Learning device, estimation device, estimation system, display device, display system, learning method, estimation method and computer program

Abstract

To express an emotion by a color reflecting an individual difference.SOLUTION: An estimation system includes a first emotion estimation unit, a learning unit, a second emotion estimation unit and a color estimation unit. The first emotion estimation unit estimates an emotion of a model generation object person on the basis of biological information obtained by measuring the model generation object person and changing according to the emotion. The learning unit uses estimated emotion information and information of a color selected by the model generation object person as the color representing the emotion when measuring the biological information to learn a color estimation model representing a correspondence between the emotion information and color information for every model generation object person. The second emotion estimation unit estimates an emotion of an estimation object person on the basis of biological information obtained by measuring the estimation object person and changing according to the emotion. The color estimation unit estimates color information corresponding to the estimated emotion information of the estimation object person using a color estimation model trained on the model generation object person who is the same as or different from the estimation object person.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a learning device, an estimation device, an estimation system, a display device, a display system, a learning method, an estimation method, and a computer program. [Background technology]

[0002] Currently, the shortage of care workers has become a social issue, and in the future, an increase in the elderly population and burnout due to emotional labor are expected to lead to an increase in the turnover rate. However, it is not easy to quickly formulate sufficient countermeasures to address this issue. One of the causes of this increase in turnover is thought to be a lack of communication in caregiving.

[0003] Conventional technologies for achieving smooth communication with elderly people include robots, voice systems, and image display systems (see, for example, Patent Documents 1 and 2), but these conventional technologies lack a process for smooth communication that takes into account the feelings and thoughts of the care recipient and the caregiver, and may not lead to care that is considerate to the care recipient. Therefore, as a technology to support smoother communication, there is one that outputs information that represents the feelings estimated based on the biological state of the person being cared for (see, for example, Patent Documents 3 and 4). [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Laid-Open No. 2008-158697 [Patent Document 2] Japanese Patent Application Laid-Open No. 2003-108362 [Patent Document 3] Japanese Patent Application Publication No. 2020-185138 [Patent Document 4] Japanese Patent Application Publication No. 2019-17499 Summary of the Invention [Problem to be solved by the invention]

[0005] In conventional technologies such as Patent Document 3, estimated emotions are expressed by colors. However, even for the same emotion, the color that expresses that emotion varies from person to person. In the conventional technologies, the same emotion is expressed using the same color regardless of the individual.

[0006] The present invention has been made in view of the above circumstances, and provides a technique that makes it possible to express emotions using colors that reflect individual differences. [Means for solving the problem]

[0007] One aspect of the present invention is a learning device that includes: an emotion estimation unit that acquires emotion information as an estimation result from an emotion estimation unit that estimates the emotion of a model generation subject based on biometric information that is obtained by measuring the model generation subject and that changes depending on the emotion; and a learning unit that learns, for each model generation subject, a color estimation model that represents the correspondence between emotion information and color information, using the estimated emotion information and information about a color selected by the model generation subject as a color that represents the emotion at the time the biometric information was measured.

[0008] One aspect of the present invention is a learning device that includes a learning unit that learns a color estimation model that represents the correspondence between biometric information and color information for each model generation subject, based on biometric information obtained by measuring the model generation subject and that changes depending on emotions, and information on a color selected by the model generation subject as a color that represents the emotion at the time the biometric information was measured.

[0009] One aspect of the present invention is an estimation device that includes an emotion estimation unit that estimates the emotion of a subject to be estimated based on biometric information obtained by measuring the subject and that changes depending on the emotion, and a color estimation unit that estimates color information corresponding to the estimated emotion information using a color estimation model that represents a correspondence between emotion information and color information, wherein the color estimation model is trained for each subject to be model generation using information on a plurality of different emotions and information on colors selected by a subject to be model generation that is the same as or different from the subject to be estimated as colors that represent each of the plurality of different emotions.

[0010] One aspect of the present invention is an estimation device that includes a color estimation unit that estimates color information corresponding to biometric information obtained by measuring an estimation subject using a color estimation model that represents the correspondence between biometric information that changes in response to emotions and color information, and the color estimation model is learned for each model generation subject using biometric information obtained by measuring a model generation subject that is the same as or different from the estimation subject at different times, and information on colors selected by the model generation subject as colors that represent emotions at each time the biometric information was measured.

[0011] In one aspect of the present invention, the above-mentioned estimation device further includes an information control unit that performs the following processes: when the estimation subject is a first estimation subject and the model generation subject is a second estimation subject, the color information estimated by the color estimation unit is displayed on a display device corresponding to the first estimation subject; and when the estimation subject is the second estimation subject and the model generation subject is the first estimation subject, the color information estimated by the color estimation unit is displayed on a display device corresponding to the second estimation subject.

[0012] In one aspect of the present invention, in the estimation device described above, the information control unit displays the color information estimated by the color estimation unit at different times on the display device in chronological order.

[0013] One aspect of the present invention is the above-mentioned estimation device, wherein the color estimation unit estimates color information corresponding to emotion information of the target using the color estimation model.

[0014] One aspect of the present invention is the above-mentioned estimation device, wherein the estimation subject is a care recipient and the model generation subject is a caregiver, or the estimation subject is a caregiver and the model generation subject is a care recipient.

[0015] One aspect of the present invention is an estimation system comprising: a first emotion estimation unit that estimates an emotion of a model generation subject based on biometric information that is obtained by measuring the subject and that changes depending on the emotion; a learning unit that learns, for each model generation subject, a color estimation model that represents a correspondence between emotion information and color information, using information about the estimated emotion and information about a color selected by the model generation subject as a color that represents the emotion at the time the biometric information was measured; a second emotion estimation unit that estimates an emotion of the estimation subject based on biometric information that is obtained by measuring the subject and that changes in response to the emotion; and a color estimation unit that estimates color information that corresponds to the estimated emotion information of the estimation subject, using the color estimation model trained on a model generation subject that is the same as or different from the estimation subject.

[0016] One aspect of the present invention is an estimation system comprising: a learning unit that learns a color estimation model representing the correspondence between biometric information and color information for each model generation subject, based on biometric information obtained by measuring the model generation subject and which changes depending on emotions, and information on a color selected by the model generation subject as a color representing the emotion at the time the biometric information was measured; and a color estimation unit that uses the color estimation model learned for a model generation subject that is the same as or different from the estimation subject to estimate color information corresponding to the biometric information obtained by measuring the estimation subject.

[0017] One aspect of the present invention is a display device that includes a process of estimating color information corresponding to emotional information estimated based on the biometric information of a subject to be estimated using a color estimation model that represents the correspondence between emotional information and color information that represents the emotion of the subject to be modeled, or a display unit that displays color information obtained by a process of estimating color information corresponding to the biometric information of the subject to be estimated using a color estimation model that represents the correspondence between the biometric information of the subject to be modeled and colors that represent the emotions of the subject to be modeled at the time the biometric information of the subject to be modeled was measured.

[0018] One aspect of the present invention is a display system including: a first display unit that displays color information corresponding to emotion information estimated based on biometric information of a first estimation subject using a color estimation model that represents a correspondence between emotion information and color information that represents the emotion of a second estimation subject, or a color information corresponding to the biometric information of the first estimation subject using a color estimation model that represents a correspondence between the biometric information of the second estimation subject and colors that represent the emotion of the second estimation subject at the time the biometric information of the second estimation subject was measured; and a second display unit that displays the color information corresponding to emotion information estimated based on the biometric information of the second estimation subject using a color estimation model that represents a correspondence between emotion information and color information that represents the emotion of the first estimation subject, or a color information corresponding to the biometric information of the second estimation subject using a color estimation model that represents a correspondence between the biometric information of the first estimation subject and colors that represent the emotion of the first estimation subject at the time the biometric information of the first estimation subject was measured.

[0019] One aspect of the present invention is a learning method having an acquisition step of acquiring emotional information obtained by measuring a model generation subject and estimating the emotion of the model generation subject based on biometric information that changes depending on the emotion, and a learning step of learning a color estimation model that represents the correspondence between emotional information and color information for each model generation subject, using the estimated emotional information and information on a color selected by the model generation subject as a color that represents the emotion at the time the biometric information was measured.

[0020] One aspect of the present invention is a learning method including a learning step of learning, for each model generation subject, a color estimation model that represents the correspondence between biometric information and color information, based on biometric information obtained by measuring the model generation subject and that changes depending on emotions, and information on a color selected by the model generation subject as a color that represents the emotion at the time the biometric information was measured.

[0021] One aspect of the present invention is an estimation method comprising: an emotion estimation step of estimating the emotion of a subject to be estimated based on biological information obtained by measuring the subject and which changes depending on the emotion; and a color estimation step of estimating color information corresponding to the estimated emotion information using a color estimation model that represents a correspondence between emotion information and color information, wherein the color estimation model is trained for each model generation subject using information on a plurality of different emotions and information on colors selected by a model generation subject who is the same as or different from the subject to be estimated as colors that represent each of the plurality of different emotions.

[0022] One aspect of the present invention is an estimation method that includes a color estimation step of estimating color information corresponding to biometric information obtained by measuring an estimation subject using a color estimation model that represents the correspondence between biometric information that changes in response to emotions and color information, and the color estimation model is learned for each model generation subject using biometric information obtained by measuring a model generation subject that is the same as or different from the estimation subject at different times, and information on colors selected by the model generation subject as colors that represent emotions at each time the biometric information was measured.

[0023] One aspect of the present invention is a computer program for causing a computer to function as the learning device described above.

[0024] One aspect of the present invention is a computer program for causing a computer to function as the above-described estimation device.

