JP2021056985A - Method for generating learning model, computer program, spectacle lens selection support method and spectacle lens selection support system - Google Patents

Abstract

To provide a method for generating a learning model capable of expecting support to select a spectacle lens by estimating a specification of a spectacle lens suitable for a user, a computer program, a spectacle lens selection support method and a spectacle lens selection support system.SOLUTION: According to this embodiment, a spectacle lens selection support system includes a storage part for storing a learning model for outputting specification information of a spectacle lens in the case of inputting user's attribute information, measurement information related to user's eyes and spectacle use information by the user, a user information acquisition part for acquiring the attribute information, the measurement information and the use information, and a specification information acquisition part for inputting the attribute information, the measurement information and the use information acquired by the user information acquisition part to the learning model and acquiring the specification information outputted by the learning model.SELECTED DRAWING: Figure 1

Description

The present invention relates to a learning model generation method, a computer program, a spectacle lens selection support method, and a spectacle lens selection support system that assist a user in selecting a spectacle lens.

In addition to lenses that simply correct the refractive power of the eye, such as myopia, hyperopia, and astigmatism, multifocal lenses are widely used as lenses for spectacles to compensate for the decline in accommodation power due to aging. Has been done. Multifocal lenses have a refractive power that is insufficient for near vision added to the lower part of the lens as an additional refractive power, and progressive power lenses that continuously change the refractive power so that there is no visible boundary have become widespread. ing.

Furthermore, the progressive power lens has a perspective design that is suitable for walking and driving by ensuring a wide field of view at the upper part for far vision, and an area for intermediate distance and near vision according to desk work and work at hand. Various types of lenses are provided so that they can be selected according to the purpose of use, such as a medium-near design that secures a wide range and is suitable for indoor use. While there are so many options available, few users have an accurate understanding of their characteristics (merits and demerits). When making eyeglasses, it is common that the staff of the store presents lens candidates from the information heard from the user, and the user decides and purchases the lens from the presented candidates.

In Patent Document 1, evaluation data that defines the individual suitability of a plurality of types of progressive multifocal lenses with respect to the age and wearing environment of the presbyopia is prepared in advance, and the age and wearing environment of the presbyopia are determined. The method of selecting a progressive multifocal lens is to obtain information, obtain the individual suitability of the progressive multifocal lens for this information from the evaluation data, and select the optimal progressive multifocal lens for the individual with presbyopia from the obtained suitability. Proposed.

In Patent Document 2, design parameters are stored in a database for each of a plurality of types according to the degree of addition of the lens, and the maximum far point distance and the maximum near point distance when the lens selected from the database is worn are calculated. A progressive power lens selection device has been proposed in which the maximum far point distance and the maximum near point distance are displayed side by side on a display device with respect to the selected lens.

Japanese Unexamined Patent Publication No. 3-206417 Japanese Unexamined Patent Publication No. 2013-52095

When selecting a lens, candidates are narrowed down according to the rules established based on the experience of the store staff, which lens is suitable according to the user's needs, age, refraction error, and the like. Due to its optical principle, there is no universal lens for any usage environment (distance, purpose, etc.) for the progressive power lens as a measure against presbyopia with reduced accommodation power, and there is a trade-off between advantages and disadvantages. It is established by turning off, and if you make a mistake in selection, it may cause discomfort / discomfort during use and dissatisfaction with the appearance. Knowledge and experience for lens selection are inherited by staff education, but it is not easy to formalize everything properly, and there are cases where some of them end up as personal skills.

The method for selecting a progressive multifocal lens described in Patent Document 1 is a rule-based selection method for selecting a lens according to a predetermined rule. When there are multiple parameters as factors for selecting a lens, there is a problem that it is difficult to determine the optimum rule in advance because the degree of consideration of the evaluation results of the multiple parameters greatly affects the lens selection result. .. The progressive power lens selection device described in Patent Document 2 assists the user in selecting a lens by displaying a plurality of sets of a maximum far point distance and a maximum near point distance side by side. There is a risk that users who do not have it will be guided to the wrong selection.

The present invention has been made in view of such circumstances, and an object of the present invention can be expected to support selection of spectacle lenses by inferring specifications of spectacle lenses suitable for the user. The purpose of the present invention is to provide a learning model generation method, a computer program, a spectacle lens selection support method, and a spectacle lens selection support system.

The learning model generation method according to the present invention acquires teacher data in which the user's attribute information, the measurement information related to the user's eye, the usage information of the spectacles by the user, and the specification information of the spectacle lens are associated with each other. Using the acquired teacher data, a learning model that outputs the specification information when the attribute information, the measurement information, and the usage information are input is generated.

In addition, the computer program according to the present invention acquires the user's attribute information, the measurement information related to the user's eye, and the usage information of the glasses by the user to the computer, and obtains the attribute information, the measurement information, and the usage information. A process of inputting the acquired attribute information, the measurement information, and the application information to the learning model that outputs the specification information of the lens of the eyeglass when input is executed, and acquiring the specification information output by the learning model is executed. Let me.

Further, the spectacle lens selection support method according to the present invention acquires user attribute information, measurement information related to the user's eye, and usage information of the spectacles by the user, and obtains the attribute information, the measurement information, and the usage information. The acquired attribute information, the measurement information, and the application information are input to the learning model that outputs the specification information of the lens of the spectacle when input is input, and the specification information output by the learning model is acquired.

Further, the spectacle lens selection support system according to the present invention is a learning model that outputs specification information of the spectacle lens when the attribute information of the user, the measurement information related to the user's eye, and the usage information of the spectacle by the user are input. A storage unit that stores the attribute information, a user information acquisition unit that acquires the attribute information, the measurement information, and the usage information, and the learning model that stores the attribute information, the measurement information, and the usage information acquired by the user information acquisition unit. It is provided with a specification information acquisition unit that inputs to and acquires the specification information output by the learning model.

In the case of the present invention, it can be expected that the specifications of the spectacle lens suitable for the user can be inferred.

It is a schematic diagram for demonstrating the outline of the spectacle lens selection support system which concerns on this Embodiment. It is a block diagram which shows the structure of the server apparatus which concerns on this embodiment. It is a schematic diagram which shows one configuration example of a user information DB. It is a schematic diagram for demonstrating the configuration example of the specification prediction model. It is a schematic diagram which shows the structural example of the classification model of the specification prediction model. It is a schematic diagram which shows the structural example of the classification model of the specification prediction model. It is a schematic diagram for demonstrating another configuration example of a specification prediction model. It is a block diagram which shows the structure of the store terminal apparatus which concerns on this embodiment. It is a schematic diagram for demonstrating the structure of the inspection apparatus which concerns on this embodiment. It is a block diagram which shows the structure of the tablet type terminal apparatus which concerns on this embodiment. It is a schematic diagram which shows an example of the input reception screen of the attribute information and use information displayed by a tablet type terminal device. It is a flowchart which shows the procedure of the process performed by the tablet type terminal apparatus which concerns on this embodiment. It is a schematic diagram which shows an example of the measurement information display screen which an inspection apparatus displays. It is a flowchart which shows the procedure of the process performed by the store terminal device which concerns on this Embodiment. It is a flowchart which shows the procedure of the specification guessing process performed by the server apparatus which concerns on this Embodiment. It is a schematic diagram which shows an example of the specification information display screen which a store terminal device displays. It is a flowchart which shows the procedure of the generation processing of the specification prediction model performed by the server apparatus which concerns on this Embodiment. It is a flowchart which shows the procedure of the relearning process of the specification prediction model performed by the server apparatus which concerns on this embodiment.

A specific example of the spectacle lens selection support system according to the embodiment of the present invention will be described below with reference to the drawings. It should be noted that the present invention is not limited to these examples, and is indicated by the scope of claims, and is intended to include all modifications within the meaning and scope equivalent to the scope of claims.

