Apparatus that generates teacher data for identifying object, apparatus that produces trained model for identifying object, object identifying apparatus, method that generates teacher data for identifying object, method that produces trained model for identifying object, object identifying method, program that generates teacher data for identifying object, program that produces trained model for identifying object, object identifying program, and storage medium

Abstract

To provide an apparatus that generates teacher data for identifying an object capable of easily generating teacher data for identifying the object.SOLUTION: An apparatus that generates teacher data for identifying an object of the present disclosure comprises: an object information acquisition unit for acquiring an object image obtained by capturing an image of the object; a voice data acquisition unit for acquiring voice data including voice for vocal confirmation regarding the object; and a teacher data output unit for outputting the voice data as a correct label for the object image and outputting a combination of the object image and the voice data as teacher data for identifying the object.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present disclosure relates to a device for generating teacher data for object identification, a device for producing a trained model for object identification, an object identification device, a method for generating teacher data for object identification, a method for producing a trained model for object identification, an object identification method, a program for generating teacher data for object identification, a program for producing a trained model for object identification, an object identification program, and a recording medium. [Background technology]

[0002] In recent years, the ready-meal industry, which provides so-called ready-meals such as boxed lunches and side dishes, has been attracting attention. In the ready-meal industry, various cooked dishes need to be produced by unspecified employees, so an appropriate cooking support system is required from the viewpoint of quality control. For example, Patent Document 1 describes a cooking support method in a cooking support system in which a server device that stores information related to cooking recipes and a terminal device having a display are connected via a network. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2017-068785 A Summary of the Invention [Problem to be solved by the invention]

[0004] The system described in Patent Document 1 is only capable of displaying recipes for food products, and is unable to confirm whether the cook has used appropriate ingredients and tools. Therefore, it is conceivable to use image recognition by AI to identify finished products. However, in order to identify products using image recognition by AI, it is necessary to prepare a large amount of training data, which is costly and time-consuming. Such issues are not limited to the manufacture of food products, but are similar in various fields of product manufacturing.

[0005] Therefore, the present disclosure aims to provide a teacher data generation device for object identification that can easily generate teacher data for object identification, a trained model manufacturing device for object identification, an object identification device, a teacher data generation method for object identification, a trained model manufacturing method for object identification, an object identification method, a teacher data generation program for object identification, a trained model manufacturing program for object identification, an object identification program, and a recording medium. [Means for solving the problem]

[0006] In order to achieve the above object, the present disclosure relates to a device for generating training data for object identification, The apparatus includes an object information acquisition unit, a voice data acquisition unit, and a teacher data output unit, The object information acquisition unit acquires an object image obtained by capturing an image of an object, The voice data acquisition unit acquires voice data including a voice for confirming the object, The training data output unit outputs the voice data as a correct label for the object image, and outputs a combination of the object image and the voice data as training data for object identification.

[0007] The object identification trained model creating device of the present disclosure comprises: The system includes a training data acquisition unit and a trained model generation unit, The teacher data acquisition unit acquires, as teacher data for object identification, teacher data for object identification output by the teacher data generation device for object identification of the present disclosure, The trained model generation unit generates, through machine learning using the training data for object identification, an object identification model that outputs the type of object contained in the object image when the object image is input, as a trained model.

[0008] The object identification device of the present disclosure is The present invention includes an object information acquisition unit and an object identification unit, The object information acquisition unit acquires an object image obtained by capturing an image of an object, the object identification unit inputs the object image into an object identification model, identifies a type of object included in the object image, and outputs the identified type of object; The object identification model is The trained model is at least one of an object identification model that outputs the type of object contained in an object image when an object image is input by machine learning using the training data for object identification output by the training data generation device for object identification disclosed herein, and a trained model manufactured by the trained model manufacturing device for object identification disclosed herein.

[0009] The method for generating training data for object identification according to the present disclosure includes: The method includes an object information acquiring step, a voice data acquiring step, and a teacher data output step, The object information acquisition step acquires an object image by capturing an image of the object, The voice data acquisition step acquires voice data including a voice for confirming the object, The training data output step outputs the voice data as a correct label for the object image, and outputs a combination of the object image and the voice data as training data for object identification.

[0010] The method for creating a trained model for object identification disclosed herein includes the steps of: The method includes a training data acquisition process and a trained model generation process. The teacher data acquisition step acquires, as teacher data for object identification, teacher data for object identification output by the method for generating teacher data for object identification disclosed herein; The trained model generation process generates, through machine learning using the training data for object identification, an object identification model that outputs the type of object contained in the object image when the object image is input, as a trained model.

[0011] The object identification method of the present disclosure includes: The method includes an object information acquisition step and an object identification step, The object information acquisition step acquires an object image by capturing an image of the object, The object identification step includes inputting the object image into an object identification model, identifying a type of object included in the object image, and outputting the identified type of object; The object identification model is The trained model is at least one of an object identification model that outputs the type of object contained in an object image when an object image is input by machine learning using training data for object identification output by the method for generating training data for object identification disclosed herein, and a trained model produced by the method for producing a trained model for object identification disclosed herein.

[0012] The object identification teacher data generation program of the present disclosure includes: The method includes a step of acquiring object information, a step of acquiring voice data, and a step of outputting teacher data, The object information acquisition step acquires an object image by capturing an image of the object; The voice data acquisition step acquires voice data including a voice for confirming the object, the teacher data output step outputs the voice data as a correct label for the object image, and a combination of the object image and the voice data as teacher data for object identification; The present invention relates to a program for generating training data for object identification, which program causes a computer to execute each of the above procedures.

[0013] The object identification trained model creation program of the present disclosure includes: This includes the procedure for acquiring training data and the procedure for generating a trained model. The teacher data acquisition step includes acquiring, as teacher data for object identification, teacher data for object identification output by the teacher data generation program for object identification of the present disclosure; the trained model generation step generates, as a trained model, an object identification model that outputs a type of object included in an object image when the object image is input, by machine learning using the object identification training data; A program for creating a trained model for object identification that causes a computer to execute each of the above procedures.