[0025] One aspect of the present invention is a computer program for causing a computer to function as the above-described estimation system. [Effects of the Invention]

[0026] The present invention makes it possible to express emotions using colors that reflect individual differences. [Brief explanation of the drawings]

[0027] [Figure 1]1 is a schematic block diagram showing a system configuration of an estimation system according to a first embodiment of the present invention. [Figure 2] 1 is a schematic block diagram illustrating an example of the functional configuration of a learning device according to a first embodiment. [Figure 3] 1 is a schematic block diagram illustrating an example of the functional configuration of an estimation device according to a first embodiment. [Figure 4] 1 is a schematic block diagram illustrating an example of a functional configuration of a display device according to a first embodiment. [Figure 5] 5 is a flowchart showing a specific example of processing by the learning device according to the first embodiment. [Figure 6] 4 is a flowchart showing a specific example of processing by the estimation device according to the first embodiment. [Figure 7] FIG. 1 is a diagram illustrating an example of use of an estimation system according to a first embodiment. [Figure 8] FIG. 1 is a diagram illustrating an example of use of an estimation system according to a first embodiment. [Figure 9] FIG. 10 is a diagram illustrating a modified example of the estimation device according to the first embodiment. [Figure 10] FIG. 10 is a diagram illustrating a modified example of the estimation device according to the first embodiment. [Figure 11] FIG. 10 is a schematic block diagram showing the system configuration of an estimation system according to a second embodiment. [Figure 12] FIG. 10 is a schematic block diagram illustrating an example of the functional configuration of a learning device according to a second embodiment. [Figure 13] FIG. 10 is a schematic block diagram illustrating an example of the functional configuration of an estimation device according to a second embodiment. [Figure 14] 10 is a flowchart showing a specific example of processing by the learning device according to the second embodiment. [Figure 15] 10 is a flowchart showing a specific example of processing by the estimation device according to the second embodiment. [Figure 16] FIG. 1 is a diagram illustrating an outline of a hardware configuration example of an information processing device applied to first and second embodiments. DETAILED DESCRIPTION OF THE INVENTION

[0028] One of the objectives of the present invention is to address social issues such as a shortage of caregivers, an increasing elderly population, and rising turnover rates due to burnout. This objective, in particular, includes supporting the resolution of communication discord in caregiving. Conventional caregiving techniques have made it difficult to communicate with care recipients while fully considering the emotions of the care recipient and the caregiver's situation. To solve this problem, the present embodiment utilizes a personal color palette for each care recipient. The personal color palette is individual information that indicates the correspondence between emotions and colors.

[0029] Specifically, the estimation system according to this embodiment acquires data obtained by measuring the biological activity of the care recipient and estimates information related to the care recipient's emotion based on the acquired data. Furthermore, the care recipient selects a color associated with the estimated emotion. The estimation system learns a color estimation model representing the correspondence between emotion and color for each care recipient based on the information related to the estimated emotion and the data on the selected color. This color estimation model corresponds to a personal color palette. The color estimation model can estimate a color corresponding to any emotion in the personal color palette. The color estimation model can be used, for example, not only as a means for converting the emotion of the care recipient into a color, but also as a means for recommending the optimal color environment for the care recipient.

[0030] The estimation system according to this embodiment can also be used to facilitate emotional transmission and communication between caregivers and care recipients. It has been shown that caregivers with high perspective-taking skills are less likely to experience a decline in quality of life (QoL). Perspective-taking is the mental process of shifting one's current perspective to a different position or location (another person's perspective) and inferring the thoughts, feelings, or scenery that the person in that perspective might have. The estimation system according to this embodiment converts the care recipient's emotions into colors and displays them. This allows caregivers to visually grasp the care recipient's emotions and respond and communicate appropriately. Furthermore, caregivers can monitor the care recipient's emotions, promote perspective-taking and empathic understanding, and then provide the necessary support and care.

[0031] Because colors corresponding to emotions vary from person to person, personal color palettes also vary from person to person. Therefore, the colors estimated by the estimation system do not directly reveal specific emotions, protecting the privacy of the care recipient. On the other hand, because colors corresponding to emotions are generally universal, they are more expressive than words and can be intuitively understood by, for example, foreigners. Therefore, non-verbal communication through color can promote emotional care. Furthermore, by looking at the time series of estimated colors, it is possible to grasp, for example, the emotional ups and downs throughout the day.

[0032] Furthermore, color estimation models can be generated not only for care recipients but also for caregivers. For example, by observing the time series changes over the course of a day in the colors estimated by the generated color estimation model as colors representing the caregiver's emotions, staff and colleagues at the nursing facility can understand the caregiver's psychological state while performing their duties and can provide encouragement, etc. Furthermore, by using the caregiver's color estimation model to estimate the colors representing the care recipient's emotions, the caregiver can understand the care recipient's emotions as their own.

[0033] In addition, the estimation system according to this embodiment is also practical for coordinating a care environment. For example, the estimation system obtains color information corresponding to a target emotion, such as calmness or happiness, by inputting target emotion information into a color estimation model instead of an emotion estimated based on biological measurement data of the care recipient. The obtained color is incorporated into the living environment, such as the walls and floors of the care recipient's room. In this way, by proposing an optimal color environment tailored to the care recipient's ideal emotion, it can be used for interior materials in living spaces and facilities, and is expected to contribute to promoting comfort and stable emotions in the care recipient. Note that the estimation system according to this embodiment can also be applied to any field other than care. Detailed embodiments of the estimation system are described below.

[0034] (First embodiment) FIG. 1 is a schematic block diagram showing the system configuration of an estimation system 100 according to a first embodiment of the present invention. The estimation system 100 is a data platform that estimates colors related to emotions and is used to estimate colors representing the emotions of a target person based on biometric information obtained by measuring the target person's biometrics. To this end, the estimation system 100 performs emotion estimation and constructs a personal color palette. The estimation system 100 monitors the target person's emotions based on their biometric information, estimates colors representing the emotions obtained through monitoring based on the constructed personal color palette, and communicates the colors to other people. The estimation system 100 also uses the personal color palette to suggest colors for interior materials and other items that suit the target person.

[0035] The estimation system 100 includes a measurement device 10, a learning device 20, an estimation device 30, and a display device 40. The estimation system 100 may include one or more measurement devices 10 and one or more display devices 40. The measurement device 10, the estimation device 30, and the display device 40 are communicatively connected via a network 70 or by a communication cable or the like. The learning device 20 and the estimation device 30 may be communicatively connected via the network 70. The estimation device 30 and the display device 40 are communicatively connected via the network 70, but may also be connected by a communication cable or the like. The network 70 may be a network using wireless communication, a network using wired communication, or a network combining wireless communication and wired communication. The network 70 may be configured using a public network such as the Internet, a private network such as a local area network (LAN), or a combination of a public network and a private network. The network 70 may be configured by combining multiple networks.

[0036] The measurement device 10 is a sensor that measures a predetermined type of biometric information of the subject and outputs measurement information indicating the measured biometric information. The biometric information measured by the measurement device 10 may be any type of biometric information that changes depending on emotions. Examples of biometric information measured include, but are not limited to, brain waves, pulse rate, body temperature, sweating, respiratory rate, blood pressure, blood oxygen saturation, posture, and voice. Furthermore, although using biometric information that can be measured non-invasively reduces the burden on the subject, biometric information that can be measured invasively may also be used.

[0037] The measuring device 10 can be any conventional sensor that measures biological information. If the type of biological information is brain waves, the measuring device 10 is, for example, an brain wave sensor. If the type of biological information is pulse, the measuring device 10 is, for example, a pulse sensor. If the type of biological information is body temperature, the measuring device 10 is, for example, a temperature sensor that measures the surface temperature of the human body. If the type of biological information is posture, the measuring device 10 is, for example, a computer device that measures posture based on an image of the person to be estimated. If the type of biological information is voice, the measuring device 10 is, for example, a microphone.

[0038] The measurement information includes the measurement time, the measurement value obtained by measuring the subject at the measurement time, and information on the type of biological information being measured. Furthermore, the measurement device 10 may add a personal ID, which is personal identification information for identifying the subject, to the measurement information before transmitting it. The identification information of the measurement device 10 may be used as the personal ID or information on the type of biological information.

[0039] FIG. 2 is a schematic block diagram showing a specific example of the functional configuration of learning device 20 according to the first embodiment. Only functional blocks related to this embodiment are shown in FIG. 2. Learning device 20 is configured using an information processing device such as a personal computer or a server device. Learning device 20 includes a communication unit 21, a sensor input interface 22, an input unit 23, a display unit 24, a memory unit 25, and a control unit 26.

[0040] The communication unit 21 is a communication device. The communication unit 21 may be configured as, for example, a network interface. The communication unit 21 communicates data with other devices via the network 70 in accordance with the control of the control unit 26. The communication unit 21 may be a device that performs wireless communication or a device that performs wired communication.

[0041] The sensor input interface 22 is connected to the measurement device 10 via a communication cable. The sensor input interface 22 receives measurement information transmitted from the measurement device 10. Note that when the measurement information is received via the network 70, the learning device 20 does not need to have the sensor input interface 22.

[0042] The input unit 23 is a keyboard, a mouse, a button, a touch panel, etc., and receives information input by a user's operation. Information specifying a color corresponding to an emotion is input via the input unit 23.

[0043] Display unit 24 outputs information in a form that can be recognized by the user. Display unit 24 may be an image display device such as a liquid crystal display or an organic EL (Electro Luminescence) display. Display unit 24 may also be an interface for connecting an image display device to learning device 20. In this case, display unit 24 generates a video signal for displaying image data and outputs the video signal to the image display device connected to it. Display unit 24 may also be configured as a touch panel integrated with input unit 23.

