JP2021108918A - Aroma information providing device, aroma information providing method, aroma information providing program, and aroma diffuser - Google Patents

Abstract

To provide aroma information suitable of all the users existing in a predetermined space on the basis of the mental conditions of all the users.SOLUTION: An aroma information providing device 1 comprises: an acquisition unit for acquiring brain wave information of a user; an analysis unit for analyzing the brain wave information acquired by the acquisition unit; an estimation unit for estimating an aroma suitable to the user, based on a result of the analysis by the analysis unit; and an output unit for outputting aromatic information indicating the aroma information estimated by the estimation unit.SELECTED DRAWING: Figure 1

Description

The present invention relates to a scent information providing device, a scent information providing method, a scent information providing program, and a scent diffuser. Specifically, the present invention relates to a device that analyzes brain waves of a plurality of users and provides information on scents that are preferable to humans existing in a predetermined space, a method of providing the scent, a program to be provided, and a diffuser that provides the scent.

Conventionally, as in Patent Document 1 below, a scent generator that mixes and produces a plurality of types of scents is known.

JP-A-2019-170692

In the scent generator as in Patent Document 1, a scent containing a plurality of scents is emitted according to the user's comfort level and achievement level. However, in recent years, it has been desired to provide an optimum scent for all a plurality of users existing in a predetermined space.

Therefore, the present invention provides a scent information providing device, a scent information providing method, and a scent information providing that can provide optimum scent information to all a plurality of users existing in a predetermined space based on the mental state of each user. It is intended to provide programs and scent diffusers.

In order to solve the above problems, the scent information providing device of the present invention is based on an acquisition unit that acquires user's brain wave information, an analysis unit that analyzes the brain wave information acquired by the acquisition unit, and an analysis result by the analysis unit. It also includes an estimation unit that estimates a scent suitable for the user, and an output unit that outputs fragrance information indicating the scent information estimated by the estimation unit.

The acquisition unit may acquire the electroencephalogram information of a plurality of users, and the output unit may output the fragrance information indicating one scent information suitable for all the users from the electroencephalogram information of the plurality of users.

The scent information providing device of the present invention includes a storage unit that stores brain wave information and a scent learning model that learns the relationship between the content of the brain wave information and the scent, and the analysis unit performs sensitivity analysis on the brain wave information. Sensitivity information, which is information that quantifies the user's sensibilities, may be generated, and the estimation unit may use the sensation information as input and use a scent learning model to estimate a scent suitable for the user.

The fragrance information providing device of the present invention includes a storage unit that stores sensitivity information that is information obtained by quantifying the user's sensitivity, and the estimation unit is suitable for the user according to the difference from the threshold value by inputting the sensitivity information. The scent may be estimated.

The scent information providing device of the present invention includes a storage unit that stores sensibility information, which is information obtained by quantifying the user's sensibility, and the estimation unit inputs the sensibility information and uses a scent learning model to scent suitable for the user. May be estimated.

The user's sensibility includes a plurality of sensibilities, the analysis unit generates quantified information for each of the user's multiple sensibilities as sensibility information indicating the user's sensibility, and the estimation unit is among the user's multiple sensibilities. A scent suitable for the user may be estimated using a scent learning model by inputting a high numerical sensitivity.

The analysis unit may generate sensibility information indicating sensibilities exceeding a predetermined threshold value for each numerical value of the plurality of sensibilities of the user.

The user's sensibility includes a plurality of sensibilities, the analysis unit generates quantified information for each of the user's multiple sensibilities as sensibility information indicating the user's sensibility, and the estimation unit is among the user's multiple sensibilities. A scent suitable for the user may be estimated using a scent learning model by inputting a characteristic having a low numerical value.

The analysis unit may generate sensibility information indicating sensibilities equal to or less than a predetermined threshold value for each numerical value of the plurality of sensibilities of the user.

The analysis unit may generate sensibility information indicating a predetermined number of user sensibilities from the lowest numerical value for each of the plurality of user sensibilities.

The output unit may further output sensitivity information indicating the user's sensitivity.

The scent information providing device of the present invention further includes a reception unit that receives input of sensitivity information about the user from the user, and a storage unit that holds a first conversion table that links the relationship between the user's sensitivity value and the scent. The estimation unit may further estimate the scent suitable for the user by adding the user's sensitivity information received by the reception unit and the scent information specified by using the first conversion table. ..

The first conversion table includes a first sub-table in which the sensibility value and the rank are associated with each other, and a second sub-table in which the relationship between the rank and the scent is associated with each other. It may be specified and the scent corresponding to the specified rank may be specified.

The reception unit may provide the user with a user interface that indicates options indicating the sensibility contents as a plurality of sensibility values.

Further, in order to solve the above problem, the scent information providing method for providing the scent information of the present invention analyzes the acquisition step of acquiring the user's brain wave information from the user and the user's brain wave information acquired in the acquisition step. The analysis step includes an estimation step for estimating a scent suitable for the user based on the analysis result in the analysis step, and an output step for outputting fragrance information indicating the scent information estimated in the estimation step.

Further, in order to solve the above problems, the scent diffuser that provides the scent of the present invention has an acquisition unit that acquires the user's brain wave information, an analysis unit that analyzes the brain wave information acquired by the acquisition unit, and an analysis unit. It is provided with an estimation unit that estimates a scent suitable for the user based on the result of the above, and an ejection unit that ejects a fragrance based on the scent information estimated by the estimation unit.

Further, in order to solve the above problems, the scent information providing program that provides the scent of the present invention has an acquisition function for acquiring the user's brain wave information and an analysis function for analyzing the brain wave information acquired by the acquisition function. And, based on the result of the analysis by the analysis function, the estimation function for estimating the scent suitable for the user and the output function for outputting the fragrance based on the scent information estimated by the estimation function are realized.

According to the scent information providing device, the scent information providing method, the scent information providing program, and the scent diffuser of the present invention, the scent information can be obtained by analyzing the brain waves of a plurality of users existing in a predetermined space. Therefore, it is possible to provide customized optimum scent information to all of a plurality of users based on the mental state of all of a plurality of users existing in a predetermined space.

It is a figure which shows the whole structure which provides the scent information of 1st Embodiment of this invention. It is a figure which shows the structural example of the scent information providing apparatus of 1st Embodiment of this invention. It is a figure which shows the structural example of the diffuser of 1st Embodiment of this invention. It is a figure which shows the structural example of the electroencephalograph of 1st Embodiment of this invention. (A) is a figure which shows the data (the table which shows the correspondence between the numerical value of sensitivity and rank) of 1st Embodiment of this invention. (B) is a diagram showing data (a table showing the correspondence between sensibilities and ranks and scents) of the first embodiment of the present invention. It is a sequence diagram which shows the exchange between the apparatus of 1st Embodiment of this invention. It is a flowchart which shows the operation of the electroencephalograph of 1st Embodiment of this invention. It is a flowchart which shows the operation of the scent information providing apparatus of 1st Embodiment of this invention. It is a flowchart which shows the operation of the diffuser of 1st Embodiment of this invention. It is a figure which shows the usage form which the user of the 1st Embodiment of this invention wears an electroencephalograph. It is a figure showing the screen for selecting which one to display from the graph which changed the signal measured by the electroencephalograph of 1st Embodiment of this invention to the user's sensibility. It is a graph which shows the change of sensitivity estimated from the electroencephalogram of the 1st embodiment of this invention. It is a sensitivity graph image derived from the electroencephalogram of the first embodiment of the present invention. It is a block diagram which shows the structural example of the diffuser of 2nd Embodiment of this invention. FIG. 5 is a structural diagram in which the diffuser of the second embodiment of the present invention injects a plurality of types of scents. It is a flowchart which shows the operation of the diffuser of 2nd Embodiment of this invention.