[0014] The object identification program of the present disclosure is The method includes an object information acquisition step and an object identification step, The object information acquisition step acquires an object image by capturing an image of the object; The object identification step includes inputting the object image into an object identification model, identifying a type of object included in the object image, and outputting the identified type of object; The object identification model is At least one of a trained model that outputs a type of object included in an object image when the object image is input by machine learning using the object identification teacher data output by the object identification teacher data generation program of the present disclosure, and a trained model produced by the object identification trained model production program of the present disclosure, The present invention is an object identification program for causing a computer to execute each of the above procedures. Effect of the Invention

[0015] According to the present disclosure, training data for object identification can be easily generated. [Brief description of the drawings]

[0016] [Figure 1] FIG. 1 is a block diagram showing a configuration of an example of an object identification teacher data generating device according to the present disclosure. [Diagram 2] FIG. 2 is a block diagram showing an example of a hardware configuration of the object identification teacher data generating device of the present disclosure. [Diagram 3] FIG. 3 is a flowchart showing an example of processing in the object identification teacher data generating device of the present disclosure. [Figure 4] FIG. 4 is a block diagram showing a configuration of an example of an object classification trained model generating device according to the present disclosure. [Diagram 5]FIG. 5 is a block diagram showing an example of a hardware configuration of the object identification teacher data generating device of the present disclosure. [Figure 6] FIG. 6 is a flowchart showing an example of processing in the object identification trained model generating device of the present disclosure. [Figure 7] FIG. 7 is a block diagram showing a configuration of an example of an object identification device according to the present disclosure. [Figure 8] FIG. 8 is a block diagram showing an example of a hardware configuration of an object identification device according to the present disclosure. [Figure 9] FIG. 9 is a flowchart showing an example of processing in the object identification device of the present disclosure. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0017] In the present disclosure, the term "product" is not particularly limited and includes, for example, various items produced by manufacturing operations. Specific examples of the product include cooked food (dishes), boxed lunches, etc.

[0018] Hereinafter, an embodiment of the present disclosure will be described with reference to the drawings. The present disclosure is not limited to the following embodiment. In each of the drawings, the same parts are given the same reference numerals. In addition, the description of each embodiment can be mutually incorporated unless otherwise specified, and the configurations of each embodiment can be combined unless otherwise specified.

[0019] [Embodiment 1] Fig. 1 is a block diagram showing an example of the configuration of a teacher data generation device 10 for object identification (hereinafter also referred to as "the device") according to this embodiment. As shown in Fig. 1, the device 10 includes an object information acquisition unit 11, a voice data acquisition unit 12, and a teacher data output unit 13. In addition, the device 10 may include, for example, an input unit, an output unit, a display unit, and / or a storage unit, although these are not shown.

[0020] 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. Furthermore, the present device 10 may be, for example, a personal computer (PC, for example, desktop type or notebook type) in which the program of the present disclosure is installed, a smartphone, a tablet terminal, etc. The present device 10 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.

[0021] 2 illustrates a block diagram of a hardware configuration of the present device 10. The present 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, an output device 106, a communication device 107, etc. Each part of the present device 10 is connected to each other via the bus 103 by each interface (I / F).

[0022] The central processing unit 101 cooperates with other components through a controller (such as a system controller or an I / O controller) and controls the entire device 10. In the device 10, the central processing unit 101 executes, for example, the program disclosed herein and other programs, and also reads and writes various information. Specifically, for example, the central processing unit 101 functions as an object information acquisition unit 11, a voice data acquisition unit 12, and a teacher data output unit 13. The device 10 may include, as a computing device, other computing devices such as a CPU, a GPU (Graphics Processing Unit), an APU (Accelerated Processing Unit), or a combination of these.

[0023] The bus 103 can also be connected to, for example, an external device. Examples of the external device include a user terminal, 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.

[0024] 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 disclosure, 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).

[0025] 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 operating program including the program of the present disclosure. The storage device 104 may be, for example, a combination of a recording medium and a drive that reads and writes from 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 the recording medium and the drive are integrated, or a solid state drive (SSD).

[0026] In the present device 10, the memory 102 and the storage device 104 can 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. In this case, the memory 102 and the storage device 104 may store, for example, the above-mentioned information of the user of the present device. Note that at least a part 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 distributed and stored in a plurality of terminals using a block chain technology or the like. In addition, when the present device 10 includes the storage unit, for example, the memory 102 and the storage device 104 function as the storage unit. The storage unit can store, for example, an object image, voice data, object identification teacher data, object attribute information, an object identification model, and the like.

[0027] The device 10 further includes, for example, an input device 105 and an output device 106. The input device 105 may be, for example, a pointing device such as a touch panel, a track pad, or a mouse; a keyboard; an imaging means such as a camera or a scanner; a card reader such as an IC card reader or a magnetic card reader; and an audio input means such as a microphone. The output device 106 may be, for example, a display device such as an LED display or a liquid crystal display; an audio output device such as a speaker; a printer; or the like. In the first embodiment, the input device 105 and the output device 106 are configured separately, but the input device 105 and the output device 106 may be configured as an integrated device, such as a touch panel display.

[0028] Next, an example of the method for generating training data for object identification according to this embodiment will be described with reference to the flowchart of Fig. 3. The method for generating training data for object identification according to this embodiment is carried out as follows, for example, using the device 10 of Fig. 1 or Fig. 2. Note that the method for generating training data for object identification according to this embodiment is not limited to use of the device 10 of Fig. 1 or Fig. 2.

