estimated flowering information and the second evaluation information. The description of each layer of the neural network can be used by appropriately reinterpreting it in accordance with the description of the neural network.
[0064] In addition, the trained model generating unit 22 may generate a fireworks evaluation model by machine learning using teacher data that links the flowering information and the second evaluation information. As described above, the fireworks evaluation model is a model that generates and outputs second evaluation information corresponding to the flowering information when the flowering information is input. In other words, it can be said that the device 20 constructs a neural network that receives the flowering information as an input and outputs the second evaluation information. In the neural network, the input layer receives the flowering information, and the output layer outputs the second evaluation information, and the neural network is the same as described above. The description of each layer of the neural network can be used by appropriately reinterpreting it in accordance with the description of the neural network.
[0065] Each trained model generated by this embodiment is used, for example, in the fireworks development support device 10 described in the above-described embodiment 1. This makes it possible to generate estimated flowering information, estimated configuration information, and evaluation models, and to support the development of fireworks and knowledge acquisition.
[0066] [Embodiment 3] The program of the present embodiment is a program for causing a computer to execute each step of the method of the present invention as a procedure. In the present invention, the "procedure" may be read as "processing." The program of the present embodiment may be recorded, for example, in a computer-readable recording medium. The recording medium is not particularly limited, and examples thereof include a read-only memory (ROM), a hard disk (HD), and an optical disk.
[0067] Although the present invention has been described above with reference to the embodiment, the present invention is not limited to the above embodiment. Various modifications that can be understood by those skilled in the art can be made to the configuration and details of the present invention within the scope of the present invention.
[0068] <Additional Notes> Some or all of the above embodiments may be described as follows, but are not limited to the following supplementary notes. (Appendix 1) An acquisition unit, a generation unit, and an output unit, The acquisition unit acquires at least one of planned configuration information indicating a configuration of the planned firework ball and planned blooming information indicating a blooming state of the planned firework ball, When the concept configuration information is acquired, The generation unit inputs the acquired concept configuration information into a flowering information estimation model to generate estimated flowering information corresponding to the concept configuration information, When the above-mentioned concept development information is acquired, The generation unit inputs the acquired conception information into a composition information estimation model to generate estimated composition information of the firework ball corresponding to the conception information; The output unit outputs at least one of the estimated flowering information and the estimated configuration information. (Appendix 2) The flowering information estimation model and the configuration information estimation model are a same trained model. (Appendix 3) 3. The fireworks development support device according to claim 1, wherein the generation unit inputs the acquired concept configuration information into the flowering information estimation model to generate the estimated flowering information and first evaluation information for the estimated flowering information. (Appendix 4) Further, a learning processing unit is included, The fireworks development support device according to any one of appendices 1 to 3, wherein the learning processing unit constructs at least one of the flowering information estimation model and the composition information estimation model by machine learning using teacher data that links actual composition information and actual flowering information of firework balls. (Appendix 5) The acquisition unit further acquires second evaluation information for the flowering information, The fireworks development support device according to claim 4, wherein the learning processing unit constructs the flowering information estimation model by machine learning using teacher data that links the flowering information and the second evaluation information. (Appendix 6) Further, an evaluation information generating unit is included, the evaluation information generation unit inputs flowering information into a fireworks evaluation model to generate second evaluation information corresponding to the flowering information; The acquisition unit acquires the generated second evaluation information, The learning processing unit constructs the flowering information estimation model by machine learning using teacher data in which the flowering information and the second evaluation information are linked. 6. The fireworks development support device according to claim 5. (Appendix 7) An acquisition unit and a trained model generation unit are included, The acquisition unit acquires teacher data linking actual configuration information and actual flowering information of the fireworks ball, the trained model generation unit generates and outputs estimated flowering information corresponding to concept configuration information when concept configuration information of a planned firework ball is input by machine learning using the teacher data; and A configuration information estimation model that generates and outputs estimated configuration information corresponding to the conception information when conception information of a planned fireworks ball is input; A trained model generation device that generates at least one of the trained models. (Appendix 8) The acquisition unit acquires teacher data linking the blooming information of the fireworks ball with the second evaluation information, The trained model generation device of claim 7, wherein the trained model generation unit generates a flowering information estimation model that generates and outputs the estimated flowering information and first evaluation information corresponding to the estimated flowering information when the concept configuration information is input, by machine learning using the teacher data. (Appendix 9) The method includes an acquisition step, a generation step, and an output step, The acquiring step acquires at least one of design configuration information indicating a configuration of the designed firework ball and design blooming information indicating a blooming state of the designed firework ball, When the concept configuration information is acquired, The generating step inputs the acquired concept configuration information into a flowering information estimation model to generate estimated