JP6716123B1 - Festival method, computer program, and festival screening device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a ritual method, a computer program, and a ritual examination apparatus capable of appropriately judging whether or not a ritual is properly performed by a correct ritual subject. SOLUTION: Based on the information of the ritual object, the information of the ritual person, the status of the ritual of the ritual person, the acquired information of the ritual object, the information of the ritual person, and the situation. Then, it is determined whether or not the ritual is correct, and when the ritual is determined to be correct, payment processing of the cost of the ritual to the ritual is executed. [Selection diagram] Fig. 17

Description

The present invention relates to a festival method, a computer program, and a festival screening device.

There are many people who are worried about the absence of family members or children living abroad because there is no one to memorialize or keep a grave after their death.
After the contractor's death, there is a contract in which a funeral company or the like handles events up to the funeral.
In addition, in Patent Document 1, it is reserved in advance to manage and execute Buddhist events on behalf of the Danke, and if a cemetery becomes an unrelated cemetery, the cemetery will be surrendered 32 years later, and it will be transferred to an unreserved tower etc. for a permanent memorial service. The invention of Danke management and graveyard/gravestone management system is disclosed.

JP, 2001-236425, A

In the case of the invention according to Patent Document 1, an insurance company receives an insurance premium from a beneficiary, and a religious corporation such as a temple manages and executes Buddhist events on behalf of the Danke. However, there is a problem that the monk cannot check whether or not the memorial service according to the type has been done properly on time and without taking a nap.

The present invention has been made in view of such circumstances, and a ritual method, a computer program, and a ritual examination device capable of satisfactorily determining whether or not a ritual is properly performed by a correct ritual subject to a correct ritual target person. The purpose is to provide.

A ritual method according to an aspect of the present invention, a computer, to obtain information on the object of the ritual, to obtain information of the ritual, to obtain a video image of the ritual in the ritual through a camera, the ritual in the ritual. When a video image of a person is input, a learning model that outputs correctness information indicating whether the ritual is right or wrong is input by inputting a video image of the ritual person during the ritual through the camera, and is output from the learning model. Based on the correctness information, the acquired information on the ritual object, and the information on the ritual person, a process of determining whether the ritual is correct or not is executed.

According to the above configuration, since the right or wrong of the ritual is determined based on the information of the acquired ritual object (the person who is enshrined) and the ritual, and the status of the ritual, the correct ritual person falls asleep and is ambiguous with respect to the correct ritual object. There is no abnormal behavior such as bullshit, hollowing out, etc., and it is possible to satisfactorily determine whether or not the ritual was performed correctly.

In the above ritual method, as the status of the ritual, obtain a video image of the ritual, when the video image of the ritual is input, to the learning model that outputs the right or wrong of the ritual, the obtained image, You may input and output the right or wrong of the ritual and judge the right or wrong of the ritual.

According to the above configuration, it can be automatically and accurately determined based on the image whether or not the festival has been properly performed.

In the above ritual method, as the status of the ritual, acquires the voice of the ritual, accepts the voice of the ritual, and inputs the acquired voice to a learning model that outputs the right or wrong of the ritual, and the ritual. The correctness may be output and the correctness may be determined.

According to the above configuration, it can be automatically and accurately determined based on the voice whether or not the festival has been properly performed.

In the above-mentioned ritual method, if the learning model is determined to be positive, the verification result of the ritual by the verifier is accepted, and when it is determined to be positive by the verification result, the ritual is determined to be positive. Good.

According to the above configuration, the accuracy of ritual determination is further improved.

In the above ritual method, the verifier may verify the statement describing the payment through the terminal device.

In Patent Document 1, the fee for the ritual service is deducted from the insurance money after the ritual service (service) is completed, but it is not possible to confirm whether or not the deduction process has been performed without fraud.
According to the ritual method described above, it is possible to determine whether or not the payment of the ritual cost is correct.

In the above ritual method, the ritual object may be specified by a QR code (registered trademark) or an IC tag attached to the tile.

According to the above configuration, it is possible to reliably identify the object of the ritual.

In the above ritual method, when the ritual is completed and it is determined that the ritual is positive, a smart contract for paying the ritual cost of the ritual may be recorded in the blockchain.

According to the above configuration, the smart contract process can be executed in the block chain, and the payment process can be automatically executed when the transaction is approved.

In the above ritual method, the information that the ritual is finished and is determined to be positive may be transmitted to the block chain, and the processing of the smart contract may be executed.

According to the above configuration, the processing of the smart contract is executed, triggered by the transmission of the information that the ritual is finished and that it is determined to be positive.

In the above ritual method, the information may include information that the statement describing the payment is determined to be positive.

According to the above-mentioned composition, it can be verified whether payment of the cost of the ritual is correct.

In the above ritual method, from the wallet address of the terminal that manages the ritual object, send a transaction including a virtual currency amount that is a deposit to the ritual to the smart contract address to the blockchain, and the ritual ends, and If it is determined that the virtual contract is performed, a transaction including a part of the virtual currency amount of the deposited virtual currency amount addressed to the worshiper's wallet address with the fulfillment of the smart contract may be transmitted to the blockchain.

According to the above configuration, it is possible to easily transmit the cost of the ritual.

A computer program according to an aspect of the present invention obtains information on the rituals, obtains information on the ritual objects attached to the tiles, acquires the status of the rituals of the rituals, and information on the rituals, The computer is caused to execute processing for outputting information on the object of worship and the status of the worship.

According to the above configuration, it is possible to output the information of the ritual, the information of the ritual and the image of the ritual on the mobile terminal of the ritual and to examine the ritual by the external device.

In the above computer program, when the ritual of the ritual is determined to be positive, information about the cost obtained by the ritual may be displayed.

The ritual can check the cost of the ritual when it is determined that the ritual was done correctly.

A ritual examination device according to an aspect of the present invention includes a first obtaining unit that obtains information about a ritual object, a second obtaining unit that obtains information about a ritual, and a third obtaining unit that obtains the status of the ritual of the ritual. An acquisition unit, information about the ritual object, information about the ritual person, and a determination unit that determines whether the ritual is correct based on the situation, and when the determination unit determines positive, the And a payment processing unit that executes payment processing of the cost of the festival.

According to the above configuration, since the right or wrong of the ritual is judged based on the information of the acquired ritual object (the person who is enshrined) and the ritual, and the status of the ritual, the right person of the ritual is supposed to fall asleep etc. There is no abnormal behavior, and it is possible to satisfactorily determine whether or not the ritual was performed correctly.

In the above-mentioned ritual examination device, the third acquisition unit acquires an image of the image of the ritual, and the determination unit accepts an image of the image of the ritual, and outputs a learning model for outputting the right or wrong of the ritual, The correctness may be determined by inputting the acquired image and outputting the correctness of the festival.

According to the present invention, after identifying the object of worship and the person who worships the ritual, the right or wrong of the worship is determined based on the acquired worship status, so that the right person of worship does not have an abnormal behavior such as dozing. , It is possible to satisfactorily determine whether or not the ritual was performed correctly.