<System overview>
FIG. 1 is a schematic diagram for explaining the outline of the spectacle lens selection support system according to the present embodiment. The spectacle lens selection support system according to the present embodiment includes a server device 1, a plurality of store terminal devices 3, a plurality of inspection devices 4, and a plurality of tablet-type terminal devices 5. The store terminal device 3, the inspection device 4, and the tablet-type terminal device 5 are devices provided in a store that sells eyeglasses, and a plurality of devices may be provided in one store. On the other hand, the server device 1 is a device provided in a place different from the store, and the server device 1 and each store terminal device 3 can communicate with each other via a network N such as the Internet. However, the server device 1 may be provided in any store such as a head office.

The store terminal device 3 provided in each store can communicate with the inspection device 4 and the tablet type terminal device 5 in the store. For example, the store terminal device 3 and the inspection device 4 are connected via a communication cable and communicate via the communication cable. Further, for example, the store terminal device 3 and the tablet type terminal device 5 communicate by wireless communication such as a wireless LAN (Local Area Network). The inspection device 4 is a device that performs a refraction and visual acuity test of a user who visits a store to purchase eyeglasses. The inspection device 4 measures, for example, the refraction error (refractive power) of the user's eye, near vision, and the like, and transmits the measurement information related to the user's eye to the store terminal device 3. The tablet-type terminal device 5 performs a process of accepting an input of information from the user. The tablet-type terminal device 5 receives, for example, attribute information such as the age and gender of the user and usage information regarding the purpose of use and usage environment of the glasses that the user intends to purchase from the user and transmits the information to the store terminal device 3. ..

The store terminal device 3 is, for example, a device operated by a store clerk. The store terminal device 3 acquires measurement information related to the user's eyes from the inspection device 4, and also acquires user attribute information and user's eyeglass usage information from the tablet-type terminal device 5. The store terminal device 3 transmits the acquired measurement information, attribute information, and usage information as a set to the server device 1. After that, the store terminal device 3 displays the specification information of the lens suitable for the user estimated by the server device 1 on a display or the like, and proposes a lens of eyeglasses suitable for the user.

The server device 1 is a process of estimating the specifications of the spectacle lens suitable for this user (for example, the type of lens and the addition power) based on the user's attribute information, measurement information, and usage information acquired from the store terminal device 3. I do. The server device 1 transmits the lens specification information, which is the estimation result, to the store terminal device 3. The server device 1 according to the present embodiment includes a specification prediction model, which is a learning model learned by machine learning. This specification prediction model has been learned in advance so as to output the specification information of the spectacle lens in response to the input of the user's attribute information, measurement information, and application information. Using this specification prediction model, the server device 1 inputs the user attribute information, measurement information, and usage information acquired from the store terminal device 3 into the specification prediction model, and acquires the lens specification information output by the specification prediction model. By doing so, we infer the use of eyeglass lenses suitable for the user.

Further, the server device 1 performs a specification prediction model generation (learning) process and a re-learning process. In the first learning process of the specification prediction model, the server device 1 uses, for example, teacher data created based on information on eyeglasses that have been proposed and sold based on the judgment of a store clerk, etc. Performs the learning process of the prediction model. The server device 1 performs learning processing of the specification prediction model using the teacher data in which the user's attribute information, measurement information, and usage information are associated with the specification information of the lens of the spectacles.

Further, the store terminal device 3 presents the specification information of the spectacle lens estimated by the server device 1 to the user, and finally transmits the information regarding the spectacle lens selected and purchased by the user to the server device 1. The information transmitted by the store terminal device 3 at this time may include, for example, user attribute information, measurement information, and usage information, and lens specification information associated therewith. Further, the store terminal device 3 may acquire tracking information (for example, whether or not there is a return or exchange, the user's impression after use, etc.) about the user who purchased the eyeglasses and transmit it to the server device 1. The server device 1 accumulates information from the store terminal device 3 in a database, generates teacher data based on the accumulated information, and retrains the specification prediction model using the generated teacher data.

In the present embodiment, the store terminal device 3, the inspection device 4, and the tablet terminal device 5 are provided as individual devices in each store, but the present invention is not limited to this. For example, the functions of the store terminal device 3, the inspection device 4, and the tablet terminal device 5 may be integrated into one device. For example, the store terminal device 3 may be configured to accept input of user attribute information, usage information, and the like. In this case, the tablet type terminal device 5 may not be provided in the store. Further, for example, the functions of the store terminal device 3 and the inspection device 4 may be integrated into one device, or the functions of the inspection device 4 and the tablet type terminal device 5 may be integrated into one device, for example. Further, the store terminal device 3 and the inspection device 4 may be not a device operated by a store clerk but a device operated by a user. The tablet-type terminal device 5 may be a device operated by a store clerk instead of a device operated by the user.

Further, a plurality of stores may not exist and there may be one store. In this case, the store terminal device 3 may have the function of the server device 1. That is, the store terminal device 3 may perform the estimation processing of the specification information of the lens using the specification prediction model. However, in this case, the store terminal device 3 may perform the learning process of the specification prediction model, and the device different from the store terminal device 3 may perform the learning process. The functions of the server device 1, the store terminal device 3, the inspection device 4, and the tablet terminal device 5 may be integrated into one device.

A user who visits a store to purchase eyeglasses first inputs attribute information such as his / her age and gender and usage information such as the purpose of use and environment of use of the eyeglasses to be purchased on the tablet terminal device 5. To do. The tablet-type terminal device 5 transmits the input attribute information and usage information to the store terminal device 3. The user is then subjected to a refraction and visual acuity test by the testing device 4. The inspection by the inspection device 4 may be carried out under the initiative of a store clerk, or may be carried out by the user operating the inspection device 4 by himself / herself. In the present embodiment, the inspection device 4 at least measures the refraction error and / or near vision of the user's eye, and transmits the measurement result to the store terminal device 3. The store terminal device 3 transmits the user attribute information and usage information acquired from the tablet terminal device 5 and the measurement information acquired from the inspection device 4 as a set to the server device 1.

The server device 1 acquires user attribute information, usage information, and measurement information from the store terminal device 3, and inputs the acquired information into a trained specification prediction model. The specification prediction model outputs specification information such as the type of spectacle lens and the addition power according to the input attribute information, application information, and measurement information. The server device 1 acquires the specification information output by the specification prediction model and transmits it to the store terminal device 3. The store terminal device 3 that has received the specification information from the server device 1 displays information such as the type of spectacle lens and the addition power estimated by the specification prediction model as suitable for the user, and displays the information such as the addition power of the spectacles to the user. We propose lens specifications. Based on the specification information displayed on the store terminal device 3, the user can try on eyeglasses corresponding to this specification and select the lens of the eyeglasses to be purchased.

The server device 1 stores and stores the attribute information, the usage information, and the measurement information acquired from the store terminal device 3 in a database in association with the specification information output by the specification prediction model with respect to this information. Further, the store terminal device 3 acquires the specification information of the lens of the spectacles finally purchased by the user based on the specification information from the server device 1, and transmits the acquired specification information to the server device 1. Further, the store terminal device 3 acquires tracking information such as that the user who purchased the eyeglasses wants to replace the lens at a later date, returns the eyeglasses, and the user feels the use of the eyeglasses, and transmits the tracking information to the server device 1. .. The server device 1 stores this information in a database.

After that, the server device 1 creates teacher data based on the information accumulated in the database every time a predetermined period (for example, one week or one month) elapses, and uses the created teacher data to create a specification prediction model. Perform relearning process. The teacher data may be created automatically by the server device 1, or may be created by the administrator who manages the server device 1.