[0044] The storage unit 25 is configured using a storage device such as a magnetic hard disk drive or a semiconductor storage device. The storage unit 25 stores data used by the control unit 26. The storage unit 25 may function as, for example, a measurement information storage unit 251, a feature information storage unit 252, a teacher data storage unit 253, and a trained model storage unit 254.

[0045] The measurement information storage unit 251 stores measurement information used for the learning process of the color estimation model executed in the learning device 20.

[0046] The feature amount information storage unit 252 stores feature amount information. The feature amount information is biometric information used to estimate emotions. The feature amount information indicates the type of feature amount (biometric information) and the value of the feature amount. Measurement information may be used as the feature amount information as is, or the feature amount information may be calculated by performing a predetermined calculation on one or more measurement values obtained from one or more pieces of measurement information. The feature amount information is used as an explanatory variable. Note that when measurement information is used as the feature amount information as is, the storage unit 25 does not need to have the feature amount information storage unit 252.

[0047] The teacher data storage unit 253 stores teacher data that associates estimated emotion information indicating an emotion estimated based on feature information with color information indicating a color selected by the subject as a color corresponding to that emotion.

[0048] The trained model storage unit 254 stores a color estimation model trained for each individual through a training process using the training data stored in the training data storage unit 253. The color estimation model is a personal color palette that represents the correspondence between information expressing emotions and color information. The trained color estimation model is also referred to as a trained model.

[0049] The control unit 26 is configured using a processor such as a CPU (Central Processing Unit) or a GPU (Graphics Processing Unit) and a memory. The control unit 26 functions as an information control unit 261, a feature calculation unit 262, an emotion estimation unit 263, a teacher data generation unit 264, and a learning unit 265 by the processor executing a program. Note that all or part of the functions of the control unit 26 may be realized using hardware such as an ASIC (Application Specific Integrated Circuit), a PLD (Programmable Logic Device), or an FPGA (Field Programmable Gate Array). The above program may be recorded on a computer-readable recording medium. Examples of computer-readable recording media include portable media such as a flexible disk, a magneto-optical disk, a ROM (Read Only Memory), a CD-ROM (Compact Disc Read Only Memory), and a semiconductor storage device (e.g., a Solid State Drive (SSD)), as well as storage devices such as a hard disk or semiconductor storage device built into a computer system. The above program may be transmitted via a telecommunications line.

[0050] The information control unit 261 controls the input and output of information. For example, the information control unit 261 acquires measurement information from another device (e.g., the measuring device 10) and records it in the measurement information storage unit 251. Furthermore, for example, the information control unit 261 transmits a trained model stored in the trained model storage unit 254 to another device (e.g., the estimation device 30). Such exchange of information between the information control unit 261 and another device may be performed, for example, through communication by the communication unit 21 or the sensor input interface 22.

[0051] The feature calculation unit 262 generates feature information used by the emotion estimation unit 263 to estimate emotions by performing a predetermined calculation on the measurement values indicated by the measurement information acquired by the information control unit 261. For example, the feature information is input data for an emotion estimation model used by the emotion estimation unit 363 to estimate emotions. When the measurement information is electroencephalogram (EEG) measurement data, the feature information may be, for example, an EEG state indicated by the difference between "Meditation," which indicates a level of relaxation based on alpha waves, and "Attention," which indicates a level of concentration based on beta waves. When the measurement information is pulse measurement data, the feature information may be, for example, heart rate fluctuations expressed by an index such as pNNx. pNN50 is the percentage of heartbeats in which the interval between adjacent heartbeat waveform peaks exceeds 50 milliseconds among 30 consecutive heartbeats. When the measurement information is voice measurement data, the biometric feature information may be, for example, a spectral outline, spectral power, frequency, or voice recognition result obtained from the voice measurement data. The feature calculation unit 262 associates the calculated feature information with the personal ID read from the measurement information and writes the information to the feature information storage unit 252. When the biometric information set in the measurement information is used as a feature, the feature calculation unit 262 may store the measurement information as is as feature information in the feature information storage unit 252.

[0052] The emotion estimation unit 263 obtains estimated emotion information representing an estimated emotion based on one or more feature quantities indicated by one or more types of feature quantity information. Any existing technology can be used for the estimation. For example, the emotion estimation unit 263 inputs the feature quantities indicated by the feature quantity information into any existing emotion estimation model and obtains estimated emotion information as an output. The emotion estimation unit 263 outputs the estimated emotion information to the training data generation unit 264. For example, the estimated emotion information includes multiple emotion values corresponding to the amounts of multiple types of emotions arranged on a Russell circumplex model. In this case, for example, the technology described in Patent Document 3 can be used as the emotion estimation model. The technology described in Patent Document 3 uses electroencephalogram state and heart rate fluctuations as feature quantities to obtain a position on the Russell circumplex model. The position on the Russell circumplex model indicates values on the emotion coordinate axes corresponding to "surprise-sleepy," "excitement-melancholy," "joy-sadness," and "relaxation-anger," respectively.

[0053] The teacher data generation unit 264 acquires information about the color selected by the subject as the color representing the emotion at the time of measuring the biological information. For example, the subject selects a color representing the emotion from the colors displayed on the display unit 24 using the input unit 23. The teacher data generation unit 264 writes teacher data that associates the estimated emotion information, information about the selected color, and the subject's personal ID into the teacher data storage unit 253.

[0054] The learning unit 265 executes a learning process for the color estimation model using the training data stored in the training data storage unit 253. Any machine learning method can be used to learn the color estimation model. Specific examples of the learning process include machine learning such as supervised learning using a neural network. The learning unit 265 generates a color estimation model for outputting color information corresponding to input emotional information, for example, by performing supervised learning. The color information is information that expresses color numerically using, for example, RGB, L*a*b*, or the like. The learning unit 265 records the generated color estimation model as a trained model in the trained model storage unit 254. The trained model obtained by the learning unit 265 may be transmitted to the estimation device 30 and recorded in an estimation model storage unit 351 of the estimation device 30, which will be described later.

[0055] FIG. 3 is a schematic block diagram showing a specific example of the functional configuration of the estimation device 30 according to the first embodiment. In FIG. 3, only functional blocks related to this embodiment are shown. The estimation device 30 is configured using an information processing device such as a personal computer or a server device. The estimation device 30 includes a communication unit 31, a sensor input interface 32, an input unit 33, a display unit 34, a storage unit 35, and a control unit 36.

[0056] The communication unit 31 is a communication device. The communication unit 31 may be configured as, for example, a network interface. The communication unit 31 communicates data with other devices via the network 70 in accordance with the control of the control unit 36. The communication unit 31 may be a device that performs wireless communication or a device that performs wired communication.

[0057] The sensor input interface 32 is connected to the measurement device 10 via a communication cable. The sensor input interface 32 receives measurement information transmitted from the measurement device 10. Note that when measurement information is received via the network 70, the estimation device 30 does not need to have the sensor input interface 22.

[0058] The input unit 33 is a keyboard, mouse, button, touch panel, etc., and receives information input by user operation. The input unit 33 inputs the personal ID of the color estimation model used for estimation and emotion information indicating the emotion of the target.

[0059] The display unit 34 outputs information in a form that can be recognized by the user. The display unit 34 may be, for example, an image display device such as a liquid crystal display or an organic EL display. The display unit 34 may be an image display device or an interface for connecting the display device 40 to the estimation device 30. In this case, the display unit 34 generates a video signal for displaying image data and outputs the video signal to the image display device or the display device 40 connected to the display unit 34. The display unit 34 may be configured as a touch panel integrated with the input unit 33.

[0060] The storage unit 35 is configured using a storage device such as a magnetic hard disk drive or a semiconductor storage device. The storage unit 35 stores data used by the control unit 36. The storage unit 35 may function as, for example, an estimation model storage unit 351 and an estimation result storage unit 352.

[0061] The estimation model storage unit 351 stores a color estimation model used by the color estimation unit 364 when performing estimation processing. The color estimation model may be, for example, a learned model generated in advance by a learning process. Such learning processing may be performed by, for example, another device (e.g., the learning device 20) or by the device itself (the estimation device 30). The color estimation model does not necessarily have to be generated by a learning process. The color estimation model may be configured using, for example, a lookup table that associates emotion information with color information, or may be configured in another manner. The estimation result storage unit 352 stores estimation result information that associates estimation times with color information estimated using the color estimation model.

[0062] The control unit 36 is configured using a processor such as a CPU or a GPU, and a memory. The control unit 36 functions as an information control unit 361, a feature calculation unit 362, an emotion estimation unit 363, and a color estimation unit 364 by the processor executing a program. Note that all or part of the functions of the control unit 36 may be realized using hardware such as an ASIC, a PLD, or an FPGA. The above program may be recorded on a computer-readable recording medium. Examples of computer-readable recording media include portable media such as a flexible disk, a magneto-optical disk, a ROM, a CD-ROM, and a semiconductor storage device (e.g., an SSD), and storage devices such as a hard disk or semiconductor storage device built into a computer system. The above program may be transmitted via a telecommunications line.

[0063] The information control unit 361 controls the input and output of information. For example, the information control unit 361 acquires measurement information of the estimation subject from another device such as the measuring device 10. Furthermore, for example, the information control unit 361 transmits color information indicating the estimation result obtained by the color estimation unit 364 to another device such as the display device 40. Such exchange of information between the information control unit 361 and another device may be performed, for example, through communication by the communication unit 31 or the sensor input interface 32. Furthermore, the information control unit 361 may cause the display unit 34 to display the color information indicating the estimation result.