[0029] The object information acquisition unit 11 acquires an object image obtained by capturing an image of an object (S1, object information acquisition step). The object image may be, for example, a still image or a video. When the object image is a video, the video may include, for example, audio data described later. The object information acquisition unit 11 may acquire the object image from, for example, an imaging device that captured the object (for example, a camera, a video camera, an information processing terminal with a camera function (for example, a smartphone, a tablet terminal, a notebook computer, etc.)), or may acquire the object image from a recording medium that stores the object image (for example, various storage devices, server devices, databases, etc.), or may acquire the object image captured by an input device (for example, an imaging device) 105 of the device 10. The object information acquisition unit 11 may further acquire object attribute information of the object, for example. The object attribute information is not particularly limited as long as it is associated information related to the object, and any information may be used. A specific example of the object attribute information is information including at least one selected from the group consisting of the weight, contents, size, shape, and material of the object. The object information acquisition unit 11 may acquire the object attribute information by performing image analysis of the object image, or may acquire the object attribute information from an external sensor. Specific examples of the external sensor include a weight sensor such as a weighing scale, a length measurement sensor, a shape measurement sensor, and a material discrimination sensor. The object information acquisition unit 11 may record the acquired object image and object attribute information in a storage unit of the device 10.

[0030] The voice data acquisition unit 12 acquires voice data including a voice confirmation voice related to the object (S2, voice data acquisition step). The voice data acquisition unit 12 may acquire the voice data from, for example, a microphone (including a smartphone, tablet terminal, notebook computer, etc. equipped with a microphone) set at a manufacturing site of the object, or may acquire the voice data from a recording medium (for example, various storage devices, server devices, databases, etc.) in which the voice data is saved, or may acquire the voice data acquired by an input device (for example, a microphone) 105 of the device 10. When acquiring the voice data from the recording medium, it is preferable that the voice data and the object image are information acquired simultaneously in the recording medium.

[0031] The teacher data output unit 13 outputs the voice data as a correct label for the object image, and the combination of the object image and the voice data as teacher data for object identification (S3, teacher data output step). The teacher data output unit 13 may, for example, directly use the voice data as a correct label, or may analyze the voice data to extract the name of the object contained in the voice data and use the extracted name as the correct label. The output may be, for example, output (storage) to the memory 102 or storage device 104 of the device 10, or output to an external device via a communication line network. The external device may, for example, be an external storage device, or a device that uses the teacher data for object identification generated by the device 10, specifically, a device for producing a trained model for object identification or a device for object identification disclosed below.

[0032] The object identification training data generating device of the present disclosure can acquire an object image obtained by capturing an object using an object information acquiring unit, acquire voice data including a voice confirmation voice for the object using an audio data acquiring unit, and output a combination of the object image and the audio data as training data for object identification using the audio data as a correct answer label for the object image using a training data output unit. Therefore, according to the object identification training data generating device of the present disclosure, it is possible to generate training data for object identification without manually assigning a correct answer label to the object image. Therefore, according to the present disclosure, it is possible to easily generate training data for object identification.

[0033] [Embodiment 2] Fig. 4 is a block diagram showing an example of a configuration of a trained model manufacturing device 20 for object identification (hereinafter also referred to as "this device") according to this embodiment. As shown in Fig. 4, this device 20 includes a teacher data acquisition unit 21 and a trained model generation unit 22. In addition, this device 20 may include, for example, an input unit, an output unit, a display unit and / or a storage unit, although these are not shown.

[0034] The device 20 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 20 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 20 may be incorporated into a server as a system, for example. Furthermore, the present device 20 may be, for example, a personal computer (PC, for example, desktop type or notebook type) in which the program of the present disclosure is installed, a smartphone, a tablet terminal, etc. The present 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.

[0035] 5 illustrates a block diagram of a hardware configuration of the present device 20. The present device 20 includes, for example, a central processing unit (CPU, GPU, etc.) 201, a memory 202, a bus 203, a storage device 204, an input device 205, an output device 206, a communication device 207, etc. Each part of the present device 20 is connected to each other via a bus 103 by each interface (I / F).

[0036] The central processing unit 201 cooperates with other components through a controller (such as a system controller or an I / O controller) and controls the entire device 20. In the device 20, the central processing unit 201 executes, for example, the program disclosed herein and other programs, and also reads and writes various information. Specifically, for example, the central processing unit 201 functions as a teacher data acquisition unit 21 and a trained model generation unit 22. The device 20 may include, as a computing device, other computing devices such as a CPU, a GPU (Graphics Processing Unit), an APU (Accelerated Processing Unit), or a combination of these.

[0037] The bus 203 can also be connected to, for example, an external device. Examples of the external device include a user terminal, 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 20 can be connected to an external network (the above-mentioned communication line network) by, for example, a communication device 207 connected to the bus 203, and can also be connected to other devices via the external network.

[0038] The memory 202 may be, for example, a main memory (primary storage device). When the central processing unit 201 performs processing, the memory 202 reads various operation programs, such as the program of the present disclosure, stored in the storage device 204 described below, and the central processing unit 201 receives data from the memory 202 and executes the program. The main memory may be, for example, a RAM (random access memory). The memory 202 may also be, for example, a ROM (read only memory).

[0039] The storage device 204 is also referred to as a so-called auxiliary storage device, for example, in contrast to the main memory (primary storage device). As described above, the storage device 204 stores an operating program including the program of the present disclosure. The storage device 204 may be, for example, a combination of a recording medium and a drive that reads and writes from the recording medium. The recording medium is not particularly limited, and may be, for example, an internal or external type, such as a HD (hard disk), CD-ROM, CD-R, CD-RW, MO, DVD, flash memory, and memory card. The storage device 204 may be, for example, a hard disk drive (HDD) in which the recording medium and the drive are integrated, and a solid state drive (SSD).

[0040] In the present device 20, the memory 202 and the storage device 204 can 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 20, and information used when the present device 20 executes processing. In this case, the memory 202 and the storage device 204 may store, for example, the above-mentioned information of the user of the present device. Note that at least a part of the information may be stored, for example, in an external server other than the memory 202 and the storage device 204, or may be distributed and stored in a plurality of terminals using a block chain technology or the like. In addition, when the present device 20 includes the storage unit, for example, the memory 202 and the storage device 204 function as the storage unit. The storage unit can store, for example, an object image, voice data, object identification teacher data, object attribute information, an object identification model, and the like.