<Device configuration>
FIG. 2 is a block diagram showing a configuration of the server device 1 according to the present embodiment. The server device 1 according to the present embodiment includes a processing unit 11, a storage unit (storage) 12, a communication unit (transceiver) 13, and the like. In the present embodiment, it is assumed that the processing is performed by one server device 1, but a plurality of server devices 1 may be distributed to perform the processing.

The processing unit 11 is configured by using an arithmetic processing unit such as a CPU (Central Processing Unit), an MPU (Micro-Processing Unit), or a GPU (Graphics Processing Unit). The processing unit 11 reads and executes the server program 12a stored in the storage unit 12 to estimate the specifications of the spectacle lens suitable for the user, and obtains information related to the specifications of the spectacle lens selected by the user. Various processes such as a process of accumulating and a process of learning a specification prediction model based on the accumulated information are performed.

The storage unit 12 is configured by using a large-capacity storage device such as a hard disk. The storage unit 12 stores various programs executed by the processing unit 11 and various data required for processing by the processing unit 11. In the present embodiment, the storage unit 12 stores the server program 12a executed by the processing unit 11 and the specification prediction model 12b for estimating the specifications of the lenses of the spectacles. Further, the storage unit 12 is provided with a user information DB (database) 12c that stores information related to the specifications of the lenses of the spectacles selected by the user. Details of the specification prediction model 12b and the user information DB 12c will be described later.

In the present embodiment, the server program 12a is provided in a mode of being recorded on a recording medium 99 such as a memory card or an optical disk, and the server device 1 reads the server program 12a from the recording medium 99 and stores it in the storage unit 12. However, the server program 12a may be written to the storage unit 12 at the manufacturing stage of the server device 1, for example. Further, for example, in the server program 12a, the server device 1 may acquire what is distributed by another remote server device or the like by communication. For example, in the server program 12a, the writing device may read what has been recorded on the recording medium 99 and write it in the storage unit 12 of the server device 1. The server program 12a may be provided in the form of distribution via the network, or may be provided in the form of being recorded on the recording medium 99.

The communication unit 13 communicates with various devices via a network N including the Internet and a mobile phone communication network. In the present embodiment, the communication unit 13 communicates with a plurality of store terminal devices 3 via the network N. The communication unit 13 transmits the data given by the processing unit 11 to another device, and gives the data received from the other device to the processing unit 11.

The storage unit 12 may be an external storage device connected to the server device 1. Further, the server device 1 may be a multi-computer including a plurality of computers, or may be a virtual machine virtually constructed by software. Further, the server device 1 is not limited to the above configuration, and may include, for example, a reading unit that reads information stored in a portable storage medium, an input unit that accepts operation input, a display unit that displays an image, and the like. ..

Further, in the processing unit 11 of the server device 1 according to the present embodiment, the processing unit 11 reads out and executes the server program 12a stored in the storage unit 12, thereby causing the user information acquisition unit 11a and the lens specification estimation unit 11b. , The specification information transmission unit 11c, the learning processing unit 11d, and the like are realized as software-like functional units. It should be noted that these functional parts are functional parts related to the process of estimating the specifications of the lenses of the spectacles, and the illustration and description of other functional parts will be omitted.

The user information acquisition unit 11a performs a process of acquiring information such as user attribute information, usage information, and measurement information from the store terminal device 3 by communicating with the store terminal device 3 in the communication unit 13. .. As described above, in the spectacle lens selection support system according to the present embodiment, the attribute information and application information input by the user in the tablet terminal device 5 and the measurement information about the user's inspection result performed by the inspection device 4 Is acquired by the store terminal device 3. The user information acquisition unit 11a acquires these information acquired by the store terminal device 3 from the store terminal device 3 by communication. The user information acquisition unit 11a stores the acquired user attribute information, usage information, measurement information, and the like in the user information DB 12c of the storage unit 12. Further, the user information acquisition unit 11a acquires information regarding the selection result of the eyeglasses by the user and the presence or absence of subsequent inquiries or returns from the user who purchased the eyeglasses from the store terminal device 3 and stores the information in the user information DB 12c.

The lens specification estimation unit 11b performs a process of estimating the specifications of the spectacle lens suitable for this user based on the user's attribute information, application information, and measurement information acquired by the user information acquisition unit 11a. In the present embodiment, the lens specification estimation unit 11b estimates the lens specifications using the machine-learned specification prediction model 12b stored in the storage unit 12. The lens specification estimation unit 11b inputs the attribute information, application information, and measurement information acquired by the user information acquisition unit 11a into the specification prediction model 12b, and acquires the lens specification information output by the specification prediction model 12b in response to this input. By doing so, the specifications of the spectacle lens suitable for the user are inferred.

The specification information transmission unit 11c transmits the lens specification information estimated by the lens specification estimation unit 11b to the store terminal device 3 which is the source of the user's attribute information, application information, and measurement information as the estimation result of the lens specifications. I do.

The learning processing unit 11d performs learning processing of the specification prediction model 12b that estimates the specifications of the spectacle lens suitable for the user. The learning processing unit 11d performs learning processing using teacher data created in advance by the administrator or developer of the system at the first learning stage of the specification prediction model 12b, that is, at the stage of generating the specification prediction model 12b. The teacher data used at this time can be created based on the specification information of the spectacle lens sold in the past at a store selling spectacles and the information about the user who purchased the spectacle lens. Further, the learning processing unit 11d may relearn the specification prediction model 12b using the data accumulated by the operation of this system. A device other than the server device 1 may perform the learning process of the specification prediction model 12b. In this case, the server device 1 does not have to include the learning processing unit 11d.

FIG. 3 is a schematic diagram showing a configuration example of the user information DB 12c. The user information DB 12c provided in the storage unit 12 of the server device 1 according to the present embodiment includes "user name", "attribute information", "use information", "measurement information", "guess specification information", and "purchase". Information such as "specification information" and "tracking information" is stored in association with each other. The "user name" may be, for example, the name (real name) of the user who visited the store, or may be identification information such as a user ID given to the user. The "user name" may be any information as long as it can uniquely identify the user.

The "attribute information" is information about the user's individual, for example, information such as the user's "age" and "gender". In addition, the "attribute information" may include various information such as occupation, hobbies, and the presence or absence of a chronic disease. In the present embodiment, the "attribute information" is information input by the user himself / herself using the tablet type terminal device 5 in the store.

The "use information" is information about the use of the eyeglasses that the user intends to purchase, and may include, for example, information such as "distance", "driving", and "indoor / outdoor". "Distance" is the distance to the object to be viewed with eyeglasses, for example, a long distance that emphasizes the use of eyeglasses outdoors, a medium distance that emphasizes watching TV indoors or working with a personal computer, etc. In addition, information such as a short distance that emphasizes smartphone operation or reading can be set. "Driving" is information indicating whether or not the user uses the glasses for driving a car. “Indoor / outdoor” is information indicating whether the user is likely to use the eyeglasses indoors or outdoors. In the present embodiment, the "use information" is information input by the user himself / herself using the tablet type terminal device 5 in the store.

The "measurement information" is information based on the result of the user's eye inspection performed using the inspection device 4 at the store, for example, the "refraction error (refraction force)" and "near vision" of the user's eye. Information can be included. The "refractive error" is information on changes in the refractive power of the eye with respect to the surrounding direction with respect to the three directional directions of, for example, the horizontal direction, the vertical direction, and the 45-degree azimuth direction of the eye. "Near vision" is the visual acuity when looking at something close to, for example, about 30 cm. In addition, the "measurement information" may include various information such as distance vision, interpupillary distance, and eye position.