It is a schematic diagram which shows an example of a structure of the memorial examination system which concerns on Embodiment 1. It is a block diagram which shows an example of a structure of a terminal. It is a block diagram which shows an example of a structure of a terminal. It is a block diagram showing an example of composition of a memorial examination device. It is an explanatory view showing an example of a record layout of a beneficiary DB. It is an explanatory view showing an example of a record layout of memorial service DB. It is explanatory drawing which shows an example of the record layout of memorial DB. It is explanatory drawing which shows an example of the record layout of image DB. It is an explanatory view showing an example of a record layout of a voice DB. It is an explanatory view showing an example of a record layout of statement DB. It is explanatory drawing which shows an example of a record layout of examination related DB. It is a schematic diagram which shows an example of a structure of an image examination model. It is a schematic diagram which shows the process performed in a convolutional layer. It is a schematic diagram which shows the process performed in a pooling layer. 7 is a flowchart illustrating an example of a processing procedure of an image examination model generation processing by a control unit. It is a schematic diagram which shows the structure of a voice examination model. It is a flow chart which shows a procedure of examination processing of a picture and a sound by a control part. It is explanatory drawing which shows an example of the screen displayed on the display panel of a terminal. It is an explanatory view showing an example of a display screen displayed on a display panel. It is an explanatory view showing an example of a display screen displayed on a display panel. It is an explanatory view showing an example of a display screen displayed on a display panel. It is a flow chart which shows a procedure of processing which acquires an examination result from a terminal by a control part. It is a flow chart which shows a procedure of processing which acquires an examination result from a terminal by a control part. It is explanatory drawing which shows the specification in which the fee of examination A and B was additionally written. It is explanatory drawing which shows an example of the display screen of the display panel of a terminal. It is a schematic diagram which shows an example of a structure of the memorial examination system which concerns on Embodiment 2. It is explanatory drawing which shows an example of transmission of virtual currency amount. It is a flowchart which shows the procedure of the establishment process of a smart contract. It is explanatory drawing which shows an example of the display screen of the display panel of a terminal.

Embodiments of the present invention will be described below with reference to the drawings.
(Embodiment 1)
Hereinafter, a case where the ritual is a memorial service for a Buddhist event and the memorial service is a Buddhist monk will be described as an example. The cases of beneficiaries include spouses when heirs do not exist, even when heirs exist, all are married, are overseas, and do not want to burden the heirs. The ritual method according to the present embodiment is useful when there is a case where it is not desired to be enshrined together and the bereaved cannot monitor the memorial service by the memorial service.
FIG. 1 is a schematic diagram showing an example of the configuration of the memorial examination system according to the first embodiment. In the memorial examination system, the memorial examination apparatus 1 of the operation management company, the terminal 2 of the beneficiary, the terminal 3 of the staff A of the trust company as the service providing company, the terminal 4 of the staff B of the operation management company, and the staff A and the staff. The terminal 7 of the supervisor C of B is connected via the network N such as the Internet. The beneficiary contracts with a trust company as an agent before birth, a memorial service after death, and the memorial examination device 1 executes the memorial examination of the memorial service and the payment processing of the memorial service. The service providing company may be an insurance company. The memorial examination device 1 can provide an automatic examination of a memorial service as a cloud service. The fairness of the examination of the memorial service is maintained by the examination by the staff A, the staff B, and the supervisor C who belong to another company. The number of staff is not limited to two, and may be one or three or more. The supervisor C may be omitted. Note that the memorial service provider cannot conclude a contract with the service providing company according to the present embodiment.

The memorial examination device 1 examines whether or not the memorial service is correct based on the memorial video of the memorial service (monk), and when the memorial service is determined to be positive (hereinafter, also referred to as OK), the memorial service video and the memorial service A statement about the fee and the operation and management fee of the operation management company is sent to the staffs A and B, and the result of the determination of whether or not the staff A and B is right and wrong and the result of confirmation of the content of the statement are received. Further, the judgment of the staffs A and B is monitored. The examination of the staffs A and B may be conducted by the examination and monitoring staff C, or may be carried out by automatic examination like the memorial examination device 1. FIG. 1 shows a state in which the examination and monitoring staff C acquires the state in which the staff A judges the memorial video with the camera 65 and monitors the judgment of the staff A. Instead of the camera 65, the self-shooting (monitoring) cameras 37 and 47 built in the terminals 3 and 4 may be used. When the judgment result and the confirmation result are OK, the memorial examination device 1 remits a fee to the memorial service person, and performs payment processing of the fees of the staffs A and B and the operation management fee. Hereinafter, the case where the inspection and monitoring staff C is omitted will be described.
The memorial examination device 1 acquires information regarding the presence/absence of an abnormality in the image included in the acquired video using the image examination model 159 described later, and the information regarding the presence/absence of abnormality in the sound included in the acquired image using the audio examination model 160. To get

FIG. 2 is a block diagram showing an example of the configuration of the terminal 2. The terminal 2 includes a control unit 21, a main storage unit 22, a communication unit 23, an operation unit 24, a display panel 25, an auxiliary storage unit 26, a microphone 27, an imaging unit 28, a GPS reception unit 29, and a clock unit that control the entire apparatus. 30 is provided. The terminal 2 can be composed of, for example, a smartphone, a desktop computer, a notebook personal computer, a tablet, or the like. The control unit 21 can be configured by a CPU (Central Processing Unit), a ROM (Read Only Memory), a RAM (Random Access Memory), and the like. The control unit 21 may include a GPU (Graphics Processing Unit).

The main storage unit 22 is a temporary storage area such as an SRAM (Static Random Access Memory), a DRAM (Dynamic Random Access Memory), and a flash memory, and temporarily stores data required for the control unit 21 to execute arithmetic processing. Remember.
The communication unit 23 has a function of communicating with the memorial examination device 1, the terminal 3, and the terminal 4 via the network N, and can transmit and receive required information. The communication unit 23 transmits status information indicating the status of the memorial service such as the memorial service information, the memorial target information, the position information, and the video of the memorial service to the memorial examination device 1.

The operation unit 24 includes, for example, a touch panel, a hardware keyboard, a mouse, and the like, and can operate icons and the like displayed on the display panel 25 and input characters and the like.

The display panel 25 can be configured by a liquid crystal panel, an organic EL (Electro Luminescence) display panel, or the like. The control unit 21 performs control for displaying necessary information on the display panel 25. The control unit 21 displays, for example, information on whether or not the memorial service is based on the status information. In this case, the memorial examination device 1 automatically determines whether or not the prescribed examination conditions have passed.

The auxiliary storage unit 26 is a large-capacity memory, a hard disk, or the like, and stores a program necessary for the control unit 21 to execute processing, and an application program (hereinafter referred to as an application program) 261. In addition, the auxiliary storage unit 26 includes information related to the memorial service, video data acquired from the imaging unit 28, data related to processing performed by the terminals 3 and 4 (for example, data of examination results), and data received from the memorial examination device 1. Etc. are memorized.