[0041] The device 20 further includes, for example, an input device 205 and an output device 206. The input device 205 may be, for example, a pointing device such as a touch panel, a track pad, or a mouse; a keyboard; an imaging means such as a camera or a scanner; a card reader such as an IC card reader or a magnetic card reader; and an audio input means such as a microphone. The output device 206 may be, for example, a display device such as an LED display or a liquid crystal display; an audio output device such as a speaker; a printer; or the like. In the first embodiment, the input device 205 and the output device 206 are configured separately, but the input device 205 and the output device 206 may be configured as an integrated device, such as a touch panel display.

[0042] Next, an example of the method for producing a trained model for object identification according to this embodiment will be described with reference to the flowchart of Fig. 6. The method for producing a trained model for object identification according to this embodiment is carried out as follows, for example, using the device 20 of Fig. 4 or Fig. 5. Note that the method for producing a trained model for object identification according to this embodiment is not limited to use of the device 20 of Fig. 4 or Fig. 5.

[0043] First, the teacher data acquisition unit 21 acquires the teacher data for object identification output by the device for generating teacher data for object identification disclosed herein as the teacher data for object identification (S11, teacher data acquisition step). The teacher data acquisition unit 21 may acquire the teacher data for object identification from the device for generating teacher data for object identification disclosed herein via the communication line network, for example, or may acquire the teacher data for object identification from an external storage device in which the teacher data for object identification is stored.

[0044] Next, the trained model generating unit 22 generates an object identification model that outputs the type of object included in the object image when the object image is inputted, as a trained model, by machine learning using the object identification teacher data (S12, learning step). The machine learning is not particularly limited, and is, for example, machine learning using a neural network such as a convolution neural network (CNN), a support vector machine (SVM), a Bayesian network, a regression tree, etc. The machine learning using the CNN is not particularly limited, and is, for example, semantic segmentation (IS), instance segmentation (IS), single shot detector (SSD), weighted single shot detector (WSSD), etc. In addition, the trained model generating unit 22 may generate a trained model (derived model) by retraining, for example, using the object identification teacher data and a trained model that has already been generated. Furthermore, the trained model generation unit 22 may generate a trained model obtained by transfer learning using a trained model generated using the training data for object identification, or may generate the trained model by model compression of the trained model generated using the training data for object identification.

[0045] The trained model generated by the present embodiment is used, for example, in an object identification device described later, which enables identification of the type of object using an object image obtained by capturing an object.

[0046] [Embodiment 3] Fig. 7 is a block diagram showing an example of the configuration of an object identification device 30 (hereinafter also referred to as "the device") according to this embodiment. As shown in Fig. 7, the device 30 includes an object information acquisition unit 31 and an object identification unit 32. Furthermore, the device 30 may include, for example, an input unit, an output unit, a display unit and / or a storage unit, although these are not shown.

[0047] The device 30 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 30 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 (Ad Hoc communication), an infrastructure (infrastructure communication), and an indirect communication via an access point. The device 30 may be incorporated into a server as a system, for example. Furthermore, the present device 30 may be, for example, a personal computer (PC, for example, desktop type or notebook type) in which the program of the present disclosure is installed, a smartphone, a tablet terminal, etc. The present device 30 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.

[0048] 8 illustrates a block diagram of a hardware configuration of the present device 30. The present device 30 includes, for example, a central processing unit (CPU, GPU, etc.) 301, a memory 302, a bus 303, a storage device 304, an input device 305, an output device 306, a communication device 307, etc. Each part of the present device 30 is connected to each other via the bus 303 by each interface (I / F).

[0049] The central processing unit 301 cooperates with other components through a controller (such as a system controller or an I / O controller) and is responsible for overall control of the present device 30. In the present device 30, the central processing unit 301 executes, for example, the program of the present disclosure and other programs, and also reads and writes various information. Specifically, for example, the central processing unit 301 functions as an object information acquisition unit 31 and an object identification unit 32. The present device 30 may include, as a computing device, other computing devices such as a CPU, a GPU (Graphics Processing Unit), an APU (Accelerated Processing Unit), or a combination of these.

[0050] The bus 303 can also be connected to, for example, an external device. Examples of the external device include a user terminal, 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 30 can be connected to an external network (the above-mentioned communication line network) by, for example, a communication device 307 connected to the bus 303, and can also be connected to other devices via the external network.

[0051] The memory 302 may be, for example, a main memory (primary storage device). When the central processing unit 301 performs processing, the memory 302 reads various operation programs, such as the program of the present disclosure, stored in the storage device 304 described below, and the central processing unit 301 receives data from the memory 302 and executes the program. The main memory may be, for example, a RAM (random access memory). The memory 302 may also be, for example, a ROM (read only memory).

[0052] The storage device 304 is also called an auxiliary storage device, for example, in contrast to the main memory (primary storage device). As described above, the storage device 304 stores an operating program including the program of the present disclosure. The storage device 304 may be, for example, a combination of a recording medium and a drive that reads and writes from 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, and memory card. The storage device 304 may be, for example, a hard disk drive (HDD) in which the recording medium and the drive are integrated, and a solid state drive (SSD).

[0053] In the present device 30, the memory 302 and the storage device 304 can 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 30, and information used when the present device 30 executes processing. In this case, the memory 302 and the storage device 304 may store, for example, the above-mentioned information of the user of the present device. Note that at least a part of the information may be stored, for example, in an external server other than the memory 302 and the storage device 304, or may be distributed and stored in a plurality of terminals using blockchain technology or the like. In addition, when the present device 30 includes the storage unit, for example, the memory 302 and the storage device 304 function as the storage unit. The storage unit can store, for example, an object image, voice data, object identification teacher data, object attribute information, an object identification model, and the like.