The "guess specification information" is information related to the specifications of the spectacle lens estimated by the server device 1 based on the user's attribute information, application information, and measurement information, and includes information such as "lens type" and "addition power". Can be included. As the "lens type", for example, a lens type such as a single focus lens, a bifocal type, a bifocal type that emphasizes medium / near vision, and an indoor type can be set. "Additional power" is the difference between the power of the lens for seeing far and the power of the lens for seeing near in the bifocal type (progressive power lens) eyeglasses, and the type of lens is bifocal type. If, the value of the addition frequency is set.

The "purchase specification information" is information regarding the specifications of the lenses of the spectacles actually purchased by the user, and may include information of the same items as the "guess specification information". When the user purchases eyeglasses having the lens specifications estimated by the server device 1, the same information is set in the "purchase specification information" and the "guess specification information". In the present embodiment, the "purchase specification information" is, for example, information input by a store clerk using the store terminal device 3.

The "tracking information" is subsequent information about the user who purchased the eyeglasses, and may include, for example, information such as whether or not the purchased eyeglasses are returned or exchanged, the user's impression after use, and whether or not there is a complaint. .. Further, in the "tracking information", for example, it is judged that the lens specification information inferred to this user was correct because the user did not request replacement or the like even after a predetermined period of time had passed after the purchase. That fact may be stored as tracking information. In the present embodiment, the "tracking information" is, for example, information input by a store clerk using the store terminal device 3.

The server device 1 stores and stores the information acquired from the store terminal device 3 of each store in the user information DB 12c.

FIG. 4 is a schematic diagram for explaining a configuration example of the specification prediction model 12b. The specification prediction model 12b included in the server device 1 according to the present embodiment is a learning model in which learning processing such as machine learning or deep learning has been performed in advance. The learning model performs a predetermined operation on the input value and outputs the operation result, and the storage unit 12 stores data such as the coefficient and the threshold of the function that defines this operation as the specification prediction model 12b. To. The specification prediction model 12b according to the present embodiment is a machine in advance so as to output the type and addition power of the eyeglass lens suitable for the user as the specification information in response to the input of the user's attribute information, application information and measurement information. Learning is being done.

Further, in the present embodiment, the specification prediction model 12b includes two learning models, a classification model 101 and a regression model 102. The classification model 101 is a learning model that outputs the type of spectacle lens suitable for the user in response to the input of the user's attribute information, application information, and measurement information. The classification model 101 classifies and predicts the types of lenses having a high degree of conformity with respect to the user based on the input concessions. Further, the regression model 102 is a learning model that outputs the addition power of the lens of the spectacles suitable for the user in response to the input of the attribute information, the usage information and the measurement information of the user. In the present embodiment, the regression model 102 regresses and predicts the addition power of the lens for the user who is predicted to be suitable for the bifocal type (progressive power lens) in the classification model 101. For the classification model 101 and the regression model 102, for example, a learning model of a decision tree (analysis tree and regression tree) can be adopted. However, as the classification model 101 and the regression model 102, for example, a learning model such as SVM (Support Vector Machine) or a neural network may be adopted.

In the first learning process (generation process) of the classification model 101 and the regression model 102 of the specification prediction model 12b, for example, the specification information of the spectacle lenses sold in the past at a plurality of stores and the attribute information of the user who purchased the spectacles. Teacher data for learning created based on usage information and measurement information is used. The information about the glasses sold in the past is the information about the lens selected based on the knowledge and experience of the store clerk, and the learning process is performed using the teacher data created based on this information. Therefore, the specification prediction model 12b can be a learning model that inherits the knowledge and experience of the store clerk by learning. The teacher data should be labeled positive for the lens selected for the user if there are no problems with subsequent use of the eyeglasses, and negatively labeled if there are problems. Information on the correctness of the clerk's choice may be added.

5 and 6 are schematic views showing a configuration example of the classification model 101 of the specification prediction model 12b. For example, as shown in FIG. 5, the classification model 101 has a lens type such as a single focus lens, a bifocal type, a medium-near-focused bifocal type, and an indoor type for inputting user attribute information, application information, and measurement information. It can be a learning model that outputs the goodness of fit for each. The output of the classification model 101 is output as a small numerical value from 0.00 to 1.00 for each lens type, for example, and it is presumed that the lens type for which the largest numerical value is output is the lens type suitable for this user. To. When a plurality of candidates are listed, the first candidate, the second candidate, and the like can be used in the order of this numerical value.

Further, for example, in the classification model 101, as shown in FIG. 6, a plurality of classification models 101a to 101d may be generated for each type of lens. Each classification model 101a-101d outputs the goodness of fit of this lens to the user as a decimal value from 0.00 to 1.00 for any one of the lens types.

Although not shown, the regression model 102 may also be generated as one learning model, or may be generated as a different learning model for each type of lens.

FIG. 7 is a schematic diagram for explaining another configuration example of the specification prediction model 12b. In the example shown in FIG. 7, the specification prediction model 12b has the structure of a neural network in which a plurality of neurons are interconnected. Since it is an existing technique, detailed description thereof will be omitted, but a neuron is an element that performs an operation on a plurality of inputs and outputs one value as an operation result. Neurons have information such as weighting coefficients and thresholds used in calculations. The learning model of the neural network is an input layer that accepts the input of one or more data, an intermediate layer that performs arithmetic processing on the data received by the input layer, and one or aggregating the arithmetic results of the intermediate layer. It has an output layer that outputs multiple values. The learning process of deep learning is a process of setting appropriate values for the coefficients and thresholds of each neuron constituting the neural network by using a large number of learning data given in advance. The specification prediction model 12b of this example is a trained model in which deep learning is performed using teacher data on the training model of the neural network, and is, for example, a gradient descent method, a stochastic gradient descent method, an error backpropagation method, or the like. Learning is done by the method of.

The specification prediction model 12b of this example inputs user attribute information, application information, and measurement information, and outputs a lens type and addition power that are presumed to be suitable for this user. The specification prediction model 12b may output the lens type and the addition power separately, or may output the lens type and the addition power as a set. The learning process of the specification prediction model 12b is performed using the teacher data in which the user's attribute information, application information, and measurement information are associated with the lens type and the addition power.

In addition, regardless of which learning model configuration is adopted as the specification prediction model 12b, the server device 1 uses the user's attribute information, application information, measurement information, lens type, and addition power as the system operates. The information of the above can be accumulated, and the re-learning process of the specification prediction model 12b can be performed using the accumulated information. As a result, the accuracy of prediction by the specification prediction model 12b can be improved as the system operates.

FIG. 8 is a block diagram showing the configuration of the store terminal device 3 according to the present embodiment. The store terminal device 3 according to the present embodiment includes a processing unit 31, a storage unit (storage) 32, a communication unit (transceiver) 33, a display unit (display) 34, an operation unit 35, and the like. The processing unit 31 is configured by using an arithmetic processing unit such as a CPU or MPU. The processing unit 31 reads and executes the program 32a stored in the storage unit 32 to acquire the user's attribute information, application information, and measurement information, and the specification information of the spectacle lens suitable for the user. Performs various processes such as output process.

The storage unit 32 is configured by using, for example, a magnetic storage device such as a hard disk or a non-volatile memory element such as a flash memory. The storage unit 32 stores various programs executed by the processing unit 31 and various data required for processing by the processing unit 31. In the present embodiment, the storage unit 32 stores the program 32a executed by the processing unit 31. In the present embodiment, the program 32a is distributed by a remote server device or the like, which is acquired by the store terminal device 3 by communication and stored in the storage unit 32. However, the program 32a may be written in the storage unit 32, for example, at the manufacturing stage of the store terminal device 3. For example, in the program 32a, the store terminal device 3 may read the program 32a recorded on a recording medium such as a memory card or an optical disk and store it in the storage unit 32. For example, in the program 32a, the writing device may read what has been recorded on the recording medium and write it in the storage unit 32 of the store terminal device 3. The program 32a may be provided in a mode of distribution via a network, or may be provided in a mode recorded on a recording medium.