[First Embodiment]
[Aroma information providing device]

FIG. 1 shows an overall configuration in which scent information is provided by the scent information providing device of the present invention. The scent information providing device 1 is a device that provides information on scents. The scent information providing device 1 is a device that identifies the optimum scent for a plurality of users and determines the scent to be provided based on the brain waves of the plurality of users. The fragrance information providing device 1 is realized by an information processing device typified by a personal computer, a tablet terminal, a server device, or the like, as an example. The diffuser 2 is a device that emits a scent. The electroencephalograph 3 is a device that detects and measures the electroencephalograms of a plurality of users. The scent information providing device 1 is connected to the diffuser 2, and the electroencephalograph 3 is connected to the network 30. The scent information providing device 1 acquires the user's electroencephalogram information from the electroencephalograph 3 via the network 30. The diffuser 2 and the electroencephalograph 3 exist in a predetermined space 4 such as a workplace where a plurality of users represented by a square in the lower part of FIG. 1 exist at the same time. The electroencephalograph 3 is attached to the heads of the plurality of users, and transmits the acquired electroencephalograph information to the scent information providing device 1 via the network 30. The scent information providing device 1 accesses the network 30 and acquires the user's brain wave information. The user's brain wave information may include a mental state such as the stress level of the user, and the fragrance information providing device 1 generates sensitivity information from the brain wave information. The scent information providing device 1 can generate scent information suitable for all a plurality of users existing in a predetermined space such as a workplace from the sensibility information, and is used for scenting smoke at a live venue or its scent. Information can also be provided.

Next, each component of the scent information providing device 1 will be described with reference to FIG. The scent information providing device 1 includes a reception unit 10, a control unit 20, a storage unit 60, and an output unit 80. Each component will be described in detail below.

The reception unit 10 is a communication interface that receives user information via the network 30.

The reception unit 10 receives input of sensitivity information indicating the user's sensitivity from the user. The sensibility information is information indicating the mental state of the user according to a predetermined environment. As an example, as shown in FIG. 11, the information indicating the mental state of the user includes INTEREST (interest level), LIKE (favorite level), CONCENTRATION (concentration rate), STRESS (stress level), CALMNESS (calmness level), and the like. Is. The user inputs, for example, an interest degree of 50%, a favorite degree of 40%, and the like.

FIG. 11 is a UI (User Interface) screen for inputting information related to the user's sensitivity information. In this embodiment, as shown in FIG. 11, the choice of the scent system is displayed on the user's terminal and selected, but this is an example. For example, the user can imagine a favorite scent. The text may be directly input. In this case, a table in which the relationship between the text that may be input and the scent is associated is stored in advance, and the input text and the text that may be input that are stored in advance are stored in advance. The degree of correlation with the content may be specified, and the scent associated with the specified text having a high degree of correlation may be estimated as a favorite scent.

The control unit 20 includes an acquisition unit 40, an analysis unit 50, and an estimation unit 70, controls the information received by the reception unit 10, and uses the conversion table 61 stored in the storage unit 60 to provide user information. Is converted into scent information.

The acquisition unit 40 acquires the user's electroencephalogram information from the electroencephalograph 3 described later. The electroencephalogram information is a signal directly transmitted from the electroencephalograph 3 and is a signal obtained by measuring an electric signal generated from the user's brain with electrodes. Here, the acquisition unit 40 acquires brain wave information from the network 30, that is, online information, but acquires the brain wave information so that the brain wave information can be acquired even when the connection is disconnected from the network 30 for some reason. The unit 40 may acquire offline information from the computer. As an example, the acquisition unit 40 connects the flash memory to the information acquired in advance in the online state and the text information created by the user stored in the recording medium such as the flash memory held by the user. It may be the one to be acquired. The server acquires these offline information from files stored in the shared folder of the server or the like.

Specifically, the acquisition unit 40 accesses the network 30 and collects the user's electroencephalogram information from the electroencephalograph 3 described later. The acquisition unit 40 obtains electroencephalogram information of, for example, all employees existing in a predetermined department of the company (for example, 10 people, but the number is not limited to 10 people, and may be more or less). collect.

The analysis unit 50 analyzes the brain wave information acquired by the acquisition unit 40. Analyzing here means quantifying the sensibility from the user's brain waves. The analysis unit 50 performs electroencephalogram analysis on the electroencephalogram information and generates emotional information indicating the content of the electroencephalogram information. The sensibility information referred to here may be numerical information obtained as a result of brain wave analysis, and may be any numerical value useful for grasping the sensibility of the user, which is the mental state of the user. For example, the sensibility such as the degree of interest in the sensibility information and a numerical value from 0 to 100 may be used.

FIG. 12 is a diagram in which an electric signal measured by an electroencephalograph is converted into each sensibility and converted into a graph showing a temporal transition. However, the present invention is not limited to a graph, and may be a table or the like showing numerical values of sensibilities at each time. In the graph shown in FIG. 12, it can be seen that the LIKE (favorite degree) is 0 to 45 seconds and high in 60 to 65 seconds. On the contrary, it can be seen that the INTEREST (interest rate) has the lowest value per 55 seconds or between 60 and 65 seconds.

The user's sensibility may include a plurality of sensibilities. The plurality of sensibilities are, for example, INTEREST (interest level), LIKE (favorite level), CONCENTRATION (concentration level), STRESS (stress level), CALMNESS (calmness level), and the like. These characteristics are quantified, for example, the degree of interest is 80, the degree of liking is 30, the degree of concentration is 40, the degree of stress is 50, and the degree of calm is 60, with the maximum value being 100 and the minimum value being 0. The analysis unit 50 generates quantified information for each of the plurality of sensibilities of the user as sensibility information indicating the user's sensibilities. Specifically, the analysis unit 50 analyzes the mental state of the user from the brain wave information of the user acquired by the acquisition unit 40.

Further, the analysis unit 50 may quantify all of the plurality of sensibilities of the user, or may quantify only the sensibilities selected from the plurality of sensibilities. Then, the one with the higher numerical value is selected. The maximum value is 100, the minimum value is 0, and the analysis unit 50 analyzes the brain wave information, for example, the degree of interest is 80, the degree of liking is 30, the degree of concentration is 40, the degree of stress is 50, and the degree of calm is 30. Each of the sensitivity values of the sensitivity information is compared with the threshold value. A scent that matches the sensibilities is created by using the sensibilities that exceed the threshold value as the characteristics of the user. When the threshold value is, for example, 40, a scent suitable for a user having a high degree of interest and a high degree of stress is created. For example, the degree of interest, the degree of liking, the degree of concentration, the degree of calm, etc. are examples in which higher numerical values are preferred.

In this way, by selecting the one with a high numerical value of the user's sensibility by the analysis unit 50, the scent information providing device 1 can provide the scent information that can further emphasize the sensibilities appearing to a plurality of users. .. Emphasizing sensibility means, for example, providing scent information that can increase the degree of interest when the degree of interest is increasing, but is not limited to this example. ..

Further, the analysis unit 50 quantifies all of the plurality of sensibilities of the user, and selects the one having the lower numerical value. The maximum value is 100, the minimum value is 0, and the analysis unit 50 analyzes the brain wave information, for example, the degree of interest is 30, the degree of liking is 40, the degree of concentration is 50, the degree of stress is 60, and the degree of calm is 20. Each of the sensitivity values of the sensitivity information is compared with the threshold value. A scent that matches the sensibilities is created with the sensibilities below the threshold as the characteristics of the user. When the threshold value is, for example, 40, a scent suitable for a user with a low degree of interest is created. For example, the degree of stress is an example in which a lower numerical value is preferred.

In this way, the analysis unit 50 selects the one having a low numerical value of the user's sensibility, so that the scent information providing device 1 can provide the scent information that can reinforce the lacking sensibility of the plurality of users. .. Providing scent information that reinforces weak sensibilities means, for example, providing scent information that can enhance concentration when concentration is reduced. It is not limited.