[0064] The feature calculation unit 362 generates feature information that the emotion estimation unit 363 uses to estimate emotions by performing a predetermined calculation on the measurement values indicated by the measurement information acquired by the information control unit 361. When the feature calculation unit 362 uses the biological information set in the measurement information as a feature, the feature calculation unit 362 uses the measurement information as is as feature information. The processing performed by the feature calculation unit 362 on the measurement information is the same as the processing performed by the feature calculation unit 262 of the learning device 20 on the measurement information, and the type of feature information calculated is also the same.

[0065] The emotion estimation unit 363 obtains estimated emotion information representing an estimated emotion based on one or more feature amounts indicated by one or more types of feature amount information. The emotion estimation process performed by the emotion estimation unit 363 is the same as the emotion estimation process performed by the emotion estimation unit 263 of the learning device 20. The emotion estimation unit 363 outputs the obtained estimated emotion information to the color estimation unit 364.

[0066] Color estimation unit 364 performs estimation processing using the color estimation model stored in estimation model storage unit 351 and estimated emotion information received from emotion estimation unit 363. The estimation processing estimates color information indicating a color corresponding to the emotion indicated by the estimated emotion information. Color estimation unit 364 writes estimation result information indicating the estimation time and the estimated color information to estimation result storage unit 352. The estimation time may be the measurement time of the biological information used to estimate the estimated emotion information. Furthermore, color estimation unit 364 performs estimation processing using the color estimation model stored in estimation model storage unit 351 and emotion information indicating the emotion of the target input by input unit 33, and estimates color information indicating a color corresponding to the emotion of the target.

[0067] FIG. 4 is a schematic block diagram showing a specific example of the functional configuration of the display device 40. FIG. 4 shows only functional blocks related to this embodiment. The display device 40 is preferably a badge-type information display device with a pin or clip for attachment to clothing, an information display device with a strap for hanging from the neck, or an information display device that can be worn on the arm with a band, and is portable by the person to be presumed to be a candidate and easily visible to people other than the person to be presumed to be a candidate. However, this is not limited to these examples. For example, the display device 40 may be installed near the person to be presumed to be a candidate. Furthermore, the display device 40 may be configured using an information device such as a smartphone, tablet, personal computer, or dedicated device. The display device 40 includes a communication unit 41, a control unit 42, and a display unit 43.

[0068] The communication unit 41 is a communication device. The communication unit 41 may be configured as, for example, a network interface. The communication unit 41 communicates data with other devices via the network 70. The communication unit 41 may be a device that performs wireless communication or a device that performs wired communication.

[0069] The control unit 42 is configured using a processor such as a CPU and a memory (main storage device). The control unit 42 functions by the processor executing a program. The control unit 42 displays, on the display unit 43, the color indicated by the estimation result information received from the estimation device 30. For example, the control unit 42 may display, on the display unit 43, the color indicated by the most recent estimation result information received from the estimation device 30, or may display, on the display unit 43, in chronological order of the time of reception, information on the color indicated by each piece of estimation result information received after a predetermined time or received up to a predetermined time prior to the present.

[0070] The display unit 43 is an image display device such as a liquid crystal display or an organic EL display.

[0071] 5 is a flowchart showing a specific example of the processing of the learning device 20. The learning device 20 performs the processing of FIG. 5 for each model creation subject.

[0072] First, the learning device 20 acquires measurement information of the model creation subject (step S101). That is, one or more measurement devices 10 transmit the measurement information obtained by measuring the model creation subject to the learning device 20 via the network 70 or a communication cable. The communication unit 21 or the sensor input interface 22 of the learning device 20 receives the measurement information transmitted from the measurement device 10 and outputs it to the control unit 26. The information control unit 261 of the control unit 26 writes the acquired measurement information to the measurement information storage unit 251. If the measurement information transmitted by the measurement device 10 does not include a personal ID, the information control unit 261 adds the personal ID of the model creation subject input by the input unit 23 to the received measurement information and writes the resulting information to the measurement information storage unit 251.

[0073] The feature calculation unit 262 generates feature information based on the measurement information acquired by the information control unit 261 and writes it to the feature information storage unit 252 (step S102). If the biometric information set in the acquired measurement information is to be used as is for emotion estimation, the feature calculation unit 262 writes the measurement information to the feature information storage unit 252 as feature information. If conversion of the biometric information set in the measurement information is necessary, the feature calculation unit 262 converts it into feature information to be used for emotion estimation by performing a predetermined calculation on measurement values indicated by one or more pieces of measurement information. The feature calculation unit 262 adds the personal ID of the model creation subject to the converted feature information and writes it to the feature information storage unit 252.

[0074] The emotion estimation unit 263 reads feature amount information in which the personal ID of the model creation subject is set from the feature amount information storage unit 252, and obtains estimated emotion information based on one or more feature amounts indicated by the one or more read feature amount information (step S103). For example, the emotion estimation unit 263 inputs the feature amount information into an emotion estimation model to obtain estimated emotion information. The emotion estimation unit 263 outputs the obtained estimated emotion information to the training data generation unit 264.

[0075] The teacher data generation unit 264 generates teacher data that associates the estimated emotion information with information about the color selected by the model creation subject as the color that represents the emotion (step S104). For example, when the teacher data generation unit 264 acquires the estimated emotion information from the emotion estimation unit 263, it displays a list of colors on the display unit 24. The model creation subject selects a color that represents the emotion at the time of biological information measurement from the list of colors displayed on the display unit 24, and inputs information about the selected color via the input unit 23. The teacher data generation unit 264 writes the teacher data, which associates the estimated emotion information, the input color information, and the model creation subject's personal ID, into the teacher data storage unit 253.

[0076] The learning unit 265 determines whether a predetermined number of pieces of teacher data necessary for learning a color estimation model are stored in the teacher data storage unit 253 (step S105). The predetermined number can be set arbitrarily depending on the type of machine learning, estimation accuracy, etc. If the learning unit 265 determines that the number of pieces of teacher data does not reach the predetermined number, it repeats the processing from step S101. If the learning unit 265 determines that there is a predetermined number of pieces of teacher data, it executes a learning process using the teacher data stored in the teacher data storage unit 253 to generate a color estimation model (step S106). The learning unit 265 adds the personal ID of the model creation target person to the trained model, which is the generated color estimation model, and records it in the trained model storage unit 254.

[0077] The learning unit 265 may generate a color estimation model using training data of multiple people, and fine-tune the generated color estimation model using training data of the model creation subject. This allows learning to be performed so that color information can be estimated even for emotions that are less frequently expressed in the model creation subject, for example.

[0078] FIG. 6 is a flowchart showing a specific example of processing by the estimation device 30. First, the user inputs an estimation subject ID and a usage model ID via the input unit 33 (step S201). The estimation subject ID indicates the personal ID of the estimation subject, who is the subject whose emotions are to be estimated, i.e., the subject whose biometric information is to be measured. The usage model ID is information specifying for whom the trained color estimation model is to be used, and is indicated by the personal ID. Note that the estimation subject ID and the usage model ID may be the same or different. Also, input of the estimation subject ID may be omitted. Also, if the estimation subject ID and the usage model ID are the same, input of either the estimation subject ID or the usage model ID may be omitted.

[0079] The information control unit 361 reads out a color estimation model assigned a personal ID that matches the usage model ID input by the input unit 33 from the estimation model storage unit 351, and outputs the model to the color estimation unit 364 (step S202). If a color estimation model assigned a personal ID that matches the usage model ID is not stored in the estimation model storage unit 351, the information control unit 361 acquires a color estimation model assigned a usage model ID from the learning device 20 or the like, stores the model in the estimation model storage unit 351, and outputs the model to the color estimation unit 364.

[0080] The information control unit 361 acquires measurement information of the person to be estimated from the measuring device 10 (step S203). That is, one or more measuring devices 10 transmit the measurement information obtained by measuring the person to be estimated to the estimation device 30 via the network 70 or a communication cable. The communication unit 31 or the sensor input interface 32 of the estimation device 30 receives the measurement information transmitted from the measuring device 10 and outputs it to the information control unit 361 of the control unit 36.

[0081] The feature calculation unit 362 generates feature information to be used for emotion estimation based on the measurement information acquired by the information control unit 361 in step S203 (step S204). When the biological information set in the acquired measurement information is used for emotion estimation as is, the feature calculation unit 362 uses the measurement information as feature information. The feature calculation unit 362 outputs the feature information to the emotion estimation unit 363.

[0082] The emotion estimation unit 363 obtains estimated emotion information based on one or more feature amounts indicated by the one or more feature amount information received from the feature amount calculation unit 362 (step S205). For example, the emotion estimation unit 363 inputs the feature amount information into an emotion estimation model to obtain estimated emotion information. The emotion estimation unit 363 outputs the obtained estimated emotion information to the color estimation unit 364.

[0083] Color estimation unit 364 inputs the estimated emotion information received from emotion estimation unit 363 into the color estimation model acquired by information control unit 361 in step S202, and obtains estimated color information (step S206). Color estimation unit 364 writes estimation result information that associates the estimated time, estimated emotion information, estimated color information, and estimation target person ID into estimation result storage unit 352. The estimated time may be the time when the measurement information was acquired in step S203, or the measurement time set in the measurement information.

[0084] The information control unit 361 transmits the estimated color information obtained in step S206 to the display device 40 (step S207). Note that if there are multiple display devices 40, the information control unit 361 transmits the estimated color information as display instruction information to a display device 40 predetermined according to the estimation target person ID or a display device 40 specified by information input by the input unit 33. The control unit 42 of the display device 40 displays the color indicated by the color information received from the estimation device 30 on the display unit 43. The information control unit 361 may also cause the display unit 34 to display the color indicated by the color information.