The application program 261 stored in the auxiliary storage unit 26 may be provided by a recording medium 262 that readablely records the application program 261. The recording medium 262 is, for example, a portable memory such as a USB (Universal Serial Bus) memory, an SD (Secure Digital) card, a micro SD card, or a compact flash (registered trademark). The application program 261 recorded in the recording medium 262 is provided by communication via the communication unit 23.
Further, it may be read from the recording medium 262 using a reading device (not shown) and installed in the auxiliary storage unit 26.

The microphone 27 inputs the voice of the offering person and converts it into an electric signal.
The image pickup unit 28 includes, for example, a CCD (Charge Coupled Device) camera, and picks up an image or a video. The image or video captured by the image capturing unit 28 is processed by a processing circuit (not shown) and stored in the auxiliary storage unit 26.
The GPS receiver 29 receives radio waves from a plurality of GPS satellites and detects the position of the terminal 2.
The clock unit 30 measures time and outputs date and time information.

FIG. 3 is a block diagram showing an example of the configuration of the terminal 3. The terminal 3 includes a control unit 31, which controls the entire apparatus, a main storage unit 32, a communication unit 33, an operation unit 34, a display panel 35, an auxiliary storage unit 36, and a self-shooting (monitoring) camera 37. The terminal 3 can be composed of, for example, a desktop computer, a notebook personal computer, a tablet, a smartphone, or the like. The control unit 31 can be composed of a CPU, a ROM, a RAM, and the like. The main storage unit 32, the communication unit 33, the operation unit 34, the display panel 35, and the auxiliary storage unit 36 include the main storage unit 22, the communication unit 23, the operation unit 24, the display panel 25, and the auxiliary storage unit 26 of the terminal 2. It has a similar configuration.
Terminal 4 is constructed like the terminal 3, the control unit 41, main memory 42, a communication unit 43, Ru an operation unit 44, a display panel 45, auxiliary storage unit 46 and the self-shooting (surveillance) camera 47 ..

FIG. 4 is a block diagram showing an example of the configuration of the memorial examination device 1. Memorial examination apparatus 1 includes a control unit 11, a main storage unit 12, a communication unit 13, the language processing unit 14, auxiliary storage unit 15 and a clock unit 16 which controls the entire apparatus. The memorial examination device 1 can be configured by one or a plurality of servers.
The control unit 11 can be composed of a CPU, a ROM, a RAM, and the like. The control unit 11 may include a GPU (Graphics Processing Unit).

The main storage unit 12 is a temporary storage area such as SRAM, DRAM, flash memory, etc., and temporarily stores the data necessary for the control unit 21 to execute arithmetic processing.

A plurality of the memorial examination devices 1 may perform distributed processing.
The communication unit 13 has a function of communicating with the terminals 2, 3, and 4 via the network N, and can transmit and receive required information. Specifically, the communication unit 13 can receive the video data transmitted by the terminal 2. Further, the communication unit 13 can transmit the correctness information of the image or the sound to the terminal 2.

The language processing unit 14 has a voice recognition processing function and can convert the voice included in the video into text data (character string). In addition, since a voice or a character is usually output or displayed over a plurality of frames, the conversion processing of the text data by the language processing unit is performed not in a frame unit but in a plurality of frames.
The language processing unit 14 has a morphological analysis function and uses the dictionary data to extract a word, which is a minimum unit having meaning, from the converted text data.

The auxiliary storage unit 15 is a large-capacity memory, a hard disk, etc., and is a program necessary for the control unit 11 to execute processing, a program 151 for conducting a memorial examination described below, a person-to-be-served DB 152, a person-to-be held DB 153, a person-to-care DB 154, an image The DB 155, the voice DB 156, the specification DB 157, the examination-related DB 158, the image examination model 159, the voice examination model 160, and other data are stored.

The memorial service program (hereinafter referred to as program) 151 stored in the auxiliary storage unit 15 may be provided by a recording medium 161 in which the program 151 is readable. The recording medium 161 is, for example, a portable memory such as a USB memory, an SD card, a micro SD card, or CompactFlash (registered trademark). The program 151 recorded in the recording medium 161 is read from the recording medium 161 using a reading device (not shown) and installed in the auxiliary storage unit 15. Further, the program 151 stored in the auxiliary storage unit 15 may be provided by communication via the communication unit 13.

The beneficiary DB 152 is a database that stores information on the beneficiaries who are the subjects of memorial service. FIG. 5 is an explanatory diagram showing an example of the record layout of the beneficiary DB 152.
The beneficiary data stores a beneficiary ID column, a name column, a death date column, a year-of-death column, a memorial column, a QR code column of a tile, an agent company column, and a bereaved family or agent column. The beneficiary ID column stores the beneficiary ID for identifying the beneficiary. The name is the name of the beneficiary, the date of death is the date of death of the beneficiary, the year of enjoyment is the age of the beneficiary, the name of the beneficiary is the name of the beneficiary, and the QR code of the rank is the person being marked on rank 6. The QR code and agent are the trust companies, etc. contracted by the beneficiary. The bereaved family or agent column is the name, address, email address, phone number, and mobile phone number of the bereaved family when there is a bereaved family. And the mobile phone number.

The memorial service DB 153 is a database that stores information about memorial services.
FIG. 6 is an explanatory diagram showing an example of a record layout of the memorial service DB 153.
The memorial data stores a memorial ID column, a name column, a temple name column, an address column, a mail address column, a denomination column, and a beneficiary ID column. The offering person ID column stores the offering person ID for identifying the offering person. The name is the name of the memorial service, the temple name is the name of the temple where the memorial service is held, the address is the address of the temple, the email address is the email address of the memorial service, the denomination is the denomination to which the memorial service belongs, and the beneficiary ID is It is the above-mentioned beneficiary ID.

The memorial DB 154 is a database that stores information on memorial services of beneficiaries.
FIG. 7 is an explanatory diagram showing an example of a record layout of the memorial service DB 154.
The memorial data stores a memorial ID column and a memorial content column. The beneficiary ID column stores the above-mentioned beneficiary ID for identifying the beneficiary. The memorial content column stores the type of memorial service and the date of memorial service in chronological order for each beneficiary according to the contract.
The types of memorial service include the first seven days,..., 77 days, one hundred days,.. You may conduct a monthly memorial service every month. The contents of the memorial service include offerings in front of the grave, cleaning of the grave, funerals, heritage organization, and relics arrangement. In addition to the content of the memorial service, the cost (disbursement amount, etc.) is also stored.
In FIG. 7, the beneficiary of ID No. 1 performs a monthly memorial service, a seasonal memorial service, and a tomb cleaning after every death day after death.
In the case of the ID No. 2 beneficiaries, the memorial service begins from the first anniversary of the memorial service, and the memorial service is carried out until the 33rd memorial service and the permanent memorial service 50 years later.