The electroencephalogram information of all the users existing in the predetermined space may be analyzed, or the electroencephalogram information of a predetermined number of people (for example, 10 people) in the predetermined space may be analyzed. The storage unit 60, which will be described later, holds a conversion model (learning model) for converting the numerical value obtained as a result of the electroencephalogram analysis into the user's sensitivity information. Quantify the degree of liking, concentration, stress, calmness, etc.

In addition, as a method of estimating a scent suitable for a user using an electroencephalogram, 1) a method of estimating a scent using a learning model directly from an electroencephalogram, and 2) a sensibility is specified from the electroencephalogram, and the specified sensibility is applied. There are a method of estimating the scent using the threshold value and a method of 3) identifying the sensibility from the brain wave and estimating the scent from the specified sensibility using the learning model.

As 1), the storage unit 60 stores a scent learning model that learns the relationship between the brain wave information, the content of the brain wave information, and the scent, and learns directly from the brain wave signal without analyzing the brain wave by the analysis unit 50. The model performs estimation processing to generate sensitivity information that shows the contents of brain wave information in a graph, and the estimation unit 70 estimates a scent suitable for the user by using the scent learning model with the sensitivity information as an input. The details of the estimation unit 70 will be described later.

As 2), the storage unit 60 stores the user's sensibility, the analysis unit 50 performs sensibility analysis on the brain wave information to generate sensibility information indicating the user's sensibility, and the estimation unit 70 inputs the sensibility information. As, the scent suitable for the user is estimated using the threshold value.

As 3), the storage unit 60 stores the user's sensibility, the analysis unit 50 performs sensibility analysis on the brain wave information to generate sensibility information indicating the user's sensibility, and the estimation unit 70 inputs the sensibility information. As a scent learning model, a scent suitable for the user is estimated. As the scent learning model, the storage unit 60 stores the scent learning model in which the sensibility information and the relationship between the content of the sensibility information and the scent are learned.

Learning models can be created by supervised learning with feedback, unsupervised learning without feedback, or reinforcement learning in which feedback is given when a reinforcement signal is received (for example, when the target state is reached). Created by Alternatively, it is created by deep learning using a multi-layer neural network with four or more layers that processes an input vector and converts it into an output vector using a model inspired by the connections between neurons in the brain. May be done. In this embodiment, as an example, a learning model in which the relationship between the electroencephalogram information and the user's sensibility is learned by learning the combination of the electroencephalogram state and the sensibility and its numerical value associated with the brain wave state is used. The user's sensibility is calculated by creating and performing estimation processing by this learning model. The learning model includes a scent learning model that converts the user's sensibility into a scent and a sensibility learning model that converts the brain wave analysis result into the user's sensibility. The scent learning model and the sensitivity learning model may be realized by one learning model.

The storage unit 60 holds a first conversion table 61 that links the relationship between the user's sensitivity value report and the scent. The first conversion table 61 includes a first sub-table 62 in which the sensitivity value and the rank are associated with each other, and a second sub-table 63 in which the relationship between the rank and the scent is associated with each other. The first sub-table 62 is a table used by the estimation unit 70 to rank each of the user's sensitivity values analyzed by the analysis unit 50, that is, to determine the degree of height of the sensitivity value. As shown in FIG. 5A, in the first sub-table 62, the sensitivity information such as the degree of interest, the degree of liking, the degree of concentration, the degree of stress, the degree of calm, etc. is 50 in the second row. Numerical values (%) of 40, 30 ... Are ranked in the third column as A, B, C, .... The second sub-table 63 is a table used by the estimation unit 70 to estimate the optimum scent type from the sensibilities ranked by the first sub-table 62. As shown in FIG. 5B, the second sub-table 63 is associated with the sensitivity values and ranks of a plurality of users and the scents (herb-based, citrus-based, ...). As a result, the rank is first specified by inputting the sensitivity value, and the scent corresponding to the specified rank is specified. Specifically, as an example, when the average value of the degree of interest of a plurality of users is 50 or more and the average value of the degree of liking is also 50 or more, as can be understood from FIG. 5A, each of them. Since the sensibilities of both are A, it can be estimated from FIG. 5 (b) that the provided scent is citrus.

In this way, the storage unit 60 associates the user's sensitivity value with the rank in the first sub-table 62, and associates the user's sensitivity value with the rank and the scent in the second sub-table 63. By providing the table 61, the estimation unit 70 can accurately estimate the scent based on the first conversion table 61.

The estimation unit 70 estimates one scent suitable for a plurality of users existing in the predetermined space 4 based on the result of the analysis analyzed by the analysis unit 50 using the first conversion table 61. Estimating here means determining the optimum scent for a plurality of users in a predetermined environment. For example, if the entire user is under stress, herbs (such as peppermint) and woody materials that are said to have a stress-relieving effect are estimated. Further, for example, if the entire user is in a state where the concentration level is low, a citrus material (orange, etc.), which is said to have an effect of increasing the concentration level, is estimated. This material estimation is performed by inputting the numerical value obtained as a result of brain wave analysis into the storage unit 60, which holds a conversion model (learning model) for converting into the user's sensitivity information. Be struck. This estimation is just an example and is not limited to this.

In this way, the estimation unit 70 estimates the scent suitable for the user by inputting the sensibility information, so that the scent information providing device 1 has the scent that matches the sensibilities of all the plurality of users existing in the predetermined space 4. Information can be provided.

Further, the estimation unit 70 uses the scent learning model as an input of the sensibility information to estimate one scent suitable for all the plurality of users existing in the predetermined space 4. The estimation unit 70 further adds the user's sensitivity information received by the reception unit 10 and the scent information specified by using the first sub-table 62, and all the plurality of users existing in the predetermined space 4 are added. Estimate one scent suitable for. The estimation unit 70 is suitable for all a plurality of users existing in the predetermined space 4 by further adding the sensitivity information received by the reception unit 10 and the scent information specified by using the second sub-table 63. Estimate only one scent.

For example, when "stress level" is selected from multiple sensibility information, a specific herbal material (lavender, etc.) that is expected to relax and improve work efficiency when the stress level is high. Is estimated by the estimation unit 70. Further, for example, when "concentration ratio" is selected, the estimation unit 70 estimates a specific citrus material that makes one feel positive and is expected to improve concentration.

The sensibility values of a plurality of users are obtained by taking the average value of the sensibility values of the plurality of users. When the sensibilities are batting, that is, when a plurality of sensibilities decrease and it becomes necessary to raise them, a plurality of fragrances that raise those sensibilities are blended to generate a scent. Further, the sensibilities of a plurality of users may be simply added up to use the sensibilities having a high total value.

The output unit 80 outputs fragrance information indicating the scent information estimated by the estimation unit 70. When the estimation unit 70 uses the analysis information as input and estimates one scent suitable for all a plurality of users existing in the predetermined space 4 by using the scent learning model, the scent is estimated based on the estimated scent system. Output. The output unit 80 may further output sensitivity information indicating the user's sensitivity. What is output from the output unit 80 may be audio or an image. The output information may be graphed and visualized as shown in FIG. The output unit 80 may be a communication interface, may communicate with a computer and transmit information to the computer, or may include a monitor and display an image on the monitor. Further, the output unit 80 may be capable of either wired or wireless communication, and may be communicated by any communication protocol as long as it can communicate with the diffuser 2 and the electroencephalograph 3.

The operator of the scent information providing device may exchange the fragrance that emits the specified scent, which is stored in the diffuser 2 or the like in the workplace or the like.

The scent estimation based on the sensitivity information input in FIG. 11 and the scent estimation based on the electroencephalogram analysis may be combined, for example, at a ratio of 3: 7, a ratio of 5: 5, or a ratio of 7: 3. May be combined with. This is an example, the ratio is arbitrary, and one may be 0. The scent information can be obtained by estimating the scent suitable for all the plurality of users from the electroencephalographs 3 worn by all the plurality of users existing in the predetermined space 4 by the electroencephalogram analysis.