[0085] Furthermore, in step S207, the information control unit 361 may transmit emotion display information to the display device 40 as display instruction information, instead of or in addition to the estimated color information, for displaying the estimated emotion information obtained in step S205. The emotion display information may be, for example, text data, symbols, or a plot on a Russell circle model, but is not limited to these. When the estimated emotion information is expressed by values on each emotion coordinate axis, data indicating the correspondence between the text data or symbols used for the emotion display information and the numerical ranges represented by the values on the emotion coordinate axes is stored in advance in the storage unit 35. The information control unit 361 reads from the storage unit 35 the text data or symbols corresponding to the numerical ranges including the estimated emotion information. Alternatively, the emotion display information may be text data indicating the values on each emotion coordinate axis. The control unit 42 of the display device 40 displays the emotion display information received from the estimation device 30 on the display unit 43.

[0086] The processes of steps S201 and S202 may be performed after the processes of steps S203, S204, or S205.

[0087] Furthermore, the estimation device 30 may include the learning unit 265 of the learning device 20. In step S207, the learning unit 265 of the estimation device 30 acquires the correct color information input by the user via the input unit 33, and uses the acquired correct color information and estimated emotion information as training data to learn (update) the color estimation model.

[0088] 6 at predetermined intervals. In this case, the control unit 42 of the display device 40 may display, on the display unit 43, the colors indicated by the color information received from the estimation device 30 by the processing of step S207 within the predetermined period in chronological order of reception. The predetermined period may be, for example, a period going back a predetermined amount of time from the present or a period from a predetermined time onward. Alternatively, in step S207, the information control unit 361 may generate color display information for chronologically displaying the color information indicated by the estimation result information associated with the estimation target person ID, and transmit the generated color display information to the display device 40. Specifically, the information control unit 361 reads, from the estimation result storage unit 352, the estimation result information whose estimation time falls within the predetermined period, from the estimation result information associated with the estimation target person ID. The predetermined period may be, for example, a period going back a predetermined amount of time from the present, or any period input by the input unit 33. The information control unit 361 transmits color display information, as display instruction information, to the display device 40, for displaying the colors indicated by the color information set in the read estimation result information in order of estimation time. The control unit 42 of the display device 40 displays the chronological colors indicated by the color display information received from the estimation device 30 on the display unit 43. Instead of or in addition to the color display information, the control unit 42 may transmit emotion display information, as display instruction information, to the display device 40, for displaying the chronological estimated emotion information read from the estimation result information. The control unit 42 of the display device 40 displays the chronological emotion information indicated by the emotion display information received from the estimation device 30 on the display unit 43.

[0089] Furthermore, in step S207, the information control unit 361 may include in the color display information a color that represents a predetermined meaning, for example, a predetermined change pattern of a predetermined color corresponding to the predetermined meaning. For example, a color change pattern corresponding to the received message information is included in the color display information to be displayed on the display device 40 of the care recipient. In this case, for example, data associating the message information with the color change pattern is stored in advance in the storage unit 35. The message information may be a number identifying the message. The information control unit 361 receives information about a message to be conveyed to the caregiver from a terminal or the like at the care facility. The information control unit 361 reads out a color change pattern corresponding to the received message information from the storage unit 35 and includes the color change pattern in the color display information to be displayed on the display device 40 of the care recipient. Alternatively, the information control unit 361 may receive information about a message to be conveyed to the caregiver and information about the output conditions of the message from a terminal or the like at the care facility. The message information and the output conditions may be stored in advance in the storage unit 35 of the estimation device 30. When the output condition is satisfied, the information control unit 361 includes a color change pattern corresponding to the message information in the color display information to be displayed on the display device 40 of the care recipient. As an example, the output condition can be, but is not limited to, that a measurement value indicated by a predetermined type of biometric information of the care recipient falls outside a reference range. The caregiver understands the message based on the predetermined color pattern displayed on the display device 40 of the care recipient. In this way, highly confidential, encrypted messages can be conveyed using the color display information.

[0090] FIG. 7 is a diagram illustrating an example of how the estimation system 100 is used. The display device 40 displaying color information for care recipient A is referred to as display device 40a, and the display device 40 displaying color information for caregiver B is referred to as display device 40b. In FIG. 7, the display device 40a is attached to the clothing of care recipient A. The display device 40a displays time-series color information estimated for care recipient A. By viewing the display on display device 40a, family members and caregivers can monitor the emotions of care recipient A, which promotes perspective-taking for caregiving communication. Furthermore, the display device 40b displays time-series color information estimated for caregiver B. As shown in FIG. 7, caregiver B attaches the display device 40b to his or her clothing, or an information device such as a personal computer used in the care facility is used as display device 40b, allowing staff and managers of the care facility to monitor the emotions of caregiver B.

[0091] Furthermore, when the estimated subject ID and usage model ID are the same, the estimation system 100 can estimate and display a color that represents the emotion of the estimated subject. On the other hand, by using a usage model ID that is different from the estimated subject ID, the estimation system 100 can display a color that corresponds to the emotion of the estimated subject using a personal palette of a person different from the estimated subject. An example of this is shown in FIG. 8.

[0092] FIG. 8 is a diagram illustrating another example of use of the estimation system 100. In FIG. 8, color estimation is performed using the other person's personal palette. For example, a pair is set between care recipient A and caregiver B. A display device 40a for care recipient A and a display device 40b for caregiver B constitute a display system. By using the personal ID of care recipient A as the estimation target ID and the personal ID of caregiver B as the usage model ID, the estimation system 100 displays caregiver B's color information "red," which represents care recipient A's emotion "anxiety," on the display device 40a. This display allows caregiver B to recognize care recipient A's emotion from the color representing his or her own emotion. Furthermore, by using caregiver B's personal ID as the estimation target ID and the personal ID of care recipient A as the usage model ID, the estimation system 100 displays care recipient A's color information "blue," which represents caregiver B's emotion "joy," on the display device 40b. This display allows care recipient A to recognize caregiver B's emotion from the color representing his or her own emotion.

[0093] 6, the user may input the usage model ID and emotion information indicating the target emotion via the input unit 33, rather than inputting the estimation target person ID. In this case, the estimation device 30 does not perform steps S203 to S205. In step S206, the color estimation unit 364 of the estimation device 30 inputs the emotion information input in step S201 into the color estimation model to obtain estimated color information. In step S207, the information control unit 361 transmits the color information obtained in step S206 to the display device 40 for display, or displays it on the display unit 34. For example, if the usage model ID indicating the personal ID of care recipient A and the target emotion information indicating calmness are input in step S201, the color of care recipient A corresponding to the calm emotion is estimated. By incorporating this estimated color into the living space of the care facility, such as the walls, floors, and desks, a color environment suitable for care recipient A can be created.

[0094] In the above, a color estimation model is generated for each individual, but a color estimation model may be generated for each attribute information. In this case, the attribute information is used instead of the personal ID of the person for whom the model is to be created in the process of FIG. 5. Furthermore, the attribute information is used as the model ID used in the process of FIG. 6. It is desirable that the attribute information be such that it can classify the living environment of an individual, such as age, gender, country, and family structure. This is because colors that express emotions differ depending on the living environment in which an individual grew up.

[0095] As described above, the estimation system 100 estimates emotions based on biometric information obtained by measuring the model generation subject, and generates training data that associates the estimated emotions with colors selected by the model generation subject. The estimation system 100 executes a learning process using multiple pieces of training data generated for the model generation subject, and generates a trained model. When new biometric information of the estimation subject is acquired, the estimation system 100 acquires estimated emotion information based on the biometric information. Estimated emotions have a different correlation with colors for each individual. The estimation system 100 estimates a color corresponding to the estimated emotion using the trained model. By using biometric information such as electroencephalograms, it is possible to accurately estimate colors that represent the emotions of the estimation subject.

[0096] (Variation) In the above-described embodiment, the measurement device 10 and the estimation device 30 are configured as separate devices, but they may also be configured as an integrated device. FIG. 9 is a diagram showing a modified example of the estimation device 30 configured in this manner. In FIG. 9, the same components as those of the estimation device 30 shown in FIG. 3 are assigned the same reference numerals, and their description will be omitted. The estimation device 30 shown in FIG. 9 further includes a sensor 11. The sensor 11 of the estimation device 30 shown in FIG. 9 functions similarly to the measurement device 10. In step S203 of FIG. 6, the control unit 36 operates in response to an operation on the input unit 33 and acquires measurement information output from the sensor 11. Note that the estimation device 30 may also receive measurement information from a measurement device 10 that measures a different type of biological information from that measured by the sensor 11.

[0097] Furthermore, the measurement device 10 and the learning device 20 may be configured as an integrated device. In this case, the learning device 20 shown in FIG. 2 further includes the sensor 11 that the estimation device 30 shown in FIG. 9 has. In step S101 of FIG. 5, the control unit 26 operates in response to an operation on the input unit 23, and acquires measurement information output from the sensor 11. Note that the learning device 20 may also receive measurement information from the measurement device 10 that measures a different type of biological information from that measured by the sensor 11.