The image DB 155 is a database that stores image information of images for each memorial service.
FIG. 8 is an explanatory diagram showing an example of the record layout of the image DB 155.
The image data includes an image ID sequence, a frame sequence, and an abnormal value sequence. The image ID column stores the image ID for identifying each image. Frame No. The column is the frame number of the image with the ID. Is. Images can be stored in frame units. When the frame rate is, for example, 30 fps, the video includes 30 frames (images) per second. When images are acquired at a predetermined frame interval instead of all frames for examination, the frame No. Is memorized. The abnormal value is a probability value of abnormality described later.

The audio DB 156 is a database that stores audio information of video images for each memorial service.
FIG. 9 is an explanatory diagram showing an example of a record layout of the voice DB 156. The sound and the text are stored for each divided unit time obtained by dividing the memorial time into a predetermined number.
The voice data includes a voice ID sequence and an abnormal value sequence. The voice ID column stores voice IDs for identifying sounds and texts in each division unit time. The abnormal value is a probability value of abnormality described later.

The statement DB 157 is a database that stores information on the cost of offerings for each recipient.
FIG. 10 is an explanatory diagram showing an example of the record layout of the statement DB 157.
The statement data stores a date/time column, a memorial content column, a memorial service fee column, an operation management company fee column, an examination A fee column, an examination B fee column, and a balance column for each beneficiary. The date and time is the date and time of the memorial service, the content of the memorial service is the above-mentioned memorial service content, the memorial service cost is the service cost paid to the memorial service, the examination A fee is the examination fee by the trust company staff A, and the examination B fee is the management company staff. The examination fee and the balance by B are the memorial expenses and the balance after paying each fee. In FIG. 10, it is assumed that the beneficiary has transferred 10 million yen to the trust company as an agent.

The examination-related DB 158 is a database that stores data relating to processing performed by the terminals 3 and 4.
FIG. 11 is an explanatory diagram showing an example of the record layout of the examination-related DB 158.
The examination-related data stores a beneficiary ID column, a staff A ID column, a staff B ID column, a memorial examination column, and a statement examination column. In the memorial examination column, the examination results of the memorial services of the staffs A and B are stored. In the case of OK, ○ is stored, and in the case of NG, × is stored. In the statement examination column, examination results for the statements of staffs A and B are stored. In the case of OK, ○ is stored, and in the case of NG, × is stored.

In the present embodiment, the control unit 11 acquires a memorial service, a memorial target (servant), position information of a place where the memorial service is held, and date and time information of the memorial service, and after these information is determined to be OK. , The image is acquired, and the image and the sound in the image are input to the image examination model 159 and the sound examination model 160, respectively, and it is determined whether there is an abnormality in the offering.

The learning models of the image examination model 159 and the voice examination model 160 can use a multilayer neural network (deep learning), for example, a convolutional neural network (Convolutional Neural Network) can be used, but other machine learning can be used. You may use. The image examination model 159 automatically determines whether or not the memorial image includes abnormality information, and outputs correctness information. The voice examination model 160 automatically determines whether the memorial voice includes abnormality information and outputs correctness information. Hereinafter, details of the image examination model 159 and the sound examination model 160 will be described.

As the image examination model 159, the control unit 11 constructs (generates) a neural network that receives an image as an input and outputs information indicating whether the offering is correct or not by learning the image feature amount of the offering in the image. ..
FIG. 12 is a schematic diagram showing an example of the configuration of the image examination model 159. The image examination model 159 shown in FIG. 12 is a convolutional neural network for images, and the input layer 153a, the convolutional layer 153b, the pooling layer 153c, the convolutional layer 153d, the pooling layer 153e, the fully connected layer 153f, and the output layer 153g are They are connected in this order. It should be noted that the numbers of the convolutional layers, the pooling layers, and the total bonding layers are for convenience and are not limited to the numbers shown in FIG. Also, the layer of the activation function is omitted for convenience. A video image is input to the input layer 153a in frame units. The image examination model 159 can determine whether the image includes abnormality information on a frame-by-frame basis or on a frame-by-frame basis. That is, the output layer 153g outputs correctness information of the image on a frame-by-frame basis. An example of an image containing abnormal information is a case where a memorial service person falls asleep, disappears from a place where the memorial service is performed, looks away, stretches, or otherwise performs a different behavior from the memorial service.

The output layer 153g may be configured with two output nodes, and the probability that the image content is normal and the probability that the image content is abnormal (failure) may be output. When the softmax function is used, it is a continuous probability value (for example, a value in the range from “0” to “1”). Alternatively, the reliability of at least one of passing and failing is divided into a plurality of categories (for example, reliability=100%, 95%, 90%, 85%, 80%,... It is also possible to provide as many output nodes as possible and output the probability of each segment.
The control unit 11 acquires the probability value of abnormality of the image of each frame as an abnormal value, and comprehensively determines whether or not the memorial image is abnormal. For example, when the number of frames having an abnormality probability value of 0.5 or more is equal to or greater than the threshold value, the memorial image is abnormal and the memorial service is rejected.

Further, the NG output node of the output layer 153g may be configured by a plurality of output nodes. In this case, each node may determine the reason why the image is not OK and output the probability of each reason. The reason can be, for example, "drowsiness", "looking away", "absence", etc., but is not limited to these.

FIG. 13 is a schematic diagram showing processing performed in the convolutional layer. Input/output data of the convolutional layer is also referred to as a feature map. Input data of the convolutional layer is also referred to as an input feature map, and output data of the convolutional layer is also referred to as an output feature map. The input feature map of the convolutional layer in the first stage is an input image in frame units. The processing performed in the convolution layer (also referred to as "convolution operation") is a filter operation performed by a convolution filter (also referred to as "filter").

As shown in FIG. 13, the input feature map is 8×8 pixels. The size of the filter is 3×3 pixels. In the convolution operation, the filter feature and the corresponding feature of the input feature map are multiplied while sliding the filter window at regular intervals with respect to the input feature map, the sum is calculated, and the calculated sum is output to the output feature map. Store in the corresponding pixel of. In the example of FIG. 13, the calculation result of the area S corresponding to the filter F1 of the input feature map is stored in the pixel S1 of the output feature map. Further, the calculation result of the area S corresponding to the filter F2 of the input feature map is stored in the pixel S2 of the output feature map. Similarly, the calculation result of the area S corresponding to the filter F3 of the input feature map is stored in the pixel S3 of the output feature map. The output feature map has a size of 6×6 pixels by moving the filter one pixel at a time and performing the same calculation. Here, by using the three filters F1, F2, and F3, three output feature maps are obtained.

In the learning of the image examination model 159, as the parameters relating to the filter, for example, the value of the element of the filter, the number of filters (3 in the example of FIG. 13), the size of the filter (3×3 in the example of FIG. 13), The movement width of the filter (also referred to as “slide”, which is 1 pixel in the example of FIG. 13), padding for filling the periphery (edge region) of the input feature map with 0, and the like are optimized. The convolutional layer allows extraction of spatial features of the image.