According to such a configuration of the present embodiment, it is possible to provide customized optimum scent information to all the plurality of users based on the mental states of all the plurality of users existing in the predetermined space 4.

Next, a configuration example of the diffuser 2 will be described with reference to FIG. The diffuser 2 includes a communication unit 110, a control unit 111, a drive unit 112, and a ejection unit 113. The communication unit 110 is connected to the scent information providing device 1, communicates with the scent information providing device 1, and receives the scent information transmitted from the scent information providing device 1. The control unit 111 controls the drive of the drive unit 112. The drive unit 112 is driven by the control unit 111, and the ejection unit 113 ejects the scent based on the scent information received by the communication unit 110.

Next, the configuration of the electroencephalograph 3 will be described with reference to FIG. The electroencephalograph 3 includes a sensor 120, a communication unit 121, a storage unit 122, and a control unit 123. The sensor 120 detects the user's brain wave information. The communication unit 121 is connected to the network 30, communicates with the network 30, and transmits the electroencephalogram information detected by the sensor 120 to the network 30. The storage unit 122 stores the electroencephalogram information detected by the sensor 120. The control unit 123 controls the communication unit 121 to communicate with the network 30.

As shown in FIG. 10, the electroencephalograph 3 is a headgear worn on the user's head and used, a device that senses an electric signal of the user's brain, and a device that outputs the sensed electric signal. .. The electrosurgical meter 3 may be configured to be able to communicate with an information processing device such as a tablet terminal. In this case, the electric signal measured by the electrosurgical meter 3 may be analyzed by the tablet terminal and converted into sensitivity information. When the brain wave meter side analyzes the brain wave information and converts it into the sensitivity information, the analysis by the analysis unit 50 of the fragrance information providing device 1 becomes unnecessary, and the analysis information (sensitivity information) output by the tablet terminal is estimated by the estimation unit 70. It can also be an input. In the present embodiment, basically, the electroencephalograph 3 is measured by the sensor 120 and transmits an electric signal of the brain from the communication unit 121 to the scent information providing device 1.

FIG. 11 is a UI (User Interface) screen for inputting information related to the user's sensitivity information. In this embodiment, as shown in FIG. 11, the choice of the scent system is displayed on the user's terminal and selected, but this is an example. For example, the user can imagine a favorite scent. The text may be directly input. Basically, the scent is estimated from the brain wave information, but if there is something wrong with the headgear or if the user dislikes wearing the headgear, such an input can be used.

[How to provide scent information]
Hereinafter, the scent information providing method according to the present invention will be described with reference to FIGS. 6 to 9. In the scent information providing method according to the present invention, the user's brain wave information is analyzed by the electroencephalograph 3 attached to the user to determine the user's mental state, and the scent is created based on the mental state or the scent of the scent. This is the process of creating the original information.

FIG. 6 is a sequence diagram showing an example of exchange between each device, and is a diagram showing an example of exchange for providing scent information. The electroencephalograph 3 measures the user's electroencephalogram in step S601, and transmits the electroencephalogram information to the scent information providing device 1 in step S602. The scent information providing device 1 performs brain wave analysis in step S603, and estimates the scent from the analysis result in step S604. Then, the estimated scent information is transmitted to the diffuser 2 in step S605. The diffuser 2 acquires the scent information transmitted from the scent information providing device 1, and ejects the scent estimated in step S606.

<Operation of electroencephalograph>
FIG. 7 is a flowchart illustrating the operation of the electroencephalograph 3 in the scent information providing method of the present invention.

In the reception step S11, the sensor 120 appropriately receives the user's electroencephalogram information. The sensor 120 basically continuously performs sensing, but may perform sensing at predetermined time intervals (for example, every 1 ms). The electroencephalogram information includes selective input on the user interface and electroencephalogram information acquired by the electroencephalograph 3. The electroencephalogram information includes information that can be inferred from INTEREST (interest), LIKE (likeness), CONCENTRATION (concentration), STRESS (stress), CALMNESS (calmness), and the like. Here, the sensor 120 is an electrode and acquires brain wave information, that is, an electric signal from the user's brain.

Next, in the communication step S12, the communication unit 121 transmits the brain wave information received in the reception step S11 to the network 30.

The storage unit 122 may provide a storage step S13 for storing the electroencephalogram information received by the sensor 120.

The control unit 123 may provide the control step S14 for controlling the communication unit 121 to transmit the electroencephalogram information to the network 30.

<Aroma information providing device>
FIG. 8 is a flowchart showing the operation of the scent information providing device 1 in the scent information providing method according to the present invention.

In the reception step S110, the reception unit 10 receives the user's electroencephalogram information from the network 30.

Next, in the acquisition step S120, the acquisition unit 40 acquires the brain wave information of the user received by the reception unit 10. The electroencephalogram information is a signal directly transmitted from the electroencephalograph 3 and is a signal obtained by measuring an electric signal generated from the user's brain with electrodes. When the acquisition unit 40 acquires the brain wave information, the acquisition unit 40 transmits the acquired brain wave information to the analysis unit 50.

Next, in the analysis step S130, the analysis unit 50 analyzes the electroencephalogram information transmitted from the acquisition unit 40 in the acquisition step S120. As a method for analyzing brain wave information, a known algorithm may be used.

The analysis step S130 includes a digitization step S131 and a visualization step S132.

In the quantification step S131, the analysis unit 50 quantifies the level of the user's sensitivity element. Specifically, from the analysis of brain wave information, the levels such as the degree of interest, the degree of liking, the degree of concentration, the degree of stress, and the degree of calm are quantified as elements of the user's sensitivity.

Next, in the visualization step S132, the level of the user's sensitivity element quantified in the quantification step S131 is visualized in a graph. Specifically, the analysis unit 50 may use an existing electroencephalogram analysis algorithm to convert the electroencephalogram information acquired by the acquisition unit 40 into sensibility information obtained by quantifying various sensibilities.

The storage step S140 includes a selection step S141, a first conversion step S142, and a second conversion step S143. In the selection step S141, the optimum scent is selected based on the result analyzed in the analysis step S130. The selection step S141 may be omitted.

In the first conversion step S142, the user's electroencephalogram information is converted into the scent information by the first conversion table 61 in which the relationship between the user's electroencephalogram information and the scent information is linked, which is held in the storage unit 60. Will be done. The first conversion table 61 includes a first sub-table 62 in which the sensitivity value and the rank are associated with each other, and a second sub-table 63 in which the relationship between the rank and the scent is associated with each other. The first sub-table 62 is a table used by the estimation unit 70 to rank each of the user's sensitivity values analyzed by the analysis unit 50, that is, to determine the degree of height of the sensitivity value. As shown in FIG. 5A, in the first sub-table 62, the sensitivity information such as the degree of interest, the degree of liking, the degree of concentration, the degree of stress, the degree of calm, etc. is 50 in the second row. Numerical values (%) of 40, 30 ... Are ranked in the third column as A, B, C, ....

In the second conversion step S143, the user's sensibility information is converted into scent information by the second sub-table 63 in which the relationship between the sensibility value held in the storage unit 60 and the rank is linked. The second sub-table 63 is a table used by the estimation unit 70 to estimate the optimum scent type from the sensibilities ranked by the first sub-table 62. As shown in FIG. 5B, the second sub-table 63 is associated with the sensitivity values and ranks of a plurality of users and the scents (herb-based, citrus-based, ...). As a result, the rank is first specified by inputting the sensitivity value, and the scent corresponding to the specified rank is specified. Specifically, as an example, when the average value of the degree of interest of a plurality of users is 50 or more and the average value of the degree of liking is also 50 or more, as can be understood from FIG. 5A, each of them. Since the sensibilities of both are A, it can be estimated from FIG. 5 (b) that the provided scent is citrus.

Next, in the estimation step S150, the estimation unit 70 estimates the scent suitable for the user based on the result of the analysis in the analysis step S130. This estimation is the estimation method that does not use the learning model shown in 2) above. When the estimation method shown in 1) above is used, steps S130 to S143 can be omitted, and when the estimation method shown in 3) above is used, steps S132 to S143 above can be omitted. be able to.