The communication unit 33 communicates with various devices via a network N including the Internet and a mobile phone communication network. In the present embodiment, the communication unit 33 communicates with the server device 1, the inspection device 4, the tablet-type terminal device 5, and the like via the network N. The communication unit 33 transmits the data given by the processing unit 31 to another device, and gives the data received from the other device to the processing unit 31.

The display unit 34 is configured by using a liquid crystal display or the like, and displays various images, characters, and the like based on the processing of the processing unit 31. In the present embodiment, the display unit 34 is used to display the estimation result of the specification information of the spectacle lens suitable for the user by the server device 1.

The operation unit 35 receives an operation of a store clerk or the like, and notifies the processing unit 31 of the accepted operation. For example, the operation unit 35 receives an operation of a clerk or the like by an input device such as a touch panel provided on the surface of a mechanical button or a display unit 34. Further, for example, the operation unit 35 may be an input device such as a mouse and a keyboard, and these input devices may be configured to be removable with respect to the store terminal device 3.

Further, in the store terminal device 3 according to the present embodiment, the processing unit 31 reads and executes the program 32a stored in the storage unit 32, so that the attribute information acquisition unit 31a, the usage information acquisition unit 31b, and the measurement information acquisition unit 3 The 31c, the user information transmission unit 31d, the estimation result output unit 31e, and the like are realized in the processing unit 31 as software-like functional units. The attribute information acquisition unit 31a performs a process of acquiring attribute information such as age and gender input to the tablet terminal device 5 by the user by communicating with the tablet terminal device 5 in the communication unit 33. .. Similarly, the usage information acquisition unit 31b performs a process of acquiring usage information input to the tablet terminal device 5 by the user by communicating with the tablet terminal device 5 in the communication unit.

The measurement information acquisition unit 31c performs a process of acquiring measurement information such as the refraction error and near vision of the user measured by the inspection device 4 by communicating with the inspection device 4 by the communication unit 33. Do. The user information transmission unit 31d associates the attribute information acquired by the attribute information acquisition unit 31a, the attribute information acquired by the usage information acquisition unit 31b, and the measurement information acquired by the measurement information acquisition unit 31c with the server device 1 as user information. Performs the process of sending to. The estimation result output unit 31e acquires the specification information of the spectacle lens suitable for this user transmitted by the server device 1 with respect to the user information transmitted by the user information transmission unit 31d, and displays the acquired specification information in the display unit 34. Performs output (display) processing.

FIG. 9 is a schematic diagram for explaining the configuration of the inspection device 4 according to the present embodiment. The illustrated inspection device 4 is a device for measuring the refraction characteristic (refraction error) of the user's eye. The inspection device 4 includes a processing unit 41, an operation unit 42, measuring glasses 43, a base 46, a light emitting unit 47, a distance sensor 48, and the like. The measuring spectacles 43 has a configuration in which a disc 44 having a first pinhole and a second pinhole (not shown) formed by narrowing down the incident light and allowing the light to pass through is supported by the spectacle frame 45. The measuring glasses 43 are worn by a user who is a measurement target.

The light emitting unit 47 is, for example, a display connected to the processing unit 41, and is a first so that the light passing through the first pinhole and the second pinhole of the measuring glasses 43 is simultaneously incident on the user's eye. Emits light and a second light. The processing unit 41 is configured by using, for example, a computer, and information on the arrival position at which the first light passing through the first pinhole and the second light passing through the second pinhole reach on the retina of the user's eye. Therefore, a process of calculating the refraction characteristic (refraction error) of the eye is performed. An operation unit 42, a light emission unit 47, and a distance sensor 48 are connected to the processing unit 41.

The operation unit 42 is configured by using an input device such as a keyboard and a mouse, and accepts an operation in which the user changes the arrangement direction or the positional deviation of the light emitting portion of the light emitting unit 47. The user performs an operation of changing the position deviation amount of the light emitting portion of the light emitting unit 47 by using the operation unit 42 so that the image of the light emitting portion of the light emitting unit 47 seen through the measuring glasses 43 does not shift. The processing unit 41 adjusts the emission positions of the first light and the second light by the light emitting unit 47 in response to the user's input to the operation unit 42.

The base 46 is a base on which the user rests his / her face so that the user's eyes are arranged at a predetermined position. When the user puts his / her face on the base 46, the position of the user's eyes is positioned so as to be a predetermined position suitable for measurement. The distance sensor 48 is a sensor that measures the distance from the light emitting unit 47 to the disk 44, and may be configured by using, for example, a laser range finder. The distance sensor 48 gives the measured distance to the processing unit 41.

The processing unit 41 has an adjustment amount a of the positional deviation by the user who received the input by the operation unit 42, a distance d from the light emitting unit 47 to the disk 44 measured by the distance sensor 48, and a third formed on the disk 44. Based on the distance h between the 1st pinhole and the 2nd pinhole, the user's refractive power (refraction error) K is calculated using the following equation (1).
K = D × (haa) / h… (1)
However, D = 1000 / d

Further, the processing unit 41 can communicate with the store terminal device 3, and transmits the calculated refraction force (refraction error) of the user to the store terminal device 3. In this example, the configuration in which the inspection device 4 measures the refraction error of the user has been described, but the present invention is not limited to this. The inspection device 4 may be, for example, a device for measuring the near vision of the user, or may be a device for measuring other eye characteristics. Further, the inspection device 4 may be configured to include two or more devices such as a device for measuring the refraction error and a device for measuring the near vision, and one device may be used for the refraction error and the device. It may be a device capable of performing a plurality of types of measurements such as near vision.

FIG. 10 is a block diagram showing the configuration of the tablet-type terminal device 5 according to the present embodiment. The tablet-type terminal device 5 according to the present embodiment includes a processing unit 51, a storage unit (storage) 52, a communication unit (transceiver) 53, a display unit (display) 54, an operation unit 55, and the like. The processing unit 51 is configured by using an arithmetic processing unit such as a CPU or MPU. By reading and executing the program 52a stored in the storage unit 52, the processing unit 51 receives the input of the user's attribute information, usage information, and the like, and transmits the input-accepted information to the store terminal device 3. Perform various processes such as processing.

The storage unit 52 is configured by using, for example, a magnetic storage device such as a hard disk or a non-volatile memory element such as a flash memory. The storage unit 52 stores various programs executed by the processing unit 51 and various data required for processing by the processing unit 51. In the present embodiment, the storage unit 52 stores the program 52a executed by the processing unit 51. In the present embodiment, the program 52a is distributed by a remote server device or the like, which is acquired by the tablet terminal device 5 by communication and stored in the storage unit 52. However, the program 52a may be written in the storage unit 52, for example, at the manufacturing stage of the tablet-type terminal device 5. For example, in the program 52a, the tablet-type terminal device 5 may read the program 52a recorded on a recording medium such as a memory card or an optical disk and store it in the storage unit 52. For example, in the program 52a, the writing device may read what has been recorded on the recording medium and write it in the storage unit 52 of the tablet-type terminal device 5. The program 52a may be provided in a mode of distribution via a network, or may be provided in a mode recorded on a recording medium.

The communication unit 53 communicates with various devices via a network N including the Internet and a mobile phone communication network. In the present embodiment, the communication unit 53 communicates with the store terminal device 3 via the network N. The communication unit 53 transmits the data given by the processing unit 51 to another device, and gives the data received from the other device to the processing unit 51.

The display unit 54 is configured by using a liquid crystal display or the like, and displays various images, characters, and the like based on the processing of the processing unit 51. In the present embodiment, the display unit 54 is used to display a screen for receiving input of user attribute information, usage information, and the like.