[0054] The device 30 further includes, for example, an input device 305 and an output device 306. The input device 305 may be, for example, a pointing device such as a touch panel, a track pad, or a mouse; a keyboard; an imaging means such as a camera or a scanner; a card reader such as an IC card reader or a magnetic card reader; and an audio input means such as a microphone. The output device 306 may be, for example, a display device such as an LED display or a liquid crystal display; an audio output device such as a speaker; a printer; or the like. In the first embodiment, the input device 305 and the output device 306 are configured separately, but the input device 305 and the output device 306 may be configured as an integrated device, such as a touch panel display.

[0055] Next, an example of the object identification method of this embodiment will be described with reference to the flowchart of Fig. 9. The object identification method of this embodiment is carried out as follows, for example, using the device 30 of Fig. 7 or Fig. 8. Note that the object identification method of this embodiment is not limited to the use of the device 30 of Fig. 7 or Fig. 8.

[0056] The object information acquisition unit 31 acquires an object image by capturing an image of an object (S21, object information acquisition step). The object image may be, for example, a still image or a moving image. When the object image is a moving image, the moving image may include, for example, audio data, which will be described later. The object information acquisition unit 31 may acquire the object image from, for example, an imaging device (for example, a camera, a video camera, an information processing terminal with a camera function (for example, a smartphone, a tablet terminal, a notebook computer, etc.)) that images the object, or may acquire the object image from a recording medium (for example, various storage devices, server devices, databases, etc.) that stores the object image, or may acquire the object image captured by an input device (for example, an imaging device) 305 of the present device 30. The object information acquisition unit 31 may further acquire object attribute information of the object, for example. The object attribute information is not particularly limited as long as it is associated information related to the object, and may be any information. A specific example of the object attribute information is information including at least one selected from the group consisting of the weight, contents, size, shape, and material of the object. The object information acquisition unit 31 may acquire the object attribute information by performing image analysis on the object image, or may acquire the object attribute information from an external sensor. Specific examples of the external sensor include a weight sensor such as a weighing scale, a length measurement sensor, a shape measurement sensor, a material discrimination sensor, etc. The object information acquisition unit 31 may record the acquired object image and object attribute information in a storage unit of the present device 30, for example.

[0057] The object identification unit 32 inputs the object image into an object identification model, identifies the type of object contained in the object image, and outputs the identified type of object (S22, object identification step). The object identification model is, for example, at least one of a trained object identification model that outputs the type of object contained in the object image when the object image is input by machine learning using the object identification training data output by the object identification training data generation device 10 of the first embodiment, and a trained model manufactured by the object identification trained model manufacturing device 20 of the second embodiment.

[0058] The object identification model includes, for example, an input layer for inputting an object image, an output layer for outputting the type of the object, and at least one intermediate layer provided between the input layer and the output layer. The object identification model may be a program module that is a part of artificial intelligence software. An example of the multi-layered network is a neural network. An example of the neural network is a convolution neural network (CNN), but is not limited to CNN, and may be a trained model constructed with other learning algorithms such as a neural network other than CNN, a support vector machine (SVM), a Bayesian network, or a regression tree.

[0059] The object identification model can be generated by machine learning using, for example, teacher data generated by the object identification teacher data generation device of the present disclosure. The object identification model may be, for example, a trained model generated in advance. The trained model may be a trained model (derived model) retrained using the object identification teacher data and a trained model that has already been generated. Furthermore, the trained model may be a trained model obtained by transfer learning using a trained model generated using object identification teacher data, or a trained model generated by model compression of a trained model generated using object identification teacher data.

[0060] The object identification device disclosed herein is capable of identifying objects using an object identification model generated by machine learning using teacher data for object identification generated by the object identification teacher data generation device disclosed herein.

[0061] [Embodiment 4] The object identification teacher data generation program of this embodiment is a program for causing a computer to execute each step of the object identification teacher data generation method described above. Specifically, the object identification teacher data generation program of this embodiment is a program for causing a computer to execute an object information acquisition procedure, a voice data acquisition procedure, and a teacher data output procedure.

[0062] The object information acquisition step acquires an object image by capturing an image of the object; The voice data acquisition step acquires voice data including a voice for confirming the object, The training data output step outputs the voice data as a correct label for the object image, and a combination of the object image and the voice data as training data for object identification.

[0063] The object identification training data generation program of this embodiment can also be said to be a program that causes a computer to function as an object information acquisition procedure, a voice data acquisition procedure, and a training data output procedure.

[0064] The object identification teacher data generation program of this embodiment can be used in conjunction with the description of the object identification teacher data generation device and object identification teacher data generation method of the present disclosure. For each of the steps, for example, "step" can be read as "processing". In addition, the object identification teacher data generation program of this embodiment may be recorded, for example, on a computer-readable recording medium. The recording medium is, for example, a non-transitory computer-readable storage medium. The recording medium is, for example, a non-transitory computer-readable storage medium. The recording medium is, for example, a non-transitory computer-readable storage medium. The recording medium is not particularly limited, and examples thereof include random access memory (RAM), read-only memory (ROM), hard disk (HD), flash memory (for example, SSD (Solid State Drive), USB flash memory, SD / SDHC card, etc.), optical disk (for example, CD-R / CD-RW, DVD-R / DVD-RW, BD-R / BD-RE, etc.), magneto-optical disk (MO), floppy (registered trademark) disk (FD), etc. Furthermore, the object identification teacher data generation program of this embodiment (also referred to as, for example, a programming product or an object identification teacher data generation program product) may be in a form distributed, for example, from an external computer. The "distribution" may be, for example, distribution via a communication network, or distribution via a device connected by wire. The object identification teacher data generation program of this embodiment may be installed and executed on the device to which it is distributed, or may be executed without being installed.

[0065] [Embodiment 5] The object identification trained model creation program of this embodiment is a program for causing a computer to execute each step of the object identification trained model creation method described above. Specifically, the object identification trained model creation program of this embodiment is a program for causing a computer to execute a teacher data acquisition procedure and a trained model generation procedure.

[0066] The teacher data acquisition step includes acquiring, as teacher data for object identification, teacher data for object identification output by the object identification teacher data generation device of the present disclosure, The trained model generation procedure generates, through machine learning using the training data for object identification, an object identification model that outputs the type of object contained in the object image when the object image is input, as a trained model.