In the estimation step S150, with respect to the scent information converted in the first conversion step S142 and the second conversion step S143 based on the numerical value quantified in the quantification step S131, the numerical value is increased for the element for which a higher numerical value is desired. Estimates the expected scent, and estimates the scent that is expected to decrease for factors for which a lower value is desired. For example, when the value of interest is low, a specific citrus scent that is expected to raise the mood and make a positive feeling is estimated. In addition, for example, when the stress level is high, the scents of specific herbs (lavender essential oil, etc.) and forests (Picea glehnii, etc.) that are expected to reduce stress, soothe, and relax are estimated.

Next, in the output step S160, the output unit 80 outputs the fragrance information indicating the scent information estimated in the estimation step S150. It should be noted that the fragrance that emits the estimated scent may be additionally blended or blended.

According to such a configuration of the present embodiment, it is possible to provide customized optimum scent information to all the plurality of users based on the mental states of all the plurality of users existing in the predetermined space 4.

<Diffuser>
FIG. 9 is a flowchart showing the operation of the diffuser 2 in the scent information providing method of the present invention. The diffuser 2 retains a plurality of scents, and supplements the elements that are considered to be insufficient based on the sensory information (mental state) of the analysis result obtained by analyzing the electroencephalograph information acquired by the electroencephalograph 3. The scent is emitted based on the scent information estimated by the scent information providing device 1 as the scent for the purpose.

In the control step S21, the control unit 111 controls the drive unit 112 based on the scent information acquired by the communication unit 110 from the network 30.

Next, in the drive step S22, the drive unit 112 controlled by the control unit 111 is driven.

Next, in the ejection step S23, the ejection unit 113 ejects a scent that matches the information based on the scent information by driving the drive unit 112.

[Aroma information provision program]

In the scent information providing program according to the present invention, the function of the scent information providing device 1 of the present invention is realized in the computer.

[Second Embodiment]
FIG. 14 is a block diagram showing a configuration example of the diffuser 200 according to the second embodiment of the present invention. In the second embodiment, the function of the scent information providing device 1 in the first embodiment is communicated with the sensor 120 by IoT, and the diffuser 200 is responsible for the function, so that the diffuser 200 itself provides the scent information. Is. The diffuser 200 is a device that provides a scent and information about the scent. The diffuser 200 is a device that identifies the optimum scent for a plurality of users, determines the scent to be provided, and emits the scent based on the brain waves of the plurality of users.

Next, each component of the diffuser 200 will be described. The diffuser 200 includes a reception unit 210, a communication unit 220, a drive unit 230, a control unit 240, an acquisition unit 250, a storage unit 260, an analysis unit 270, an estimation unit 280, an ejection unit 290, and an output unit 300. Each component will be described in detail below.

The reception unit 210 is a communication interface that receives user information via the network 30.

The reception unit 210 receives input of sensitivity information indicating the user's sensitivity from the user. The user's sensibility is information indicating the user's mental state. As an example, as shown in FIG. 11, the information indicating the mental state of the user includes INTEREST (interest level), LIKE (favorite level), CONCENTRATION (concentration rate), STRESS (stress level), CALMNESS (calmness level), and the like. Is. The user inputs, for example, an interest degree of 50%, a favorite degree of 40%, and the like.

FIG. 11 is a UI (User Interface) screen for inputting information related to the user's sensitivity information. In this embodiment, as shown in FIG. 11, the choice of the scent system is displayed on the user's terminal and selected, but this is an example. For example, the user can imagine a favorite scent. The text may be directly input. In this case, a table in which the relationship between the text that may be input and the scent is associated is stored in advance, and the input text and the text that may be input that are stored in advance are stored in advance. The degree of correlation with the content may be specified, and the scent associated with the specified text having a high degree of correlation may be estimated as a favorite scent.

The communication unit 220 includes a sensor, detects the user's brain wave information, communicates with the network 30, and transmits the brain wave information to the network 30.

The drive unit 230 drives the diffuser 200 when the communication unit 220 starts communicating with the network 30.

The control unit 240 includes the acquisition unit 250, the analysis unit 270, and the estimation unit 280, controls the information received by the reception unit 210, and stores the first sub-table 262 and the second sub-table 262 and the second by the storage unit 260. User information is converted into scent information using the sub-table 263.

The acquisition unit 250 acquires the user's electroencephalogram information from the electroencephalograph 3. The electroencephalogram information is a signal directly transmitted from the electroencephalograph 3 and is a signal obtained by measuring an electric signal generated from the user's brain with electrodes. FIG. 12 is a diagram in which these electric signals measured by the electroencephalograph 3 are converted into sensibilities and converted into a graph showing a temporal transition. However, the present invention is not limited to a graph, and may be a table or the like showing numerical values of sensibilities at each time. In the graph shown in FIG. 12, it can be seen that the LIKE (favorite degree) is 0 to 45 seconds and high in 60 to 65 seconds. On the contrary, it can be seen that the INTEREST (interest rate) has the lowest value per 55 seconds or between 60 and 65 seconds. Here, the acquisition unit 250 acquires brain wave information from the network 30, that is, online text information, but the acquisition unit 250 can acquire the brain wave information even when the connection is disconnected from the network 30 for some reason. The acquisition unit 250 may acquire offline information from the computer. As an example, the acquisition unit 250 may acquire text information created by the user stored in a recording medium such as a flash memory held by the user by connecting the flash memory. The offline information may be, for example, transcription data of voice data. The server acquires these offline information from files stored in the shared folder of the server or the like.

Specifically, the acquisition unit 250 accesses the network 30 and collects the user's electroencephalogram information from the electroencephalograph 3. The acquisition unit 250 collects brain wave information of, for example, all employees (for example, 10 people, but not limited to 10 people, and may be more or less) existing in a predetermined department of the company. collect.

The analysis unit 270 analyzes the electroencephalogram information acquired by the acquisition unit 250. The analysis unit 270 performs electroencephalogram analysis on the electroencephalogram information and generates sensitivity information that shows the contents of the electroencephalogram information in a graph. The sensibility information referred to here may be numerical information obtained as a result of electroencephalogram analysis, and may be any numerical value useful for grasping the mental state of the user. For example, it may be a sensitivity such as an interest level in the electroencephalogram information and a numerical value from 0 to 100.

The user's sensibility may include a plurality of sensibilities. The plurality of sensibilities are, for example, INTEREST (interest level), LIKE (favorite level), CONCENTRATION (concentration level), STRESS (stress level), CALMNESS (calmness level), and the like. These characteristics are quantified, for example, the degree of interest is 80, the degree of liking is 30, the degree of concentration is 40, the degree of stress is 50, and the degree of calm is 60, with the maximum value being 100 and the minimum value being 0. The analysis unit 270 generates quantified information for each of the plurality of sensibilities of the user as sensibility information indicating the user's sensibilities. Specifically, the analysis unit 270 analyzes the mental state of the user from the brain wave information of the user acquired by the acquisition unit 250.

Further, the analysis unit 270 may quantify all of the plurality of sensibilities of the user, or may quantify only the sensibilities selected from the plurality of sensibilities. Then, the one with the higher numerical value is selected. With the maximum value set to 100 and the minimum value set to 0, for example, the analysis unit 270 analyzed the brain wave information and obtained it, for example, the degree of interest 80, the degree of liking 30, the degree of concentration 40, the degree of stress 50, and the degree of calmness 30. Each of the sensitivity values of the sensitivity information is compared with the threshold value. A scent that matches the sensibilities is created by using the sensibilities that exceed the threshold value as the characteristics of the user. When the threshold value is, for example, 40, a scent suitable for a user having a high degree of interest and a high degree of stress is created. For example, the degree of interest, the degree of liking, the degree of concentration, the degree of calm, etc. are examples in which higher numerical values are preferred.