[0098] In this embodiment, the learning device 20 and the estimation device 30 are configured as separate devices, but they may also be configured as an integrated device. FIG. 10 is a diagram showing a modified example of the estimation device 30 configured in this manner. In FIG. 10, the same components as those of the estimation device 30 shown in FIG. 3 are denoted by the same reference numerals, and their description will be omitted. The storage unit 35 of the estimation device 30 shown in FIG. 10 also functions as a measurement information storage unit 353, a feature information storage unit 354, and a teacher data storage unit 355. The estimation model storage unit 351 shown in FIG. 10 also functions as the trained model storage unit 254 of the learning device 20. The measurement information storage unit 353, the feature information storage unit 354, and the teacher data storage unit 355 function in the same way as the measurement information storage unit 251, the feature information storage unit 252, and the teacher data storage unit 253 of the learning device 20, respectively. The control unit 36 of the estimation device 30 shown in FIG. 10 also functions as a teacher data generation unit 365 and a learning unit 366. The information control unit 361, the feature calculation unit 362, and the emotion estimation unit 363 also respectively execute the processes of the information control unit 261, the feature calculation unit 262, and the emotion estimation unit 263 of the learning device 20. The teacher data generation unit 365 and the learning unit 366 function in the same manner as the teacher data generation unit 264 and the learning unit 265 of the learning device 20.

[0099] The measuring device 10, the learning device 20, and the estimation device 30 may be configured as an integrated device. In this case, the estimation device 30 shown in Fig. 10 further includes the sensor 11 included in the estimation device 30 shown in Fig. 9. The estimation device 30 and the display device 40 may also be configured as an integrated device.

[0100] The learning device 20 may be implemented using multiple information processing devices. For example, the learning device 20 may be implemented using a device such as a cloud. It is possible to arbitrarily determine which information processing device implements which functional unit of the learning device 20. For example, in the learning device 20, the memory unit 25 and the control unit 26 may be implemented in different information processing devices. For example, the memory unit 25 of the learning device 20 may be distributed and implemented in multiple information processing devices. For example, the information control unit 261, feature calculation unit 262, emotion estimation unit 263, teacher data generation unit 264, and learning unit 265 of the learning device 20 may be distributed and implemented in multiple information processing devices. Furthermore, the same functional unit of the learning device 20 may be realized by multiple information processing devices.

[0101] The estimation device 30 may be implemented using a plurality of information processing devices. For example, the estimation device 30 may be implemented using a device such as a cloud. Which information processing device implements which functional unit of the estimation device 30 can be determined arbitrarily. For example, in the estimation device 30, the storage unit 35 and the control unit 36 may be implemented in different information processing devices. For example, the storage unit 35 of the estimation device 30 may be distributed and implemented in a plurality of information processing devices. For example, the information control unit 361, the feature calculation unit 362, the emotion estimation unit 363, and the color estimation unit 364 of the estimation device 30 may be distributed and implemented in a plurality of information processing devices. Furthermore, the same functional unit of the estimation device 30 may be realized by a plurality of information processing devices.

[0102] (Second embodiment) In the first embodiment, a color estimation model that associates emotions with color information is used. In the second embodiment, a color estimation model that associates biological information with color information is used. The second embodiment will be described focusing on the differences from the first embodiment.

[0103] Fig. 11 is a schematic block diagram showing the system configuration of an estimation system 101 according to the second embodiment. In Fig. 11, the same components as those in the estimation system 100 according to the first embodiment shown in Fig. 1 are denoted by the same reference numerals, and their description will be omitted. The estimation system 101 shown in Fig. 11 differs from the estimation system 100 of the first embodiment shown in Fig. 1 in that it includes a learning device 201 instead of the learning device 20 and an estimation device 301 instead of the estimation device 30.

[0104] Fig. 12 is a schematic block diagram showing a specific example of the functional configuration of a learning device 201 according to the second embodiment. In Fig. 12, the same components as those in the learning device 20 according to the first embodiment shown in Fig. 2 are designated by the same reference numerals, and their description will be omitted. The learning device 201 is configured using an information processing device such as a personal computer or a server device. The learning device 201 includes a communication unit 21, a sensor input interface 22, an input unit 23, a display unit 24, a storage unit 27, and a control unit 28.

[0105] The storage unit 27 is configured using a storage device such as a magnetic hard disk drive, a semiconductor storage device, etc. The storage unit 27 may function as, for example, a measurement information storage unit 251, a feature information storage unit 252, a teacher data storage unit 271, and a trained model storage unit 272.

[0106] The training data storage unit 271 stores training data that associates feature information obtained in the same manner as in the first embodiment with color information indicating the color selected by the subject. The trained model storage unit 272 stores a color estimation model trained for each individual through a training process using the training data stored in the training data storage unit 271. The color estimation model receives feature information as input and outputs color information.

[0107] The control unit 28 is configured using a processor such as a CPU or a GPU, and a memory. The control unit 28 functions as an information control unit 261, a feature calculation unit 262, a teacher data generation unit 281, and a learning unit 282 by the processor executing a program. All or part of the functions of the control unit 28 may be realized using hardware such as an ASIC, a PLD, or an FPGA. The above program may be recorded on a computer-readable recording medium. Examples of computer-readable recording media include portable media such as a flexible disk, a magneto-optical disk, a ROM, a CD-ROM, and a semiconductor storage device (e.g., an SSD), and storage devices such as a hard disk or semiconductor storage device built into a computer system. The above program may be transmitted via a telecommunications line.

[0108] The teacher data generating unit 281 acquires information on the color selected by the subject as the color representing the subject's emotion at the time of measurement of the biological information. The teacher data generating unit 281 writes teacher data in which the feature information, the selected color information, and the subject's personal ID are associated with each other into the teacher data storage unit 271.

[0109] The learning unit 282 executes a learning process for the color estimation model using the training data stored in the training data storage unit 271. Any machine learning method can be used to learn the color estimation model. Specific examples of the learning process include machine learning such as supervised learning using a neural network. The learning unit 282 generates a color estimation model for outputting color information representing emotions estimated from input feature information, for example, by performing supervised learning. The learning unit 282 records the generated color estimation model as a trained model in the trained model storage unit 272. The trained model obtained by the learning unit 282 may be transmitted to the estimation device 301 and recorded in an estimation model storage unit 371 of the estimation device 301, which will be described later.

[0110] 13 is a schematic block diagram showing a specific example of the functional configuration of an estimation device 301 according to the second embodiment. The estimation device 301 is configured using an information processing device such as a personal computer or a server device. The estimation device 301 includes a communication unit 31, a sensor input interface 32, an input unit 33, a display unit 34, a storage unit 37, and a control unit 38.

[0111] The storage unit 37 is configured using a storage device such as a magnetic hard disk drive or a semiconductor storage device. The storage unit 37 stores data used by the control unit 38. The storage unit 37 may function as, for example, an estimation model storage unit 371 and an estimation result storage unit 352.

[0112] The estimation model storage unit 371 stores a color estimation model used by the color estimation unit 381 when performing estimation processing. The color estimation model may be configured using information on a trained model generated in advance by, for example, a learning process. Such learning processing may be performed by, for example, another device (e.g., the learning device 201) or by the device itself (the estimation device 301). The color estimation model does not necessarily have to be generated by learning processing.

[0113] The control unit 38 is configured using a processor such as a CPU or a GPU, and a memory. The control unit 38 functions as an information control unit 361, a feature calculation unit 362, and a color estimation unit 381 when the processor executes a program. Note that all or part of the functions of the control unit 38 may be realized using hardware such as an ASIC, a PLD, or an FPGA. The above program may be recorded on a computer-readable recording medium. Examples of computer-readable recording media include portable media such as a flexible disk, a magneto-optical disk, a ROM, a CD-ROM, and a semiconductor storage device (e.g., an SSD), as well as storage devices such as a hard disk or semiconductor storage device built into a computer system. The above program may be transmitted via a telecommunications line.

[0114] The color estimation unit 381 performs estimation processing using the color estimation model stored in the estimation model storage unit 371 and the feature amount information generated by the feature amount calculation unit 362. Color information is estimated by the estimation processing. The color estimation unit 381 writes estimation result information indicating the estimated time and the estimated color information into the estimation result storage unit 352.

[0115] Fig. 14 is a flowchart showing a specific example of processing by the learning device 201. The learning device 201 performs the processing of Fig. 14 for each model creation subject. In Fig. 14, the same processes as those of the learning device 20 according to the first embodiment shown in Fig. 5 are denoted by the same reference numerals, and detailed descriptions thereof will be omitted.

[0116] The measurement device 10 transmits measurement information obtained by measuring the model creation subject to the learning device 201. The information control unit 261 of the learning device 201 writes the measurement information received from the measurement device 10 to the measurement information storage unit 251 (step S101). If the measurement information transmitted by the measurement device 10 does not include a personal ID, the information control unit 261 adds the personal ID of the model creation subject input by the input unit 23 to the received measurement information and writes the result to the measurement information storage unit 251.

[0117] The feature amount calculation unit 262 generates feature amount information based on the acquired measurement information, adds the personal ID of the person to be modeled, and writes the information to the feature amount information storage unit 252 (step S102). When the biometric information set in the acquired measurement information is used as is as input to the color estimation model, the feature amount calculation unit 262 writes the received measurement information to the feature amount information storage unit 252 as feature amount information.

[0118] The teacher data generation unit 281 generates teacher data that associates feature information with information about the color selected by the model creation subject as a color that represents emotion (step S301). Information about the color selected by the model creation subject is input using the input unit 23 through a process similar to the process in step S104 of Fig. 5. The teacher data generation unit 281 writes the teacher data that associates feature information, the input color information, and the personal ID of the model creation subject into the teacher data storage unit 271.

[0119] The learning unit 282 determines whether a predetermined number of pieces of teacher data necessary for learning a color estimation model are stored in the teacher data storage unit 271 (step S105). If the learning unit 282 determines that the number of pieces of teacher data does not reach the predetermined number, it repeats the processing from step S101. If the learning unit 282 determines that the predetermined number of pieces of teacher data is present, it executes a learning process using the teacher data stored in the teacher data storage unit 271 to generate a color estimation model (step S302). The learning unit 282 adds the personal ID of the model creation target person to the trained model, which is the generated color estimation model, and records it in the trained model storage unit 272.