[0067] In addition, the trained model creation program for object identification in this embodiment can also be said to be a program that causes a computer to function as a teacher data acquisition procedure and a trained model generation procedure.

[0068] The object identification trained model manufacturing program of this embodiment can be used in conjunction with the description of the object identification trained model manufacturing device and the object identification trained model manufacturing method of the present disclosure. For each procedure, for example, the "procedure" can be read as "processing". In addition, the object identification trained model manufacturing program of this embodiment may be recorded, for example, in a computer-readable recording medium. The recording medium is, for example, a non-transitory computer-readable storage medium. The recording medium is, for example, a non-transitory computer-readable storage medium. The recording medium is, for example, a non-transitory computer-readable storage medium. The recording medium is not particularly limited, and examples thereof include random access memory (RAM), read-only memory (ROM), hard disk (HD), flash memory (for example, SSD (Solid State Drive), USB flash memory, SD / SDHC card, etc.), optical disk (for example, CD-R / CD-RW, DVD-R / DVD-RW, BD-R / BD-RE, etc.), magneto-optical disk (MO), floppy (registered trademark) disk (FD), etc. Furthermore, the trained model manufacturing program for object identification of this embodiment (also referred to as, for example, a programming product or a trained model manufacturing program product for object identification) may be in a form distributed, for example, from an external computer. The "distribution" may be, for example, distribution via a communication network, or distribution via a device connected via a wire. The trained model manufacturing program for object identification of this embodiment may be installed and executed on the device to which it is distributed, or may be executed without being installed.

[0069] [Embodiment 6] The object identification program of the present embodiment is a program for causing a computer to execute each step of the object identification method described above. Specifically, the object identification program of the present embodiment is a program for causing a computer to execute an object information acquisition procedure and an object identification procedure.

[0070] The object information acquisition step acquires an object image by capturing an image of the object; The object identification step includes inputting the object image into an object identification model, identifying a type of object included in the object image, and outputting the identified type of object; The object identification model is The trained model is an object identification model that outputs the type of object contained in an object image when the object image is input, through machine learning using the training data for object identification output by the training data generation device for object identification disclosed herein, and / or a trained model produced by the object identification device disclosed herein.

[0071] The object identification program of this embodiment can also be said to be a program that causes a computer to function as an object information acquisition procedure and an object identification procedure.

[0072] The object identification program of this embodiment can be implemented by using the description of the object identification device and the object identification method of the present disclosure. For example, the "procedure" in each of the steps can be read as "processing". The object identification program of this embodiment may be recorded in a computer-readable recording medium. The recording medium is, for example, a non-transitory computer-readable storage medium. The recording medium is, for example, a non-transitory computer-readable storage medium. The recording medium is, for example, a non-transitory computer-readable storage medium. The recording medium is not particularly limited, and examples thereof include random access memory (RAM), read-only memory (ROM), hard disk (HD), flash memory (e.g., SSD (Solid State Drive), USB flash memory, SD / SDHC card, etc.), optical disk (e.g., CD-R / CD-RW, DVD-R / DVD-RW, BD-R / BD-RE, etc.), magneto-optical disk (MO), floppy (registered trademark) disk (FD), etc. Furthermore, the object identification program of this embodiment (e.g., also referred to as a programming product or an object identification program product) may be in a form distributed from an external computer, for example. The "distribution" may be, for example, distribution via a communication network, or distribution via a device connected by wire. The object identification program of this embodiment may be installed and executed on the device to which it is distributed, or may be executed without being installed.

[0073] Although the present disclosure has been described above with reference to the embodiments, the present disclosure is not limited to the above-mentioned embodiments. Various modifications that can be understood by a person skilled in the art can be made to the configuration and details of the present disclosure within the scope of the present disclosure. Furthermore, each embodiment can be appropriately combined with other embodiments.