Further, the analysis unit 270 quantifies all of the plurality of sensibilities of the user, and selects the one having the lower numerical value. The maximum value is 100, the minimum value is 0, and the analysis unit 270 analyzes the brain wave information to obtain, for example, interest degree 30, favorite degree 40, concentration degree 50, stress degree 60, and calmness degree 20. Each of the sensitivity values of the sensitivity information is compared with the threshold value. A scent that matches the sensibilities is created with the sensibilities below the threshold as the characteristics of the user. When the threshold value is, for example, 40, a scent suitable for a user with a low degree of interest is created. For example, the degree of stress is an example in which a lower numerical value is preferred.

The electroencephalogram information of all the users existing in the predetermined space 4 may be analyzed, or the electroencephalogram information of a predetermined number of people (for example, 10 people) in the predetermined space 4 may be analyzed. The storage unit 260, which will be described later, holds a conversion model (learning model) for converting the numerical value obtained as a result of brain wave analysis into the user's sensitivity information. Quantify the degree of liking, concentration, stress, calmness, etc.

In addition, as a method of estimating a scent suitable for a user using an electroencephalogram, 1) a method of estimating a scent using a learning model directly from an electroencephalogram, and 2) a sensibility is specified from the electroencephalogram, and the specified sensibility is applied. There are a method of estimating the scent according to the difference from the threshold value and a method of 3) identifying the sensibility from the brain wave and estimating the scent from the specified sensibility using a learning model.

As 1), the storage unit 260 stores a scent learning model that learns the relationship between the brain wave information, the content of the brain wave information, and the scent, and directly learns the brain wave signal without analyzing the brain wave by the analysis unit 270. It is input to the estimation process of the model to generate the sensitivity information which is the information obtained by quantifying the user's sensitivity, and the estimation unit 280 estimates the fragrance suitable for the user by using the fragrance learning model by inputting the sensitivity information. The details of the estimation unit 280 will be described later.

As 2), the storage unit 260 stores the user's sensibility, the analysis unit 270 performs the sensibility analysis on the brain wave information, generates the sensibility information which is the information obtained by quantifying the user's sensibility, and the estimation unit 280. Inputs sensitivity information and estimates a scent suitable for the user according to the difference from the threshold value.

As 3), the storage unit 260 stores the user's sensibility, the analysis unit 270 performs the sensibility analysis on the brain wave information, generates the sensibility information which is the information obtained by quantifying the user's sensibility, and the estimation unit 280. Uses the sensibility information as input and estimates the scent suitable for the user using the scent learning model. As the scent learning model, the storage unit 260 stores the sensibility information and the scent learning model that learns the relationship between the content of the sensibility information and the scent.

Learning models can be created by supervised learning with feedback, unsupervised learning without feedback, or reinforcement learning in which feedback is given when a reinforcement signal is received (for example, when the target state is reached). Created by Alternatively, it is created by deep learning using a multi-layer neural network with four or more layers that processes an input vector and converts it into an output vector using a model inspired by the connections between neurons in the brain. May be done. In this embodiment, as an example, a learning model in which the relationship between the electroencephalogram information and the user's sensibility is learned by learning the combination of the electroencephalogram state and the sensibility and its numerical value associated with the brain wave state is used. The user's sensibility is calculated by creating and performing estimation processing by this learning model. The learning model includes a scent learning model that converts the user's sensibility into a scent and a sensibility learning model that converts the brain wave analysis result into the user's sensibility. The scent learning model and the sensitivity learning model may be realized by one learning model.

The storage unit 260 holds a first conversion table 261 that links the relationship between the user's sensitivity information and the scent. The first conversion table 261 includes a first sub-table 262 in which the sensitivity value and the rank are associated with each other, and a second sub-table 263 in which the relationship between the rank and the scent is associated with each other. The first sub-table 262 is a table used by the estimation unit 280 to rank each of the user's sensitivity values analyzed by the analysis unit 270. As shown in FIG. 5A, in the first sub-table 262, the sensitivity information such as the degree of interest, the degree of liking, the degree of concentration, the degree of stress, the degree of calm, etc. is 50 in the second column. The numerical values (%) of 40, 30 ... are lined up in the third row with the scents of herbs, citrus, and so on. In the second sub-table 263, the second sub-table 263 is a table used by the estimation unit 280 to estimate the optimum scent type from the sensibilities ranked by the first sub-table 262. As shown in FIG. 5B, the second sub-table 263 is associated with the sensitivity values and ranks of a plurality of users and the scents (herb-based, citrus-based, ...). As a result, the rank is first specified by inputting the sensitivity value, and the scent corresponding to the specified rank is specified. Specifically, as an example, when the average value of the degree of interest of a plurality of users is 50 or more and the average value of the degree of liking is also 50 or more, as can be understood from FIG. 5A, each of them. Since the sensibilities of both are A, it can be estimated from FIG. 5 (b) that the provided scent is citrus.

The estimation unit 280 estimates one scent suitable for a plurality of users existing in a predetermined space 4 based on the result of the analysis analyzed by the analysis unit 270 using the first conversion table 261. For example, if the entire user is under stress, herbs (such as peppermint) and woody materials that are said to have a stress-relieving effect are estimated. Further, for example, if the entire user is in a state where the concentration level is low, a citrus material (orange, etc.), which is said to have an effect of increasing the concentration level, is estimated. This material estimation is performed by inputting the numerical value obtained as a result of brain wave analysis into the storage unit 260, which holds a conversion model (learning model) for converting into the user's sensitivity information. Be struck. This estimation is just an example and is not limited to this.

Further, the estimation unit 280 uses the scent learning model as an input of the sensibility information to estimate one scent suitable for all the plurality of users existing in the predetermined space 4. The estimation unit 280 further adds the user's sensitivity information received by the reception unit 210 and the scent information specified by using the first conversion table 261 to all the plurality of users existing in the predetermined space 4. Estimate one scent suitable for. The estimation unit 280 is suitable for all a plurality of users existing in the predetermined space 4 by further adding the sensitivity information received by the reception unit 210 and the scent information specified by using the second conversion table 262. Estimate only one scent.

For example, when "stress level" is selected from multiple sensibility information, a specific herbal material (lavender, etc.) that is expected to relax and improve work efficiency when the stress level is high. Is estimated by the estimation unit 280. Further, for example, when "concentration ratio" is selected, the estimation unit 280 estimates a specific citrus material that makes one feel positive and is expected to improve concentration.

The sensibility values of a plurality of users are obtained by taking the average value of the sensibility values of the plurality of users. When the sensibilities are batting, that is, when a plurality of sensibilities decrease and it becomes necessary to raise them, a plurality of fragrances that raise those sensibilities are blended to generate a scent. Further, the sensibilities of a plurality of users may be simply added up to use the sensibilities having a high total value.

The ejection unit 290 ejects a scent based on the scent information estimated by the estimation unit 280. When the estimation unit 280 uses the analysis information as input and estimates one scent suitable for all a plurality of users existing in the predetermined space 4 by using the scent learning model, the scent is estimated based on the estimated scent system. Squirt.

The output unit 300 outputs the scent information estimated by the estimation unit 280. When the estimation unit 280 uses the analysis information as input and estimates one scent suitable for all a plurality of users existing in the predetermined space 4 by using the scent learning model, the scent is estimated based on the estimated scent system. Output information. The output unit 300 may output fragrance information to the information processing device communicably connected to the diffuser 200 in the form of text information or image information by communication, or may output the fragrance information to the diffuser 200 via a monitor or the like. A port that can be connected to an output device may be provided to output to the monitor in the form of display information, or a small monitor, touch panel, or the like may be provided on the diffuser 200 main body to display on the small monitor. It may be output with. The output unit 300 may further output sensitivity information, which is information that quantifies the user's sensitivity. What is output from the output unit 300 may be audio or an image. The output information is graphed and visualized as shown in FIG.