flowering information corresponding to the concept configuration information; When the above-mentioned concept development information is acquired, The generating step inputs the acquired conception information into a composition information estimation model to generate estimated composition information of the firework ball corresponding to the conception information; The fireworks development support method, wherein the output step outputs at least one of the estimated flowering information and the estimated configuration information. (Appendix 10) The fireworks development support method according to claim 9, wherein the flowering information estimation model and the configuration information estimation model are the same trained model. (Appendix 11) 11. The fireworks development support method according to claim 9, wherein the generating step inputs the acquired concept configuration information into the flowering information estimation model to generate the estimated flowering information and first evaluation information for the estimated flowering information. (Appendix 12) Further, a learning process is included, The fireworks development support method according to any one of Appendices 9 to 11, wherein the learning process step constructs at least one of the flowering information estimation model and the composition information estimation model by machine learning using training data that links actual composition information and actual flowering information of fireworks balls. (Appendix 13) The acquiring step further includes acquiring second evaluation information for the flowering information, The fireworks development support method according to claim 12, wherein the learning process constructs the flowering information estimation model by machine learning using teacher data that links the flowering information and the second evaluation information. (Appendix 14) Further, the method includes an evaluation information generating step, the evaluation information generating step inputs flowering information into a fireworks evaluation model to generate second evaluation information corresponding to the flowering information; The acquiring step acquires the generated second evaluation information, The fireworks development support method according to claim 13, wherein the learning process constructs the flowering information estimation model by machine learning using teacher data that links the flowering information and the second evaluation information. (Appendix 15) The method includes an acquisition step and a trained model generation step, The acquiring step acquires teacher data linking actual configuration information and actual blooming information of the fireworks ball, The trained model generating step generates and outputs estimated flowering information corresponding to concept configuration information of a designed firework ball by machine learning using the training data, when concept configuration information of the designed firework ball is input. A configuration information estimation model that generates and outputs estimated configuration information corresponding to the conception information when conception information of a planned fireworks ball is input; A trained model generation method for generating at least one of the trained models. (Appendix 16) The acquiring step acquires teacher data linking the blooming information of the fireworks ball with the second evaluation information, The trained model generation method of claim 15, wherein the trained model generation step generates a flowering information estimation model by machine learning using the teacher data, which generates and outputs the estimated flowering information and first evaluation information corresponding to the estimated flowering information when the concept configuration information is input. (Appendix 17) A program for causing a computer to carry out steps including obtaining, generating and outputting steps: The acquisition step includes acquiring at least one of planned configuration information indicating a configuration of the planned firework ball and planned blooming information indicating a blooming state of the planned firework ball, When the concept configuration information is acquired, The generating step includes inputting the acquired concept configuration information into a flowering information estimation model to generate estimated flowering information corresponding to the concept configuration information; When the above-mentioned concept development information is acquired, The generation step includes inputting the acquired conception information into a composition information estimation model to generate estimated composition information of the firework ball corresponding to the conception information; The output step includes outputting at least one of the estimated flowering information and the estimated configuration information. (Appendix 18) The program according to claim 17, wherein the flowering information estimation model and the configuration information estimation model are the same trained model. (Appendix 19) The program described in Appendix 17 or 18, wherein the generation step inputs the acquired concept configuration information into the flowering information estimation model to generate the estimated flowering information and first evaluation information for the estimated flowering information. (Appendix 20) Further, a learning procedure is included, The learning processing procedure is a program described in any one of Appendices 17 to 19, which constructs at least one of the flowering information estimation model and the composition information estimation model by machine learning using teacher data that links actual composition information and actual flowering information of fireworks balls. (Appendix 21) The acquisition step further includes acquiring second evaluation information for the flowering information, The program described in Appendix 20, wherein the learning process procedure constructs the flowering information estimation model by machine learning using teacher data that links the flowering information and the second evaluation information. (Appendix 22) Further, the method includes a step of generating evaluation information, the evaluation information generating step inputs flowering information into a fireworks evaluation model to generate second evaluation information corresponding to the flowering information; The obtaining step obtains the generated second evaluation information, The program described in Appendix 21, wherein the learning process procedure constructs the flowering information estimation model by machine learning using teacher data that links the flowering information and the second evaluation information. (Appendix 23) A program for causing a computer to execute a procedure including an acquisition procedure and a trained model generation procedure: The acquisition step includes acquiring teacher data that links actual configuration information and actual flowering information of the fireworks ball; The trained model generation step includes: a flowering information estimation model that generates and outputs estimated flowering information