[0074] <Additional Notes> A part or all of the above-described embodiments can be described as follows, but is not limited to the following. (Appendix 1) The apparatus includes an object information acquisition unit, a voice data acquisition unit, and a teacher data output unit, The object information acquisition unit acquires an object image obtained by capturing an image of an object, The voice data acquisition unit acquires voice data including a voice for confirming the object, The teacher data output unit outputs the combination of the object image and the voice data as teacher data for object identification using the voice data as a correct label for the object image. (Appendix 2) The teacher data output unit analyzes the voice data to extract the name of the object contained in the voice data, and sets the extracted name as the correct label. (Appendix 3) the object information acquisition unit acquires object attribute information of the object in association with the object image; 3. The device for generating teacher data for object identification described in claim 1 or 2, wherein the teacher data output unit outputs the object attribute information by linking it to the teacher data for object identification. (Appendix 4) A device for generating training data for object identification as described in Appendix 3, wherein the object attribute information includes at least one selected from the group consisting of the weight, size, shape, and material of the object. (Appendix 5) The system includes a training data acquisition unit and a trained model generation unit, The teacher data acquisition unit acquires, as the teacher data for object identification, teacher data for object identification output by a teacher data generation device for object identification described in any one of Supplementary Notes 1 to 4; The trained model generation unit is a trained model manufacturing device for object identification that generates, as a trained model, an object identification model that outputs the type of object contained in the object image when the object image is input, through machine learning using the training data for object identification. (Appendix 6) The present invention includes an object information acquisition unit and an object identification unit, The object information acquisition unit acquires an object image obtained by capturing an image of an object, the object identification unit inputs the object image into an object identification model, identifies a type of object included in the object image, and outputs the identified type of object; The object identification model is An object identification device which is at least one of a trained model that outputs the type of object contained in an object image when an object image is input by machine learning using training data for object identification output by a training data generation device for object identification described in any of Supplementary Notes 1 to 4, and a trained model manufactured by the trained model manufacturing device for object identification described in Supplementary Note 5. (Appendix 7) the object information acquisition unit acquires object attribute information of the object in association with the object image; 7. The object identification device according to claim 6, wherein the object identification unit inputs the object image and the object attribute information into the object identification model to identify the type of object contained in the object image. (Appendix 8) The method includes an object information acquiring step, a voice data acquiring step, and a teacher data output step, The object information acquisition step acquires an object image by capturing an image of the object, The voice data acquisition step acquires voice data including a voice for confirming the object, The teacher data output step of the method for generating teacher data for object identification outputs the voice data as a correct label for the object image, and outputs a combination of the object image and the voice data as teacher data for object identification. (Appendix 9) The method for generating teacher data for object identification described in Appendix 8, wherein the teacher data output process analyzes the voice data to extract the name of the object contained in the voice data, and sets the extracted name as the correct label. (Appendix 10) The object information acquisition step acquires object attribute information of the object in association with the object image, 10. The method for generating training data for object identification described in claim 8 or 9, wherein the training data output process outputs the object attribute information by linking it to the training data for object identification. (Appendix 11) A method for generating training data for object identification as described in Appendix 10, wherein the object attribute information includes at least one selected from the group consisting of the weight, size, shape, and material of the object. (Appendix 12) The method includes a training data acquisition process and a trained model generation process. The teacher data acquisition step acquires, as the teacher data for object identification, teacher data for object identification output by a teacher data generation method for object identification described in any one of Supplementary Notes 8 to 11; The trained model generation process is a method for manufacturing a trained model for object identification, which generates an object identification model as a trained model by machine learning using the training data for object identification, when the object image is input, which outputs the type of object contained in the object image. (Appendix 13) The method includes an object information acquisition step and an object identification step, The object information acquisition step acquires an object image by capturing an image of the object, The object identification step includes inputting the object image into an object identification model, identifying a type of object included in the object image, and outputting the identified type of object; The object identification model is An object identification method, comprising: at least one of a trained model that outputs the type of object contained in an object image when an object image is input, by machine learning using training data for object identification output by a method for generating training data for object identification described in any one of Appendices 8 to 11; and a trained model manufactured by a method for manufacturing a trained model for object identification described in Appendices 12. (Appendix 14) The object information acquisition step acquires object attribute information of the object in association with the object image, The object identification method according to claim 13, wherein the object identification step inputs the object image and the object attribute information into the object identification model to identify the type of object contained in the object image. (Appendix 15) The method includes a step of acquiring object information, a step of acquiring voice data, and a step of outputting teacher data, The object information acquisition step acquires an object image by capturing an image of the object; The voice data acquisition step acquires voice data including a voice for confirming the object, the teacher data output step outputs the voice data as a correct label for the object image, and a combination of the object image and the voice data as teacher data for object identification; A program for generating training data for object identification, for causing a computer to execute each of the above procedures. (Appendix 16) The teacher data generation program for object identification described in Appendix 15, wherein the teacher data output procedure analyzes the voice data to extract the name of the object contained in the voice data, and sets the extracted name as the correct label. (Appendix 17) the object information acquisition step acquires object attribute information of the object in association with the object image; The object identification teacher data generation program according to claim 15 or 16, wherein the teacher data output step outputs the object attribute information by linking it to the object identification teacher data. (Appendix 18) A training data generation program for object identification as described in Appendix 17, wherein the object attribute information includes at least one selected from the group consisting of the weight, size, shape, and material of the object. (Appendix 19) This includes the procedure for acquiring training data and the procedure for generating a trained model. The teacher data acquisition step includes acquiring, as teacher data for object identification, teacher data for object identification output by a teacher data generation program for object identification described in any one of appendices 15 to 18; the trained model generation step generates, as a trained model, an object identification model that outputs a type of object included in an object image when the object image is input, by machine learning using the object identification training data; A program for creating a trained model for object identification, for causing a computer to execute each of the above procedures. (Appendix 20) The method includes an object information acquisition step and an object identification step, The object information acquisition step acquires an object image by capturing an image of the object; The object identification step includes inputting the object image into an object identification model, identifying a type of object included in the object image, and outputting the identified type of object; The object identification model is At least one of a trained model that outputs a type of object contained in an object image when the object image is input by machine learning using training data for object identification output by a training data generation program for object identification described in any one of Supplementary Notes 15 to 18, and a trained model produced by a trained model production program for object identification described in Supplementary Note 19, An object identification program for causing a computer to execute each of the above procedures. (Appendix 21) the object information acquisition step acquires object attribute information of the object in association with the object image; 21. The object identification program according to claim 20, wherein the object identification step inputs the object image and the object attribute information into the object identification model to identify the type of object contained in the object image. (Appendix 22) The method includes a step of acquiring object information, a step of acquiring voice data, and a step of outputting teacher data, The object information acquisition step acquires an object image by capturing an image of the object; The voice data acquisition step acquires voice data including a voice for confirming the object, the teacher data output step outputs the voice data as a correct label for the object image, and a combination of the object image and the voice data as teacher data for object identification; A computer-readable recording medium having recorded thereon a training data generation program for object identification for causing a computer to execute each of the above procedures. (Appendix 23) A recording medium for generating teacher data for object identification as described in Appendix 22, wherein the teacher data output procedure analyzes the voice data to extract the name of the object contained in the voice data, and sets the extracted name as the correct label. (Appendix 24) the object information acquisition step acquires object attribute information of the object in association with the object image; A recording medium for generating training data for object identification described in Appendix 22 or 23, wherein the training data output step outputs the object attribute information by linking it to the training data for object identification. (Appendix 25) A recording medium for generating training data for object identification as described in Appendix 24, wherein the object attribute information includes at least one selected from the group consisting of the weight, size, shape, and material of the object. (Appendix 26) This includes the procedure for acquiring training data and the procedure for generating a trained model. The teacher data acquisition step includes acquiring, as teacher data for object identification, teacher data for object identification output by a teacher data generation program for object identification described in any one of appendices 15 to 18; the trained model generation step generates, as a trained model, an object identification model that outputs a type of object included in an object image when the object image is input, by machine learning using the object identification training data; A computer-readable recording medium having recorded thereon a program for creating a trained model for object identification for causing a computer to execute each of the above procedures. (Appendix 27) The method includes an object information acquisition step and an object identification step, The object information acquisition step acquires an object image by capturing an image of the object; The object identification step includes inputting the object image into an object identification model, identifying a type of object included in the object image, and outputting the identified type of object; The object identification model is At least one of a trained model that outputs a type of object contained in an object image when the object image is input by machine learning using training data for object identification output by a training data generation program for object identification described in any one of Supplementary Notes 15 to 18, and a trained model produced by a trained model production program for object identification described in Supplementary Note 19, A computer-readable recording medium having recorded thereon an object identification program for causing a computer to execute each of the above procedures. (Appendix 28) the object information acquisition step acquires object attribute information of the object in association with the object image; 28. The recording medium according to claim 27, wherein the object identification step inputs the object image and the object attribute information into the object identification model to identify the type of object contained in the object image. [Industrial Applicability]