The scent estimation based on the sensitivity information input in FIG. 11 and the scent estimation based on the electroencephalogram analysis may be combined, for example, at a ratio of 3: 7, a ratio of 5: 5, or a ratio of 7: 3. May be combined with. This is an example, the ratio is arbitrary, and one may be 0. The scent information can be obtained by estimating the scent suitable for all the plurality of users from the electroencephalographs 3 worn by all the plurality of users existing in the predetermined space 4 by the electroencephalogram analysis.

According to such a configuration of the present embodiment, it is possible to provide customized optimum scent information to all the plurality of users based on the mental states of all the plurality of users existing in the predetermined space 4.

Next, a structure showing that the diffuser 200 can inject a plurality of types of scents will be described with reference to FIG.

The diffuser 200 includes a container 201 and a ejection portion 202. A plurality of tanks 203 are housed in the container 201, and different types of fragrances are sealed in each of them. The ejection unit 202 is provided with an open / close switch (not shown), and the opening / closing is switched by the control unit (not shown), and the scent may or may not be ejected. A plurality of pipes 204 are inserted from the ejection portion 202 to the inside of the container, and one lower portion of each pipe 204 is inserted into each tank 203, sucks up the fragrance, mixes the fragrance, and ejects from the ejection portion 202. It is designed to do.

Next, the operation of the diffuser 200 will be described with reference to FIG.

In the reception step S210, the reception unit 210 receives the user's sensitivity information from the user via the network 30.

Next, in the control step S220, the control unit 240 controls the drive unit 230.

Next, in the drive step S230, the drive unit 230 drives the diffuser 200.

Next, in the acquisition step S240, the acquisition unit 250 acquires the user's brain wave information. The electroencephalograph information is a signal directly transmitted from the electroencephalograph, and is a signal obtained by measuring an electric signal generated from the user's brain with electrodes. Here, the acquisition unit 250 acquires brain wave information from the network, that is, online text information, but acquires the brain wave information so that the brain wave information can be acquired even when the connection is disconnected from the network 30 for some reason. The unit 250 may acquire offline information from the computer. As an example, the acquisition unit 250 connects the flash memory to the information acquired in advance in the online state and the text information created by the user stored in the recording medium such as the flash memory held by the user. It may be the one to be acquired. The server acquires these offline information from files stored in the shared folder of the server or the like.

Next, in the analysis step S250, the analysis unit 270 analyzes the electroencephalogram information transmitted from the acquisition unit 250 in the acquisition step S240.

The analysis step S250 includes a digitization step S251 and a visualization step S252.

In the quantification step S251, the analysis unit 270 quantifies the level of the user's sensitivity element. Specifically, from the analysis of brain wave information, the levels such as the degree of interest, the degree of liking, the degree of concentration, the degree of stress, and the degree of calm are quantified as elements of the user's sensitivity.

Next, in the visualization step S252, the level of the user's sensitivity element quantified in the quantification step S251 is visualized in a graph. Specifically, the analysis unit 270 may use an existing electroencephalogram analysis algorithm to convert the electroencephalogram information acquired by the acquisition unit 250 into sensibility information obtained by quantifying various sensibilities.

FIG. 12 is a diagram in which these electric signals measured by an electroencephalograph are converted into sensibilities and converted into a graph showing a temporal transition. In the graph shown in FIG. 12, it can be seen that the LIKE (favorite degree) is 0 to 45 seconds and high in 60 to 65 seconds. On the contrary, it can be seen that the INTEREST (interest rate) has the lowest value per 55 seconds or between 60 and 65 seconds.

The storage step S260 includes a selection step S261, a first conversion step S262, and a second conversion step S263. In the selection step S261, the optimum scent is selected based on the result analyzed in the analysis step S250. The selection step S261 may be omitted.

In the first conversion step S262, the user's sensitivity information is converted into the scent information by the first conversion table 261 that is stored in the storage unit 260 and in which the relationship between the user's sensitivity information and the scent information is linked. Will be done. The first conversion table 261 includes a first sub-table 262 in which the sensitivity value and the rank are associated with each other, and a second sub-table 263 in which the relationship between the rank and the scent is associated with each other. The first sub-table 262 is a table used by the estimation unit 280 to rank each of the user's sensitivity values analyzed by the analysis unit 270, that is, to determine the degree of height of the sensitivity value. As shown in FIG. 5A, in the first sub-table 262, the sensitivity information such as the degree of interest, the degree of liking, the degree of concentration, the degree of stress, the degree of calm, etc. is 50 in the second row. Numerical values (%) of 40, 30 ... Are ranked in the third column as A, B, C, ....

In the second conversion step S263, the user's sensibility information is converted into scent information by the second sub-table 263, which is stored in the storage unit 260 and in which the relationship between the user's sensibility information and the scent is linked. .. The second sub-table 263 is a table used by the estimation unit 280 to estimate the optimum scent type from the sensibilities ranked by the first sub-table 262. As shown in FIG. 5B, the second sub-table 263 is associated with the sensitivity values and ranks of a plurality of users and the scents (herb-based, citrus-based, ...). As a result, the rank is first specified by inputting the sensitivity value, and the scent corresponding to the specified rank is specified. Specifically, as an example, when the average value of the degree of interest of a plurality of users is 50 or more and the average value of the degree of liking is also 50 or more, as can be understood from FIG. 5A, each of them. Since the sensibilities of both are A, it can be estimated from FIG. 5 (b) that the provided scent is citrus.

Next, in the estimation step S270, the estimation unit 280 estimates the scent suitable for the user based on the result of the analysis in the analysis step S250.

Next, in the blending step S280 and the blending step S290, based on the numerical value quantified in the quantification step S251, a fragrance material that is expected to increase the numerical value is blended for the element for which a higher numerical value is desired, and the value is lower. For the elements for which the numerical value is desired, the fragrance material that is expected to decrease the numerical value is blended to prepare the fragrance. For example, when the value of interest is low, a specific citrus flavoring material that is expected to raise the mood and make a positive feeling is mixed. In addition, for example, when the stress level is high, a specific herbal (lavender essential oil, etc.) and forest-based (Picea glehnii, etc.) fragrance materials that are expected to reduce stress, soothe, and relax are mixed to mix the fragrance. do. The ratio of the fragrance material required in step S280 is adjusted according to each analysis value.

Next, in the output step S300, the output unit 300 outputs the scent information estimated by the estimation unit 280.

Next, in the ejection step S310, the ejection unit 290 outputs the scent based on the scent information output by the output unit 300.

According to such a configuration of the present embodiment, it is possible to provide customized optimum scent information to all the plurality of users based on the mental states of all the plurality of users existing in the predetermined space 4.

<Other Embodiments>
In the above embodiment, the space production service using the blended fragrance based on the correlation with the brain wave and the sensitivity evaluation according to the purpose and the scene has been described based on the user's request, but the present invention is not limited to this. Based on customer feedback, it may be used for space production services using compounded fragrances customized (improved) according to the purpose and scene, and people and the environment while evaluating brain waves and sensibilities in real time. It may be used for a service of controlling and injecting a fragrance suitable for.

The elements of the user's sensibility include, but are not limited to, the degree of interest, the degree of liking, the degree of concentration, the degree of stress, and the degree of calmness. Happiness), excitement, positiveness, fun, satisfaction, activity, excitement, joy, surprise, happiness, relaxation, refreshment, curiosity, ideals, stability, liberalism, excitement, change tolerance, self-transcendence , Self-promotion, status quo, pleasure, erotic, beautiful, cute, pretty, elegant, cool, humor, unique, cool, dandy, fashionable, urban, intimate, modern, classic, natural, transparent, indoor, Outdoors, leader, efficiency-oriented, steady, steadily, good listener, healing, aggressive, creativity, warmth, refreshing, individuality, self-expression, personality, gorgeous, gentle, refreshing, hope, diversity, expectation, anxiety , Growth motivation, approval desire, self-confidence, attractiveness, awakening, tranquility, refreshing feeling, stimulating, bright, dark, light, heavy, warm, cold, hard, soft, passionate, gentle, charming, dignity, fresh, Rui Personality, dryness, wildness, cleanliness, sexyness, seriousness, and sociability are also factors in the user's sensibility.