FIG. 14 is a schematic diagram showing processing performed in the pooling layer. The pooling layer performs a process of reducing the size of the two-dimensional feature map output from the convolutional layer. Specifically, the processing of consolidating the local areas of the image into one element is performed. For example, as shown in FIG. 14, in a 6×6 pixel feature map (output feature map), 2×2 local regions (window W) are aggregated into “4” which is the maximum value of each element. ing. It should be noted that the sliding of the window W is equal to the size of the window W, and in the example of FIG. 14, it slides by 2 pixels, so the feature map of 6×6 pixels is reduced to 3×3 pixels. With the pooling layer, for example, even if the characteristic portion is slightly deformed or displaced in the image, the characteristic portion can be extracted by absorbing the difference due to the deformation or displacement. In the present embodiment, the processing is performed for each frame, but the present invention is not limited to this. For example, the processing may be performed every 10 frames. Alternatively, a predetermined number of consecutive frames (for example, 10 frames) may be input as time-series information, and the output may be information regarding correctness (Recurrent Neural Network).

FIG. 15 is a flowchart showing an example of a processing procedure of generation processing of the image examination model 159 by the control unit 11.
The control unit 11 acquires teacher data in which a plurality of images in memorial service are associated with information indicating the correctness of each image (step S11).

The control unit 11 uses the teacher data to generate the image examination model 159 that outputs correctness information of the image when the memorial image is input (step S12). Specifically, the control unit 11 inputs the images of a plurality of persons, which is teacher data, to the input layer of the neural network, and acquires the identification result for identifying the presence/absence of abnormality in the image from the output layer. Examples of the abnormal image include images such as "dozing", "looking away", and "absent". In the case of "drowsiness", in order to distinguish from the blink, the frame No. Refers to a plurality of consecutive images. The control unit 11 compares the acquired identification result with the correct answer value (information labeled on the image) of the teacher data, and performs an operation in the intermediate layer so that the identification result output from the output layer approaches the correct answer value. The parameters used for processing are normalized. The parameters are, for example, weights between neurons (coupling coefficient), coefficients of an activation function used in each neuron, and the like. Although the method of parameter re-correction is not particularly limited, for example, the control unit 11 re-corrects various parameters using the error back propagation method. The control unit 11 stores the generated image examination model 159 in the auxiliary storage unit 15, and ends the series of processes.
In the case of RNN, a predetermined number of frames of image data are input, and a normal probability value and an abnormal probability value are output.

FIG. 16 is a schematic diagram showing the structure of the voice examination model 160.
The control unit 11 uses the sound and the text acquired by the language processing unit 14 as an input by learning the feature amount of the abnormality as the voice examination model 160, and outputs the information indicating the presence or absence of the abnormality of the memorial service. To build. The sound and the text are input every predetermined input time (the division unit time). For example, the neural network is a CNN (Convolution Neural Network), and has an input layer that receives input of sounds and texts, an output layer that outputs a result of identifying whether or not there is an abnormality, and an intermediate layer that extracts a feature amount.
In FIG. 16, the number of intermediate layers is three, but the number is not limited to this. For example, when the voice examination model 160 is CNN, the middle layer has a configuration in which convolution layers and pooling layers are alternately connected, and finally extracts the feature amount while compressing information. In FIG. 16, the description of the convolution layer and the pooling layer is omitted. The output layer has two neurons that output an identification result that identifies an abnormality in the offering, and identifies whether there is an abnormality in the offering based on the feature amount output from the intermediate layer. One neuron outputs a normal probability value, and the other neuron outputs an abnormal probability value.

Although the voice examination model 160 is described as being a CNN, the voice examination model 160 is not limited to the CNN, and for example, an RNN can be used. In the RNN, by using the intermediate layer at the previous time together with the input layer at the next time for learning, it is possible to consider time series information of text, that is, a plurality of words, or time series information of a speech waveform.

Further, without using the voice examination model 160, the text converted by the language processing unit 14 is calculated a correlation value with a sutra for a predetermined memorial stored in the auxiliary storage unit 15 in advance, and the degree of coincidence is determined. By doing so, the presence or absence of an abnormality in the offering may be determined.

The control unit 11 sets teacher data (abnormal data, normal data) in which sounds and texts input at each input time are associated with information (abnormal or normal) indicating abnormalities in each sound or text. ) Is used for learning.

The control unit 11 inputs the sound and the text, which are the teacher data, to the input layer, performs the arithmetic processing in the intermediate layer, and obtains the identification result indicating the presence or absence of the abnormality of the offering from the output layer. The identification result output from the output layer is a continuous probability value (for example, a value in the range from “0” to “1”) when the softmax function is used. The identification result may be a value (for example, a value of "0" or "1") discretely indicating the presence or absence of abnormality.

The control unit 11 compares the identification result output from the output layer with the information labeled on the voiceprint and the text in the teacher data, that is, the correct answer value, so that the output value from the output layer approaches the correct answer value. The parameters used for the arithmetic processing in the intermediate layer are normalized. The parameters are, for example, weights between neurons (coupling coefficient), coefficients of an activation function used in each neuron, and the like. Although the method of parameter re-correction is not particularly limited, for example, the control unit 11 re-corrects various parameters using the error back propagation method.

The control unit 11 performs the above processing on the sound and text included in the teacher data to generate the voice examination model 160. When the sound and the text are acquired, the control unit 11 uses the voice examination model 160 to detect whether there is an abnormality in the offering.

Abnormal sounds include the case where the voiceprint of the sound does not match the voiceprint of the memorialist, the case where the sound is not a real voice and is emitted from the player, and the abnormalities of the text do not match the text of the memorial service. There are cases, etc., and an abnormal value is calculated by combining sound and text.
For example, if the probability value of the abnormality of the sound and the text input for each input time is 0.5 or more, the number of the memorial sounds is abnormal, and the memorial is rejected.
Note that only the sound may be input to the voice examination model 160 without inputting the text.

The control unit 11 of the memorial examination device 1 may learn the image examination model 159 and the voice examination model 160 using the teacher data for each memorial service. Algorithms and parameters (for example, parameters relating to filters) included in the image examination model 159 and the voice examination model 160 are optimized in a form specialized for each memorial service to obtain the image examination model 159 and the voice examination model 160. be able to. This makes it possible to improve the accuracy of the memorial examination.

Further, when the control unit 11 determines that the acquired image does not pass the examination conditions by the image examination model 159 and the voice examination model 160, the control unit 11 outputs the reliability of the decision result so that the image examination model 159 and the voice examination model are output. The model 160 can be retrained.

Further, the control unit 11 may train the image examination model 159 so that the image examination model 159 outputs the identification information for identifying the NG frame among the plurality of frames forming the acquired video.

If the length and frame rate of the video are set in advance, and if the NG frame (successive multiple frames may be known), for example, it is easy to know how many seconds after the start of the video the frame is abnormal. can do.