The operation unit 55 accepts the user's operation and notifies the processing unit 51 of the accepted operation. For example, the operation unit 55 accepts a user's operation by an input device such as a mechanical button or a touch panel provided on the surface of the display unit 54. Further, for example, the operation unit 55 may be an input device such as a mouse and a keyboard, and these input devices may be configured to be removable with respect to the tablet type terminal device 5.

Further, in the tablet-type terminal device 5 according to the present embodiment, the processing unit 51 reads out and executes the program 52a stored in the storage unit 52, so that the display processing unit 51a, the input receiving unit 51b, the information transmitting unit 51c, etc. Is realized in the processing unit 51 as a software-like functional unit. The display processing unit 51a performs a process of displaying an input screen for receiving input of attribute information and usage information from the user on the display unit 54. The input receiving unit 51b performs a process of receiving the input of the attribute information and the usage information by the user based on the operation performed on the operation unit 55 when the input screen is displayed on the display unit 54. The information transmission unit 51c performs a process in which the input reception unit 51b transmits the attribute information and the usage information of the user who has received the input to the store terminal device 3 by the communication unit 53.

<Glasses lens selection support processing>
A user who visits a store to purchase eyeglasses first inputs attribute information such as his / her age and gender and usage information such as the purpose of use and environment of use of the eyeglasses to be purchased on the tablet terminal device 5. To do. At this time, the tablet-type terminal device 5 displays an input reception screen for attribute information and usage information on the display unit 54. FIG. 11 is a schematic diagram showing an example of an input reception screen for attribute information and usage information displayed by the tablet-type terminal device 5. At the top of the illustrated input reception screen, below the message "Please answer the following questions", there is an input box that accepts name input, an input box that accepts age input, and either male or female as gender. There is a radio button (option button) that accepts the selection of. Further below, there is a radio button that accepts the selection of whether or not you have used eyeglasses, along with the message "Have you ever used eyeglasses?" And the message "Have you used progressive power lenses?" In addition, there is a radio button that accepts the selection of whether or not the progressive power lens has been used. The tablet-type terminal device 5 accepts the input of the user's attribute information by the input box and the radio button provided at the upper part of the input reception screen.

At the bottom of the input reception screen, a radio button for accepting a user's choice regarding how to use the glasses and a radio button for accepting a choice as to whether or not to use the glasses for driving a car are provided. Acceptance of how to use eyeglasses is provided below the message "What kind of usage do you value?", "You can see far away when you go out", "You can see well when working on a computer or indoors" and "Books and books" Accepts selection from the three options of "I can see the smartphone screen well". When accepting whether or not to use eyeglasses for driving a car, a selection is accepted from the "Yes" or "No" options provided with the message "Are the glasses also used for driving a car?". The tablet-type terminal device 5 accepts the input of the user's usage information by the radio buttons provided on the stocks of these input reception screens.

The tablet-type terminal device 5 displays the input reception screen of FIG. 11 on the display unit 54, and reflects the input to each item based on the operation on the operation unit 55. The tablet terminal device 5 accepts the input of the final information of the user for each item of the input reception screen in response to the operation for the completion button provided at the bottom of the input reception screen, and receives the input of the user's attribute information and the user's attribute information. The usage information is transmitted to the store terminal device 3.

FIG. 12 is a flowchart showing a procedure of processing performed by the tablet-type terminal device 5 according to the present embodiment. The display processing unit 51a of the processing unit 51 of the tablet-type terminal device 5 according to the present embodiment displays the input reception screen shown in FIG. 11 on the display unit 54 (step S1). The input reception unit 51b of the processing unit 51 determines whether or not the user has completed the input of information according to whether or not the operation for the completion button on the input reception screen has been performed (step S2). If the input is not completed (S2: NO), the processing unit 51 returns to step S1 and continues to display the input reception screen.

When the input of information by the user is completed (S2: YES), the input reception unit 51b acquires the information input in each item of the input reception screen (step S3). The information transmission unit 51c of the processing unit 51 transmits the information received in step S3 to the store terminal device 3 as user attribute information and usage information (step S4), and ends the process.

The user who has completed the input of the attribute information and the usage information in the tablet type terminal device 5 measures the refraction error, the near vision, and the like by the inspection device 4. The inspection by the inspection device 4 may be carried out under the initiative of a store clerk, or may be carried out by the user operating the inspection device 4 by himself / herself. For example, the user measures the refraction error with the inspection device 4 having the configuration shown in FIG. The user wears the measuring spectacles 43 of the inspection device 4, puts his / her face on the base 46, and sees the light emitting portion 47 from the first pinhole and the second pinhole formed in the disk 44 of the measuring spectacles 43. .. The user performs an operation of adjusting the positional deviation of the light emission image based on the first light and the second light emitted from the light emitting unit 47 by the operation unit 42. The inspection device 4 calculates the refraction error based on the adjustment amount by the user, and displays the measurement result on a display or the like.

FIG. 13 is a schematic view showing an example of the measurement information display screen displayed by the inspection device 4. On the illustrated measurement information display screen, the message "Your refraction error measurement result" is displayed at the top, and the refraction error measurement results of the right eye and the left eye are displayed below it. In this example, as the measurement results of the refraction error, for example, the measurement results of the user's spherical power S, cylindrical (astigmatism) power C, and astigmatism axis AX are shown for each of the right eye and the left eye.

After finishing the measurement of the refraction error of the user, the inspection device 4 displays the measurement information display screen of FIG. 13 on a display or the like. When the inspection device 4 receives an operation on the OK button provided at the bottom of the measurement information display screen, the inspection device 4 transmits the measurement result to the store terminal device 3 as the user's measurement information.

FIG. 14 is a flowchart showing a procedure of processing performed by the store terminal device 3 according to the present embodiment. The attribute information acquisition unit 31a and the usage information acquisition unit 31b of the processing unit 31 of the store terminal device 3 according to the present embodiment acquire the attribute information and the usage information transmitted from the tablet terminal device 5 (step S11). Next, the measurement information acquisition unit 31c of the processing unit 31 acquires the measurement information transmitted from the inspection device 4 (step S12). The user information transmission unit 31d of the processing unit 31 transmits the attribute information and usage information acquired in step S11 and the measurement information acquired in step S12 to the server device 1 (step S13).

The estimation result output unit 31e of the processing unit 31 determines whether or not the specification information of the glasses transmitted by the server device 1 has been received in response to the information transmission in step S13 (step S14). When the specification information is not received (S14: NO), the estimation result output unit 31e waits until the specification information is received from the server device 1. When the specification information of the spectacles is received from the server device 1 (S14: YES), the estimation result output unit 31e displays the received specification information on the display unit 34 as a candidate for the lens of the spectacles suitable for the user (step S15). ..

After that, the processing unit 31 determines whether or not the user has selected the specification information of the lens displayed as the candidate as the specification of the eyeglasses for which the candidate is purchased (step S16). When the displayed candidate lens is not selected by the user (S16: NO), the processing unit 31 acquires the specification information of the next candidate lens (step S17), returns the process to step S15, and then returns to step S15. Display specification information about candidates for. When the displayed candidate lens is selected by the user (S16: YES), the processing unit 31 ends the processing. If the specification information received from the server device 1 in step S14 includes lens specifications for a plurality of candidates, the processing unit 31 acquires the specifications of the next candidate from the received specification information. Can be done. On the other hand, when the specification information received from the server device 1 does not include a plurality of candidates, the processing unit 31 requests the server device 1 to transmit the specification information of the lens as the next candidate, and this request. The server device 1 transmits the specification information of the next candidate to the store terminal device 3 according to the above.

FIG. 15 is a flowchart showing a procedure of specification estimation processing performed by the server device 1 according to the present embodiment. The user information acquisition unit 11a of the processing unit 11 of the server device 1 according to the present embodiment acquires the user attribute information, usage information, and measurement information transmitted by the store terminal device 3 (step S21). The user information acquisition unit 11a stores the information acquired in step S21 in the user information DB 12c of the storage unit 12 (step S22).