In the above embodiment, the user's brain wave information is assumed to be a waveform, but this may be voice information or image information. In the case of voice information, the content of the voice information may be converted into text information by voice recognition and converted into fragrance information by using the method shown in the above embodiment, or voice may be input into fragrance information. A learning model to be converted may be created, and the voice may be converted into fragrance information by estimation by the learning model. Also, in the case of an image, if a character string is included in the image, the character string is analyzed by image recognition, converted into text information, and converted into fragrance information by the method shown in the above embodiment. You may do it. Alternatively, a learning model that learns the relationship between the image and the scent may be prepared, and the image may be input and configured to be converted into scent information.

Further, in the above embodiment, the sensitivity value is specified from the user's brain wave information, but the sensitivity value may be specified from other biological information such as pulse, heart rate, respiratory rate, and blood pressure. When the sensitivity value is specified from the pulse or heartbeat, a sphygmomanometer may be used instead of the electroencephalograph. In that case, a learning model that estimates the user's sensibility from the acquired pulsation intensity and the pulsation interval may be created, and the pulse and heartbeat information may be converted into fragrance information by the estimation by the learning model.

In the scent diffuser, the fragrance is stored as a liquid and is sprayed as a mist from the ejection part, but it can also be applied as an aroma oil and can also be used as a perfume. Further, as a material for providing the scent according to the present invention, the liquid is an example, and it may be stored as a solid or a gas. When provided as an individual, it may be used as a scent for soap as an example.

It can also be used not only in the workplace but also in public places such as concert halls, museums, movie theaters, and stadiums. In these cases, it is possible to create a scent suitable for the sensibilities of the entire audience. It can also be used in medical facilities such as hospitals, elderly homes, and long-term care facilities. Target users in medical facilities are, for example, patients, visitors (visitors), doctors, nurses, etc., and target users in elderly homes and long-term care facilities are, for example, residents and caregivers who are cared for. , Visitors, etc.

Further, various modifications described above may be selected and combined as appropriate, or other modifications may be applied.

1 Fragrance information providing device 2 Diffuser 3 Brain wave meter 4 Predetermined space 10 Reception unit 20 Control unit 30 Network 40 Acquisition unit 50 Analysis unit 60 Storage unit 70 Estimating unit 80 Output unit 61 1st conversion table 62 1st sub-table 63 2nd Sub-table 110 Communication unit 111 Control unit 112 Drive unit 113 Ejection unit 120 Sensor 121 Communication unit 122 Storage unit 123 Control unit 200 Diffuser 201 Container 202 Ejection unit 203 Tank 204 Tube 210 Reception unit 220 Communication unit 230 Drive unit 240 Control unit 250 Acquisition Unit 260 Storage unit 261 First conversion table 262 First sub-table 263 Second sub-table 270 Analysis unit 280 Estimating unit 290 Ejecting unit 300 Output unit

Claims (18)

The acquisition unit that acquires the user's brain wave information,
An analysis unit that analyzes the brain wave information acquired by the acquisition unit,
An estimation unit that estimates a scent suitable for the user based on the result of analysis by the analysis unit, and an estimation unit.
A scent information providing device including an output unit that outputs fragrance information indicating scent information estimated by the estimation unit. The acquisition unit acquires brain wave information of a plurality of users and obtains brain wave information.
The scent information providing device according to claim 1, wherein the output unit outputs fragrance information indicating one scent information suitable for all users from the brain wave information of the plurality of users. The scent information providing device is
It is equipped with a storage unit that stores brain wave information and a scent learning model that learns the relationship between the contents of the brain wave information and scent.
The analysis unit performs sensitivity analysis on the brain wave information to generate sensitivity information which is information obtained by quantifying the user's sensitivity.
The scent information providing device according to claim 1 or 2, wherein the estimation unit estimates a scent suitable for the user by using the scent learning model with the sensitivity information as an input. The scent information providing device is
Equipped with a storage unit that stores sensitivity information, which is information that quantifies the user's sensitivity.
The scent information providing device according to claim 1 or 2, wherein the estimation unit estimates a scent suitable for the user according to a difference from a threshold value by inputting the sensitivity information. The scent information providing device is
Equipped with a storage unit that stores sensitivity information, which is information that quantifies the user's sensitivity.
The scent information providing device according to claim 1 or 2, wherein the estimation unit estimates a scent suitable for the user by using the scent learning model with the sensitivity information as an input. The user's sensibilities include a plurality of sensibilities.
The analysis unit generates quantified information for each of the plurality of sensibilities of the user as sensibility information indicating the sensibilities of the user.
3. Fragrance information providing device. The scent information providing device according to claim 6, wherein the analysis unit generates sensibility information indicating sensibilities exceeding a predetermined threshold value for each numerical value of the plurality of sensibilities of the user. The scent information providing device according to claim 6, wherein the analysis unit generates sensibility information indicating a predetermined number of user sensibilities from the highest numerical value for each of the plurality of sensibilities of the user. The user's sensibilities include a plurality of sensibilities.
The analysis unit generates quantified information for each of the plurality of sensibilities of the user as sensibility information indicating the sensibilities of the user.
The scent according to claim 6, wherein the estimation unit estimates a scent suitable for the user by using the scent learning model by inputting a characteristic having a low numerical value among a plurality of sensibilities of the user. Information providing device. The scent information providing device according to claim 9, wherein the analysis unit generates sensibility information indicating sensibilities equal to or less than a predetermined threshold value for each numerical value of the plurality of sensibilities of the user. The scent information providing device according to claim 9, wherein the analysis unit generates sensibility information indicating a predetermined number of user sensibilities from the lowest numerical value for each of the plurality of sensibilities of the user. The scent information providing device according to any one of claims 4 to 11, wherein the output unit further outputs sensitivity information indicating the user's sensitivity. The scent information providing device further
A reception unit that accepts input of sensitivity information about the user from the user,
It is provided with a storage unit that holds a first conversion table that links the relationship between the user's sensitivity value and the scent.
The estimation unit is further characterized by adding the user's sensitivity information received by the reception unit and the scent information specified by using the first conversion table to estimate the scent suitable for the user. The scent information providing device according to any one of claims 1 to 12. The first conversion table is
The first sub-table in which the sensitivity value and the rank are associated with each other
Includes a second sub-table that associates the relationship between rank and scent.
The scent information providing device according to claim 13, wherein the rank is first specified by inputting the sensitivity value, and then the scent corresponding to the specified rank is specified. The scent information providing device according to claim 13, wherein the reception unit provides the user with a user interface that indicates options indicating the sensitivity contents as a plurality of sensitivity values. The acquisition step to acquire the user's brain wave information,
An analysis step for analyzing the user's electroencephalogram information acquired in the acquisition step, and
An estimation step for estimating a scent suitable for the user based on the analysis result in the analysis step, and an estimation step.
A scent information providing method including an output step for outputting fragrance information indicating scent information estimated in the estimation step. The acquisition unit that acquires the user's brain wave information,
An analysis unit that analyzes the brain wave information acquired by the acquisition unit,
An estimation unit that estimates a scent suitable for the user based on the result of analysis by the analysis unit, and an estimation unit.
A scent diffuser including an ejection unit that ejects a fragrance based on the scent information estimated by the estimation unit. On the computer
The acquisition function to acquire the user's brain wave information,
An analysis function that analyzes the brain wave information acquired by the acquisition function, and
An estimation function that estimates a scent suitable for the user based on the result of analysis by the analysis function, and an estimation function.
A scent information providing program that realizes an output function that outputs a fragrance based on the scent information estimated by the estimation function.

Images