[0120] The learning unit 282 may generate a color estimation model using training data from multiple people, and fine-tune the generated color estimation model using the training data from the person for whom the model is to be created.

[0121] Fig. 15 is a flowchart showing a specific example of the processing of the estimating device 301. In Fig. 15, the same processes as those of the estimating device 30 according to the first embodiment shown in Fig. 6 are denoted by the same reference numerals, and detailed description thereof will be omitted.

[0122] First, the user inputs the estimation target person ID and the usage model ID via the input unit 33 (step S201). The information control unit 361 reads out a color estimation model assigned a personal ID that matches the usage model ID input via the input unit 33 from the estimation model storage unit 371, and outputs the color estimation model to the color estimation unit 381 (step S401). If a color estimation model assigned a personal ID that matches the usage model ID is not stored in the estimation model storage unit 371, the information control unit 361 acquires a color estimation model assigned a usage model ID from the learning device 201 or the like, stores it in the estimation model storage unit 371, and outputs it to the color estimation unit 381.

[0123] The information control unit 361 acquires the measurement information of the estimation subject transmitted from the measurement device 10 (step S203). The feature amount calculation unit 362 generates feature amount information based on the measurement information acquired in step S203 (step S204). When the biometric information set in the acquired measurement information is used as feature amount information to be input into the color estimation model, the feature amount calculation unit 362 uses the received measurement information as feature amount information as is. The feature amount calculation unit 362 outputs the feature amount information to the color estimation unit 381.

[0124] The color estimation unit 381 inputs the feature information into the color estimation model acquired by the information control unit 361 in step S401, and obtains estimated color information (step S402). The color estimation unit 381 writes estimation result information, which associates the estimation time, estimated color information, and estimation target person ID, into the estimation result storage unit 352. The information control unit 361 transmits the estimated color information obtained in step S402 to the display device 40 (step S207). The control unit 42 of the display device 40 displays the color indicated by the received color information on the display unit 43. The information control unit 361 may also cause the display unit 34 to display the color indicated by the color information.

[0125] The estimation device 301 may include the learning unit 282 of the learning device 201. In step S207, the learning unit 282 of the estimation device 301 acquires the correct color information input by the user via the input unit 33, and uses the acquired correct color information and feature amount information as training data to learn (update) the color estimation model.

[0126] 15 at predetermined time intervals. As in the first embodiment, the control unit 42 of the display device 40 displays the colors indicated by the color information received from the estimation device 301 within a predetermined period on the display unit 43 in chronological order in the order of reception. Alternatively, in step S207, the information control unit 361 may transmit color display information to the display device 40, causing the color information indicated by the estimation result information associated with the estimation target person ID to be displayed in chronological order, as in the first embodiment.

[0127] As in the first embodiment, the estimation system 101 may generate a color estimation model for each attribute information. In this case, the attribute information is used instead of the personal ID of the person for whom the model is to be created in the process of Fig. 14. Furthermore, the attribute information is used as the model ID to be used in the process of Fig. 15.

[0128] Also, as in Figure 8, by using the personal ID of care recipient A as the estimated subject ID and the personal ID of caregiver B as the usage model ID, the estimation device 301 can display color information of caregiver B that represents the emotions of care recipient A on display device 40a, and by using the personal ID of caregiver B as the estimated subject ID and the personal ID of care recipient A as the usage model ID, the estimation device 301 can display color information of care recipient A that represents the emotions of caregiver B on display device 40b.

[0129] (Variation) In the above-described embodiment, the measurement device 10 and the estimation device 301 are configured as separate devices, but they may also be configured as an integrated device. In this case, the estimation device 301 shown in FIG. 13 is configured to further include the sensor 11 included in the estimation device 30 shown in FIG. 9. In step S203 of FIG. 15, the control unit 38 operates in response to an operation on the input unit 33, and acquires measurement information output from the sensor 11. The estimation device 301 may also receive measurement information from a measurement device 10 that measures a different type of biological information from that measured by the sensor 11.

[0130] Furthermore, the measurement device 10 and the learning device 201 may be configured as an integrated device. In this case, the learning device 201 shown in FIG. 12 is configured to further include the sensor 11 included in the estimation device 30 shown in FIG. 9. In step S101 of FIG. 14, the control unit 28 acquires measurement information output from the sensor 11. The learning device 201 may further receive measurement information from the measurement device 10 that measures a different type of biological information from that measured by the sensor 11.

[0131] In this embodiment, the learning device 201 and the estimation device 301 are configured as separate devices, but they may also be configured as an integrated device. In this case, the estimation model storage unit 371 of the estimation device 301 shown in FIG. 13 also functions as the trained model storage unit 272 of the learning device 201 shown in FIG. 12. Furthermore, the storage unit 37 functions in the same manner as the measurement information storage unit 251, the feature information storage unit 252, and the teacher data storage unit 271 of the learning device 201. In addition, the information control unit 361 and the feature calculation unit 362 of the estimation device 301 shown in FIG. 13 also perform the processes of the information control unit 261 and the feature calculation unit 262 of the learning device 201 shown in FIG. 12, respectively. Furthermore, the control unit 38 functions in the same manner as the teacher data generation unit 281 and the learning unit 282 of the learning device 201 shown in FIG. 12. Furthermore, the estimation device 301 may be configured as an integrated device by including the sensor 11 included in the estimation device 30 shown in Fig. 9, thereby integrating the measurement device 10, the learning device 201, and the estimation device 301. Furthermore, the estimation device 301 and the display device 40 may be integrated into one device.

[0132] The learning device 201 may be implemented using a plurality of information processing devices. For example, the learning device 201 may be implemented using a device such as a cloud. It is possible to arbitrarily determine which information processing device implements which functional unit of the learning device 201. Furthermore, the same functional unit of the learning device 201 may be realized by a plurality of information processing devices. Similarly, the estimation device 301 may be implemented using a plurality of information processing devices. For example, the estimation device 301 may be implemented using a device such as a cloud. It is possible to arbitrarily determine which information processing device implements which functional unit of the estimation device 301. Furthermore, the same functional unit of the estimation device 301 may be realized by a plurality of information processing devices.

[0133] According to the above-described embodiment, the estimation systems 100 and 101 can, for example, select an appropriate color based on the emotion of the care recipient and provide information useful for communication with the caregiver. Furthermore, by applying advanced technologies, such as using electroencephalogram (EEG) measurements for emotion estimation, it is possible to improve and streamline communication in care settings. Furthermore, the estimation systems 100 and 101 are suitable for use in various areas related to care, such as emotion monitoring and adjustment of the care environment. The present embodiment is expected to bring various benefits to care settings, such as reducing the burden on caregivers and improving the quality of care.

[0134] FIG. 16 is a diagram illustrating an outline of an example hardware configuration of an information processing device 90 applied to the first and second embodiments. The information processing device 90 includes a processor 91, a main storage device 92, a communication interface 93, an auxiliary storage device 94, an input / output interface 95, and an internal bus 96. The processor 91, the main storage device 92, the communication interface 93, the auxiliary storage device 94, and the input / output interface 95 are communicably connected to each other via the internal bus 96. The information processing device 90 may be applied to, for example, the learning devices 20 and 201 and the estimation devices 30 and 301. In this case, for example, the communication units 21 and 31 may be configured using the communication interface 93. For example, the memory units 25 and 27 and the memory units 35 and 37 may be configured using the auxiliary storage device 94. Furthermore, the control units 26 and 28 and the control units 36 and 38 may be configured using the processor 91 and the main storage device 92.

[0135] According to the embodiment described above, the estimation system includes a first emotion deduction unit, a learning unit, a second emotion deduction unit, and a color deduction unit. The estimation system corresponds, for example, to estimation system 100 of the embodiment. The first emotion deduction unit corresponds, for example, to emotion deduction unit 263 of the embodiment, and the second deduction emotion unit corresponds, for example, to emotion deduction unit 363 of the embodiment. When the estimation system includes a learning device and an estimation device, for example, the learning device may include the learning unit, or the estimation device may include the second emotion deduction unit and the color deduction unit. Furthermore, for example, the learning device may further include a first emotion deduction unit. The first emotion deduction unit estimates the emotion of the model generation subject based on biometric information obtained by measuring the model generation subject and that changes depending on the emotion. The learning unit trains a color estimation model representing the correspondence between emotion information and color information for each model generation subject using information on the estimated emotion and information on the color selected by the model generation subject as a color representing the emotion at the time the biometric information was measured. The second emotion estimation unit estimates the emotion of the estimation target based on biological information obtained by measuring the estimation target and that changes in response to the emotion. The color estimation unit estimates color information corresponding to the estimated emotion information of the estimation target using a color estimation model trained on a model generation target that is the same as or different from the estimation target. The color estimation unit may estimate color information corresponding to the emotion information of a target using the color estimation model.

[0136] The estimation system may also include a learning unit and a color estimation unit. The estimation system corresponds, for example, to the estimation system 101 of the embodiment. When the estimation system includes a learning device and an estimation device, for example, the learning device may include the learning unit, or the estimation device may include the color estimation unit. The learning unit learns a color estimation model representing the correspondence between biometric information and color information for each model generation subject, based on biometric information obtained by measuring the model generation subject and changing depending on emotions, and information on colors selected by the model generation subject as colors representing emotions at the time the biometric information was measured. The color estimation unit estimates color information corresponding to the biometric information obtained by measuring the estimation subject, using a color estimation model trained on a model generation subject that is the same as or different from the estimation subject.