The lens specification estimation unit 11b of the processing unit 11 inputs the user attribute information, application information, and measurement information acquired in step S21 into the specification prediction model 12b (step S23). The lens specification estimation unit 11b acquires the specification information (lens type and addition power) output by the specification prediction model 12b in response to the information input in step S23 (step S24). The lens specification estimation unit 11b stores the specification information acquired in step S24 in the user information DB 12c of the storage unit 12 as estimation specification information (step S25). The specification information transmission unit 11c of the processing unit 11 transmits the lens specification information acquired in step S24 to the store terminal device 3 (step S26), and ends the process.

FIG. 16 is a schematic view showing an example of a specification information display screen displayed by the store terminal device 3. On the illustrated specification information display screen, the message "Recommended lens for you" is displayed at the top, and below that, a single focus lens, bifocal type, medium-near-focused bifocal type, and indoor (middle-near) type. Images illustrating the shapes of the four lens types are displayed side by side in the horizontal direction. On the specification information display screen, one lens type is displayed in a selected state according to the specification information from the server device 1. In this figure, the selected state is expressed by enclosing the lens type and its description in a thick frame.

Further, on the specification information display screen, when the bifocal type (progressive power lens) is selected, the addition power is shown according to the specification information from the server device 1. In the illustrated example, a value of 1.50D is shown as the addition power of the bifocal type. In addition, on the specification information display screen, a brief explanation about the selected lens type is displayed. In the illustrated example, the sentence "It corresponds to a wide range from a distance to the hand, it is also recommended for outdoor activities" is displayed as a description of the bifocal type.

The store terminal device 3 displays a specification information display screen on the display unit 54 in which a lens type presumed to be suitable for the user is selected based on the lens specification information received from the server device 1. Further, when the bifocal type (progressive power lens) is presumed to be suitable for the store terminal device 3 in the specification information, the estimated value of the addition power included in the specification information is shown on the specification information display screen.

At the bottom of the specification information display screen, there is a button labeled "Select a recommended type of lens" and a button labeled "Select another type of lens". It is provided. When the store terminal device 3 accepts an operation for the "select a recommended type of lens" button, the lens type and addition power lens shown on the specification information display screen are selected by the user. The subsequent processing is performed as. Subsequent processing includes, for example, notifying the store clerk of the user's selection result and encouraging the provision of eyeglasses to be tried on, or displaying an estimated price for the eyeglasses of the selected lens. obtain.

Further, when the store terminal device 3 accepts an operation for the button of "select another type of lens", the store terminal device 3 displays, for example, a specification information display screen showing the specification information of the second candidate lens. The server device 1 may include a plurality of lens specification candidates in the specification information based on the numerical values output by the specification prediction model 12b. In this case, the store terminal device 3 includes the specification information received from the server device 1. The information of the included second candidate is displayed. Alternatively, the store terminal device 3 requests the server device 1 for the specification information of the second candidate lens, and in response to this request, the server device 1 infers the specifications of the second candidate lens and stores the specification information. It may be transmitted to the terminal device 3.

For example, when the specification prediction model 12b outputs a numerical value such as the degree of suitability or recommendation to the user for each lens type, the server device 1 selects the first candidate, the second candidate, and the like in descending order of the numerical value. .. The server device 1 transmits the specification information including the information of a plurality of candidates for the ones in a predetermined rank range such as the top three, or the specification information for the one whose goodness-of-fit value exceeds a predetermined threshold value. Send information including information on multiple candidates. When the specification information received from the server device 1 includes information on a plurality of candidates, the store terminal device 3 presents the lens specifications to the user on the specification information display screen in order from the highest ranked candidate, and the user. If this lens is not selected, the specifications of the next candidate lens are presented to the user on the same specification information display screen. The store terminal device 3 may present the specification information of a plurality of candidates to the user, for example, on the specification information display screen, and in this case, the plurality of candidates may be ranked and displayed.

The user can select a lens based on the specifications recommended on the specification information display screen of the store terminal device 3, and try on, for example, eyeglasses having this lens. The user decides to purchase these spectacles when he / she is satisfied with the spectacles he / she has tried on, and when he / she is not satisfied with the spectacles he / she has tried on, he / she repeats trying on the spectacles by changing the lens specifications such as the second candidate, the third candidate, and so on. Can be done.

<Learning process>
FIG. 17 is a flowchart showing a procedure for generating the specification prediction model 12b performed by the server device 1 according to the present embodiment. The learning processing unit 11d of the processing unit 11 of the server device 1 according to the present embodiment contains user attribute information, usage information, measurement information, and spectacle lens specification information created in advance by the designer of the system or the like. Acquire the associated teacher data (step S31). The learning processing unit 11d uses the teacher data acquired in step S31 to perform supervised learning processing on a learning model prepared in advance (step S32). The learning processing unit 11d stores the learned learning model learned by the learning process in step S32 in the storage unit 12 as the specification prediction model 12b (step S33), and ends the generation process.

FIG. 18 is a flowchart showing the procedure of the re-learning process of the specification prediction model 12b performed by the server device 1 according to the present embodiment. The processing unit 11 of the server device 1 according to the present embodiment acquires the specification information of the lens of the spectacles finally purchased by the user from the store terminal device 3, and uses the acquired information as the purchase specification information of the storage unit 12. It is stored in the user information DB 12c (step S41). Further, the processing unit 11 acquires tracking information regarding the presence or absence of inquiries, returns, exchanges, complaints, etc. from the user who purchased the eyeglasses from the store terminal device 3, and stores the acquired tracking information in the user information DB 12c of the storage unit 12. Store (step S42). The processing unit 11 determines whether or not the accumulation of a sufficient amount of information for re-learning has been completed in the user information DB 12c (step S43). When the accumulation of sufficient information is not completed (S43: NO), the processing unit 11 returns to step S41 and continues to accumulate the information.

When the accumulation of sufficient information is completed (S43: YES), the learning processing unit 11d of the processing unit 11 creates teacher data for re-learning based on the information stored in the user information DB 12c (step S44). .. At this time, the learning processing unit 11d can create teacher data by associating the user attribute information, the usage information, and the measurement information stored in the user information DB 12c with the purchase specification information, for example. Further, for example, when the guessing specification information and the purchasing specification information are different from each other, the learning processing unit 11d may attach a negative label to the guessing specification information and use it as teacher data. Further, for example, the learning processing unit 11d may attach a negative label to the purchase specification information returned or exchanged based on the tracking information and use it as teacher data.

The learning processing unit 11d reads out the specification prediction model 12b stored in the storage unit 12 (step S45). The learning processing unit 11d relearns the specification prediction model 12b read in step S45 using the teacher data created in step S44 (step S46). The learning processing unit 11d updates the specification prediction model 12b by storing the specification prediction model 12b relearned in step S46 in the storage unit 12 (step S47), and ends the processing.

<Summary>
In the spectacle lens selection support system according to the present embodiment having the above configuration, the server device 1 acquires the user's attribute information, application information, and measurement information from the store terminal device 3, and the acquired information is obtained from the specification prediction model 12b. By acquiring the specification information of the spectacle lens output by the specification prediction model 12b by inputting to, the specification of the spectacle lens suitable for the user is predicted. The specification prediction model 12b is a learning model generated by pre-learning using teacher data in which user attribute information, application information, and measurement information are associated with spectacle lens specification information. As a result, the spectacle lens selection support system can infer the specifications of the spectacle lens suitable for the user and present it to the user, and can be expected to assist the user in selecting the spectacle lens.