Further, the control unit 11 may train the image examination model 159 so that the image examination model 159 outputs the image information in the NG frame among the plurality of frames forming the acquired video. For example, when the memorial person is dozing, a thumbnail (image information) of the image can be output. If the voice or text is abnormal, the voice or text can be output. As a result, the abnormal portion of the voice data can be grasped.

The image examination model 159 and the sound examination model 160 may be stored in the terminal 2 instead of being stored in the memorial examination device 1, and the memorial service may be examined by the terminal 2.
Further, the examination of the image and the examination of the sound may be performed not by the two learning models individually but by one learning model. In this case, the contents of the image are compared with the sound, and it is determined whether or not an abnormality is included.

Hereinafter, the examination process of the memorial service by the memorial examination device 1 will be described in detail.
The memorial person arranges the terminal 2 toward himself so that the situation of the memorial service can be imaged by the terminal 2.
FIG. 17 is a flowchart showing the procedure of the image and sound examination processing by the control unit 11. The control unit 11 may notify the terminal 2 that the predetermined memorial time has come before the examination process described below. In addition, when the beneficiary has a bereaved family, the bereaved family may instruct the terminal 2 about the start time of a predetermined memorial service. If there is a survivor, the memorial contents (a series of memorial video and audio) are transmitted to the survivor. The offering may start at sunrise, but it should be finished by sunset, and more preferably in the morning.
The control unit 21 reads the application program 261 when the memorial service person starts the application (S201). When the application is started, the screen shown in FIG. 18 is displayed on the display panel of the terminal 2. On the screen, buttons for "Send memorial information", "Send memorial information", "Send location information", "Send date and time information", "Get video", "Send video", and "Error" are displayed. .. After the beneficiary has taken a picture of himself/herself with the image pickup unit 28 of the terminal 2 or has given a name to the microphone 27 of the terminal 2, the control unit 21 taps the “send person information” button, and the control unit 21 is The information is transmitted to the memorial examination device 1 (S202).
The control unit 11 receives the offering information (S101).

The control unit 21 transmits the memorial object information to the memorial examination device 1 by tapping the “send memorial object information” button by the image of the QR code attached to the tile 6 by the image memoir 28 read by the memorial person ( S203).
The control unit 11 receives the offering information (S102).
When the memorial staff taps the "positional information transmission" button, the control unit 21 acquires the positional information from the GPS receiving unit 29 and transmits the positional information to the memorial examination device 1 (S204).
The control unit 11 receives the position information (S103).
When the memorial staff taps the "send date and time information" button, the control unit 21 acquires the date and time information from the clock unit 30 and sends the date and time information to the memorial examination device 1 (S205).
The control unit 11 receives the date and time information (S104).

The control unit 11 determines whether all the information of the memorial service information, the memorial service information, the position information, and the date and time information is OK based on the beneficiary DB 152, the memorial service DB 153, and the memorial service DB 154 (S105). ). The control unit 11 confirms that the beneficiary corresponding to the memorial object is stored in the beneficiary DB 152 and the beneficiary is stored in the beneficiary DB 153. Then, it is determined whether the address and temple name obtained from the position information match the contents of the memorial service DB 153, and whether the date and time match the date and time of the memorial service content of the memorial DB 154. For example, in the case of monthly memorial service, when the date and time is within ±3 days from the month of death, and when the memorial service is within ±2 months, it is determined that the date and time match the content of the memorial service. The numerical value is not limited to this case. When the control unit 11 determines that it is not OK (S105: NO), it sends an error message to the terminal 2 (S106) and ends the process. On the display panel 25 of the terminal 2, the “error” button flashes. When the mourner taps the "error" button, as shown in FIG. 19, the "monger", "servant", "date and time", "position", and "time" buttons are displayed, and NG The reason button flashes. For example, if the "date and time" is incorrect, the "date and time" button will blink.

When determining that all the information is OK (S105: YES), the control unit 11 transmits a video transmission instruction to the terminal 2 (S107).
When the memorial service person taps the "image acquisition" button, the imaging unit 28 starts imaging the memorial service, and the control unit 21 acquires an image (S206).
When the memorial staff taps the "video transmission" button, the control unit 21 transmits the video to the memorial examination device 1 (S207). During the memorial service, images are transmitted. The imaging ends when the memorial staff taps the "image acquisition" button again. The memorial service person may transmit the image after the memorial service is completed and the imaging is completed.
The control unit 11 receives the video (S108).

The control unit 11 determines whether or not the offering time measured by the clock unit 30 is OK according to the contents of the offering (S109). When the control unit 11 determines that it is not OK (S109: NO), the process proceeds to S106, and the process ends. The control unit 11 causes the “error” to blink on the display panel 25, and causes the “time” button to blink when the memorial staff taps the “error” button.

When the control unit 11 determines that the result is OK (S109: YES), the image in frame units extracted from the video is input to the image examination model 159 (S110).
The control unit 11 determines whether or not the image examination result output by the image examination model 159 is OK (S111). The control unit 11 compares the abnormal value output by the image examination model 159 with a threshold value and determines whether the examination result is OK.
When the control unit 11 determines that the image examination result is not OK (S111: NO), the control unit 11 transmits an instruction to redo the memorial service to the terminal 2 (S112), and displays NG evidence on the display panel 25 of the terminal 2 (S113). ), the processing ends.

FIG. 20 is an explanatory diagram showing an example of a display screen displayed on the display panel 25. The control unit 11 indicates the state in which the memorial person is dozing, together with the reason for the image ID and NG. The control unit 11 reads out images corresponding to the number of frames input to the image examination model 159 from the image DB 155, and sequentially displays the images on the display panel 25. The control part 11 displays the image (two continuous images in FIG. 20) in which the memorial person is dozing with the elapsed time. It is distinguished from blinking because the eyes are closed over multiple images. On the screen, buttons for “accept” and “not allow” are also displayed, and the memorial staff can tap any button to input.

When the control unit 11 determines that the image examination result is OK (S111: YES), the sound and the text are input to the voice examination model 160 (S114).
The control unit 11 determines whether or not the examination result of the voice output by the voice examination model 160 is OK (S115). The control unit 11 compares the abnormal value output by the voice examination model 160 with a threshold value to determine whether the examination result is OK. When the control unit 11 determines that the voice examination result is not OK (S115: NO), the process proceeds to S112. In S113, for example, when the content of the text is abnormal, as shown in FIG. 21, the normal text and the actual text are written together with the voice ID.
When the control unit 11 determines that the voice examination result is OK (S115: YES), the process ends.

Note that in the flowchart of FIG. 17, the audio is examined after the image of the video is examined, but the present invention is not limited to this case, and the image may be examined after the audio is examined. You can go.