corresponding to concept configuration information when concept configuration information of a planned firework ball is input by machine learning using the training data; A configuration information estimation model that generates and outputs estimated configuration information corresponding to the conception information when conception information of a planned fireworks ball is input; Generate at least one of the trained models. (Appendix 24) The acquisition step includes acquiring teacher data linking the blooming information of the fireworks ball with the second evaluation information; The learned model generation procedure is a program described in Appendix 23, which generates a flowering information estimation model by machine learning using the teacher data, which generates and outputs the estimated flowering information and first evaluation information corresponding to the estimated flowering information when the concept configuration information is input. (Appendix 25) A computer-readable recording medium having a program according to any one of appendices 17 to 24 recorded thereon. [Industrial Applicability]
[0069] According to the present invention, it is possible to design the configuration of a firework ball from a launch simulation of the conceived firework ball and the blooming state of the conceived firework ball. Therefore, the present invention is useful for acquiring knowledge about the development and manufacturing of firework balls, for example. [Explanation of symbols]
[0070] 1. Fireworks Information Database 2. Evaluation information database 10 Fireworks development support device 11 Acquisition Department 12 Generation part 121 Estimated flowering information generation section 122 Estimated configuration information generation unit 13 Output section 14 Evaluation information generation unit 15 Learning processing section 20 Trained model generator 21 Acquisition Department 22 Trained model generation unit 101 Central Processing Unit 102 Memory 103 Bus 104 Storage device 105 Input Device 106 Display device 107 Communication Devices
Claims
1. An acquisition unit, a generation unit, and an output unit, The acquisition unit acquires at least one of planned configuration information indicating a configuration of the planned firework ball and planned blooming information indicating a blooming state of the planned firework ball, When the concept configuration information is acquired, The generation unit inputs the acquired concept configuration information into a flowering information estimation model to generate estimated flowering information corresponding to the concept configuration information and first evaluation information for the estimated flowering information; When the above-mentioned concept development information is acquired, The generation unit inputs the acquired conception information into a composition information estimation model to generate estimated composition information of the firework ball corresponding to the conception information; The output unit outputs at least one of the estimated flowering information and the estimated configuration information.
2. Further, a learning processing unit is included, The fireworks development support device according to claim 1 , wherein the learning processing unit constructs at least one of the flowering information estimation model and the composition information estimation model by machine learning using teacher data that links actual composition information and actual flowering information of firework balls.
3. The acquisition unit further acquires second evaluation information for the flowering information, The fireworks development support device according to claim 2 , wherein the learning processing unit constructs the flowering information estimation model by machine learning using teacher data in which the flowering information and the second evaluation information are linked.
4. Further, an evaluation information generating unit is included, the evaluation information generation unit inputs flowering information into a fireworks evaluation model to generate second evaluation information corresponding to the flowering information; The acquisition unit acquires the generated second evaluation information, The learning processing unit constructs the flowering information estimation model by machine learning using teacher data in which the flowering information and the second evaluation information are linked.
4. The fireworks development support device according to claim 3.
5. The method includes an acquisition step, a generation step, and an output step, The acquiring step acquires at least one of design configuration information indicating a configuration of the designed firework ball and design blooming information indicating a blooming state of the designed firework ball, When the concept configuration information is acquired, The generating step inputs the acquired concept configuration information into a flowering information estimation model to generate estimated flowering information corresponding to the concept configuration information and first evaluation information for the estimated flowering information; When the above-mentioned concept development information is acquired, The generating step inputs the acquired conception information into a composition information estimation model to generate estimated composition information of the firework ball corresponding to the conception information; The fireworks development support method, wherein the output step outputs at least one of the estimated flowering information and the estimated configuration information.
6. A program for causing a computer to execute steps including an acquisition step, a generation step, and an output step: The acquisition step includes acquiring at least one of planned configuration information indicating a configuration of the planned firework ball and planned blooming information indicating a blooming state of the planned firework ball, When the concept configuration information is acquired, The generating step includes inputting the acquired concept configuration information into a flowering information estimation model to generate estimated flowering information corresponding to the concept configuration information and first evaluation information for the estimated flowering information; When the above-mentioned concept development information is acquired, The generation step includes inputting the acquired conception information into a composition information estimation model to generate estimated composition information of the firework ball corresponding to the conception information; The output step includes outputting at least one of the estimated flowering information and the estimated configuration information.
Citation Information
Patent Citations
Method of selectively receiving display object, storage medium and program execution device
JP2001155188A
Imaging device, imaging method, and program of imaging device
JP2015178985A
Firework ball
JP2016020753A
Learning image generation apparatus, learning image generation method, image recognition device, and image recognition method
JP2019191973A
Light display method and fireworks shell
JP2019207395A