Further, in the spectacle lens selection support system according to the present embodiment, the attribute information input to the specification prediction model 12b includes the user's age information, and the measurement information includes the measurement result of the refraction error of the eyes. Previous studies have shown that the user's age can cause presbyopia, and that there is a correlation between the user's visual acuity and refraction error, and the user's age and refraction error are used as input information. By doing so, the spectacle lens selection support system can be expected to realize accurate estimation of lens specifications using the specification prediction model 12b.

Further, in the spectacle lens selection support system according to the present embodiment, the store terminal device 3 acquires the measurement result of the refraction error from the inspection device 4 capable of measuring the refraction error of the user, and the refraction error is obtained from the store terminal device 3. The server device 1 acquires the measurement information of. As a result, the server device 1 can easily acquire the measurement information of the refraction error of the user and use it for estimating the specifications of the lens.

Further, in the spectacle lens selection support system according to the present embodiment, the specification information output by the specification prediction model 12b includes information on the type of spectacle lens and the addition power of the lens. As a result, the user can know the type of spectacle lens and the power of addition that are suitable for himself / herself, and can easily select the spectacle lens.

Further, in the spectacle lens selection support system according to the present embodiment, the user's attribute information, application information and measurement information, the prediction specification information predicted by the specification prediction model 12b, and the purchase specifications finally selected and purchased by the user are used. The server device 1 stores and stores the information and the tracking information about the glasses purchased by the user in the user information DB 12c, generates teacher data based on the accumulated information, and relearns the specification prediction model 12b. I do. As a result, the spectacle lens selection support system can be expected to update the specification prediction model 12b based on the information accumulated by the operation of the system and improve the accuracy of estimating the spectacle lens suitable for the user.

In the present embodiment, the spectacle lens selection support system is configured to include a plurality of devices such as a server device 1, a store terminal device 3, an inspection device 4, and a tablet-type terminal device 5. Not limited. For example, one device may perform the processing performed by the server device 1, the store terminal device 3, the inspection device 4, and the tablet terminal device 5 in the present embodiment. The device configuration shown in the present embodiment and the division of roles of each device can be changed as appropriate. Further, the configuration of the inspection device 4 for measuring the refraction error shown in FIG. 9 is an example, and the present invention is not limited to this, and the inspection device 4 may perform any measurement. The screens shown in FIGS. 11, 13 and 16 are examples, and are not limited thereto.

Further, in the present embodiment, the store terminal device 3 performs the display processing of the specification information display screen shown in FIG. 16, but the present invention is not limited to this, and the display processing of the specification information display screen is performed by the server device 1. May go. That is, the server device 1 generates image data or the like for displaying the specification information display screen shown in FIG. 16 based on the specification information output by the specification prediction model 12b and transmits the data or the like to the store terminal device 3. The specification information display screen may be displayed on the display unit 34 based on the data received by the store terminal device 3 that has received the above.

The embodiments disclosed this time should be considered as exemplary in all respects and not restrictive. The scope of the present invention is indicated by the scope of claims, not the above-mentioned meaning, and is intended to include all modifications within the meaning and scope equivalent to the scope of claims.

1 Server device 3 Store terminal device 4 Inspection device 5 Tablet type terminal device 11 Processing unit 11a User information acquisition unit 11b Lens specification estimation unit (specification information acquisition unit)
11c Specification information transmission unit 11d Learning processing unit 12 Storage unit 12a Server program (computer program)
12b Specification prediction model (learning model)
12c User information DB
13 Communication unit 31 Processing unit 31a Attribute information acquisition unit 31b Usage information acquisition unit 31c Measurement information acquisition unit 31d User information transmission unit 31e Guessing result output unit 32 Storage unit 32a Program (computer program)
33 Communication unit 34 Display unit 35 Operation unit 41 Processing unit 42 Operation unit 43 Measuring glasses 44 Disc 45 Eyeglass frame 46 Base 47 Light emitting unit 48 Distance sensor 51 Processing unit 51a Display processing unit 51b Input reception unit 51c Information transmission unit 52 Memory 52a program

Claims (19)

Acquire teacher data in which the user's attribute information, the measurement information related to the user's eye, the usage information of the spectacles by the user, and the specification information of the spectacle lens are associated with each other.
Using the acquired teacher data, a learning model that outputs the specification information when the attribute information, the measurement information, and the usage information are input is generated.
How to generate a learning model. The attribute information includes the age of the user.
The measurement information includes the measurement result of the refraction error of the user's eye.
The method for generating a learning model according to claim 1. The specification information includes the type of spectacle lens.
The method for generating a learning model according to claim 1 or 2. The specification information includes the addition power of the spectacle lens.
The method for generating a learning model according to any one of claims 1 to 3. The teacher data in which the attribute information, the measurement information, and the usage information of the user input to the generated learning model are associated with the tracking information related to the lens selected by the user is acquired.
Using the acquired teacher data, the learning model is regenerated.
The method for generating a learning model according to any one of claims 1 to 4. On the computer
Acquire user attribute information, measurement information related to the user's eyes, and usage information of eyeglasses by the user.
The acquired attribute information, the measurement information, and the usage information are input to the learning model that outputs the specification information of the lens of the spectacles when the attribute information, the measurement information, and the usage information are input.
A computer program that executes a process of acquiring the specification information output by the learning model. The attribute information includes the age of the user.
The measurement information includes the measurement result of the refraction error of the user's eye.
The computer program according to claim 6. Obtaining the measurement result from a refraction error measuring device that measures the refraction error of the user's eye.
The computer program according to claim 7. The specification information includes the type of spectacle lens.
The computer program according to any one of claims 6 to 8. The specification information includes the addition power of the spectacle lens.
The computer program according to any one of claims 6 to 9. The attribute information, the measurement information, and the usage information input to the learning model, the specification information output by the learning model, and the information related to the lens selected by the user are stored in a database in association with each other.
The computer program according to any one of claims 6 to 10. The computer program according to any one of claims 6 to 11, which displays a screen for presenting the specifications of eyeglasses to the user based on the specification information acquired from the learning model. The computer program according to claim 12, wherein an image schematically showing a type of a lens of eyeglasses and an addition power of the lens are presented in association with each other on the screen. Accepting the selection of approval or disapproval regarding the specifications of the glasses presented on the screen,
The computer program according to claim 12 or 13, which displays a screen for presenting another specification when the option of rejecting the specification is accepted. On the computer
Acquire user attribute information, measurement information related to the user's eyes, and usage information of eyeglasses by the user.
When the acquired attribute information, the measurement information, and the usage information are input to the learning model that outputs the specification information of the lens of the spectacle when the attribute information, the measurement information, and the usage information are input, the said Acquire the specification information output by the learning model and
Based on the acquired specification information, a screen for presenting the specifications of the glasses to the user is displayed.
A computer program that executes processing. The computer program according to claim 15, wherein an image schematically showing a type of a lens of spectacles and an addition power of the lens are presented in association with each other on the screen. Accepting the selection of approval or disapproval regarding the specifications of the glasses presented on the screen,
The computer program according to claim 15 or 16, which displays a screen for presenting another specification when the option of rejecting the specification is accepted. Acquire user attribute information, measurement information related to the user's eyes, and usage information of eyeglasses by the user.
The acquired attribute information, the measurement information, and the usage information are input to the learning model that outputs the specification information of the lens of the spectacles when the attribute information, the measurement information, and the usage information are input.
Acquire the specification information output by the learning model.
Eyeglass lens selection support method. A storage unit that stores a learning model that outputs the specification information of the spectacle lens when the user's attribute information, the measurement information related to the user's eye, and the usage information of the spectacles by the user are input.
A user information acquisition unit that acquires the attribute information, the measurement information, and the usage information, and
A spectacle lens selection including a specification information acquisition unit that inputs the attribute information, the measurement information, and the usage information acquired by the user information acquisition unit into the learning model and acquires the specification information output by the learning model. Support system.

Images