When it is determined that the video examination result and the audio examination result are OK, the control unit 11 reads the statement corresponding to the person to be donated from the statement DB 157, and as shown in FIG. Write down the content of the memorial service, the memorial service fee, the management company fee, and the balance. Then, the video and the statement are transmitted to the terminal 3 of the staff A and the terminal 4 of the staff B of the operation management company. Each staff member personally examines the contents of the video and the statement, and sends the statement including the result of the video examination and the fee of the person who performed the examination to the memorial examination device 1. After confirming the staff ID, the memorial examination device 1 determines whether the examination is OK and whether the statement is OK.

22 and 23 are flowcharts showing the procedure of processing for acquiring the examination result from the terminals 3 and 4 by the control unit 11.
The control unit 11 transmits the video and the description to the terminals 3 and 4 (S121).
The control unit 31 or 41 receives the video and description (S301, S401).
The control unit 31 or 41 receives the staff ID input by the staff A or B through the operation unit 34 or 44 (S302, S402).
The control unit 31 or 41 transmits the staff ID (S303, S403).
The control unit 11 receives the staff ID (S122).

The control unit 11 determines whether the ID of the staff A is OK based on the examination-related DB 158 (S123). When determining that the ID of the staff A is not OK (S123: NO), the control unit 11 transmits an error message to the terminal 3 (S124) and ends the process.
When the control unit 11 determines that the ID of the staff A is OK (S123: YES), the control unit 11 determines whether the ID of the staff B is OK (S125). When the control unit 11 determines that the ID of the staff B is not OK (S125: NO), it sends an error message to the terminal 4 (S126), and ends the process.

When the control unit 11 determines that the ID of the staff B is OK (S125: YES), the result of the examination A or B of the video by the staff A or B and the confirmation result of the statement are displayed on the terminals 3 and 4. The transmission is instructed (S127).
The control unit 31 or 41 receives the examination result input by the staff A or B through the operation unit 34 or 44, and the confirmation result of the statement (S304, S404). The staff A or B inputs whether the examination result is OK or NG and the confirmation result of the specification is OK or NG. If the examination result or the confirmation result of the specification is NG, enter the reason.
The control unit 31 or 41 transmits the examination result and the confirmation result of the specification (S305, S405).

The control unit 11 receives the examination result and the description from the terminals 3 and 4 (S128).

The control unit 11 determines whether the examination result of the staff A is OK (S129). When the control unit 11 determines that the examination result of the staff A is not OK (S129: NO), the control unit 11 transmits an instruction to redo the memorial service to the terminal 2 (S130), and displays the reason for NG on the display panel 25 (S131). ), the processing ends. Reasons for NG include dozing of memorial service providers.
When the control unit 11 determines that the examination result of the staff A is OK (S129: YES), the control unit 11 determines whether the examination result of the staff B is OK (S132). When the control unit 11 determines that the examination result of the staff B is not OK (S132:NO), the process proceeds to S130.

When the control unit 11 determines that the examination result of the staff B is OK (S132: YES), the control unit 11 determines whether the statement confirmation result of the staff A is OK (S133). When the control unit 11 determines that the statement confirmation result is not OK (S133: NO), the statement is rewritten based on the reason of NG (S134), and the process proceeds to S136.
When the control unit 11 determines that the statement confirmation result is OK (S133: YES), the control unit 11 determines whether the statement confirmation result of the staff B is OK (S135). When the control unit 11 determines that the statement confirmation result is not OK (S135: NO), the process proceeds to S134.
When the control unit 11 determines that the statement confirmation result is OK (S135: YES), the fees for the examinations A and B are added to the statement (S136). FIG. 24 is an explanatory diagram showing a specification in which the fees for examinations A and B are additionally written.
The control unit 11 transmits the examination OK to the terminal 2 (S137). The remittance process to the terminal 2 is performed (S138), and the process ends.

FIG. 25 is an explanatory diagram showing an example of a display screen of the display panel 25 of the terminal 2.
The screen shows that the examination was successful, the amount of the memorial transfer fee, the transfer account, and the date and time. The memorial service cost is not limited to cash transfer to a pre-registered financial institution account, and may be paid in virtual currency. Alternatively, smartphone payment may be used.

According to the present embodiment, after determining the right or wrong of the person to be serviced and the person to be serviced, the right or wrong of the service is determined based on the acquired service status, so that the right person to be serviced, doze, etc. It is possible to satisfactorily determine whether or not the memorial service was performed correctly without any abnormal behavior.
When the image examination model 159 and the voice examination model 160 determine OK, the examination result of the memorial service by the staffs A and B is accepted, and when the examination result is OK, it is determined that the memorial service is OK. The accuracy of the examination is good.
By verifying the description, it is possible to determine whether payment of the memorial service is correct.

In addition, in the said embodiment, although the case where the examination of a video and the examination of an audio|voice is demonstrated, it is not limited to this. Only video or audio may be examined.
Further, in the above-described embodiment, the case where the image capturing unit 28 reads the QR code of the tile 6 to acquire the memorial object information is described, but the present invention is not limited to this. An IC tag may be attached to the tile 6 and the terminal 2 may be equipped with an IC tag reader, and the IC tag reader may obtain information on the IC tag.

(Embodiment 2)
FIG. 26 is a schematic diagram showing an example of the configuration of the memorial examination system according to the second embodiment. In the memorial examination system according to the second embodiment, the block chain 5 is further connected to the network N.
The block chain 5 is composed of a plurality of nodes and is a kind of distributed ledger technology or distributed network. The memorial examination device 1 may also configure a node. The block chain 5 stores data by generating a unit of data called a block at regular time intervals and connecting them like a chain. The block chain 5 is autonomously managed by using a peer-to-peer network and a distributed time stamp server. By storing the data in a block once by storing the data in a chain, the data cannot be retroactively changed. The block chain 5 may be a public type or a private type. In the present embodiment, recording of memorial service and payment of costs are implemented by the block chain 5. As the virtual currency, Ethereum, Hyperledger or the like which can implement a smart contract may be used.

The control unit 11 receives the memorial information, the memorial target information, the position information, and the video from the terminal 2, acquires the abnormal value of the image by the image examination model 159, and acquires the abnormal value of the voice by the voice examination model 160, When the examination result of the image and the sound is OK, the processing until the staff A and B receive the examination result is the same as the processing of the control unit 11 according to the first embodiment.

The control unit 11 says, "If the examination result by the image examination model 159 and the sound examination model 160, the staff A, and the staff B and the confirmation result of the statement are OK, the memorial service provider, the operation management company, the staff A, and the staff B." Send the amount of virtual currency to each wallet address as a commission." Create a smart contract code. The conditions of the smart contract are not limited to the above cases. For example, it may be “when either one of the examination results by the image examination model 159 and the voice examination model 160, the staff A, and the staff B is OK”, or the confirmation result of the specification may not be included in the conditions. ..
The control unit 11 gives the smart contract ID to the block chain 5, transmits the smart contract, and records the smart contract on the block chain 5.