[0075] According to the present disclosure, training data for object identification can be generated easily. The present disclosure is widely useful in fields that utilize object identification using images. [Explanation of symbols]

[0076] 10. Apparatus for generating training data for object identification 11 Object information acquisition unit 12 Audio data acquisition unit 13 Teacher data output section 101 Central Processing Unit 102 Memory 103 Bus 104 Storage device 105 Input Device 106 Output Device 107 Communication Devices 20. Device for creating trained models for object identification 21 Teacher Data Acquisition Department 22 Trained model generation unit 201 Central Processing Unit 202 Memory 203 Bus 204 Storage device 205 Input Device 206 Output Device 207 Communication Devices 30 Object Identification Device 31 Object information acquisition unit 32 Object Identification Unit 301 Central Processing Unit 302 Memory 303 Bus 304 Storage device 305 Input Device 306 Output Device

Claims

1. The apparatus includes an object information acquisition unit, a voice data acquisition unit, and a teacher data output unit, The object information acquisition unit acquires an object image obtained by capturing an image of an object, The voice data acquisition unit acquires voice data including a voice for confirming the object, The teacher data output unit outputs the combination of the object image and the voice data as teacher data for object identification using the voice data as a correct label for the object image.

2. 2. The device for generating training data for object identification according to claim 1, wherein the training data output unit analyzes the voice data to extract a name of the object contained in the voice data, and sets the extracted name as the correct label.

3. The system includes a training data acquisition unit and a trained model generation unit, the teacher data acquisition unit acquires, as the teacher data for object identification, the teacher data for object identification output by the device for generating teacher data for object identification according to claim 1 or 2; The trained model generation unit is a trained model manufacturing device for object identification that generates, as a trained model, an object identification model that outputs the type of object contained in the object image when the object image is input, through machine learning using the training data for object identification.

4. The present invention includes an object information acquisition unit and an object identification unit, The object information acquisition unit acquires an object image obtained by capturing an image of an object, the object identification unit inputs the object image into an object identification model, identifies a type of object included in the object image, and outputs the identified type of object; The object identification model is An object identification device which is at least one of a trained model that outputs the type of object contained in an object image when an object image is input by machine learning using training data for object identification output by the object identification training data generation device described in claim 1 or 2, and a trained model manufactured by the object identification trained model manufacturing device described in claim 3.

5. The method includes an object information acquiring step, a voice data acquiring step, and a teacher data output step, The object information acquisition step acquires an object image by capturing an image of the object, The voice data acquisition step acquires voice data including a voice for confirming the object, The teacher data output step of the method for generating teacher data for object identification outputs the voice data as a correct label for the object image, and outputs a combination of the object image and the voice data as teacher data for object identification.

6. The method includes a training data acquisition process and a trained model generation process. The teacher data acquisition step acquires, as the teacher data for object identification, teacher data for object identification output by a method for generating teacher data for object identification according to claim 5, The trained model generation process is a method for manufacturing a trained model for object identification, which generates, as a trained model, an object identification model that outputs the type of object contained in the object image when the object image is input, by machine learning using the training data for object identification.

7. The method includes an object information acquisition step and an object identification step, The object information acquisition step acquires an object image by capturing an image of the object, The object identification step includes inputting the object image into an object identification model, identifying a type of object included in the object image, and outputting the identified type of object; The object identification model is An object identification method, comprising: at least one of a trained model, which is an object identification model that outputs the type of object contained in an object image when the object image is input, by machine learning using teacher data for object identification output by the method for generating teacher data for object identification described in claim 5; and a trained model manufactured by the method for manufacturing a trained model for object identification described in claim 6.

8. The method includes a step of acquiring object information, a step of acquiring voice data, and a step of outputting teacher data, The object information acquisition step acquires an object image by capturing an image of the object; The voice data acquisition step acquires voice data including a voice for confirming the object, the teacher data output step outputs the voice data as a correct label for the object image, and a combination of the object image and the voice data as teacher data for object identification; A program for generating training data for object identification, for causing a computer to execute each of the above procedures.

9. This includes the procedure for acquiring training data and the procedure for generating a trained model. The teacher data acquisition step includes acquiring, as the teacher data for object identification, teacher data for object identification output by a teacher data generation program for object identification according to claim 8, the trained model generation step generates, as a trained model, an object identification model that outputs a type of object included in an object image when the object image is input, by machine learning using the object identification training data; A program for creating a trained model for object identification, for causing a computer to execute each of the above procedures.

10. The method includes an object information acquisition step and an object identification step, The object information acquisition step acquires an object image by capturing an image of the object; The object identification step includes inputting the object image into an object identification model, identifying a type of object included in the object image, and outputting the identified type of object; The object identification model is a trained model that outputs a type of an object included in an object image when the object image is input by machine learning using the object identification training data output by the object identification training data generation program according to claim 8, and / or a trained model produced by the object identification trained model production program according to claim 9, An object identification program for causing a computer to execute each of the above procedures.

Images