[0137] The estimation system may further include an information control unit that displays, on a display device, one or both of the color information estimated by the color estimation unit and information indicating an emotion estimated based on the biological information of the estimation subject. The information control unit may also display, on the display device, the color information estimated by the color estimation unit at different times in chronological order.

[0138] The estimation subject may be a care recipient and the model generation subject may be a caregiver. Alternatively, the estimation subject may be a caregiver and the model generation subject may be a care recipient.

[0139] The display system may have a first display unit and a second display unit. The first display unit displays color information estimated by a color estimation unit of the estimation system when the estimation subject is a first estimation subject and the model generation subject is a second estimation subject. The second display unit displays color information estimated by the color estimation unit of the estimation system when the estimation subject is a second estimation subject and the model generation subject is the first estimation subject.

[0140] The display device may have a display unit that displays, in time series, color information corresponding to emotion information estimated based on biological information of the estimation subject.

[0141] The estimation system described above uses information about the subject's emotions and information about the colors selected by the subject corresponding to those emotions to train a color estimation model for each subject, which estimates color information corresponding to each emotion. Using the trained color estimation model, the estimation system can estimate and provide colors that represent emotions in a personalized color palette for each subject. For example, by using this estimation system to present colors corresponding to the emotions of the care recipient to caregivers, it can improve communication problems in caregiving and help resolve issues such as a shortage of caregivers and an increasing elderly population.

[0142] The estimation system can display emotions intuitively based on the color selected by the subject according to their emotion, and can even suggest the optimal color environment for the subject. In this way, color can be used to communicate the emotions of caregivers and care recipients, emotions can be monitored to promote communication among caregivers, and care environment coordination can be supported, thereby improving communication discord in caregiving and supporting the provision of more effective care.

[0143] Although an embodiment of the present invention has been described above in detail with reference to the drawings, the specific configuration is not limited to this embodiment, and includes designs within the scope of the gist of the present invention. [Explanation of symbols]

[0144] 100, 101...estimation system, 10...measuring device, 11...sensor, 20, 201...learning device, 21...communication unit, 22...sensor input interface, 23...input unit, 24...display unit, 25, 27...memory unit, 251...measurement information memory unit, 252...feature information memory unit, 253, 271...teaching data memory unit, 254, 272...trained model memory unit, 26...control unit, 261...information control unit, 262...feature calculation unit, 263...emotion estimation unit, 264, 281...teaching data generation unit, 265, 282...learning unit, 30, 301...estimation device, 31...communication unit, 32...sensor input interface, 33...input unit, 34...display unit, 35, 37...memory unit 351, 371... estimation model memory unit, 352... estimation result memory unit, 353... measurement information memory unit, 354... feature information memory unit, 355... teacher data memory unit, 36, 38... control unit, 361... information control unit, 362... feature calculation unit, 363, 363... emotion estimation unit, color estimation unit 364, 365... teacher data generation unit, 366... learning unit, 40... display device, 41... communication unit, 42... control unit, 43... display unit, 90... information processing device, 91... processor, 92... main memory device, 93... communication interface, 94... auxiliary memory device, 95... input / output interface, 96... internal bus

Claims

1. a learning unit that acquires emotion information as an estimation result from an emotion estimation unit that estimates the emotion of a model generation subject based on biological information that is obtained by measuring the model generation subject and that changes depending on the emotion, and that learns a color estimation model that represents correspondence between emotion information and color information for each model generation subject, using the estimated emotion information and information on a color selected by the model generation subject as a color that represents the emotion at the time the biological information was measured; A learning device comprising:

2. a learning unit that learns, for each model generation subject, a color estimation model that represents the correspondence between biometric information and color information, based on biometric information that is obtained by measuring the model generation subject and that changes depending on emotions, and information on colors selected by the model generation subject as colors that represent emotions at the time of measuring the biometric information; A learning device comprising:

3. an emotion estimation unit that estimates an emotion of the subject to be estimated based on biological information that is obtained by measuring the subject and that changes depending on the emotion; a color estimation unit that estimates color information corresponding to the estimated emotion information using a color estimation model that represents a correspondence between emotion information and color information, the color estimation model is trained for each model generation subject using information on a plurality of different emotions and information on colors selected by a model generation subject who is the same as or different from the estimation subject as colors representing the plurality of different emotions, Estimation device.

4. a color estimation unit that estimates color information corresponding to the biometric information obtained by measuring the subject, using a color estimation model that represents correspondence between biometric information that changes in response to emotions and color information; The color estimation model is trained for each model generation subject using biometric information obtained by measuring a model generation subject who is the same as or different from the estimation subject at different times, and information on colors selected by the model generation subject as colors representing emotions at each time the biometric information is measured. Estimation device.

5. an information control unit that performs a process of displaying, on a display device corresponding to the first estimation target person, information on the color estimated by the color estimation unit when the estimation target person is a first estimation target person and the model generation target person is a second estimation target person, and a process of displaying, on a display device corresponding to the second estimation target person when the estimation target person is the second estimation target person and the model generation target person is the first estimation target person, The estimation device according to claim 3 or 4.

6. The estimation subject is a care recipient, and the model generation subject is a caregiver. Or, The estimation subject is a caregiver, and the model generation subject is a care recipient. The estimation device according to claim 3 or 4.

7. a first emotion estimation unit that estimates an emotion of a model generation subject based on biological information that is obtained by measuring the model generation subject and that changes depending on the emotion; a learning unit that uses the estimated emotion information and information on a color selected by the model generation subject as a color representing the emotion at the time of measuring the biological information to learn a color estimation model that represents a correspondence between emotion information and color information for each model generation subject; a second feeling deduction unit that deduces a feeling of the person to be inferred based on biological information that is obtained by measuring the person to be inferred and that changes in response to the feeling; a color estimation unit that estimates color information corresponding to the estimated emotion information of the estimation target person using the color estimation model trained on a model generation target person that is the same as or different from the estimation target person; An estimation system comprising:

8. a learning unit that learns, for each model generation subject, a color estimation model that represents the correspondence between biometric information and color information, based on biometric information that is obtained by measuring the model generation subject and that changes depending on emotions, and information on colors selected by the model generation subject as colors that represent emotions at the time of measuring the biometric information; a color estimation unit that estimates color information corresponding to biological information obtained by measuring the estimation subject using the color estimation model trained on a model generation subject that is the same as or different from the estimation subject; and An estimation system comprising:

9. a process of estimating color information corresponding to emotion information estimated based on biometric information of the estimation subject using a color estimation model that represents the correspondence between emotion information and color information that represents the emotion of the model generation subject, or a display unit that displays the color information obtained by a process of estimating color information corresponding to the biometric information of the estimation subject using a color estimation model that represents the correspondence between the biometric information of the model generation subject and colors that represent emotions of the model generation subject at the time of measuring the biometric information of the model generation subject; A display device comprising:

10. a first display unit that displays color information obtained by a process of estimating color information corresponding to emotion information estimated based on biometric information of a first estimated subject using a color estimation model that represents the correspondence between emotion information and color information that represents the emotion of a second estimated subject, or a process of estimating color information corresponding to the biometric information of the first estimated subject using a color estimation model that represents the correspondence between the biometric information of the second estimated subject and colors that represent emotions of the second estimated subject at the time of measuring the biometric information of the second estimated subject; a second display unit that displays color information obtained by a process of estimating color information corresponding to emotion information estimated based on the biometric information of the second estimation subject using a color estimation model that represents a correspondence between emotion information and color information that represents the emotion of the first estimation subject, or a process of estimating color information corresponding to the biometric information of the second estimation subject using a color estimation model that represents a correspondence between the biometric information of the first estimation subject and colors that represent emotions of the first estimation subject at the time of measuring the biometric information of the first estimation subject; A display system comprising:

11. an acquisition step of acquiring emotion information obtained by measuring the model generation subject and estimating the emotion of the model generation subject based on biological information that changes depending on the emotion; a learning step of learning, for each model generation subject, a color estimation model that represents a correspondence between emotion information and color information, using the estimated emotion information and information on a color selected by the model generation subject as a color that represents the emotion at the time of measuring the biological information; A learning method that has

12. a learning step of learning, for each model generation subject, a color estimation model that represents the correspondence between biometric information and color information, based on biometric information that is obtained by measuring the model generation subject and that changes depending on emotions, and information on colors selected by the model generation subject as colors that represent the emotions at the time of measuring the biometric information; A learning method that has

13. an emotion estimation step of estimating an emotion of the subject to be estimated based on biological information obtained by measuring the subject and which changes depending on the emotion; a color estimation step of estimating color information corresponding to the estimated emotion information using a color estimation model that represents a correspondence between emotion information and color information, the color estimation model is trained for each model generation subject using information on a plurality of different emotions and information on colors selected by a model generation subject who is the same as or different from the estimation subject as colors representing the plurality of different emotions, Estimation method.

14. a color estimation step of estimating color information corresponding to the biometric information obtained by measuring the subject, using a color estimation model that represents correspondence between biometric information that changes in response to emotions and color information; The color estimation model is trained for each model generation subject using biometric information obtained by measuring a model generation subject who is the same as or different from the estimation subject at different times, and information on colors selected by the model generation subject as colors representing emotions at each time the biometric information is measured. Estimation method.

15. Computer, A computer program for causing the learning device according to claim 1 or 2 to function.

16. Computer, A computer program for causing the estimation device according to claim 3 or 4 to function.

17. Computer, A computer program for causing the estimation system according to claim 7 or 8 to function.

Images