As shown in FIG. 27, the control unit 11 makes a virtual deposit as a deposit to the smart contract address (smart contract ID “01”) from the wallet address AAA of the trust company as an agent related to the beneficiary. A transaction including a currency amount (5 coins in FIG. 27) is transmitted to the block chain 5. The wallet address AAA may be the wallet address of the terminal 3 of the staff A or the terminal 4 of the staff B.
When the examination result by the image examination model 159 and the voice examination model 160, the staff A, and the staff B and the confirmation result of the specification are OK, the control unit 11 executes the smart contract processing by the block chain 5. When the transaction is approved, the transaction including the partial virtual currency amount (5 coins in FIG. 27) of the deposited virtual currency amount is transmitted to the terminal 2 addressed to the wallet address BBB of the memorial terminal 2.
Similarly, a transaction including a portion of the virtual currency amount of the deposited virtual currency amount is transmitted to the block chain 5 addressed to the wallet addresses of the terminals 3 and 4.
The wallet address is an account number for virtual currency created for each transaction, and is displayed as a character string or QR code (registered trademark). The wallet address is a mathematically derived address based on the public key cryptography used in the block chain 5. The source and destination wallet addresses are recorded in the blockchain 5 and are not tampered with as long as the system of the blockchain 5 continues to maintain them.

FIG. 28 is a flowchart showing the procedure of a smart contract establishment process.
The control unit 11 creates a code for the smart contract (S141).
The control unit 11 adds the smart contract ID and transmits the smart contract to the block chain 5 (S142).
The block chain 5 receives the smart contract (S501).
The block chain 5 records the smart contract (S502).

The control unit 11 acquires the wallet address AAA of the trust company related to the beneficiary (S143).
The control unit 11 transmits the virtual currency amount (for example, 50 coins) from the wallet address AAA to the smart contract ID “01” (S146).
The block chain 5 deposits the virtual currency amount (S503).

The control unit 11 transmits an activation trigger of the smart contract to the block chain 5 (S147). The activation trigger is, for example, information that "the examination result by the image examination model 159 and the voice examination model 160, the staff A, and the staff B and the confirmation result of the specification are OK".
The block chain 5 receives the activation trigger (S504). As the information of the specification, the content shown in FIG. 23 is converted into the virtual currency amount.
The block chain 5 executes a smart contract process (S505).
The block chain 5 records the approval of the transaction in the block (S506). As a result, it is possible to confirm the content of the transaction approval at the destination of the public key.

The block chain 5 transmits the approval of the transaction to the memorial examination device 1 and the terminal 2 (S507).
The control unit 11 receives the transaction approval (S148), and the control unit 21 of the terminal 2 receives the transaction approval (S211).
The block chain 5 transmits a part of the deposited virtual currency amount (for example, 5 coins) to the wallet address BBB5 of the memorial terminal 2 (S508), and ends the process.
Similarly, the virtual currency amount is transmitted to the wallet addresses of the staffs A and B.
The smart contract may be approved by the miner.

FIG. 29 is an explanatory diagram showing an example of a display screen of the display panel 25 of the terminal 2.
On the screen, it is described that the transaction has been approved and that XX coins have been transmitted to the wallet address BBB.

According to the present embodiment, when the image examination model 159 and the voice examination model 160, and the staffs A and B determine that the result is positive, the smart contract process is executed in the blockchain 5 and the transaction is approved. The payment process is easily executed.

The above embodiment is not restrictive. The scope of the present invention is intended to include meanings equivalent to the claims and all modifications within the scope.
In the first and second embodiments, the case where the ritual is a memorial service has been described, but the present invention is not limited to this. In Shinto, the ritual method of the present invention is also applied to the case where a priest sings a congratulatory message. It also applies to other religions at rituals after death.

1 memorial examination device 2, 3, 4 terminal 11, 21, 31 control unit 12, 22, 32 main storage unit 13, 23, 33 communication unit 24, 34 operation unit 15, 26, 36 auxiliary storage unit 16, 30 clock unit 151 Program 159 Image Screening Model 160 Voice Screening Model 5 Blockchain

Claims (11)

On the computer,
Get information about the festival,
Get the information of the worshipers,
Acquire a video of the ritual during the ritual through a camera,
When a video image of a ritual during a ritual is input, a learning model for outputting correctness information indicating whether the ritual is right or wrong is input by inputting an image of the ritual during the ritual through the camera.
Whether or not the ritual is correct is determined based on correctness information output from the learning model, information on the acquired ritual object, and information on the ritual.
A ritual method that causes processing to be performed. When it is determined to be positive, the payment processing of the cost of the ritual to the ritual is executed.
The ritual method according to claim 1. The voice of the ritual during the ritual is recorded through a microphone to obtain the voice of the ritual,
The ritual method according to claim 1, wherein the acquired voice is input to a learning model that receives the voice of the ritual and outputs the right or wrong of the ritual, and outputs the right or wrong of the ritual to determine the right or wrong. When the learning model determines positive, the verification result of the ritual by the verifier is accepted from the verifier's terminal device , and when the positive result is determined by the verification result, the ritual is determined to be positive. The method of ritual according to claim 2 or 3. The ritual method according to claim 2 , further comprising receiving a confirmation result of a statement that describes the payment process verified by a verifier . The ritual method according to any one of claims 1 to 5 , wherein a smart contract for paying the ritual's expense to the ritual is recorded in a blockchain when the ritual is determined to be positive . The ritual method according to claim 6 , further comprising transmitting information that the ritual is determined to be positive to the block chain and executing the processing of the smart contract. The ritual method according to claim 7 , wherein the information includes information that the statement describing the payment is determined to be positive. From the wallet address of the terminal that manages the ritual object, send a transaction containing the amount of virtual currency that is a deposit to the ritual to the smart contract address, to the blockchain,
When the ritual is finished and it is determined to be positive, a transaction including a part of the virtual currency amount of the virtual currency amount which is the deposit is addressed to the worship address of the ritual by the execution of the smart contract. The ritual method according to claim 7 or 8, which is transmitted to a blockchain. On the computer,
Get information about the festival,
Get the information of the worshipers,
Acquire a video of the ritual during the ritual through a camera,
When a video image of a ritual during a ritual is input, a learning model that outputs correctness information indicating whether the ritual is right or wrong is input by inputting an image of the ritual during the ritual through the camera,
Whether or not the ritual is correct is determined based on correctness information output from the learning model, information on the acquired ritual object, and information on the ritual.
A computer program that executes a process. An acquisition unit that acquires information about the festival,
A worshiper information acquisition unit that acquires worshiper information,
An image acquisition unit that acquires an image of the ritual during the ritual taken by a camera,
A learning model that outputs correctness information indicating whether the ritual is right or wrong, when a video image of the ritual during the ritual is input.
An input unit for inputting, to the learning model, an image obtained by capturing the image of the worshiper during the worship,
Correctness information output from the learning model, the acquired information about the ritual object, and a determination unit that determines whether the ritual is correct, based on the information about the ritual person.
A ritual examination device equipped with.

Images