Wearer user identification device, wearer user identification system, wearer user identification method, and wearer user identification program

Abstract

A wearer identification device according to an embodiment comprises a feature amount generation unit, a determining unit, and an identification unit. The feature amount generation unit receives, from a sensor worn on a part of the body of a user to be identified, a measurement signal corresponding to the vibration characteristics of the body part of the user measured by the sensor, and generates a feature amount representing the vibration characteristics from the measurement signal. The determining unit determines, on the basis of the magnitude of fluctuations in the feature amount generated by the feature amount generation unit, whether the state of the part on which the sensor is worn by the user is stable. The identification unit performs personal identification of the user on the basis of the feature amount generated by the feature amount generation unit, if the determining unit determines that the state of the part on which the sensor is worn is stable.

Description

[Technical field]

[0001] An embodiment of the present invention relates to a wearer user identification device, a wearer user identification system, a wearer user identification method, and a wearer user identification program. [Background technology]

[0002] There is a method called active acoustic sensing, which uses a pair of piezoelectric elements such as a piezo element, one as a speaker and the other as a microphone, to measure the vibration characteristics of an object on which the speaker and microphone are installed, and estimates the state of the object or the state of the object being grasped based on the measured vibration characteristics (see, for example, Non-Patent Document 1). This active acoustic sensing transmits sound waves in an inaudible range from the speaker toward the object, receives the vibrations propagated through the object with the microphone, and analyzes the frequency characteristics of the received signal, utilizing the fact that changes in the vibration characteristics occur due to changes in the internal structure and boundary conditions of the object on which the speaker and microphone are installed.

[0003] By using this active acoustic sensing technique, it is also possible to identify each target. [Prior art documents] [Non-patent literature]

[0004] [Non-Patent Document 1] Yuki Kubo, Kana Eguchi, Ryosuke Aoki, Shigekuni Kondo, Shozo Azuma, and Takuya Indo. "FabAuth: Printed Objects Identification Using Resonant Properties of Their Inner Structures." Proceedings of Extended Abstracts of the 2019 CHI Conference on Human Factors in Computing Systems, 2019. (CHI EA '19) Summary of the Invention [Problem to be solved by the invention]

[0005] When using active acoustic sensing to identify a user, rather than a static object, the body part to which the speaker and microphone are attached may move, unlike a static object. When the body part is moved, the internal structure of the body part changes, causing a change in the vibration characteristic value. This noise may cause the user to be unable to be identified properly.

[0006] The present invention provides a technique capable of reducing erroneous determinations when active acoustic sensing is used for personal identification of a user. [Means for solving the problem]

[0007] In order to solve the above problems, a wearing user identification device according to one aspect of the present invention includes a feature generation unit, a determination unit, and an identification unit. The feature generation unit receives a measurement signal corresponding to the vibration characteristics of the user's body part measured by the sensor from a sensor worn on the body part of the user to be identified, and generates a feature representing the vibration characteristics from the measurement signal. The determination unit determines whether the state of the part where the sensor is worn on the user is stable based on the magnitude of variation in the feature generated by the feature generation unit. When the determination unit determines that the state of the part where the sensor is worn is stable, the identification unit performs personal identification of the user based on the feature generated by the feature generation unit. The measurement signal received from the sensor is a vibration signal obtained by detecting vibrations that are applied to the user's body, which is the part where the sensor is attached, and propagated inside the body. Effect of the Invention

[0008] According to one aspect of the present invention, a technology can be provided that can reduce erroneous determinations when active acoustic sensing is used to identify a user by performing personal identification of the user only in a stable state where no significant changes occur in the vibration characteristic values. [Brief description of the drawings]

[0009] [Figure 1] FIG. 1 is a block diagram showing an example of the configuration of a wearer user identifying system including a wearer user identifying device according to an embodiment of the present invention. [Diagram 2] FIG. 2 is a plan view showing the configuration of the measurement unit worn by the user. [Diagram 3] FIG. 3 is a schematic diagram showing a state in which the measurement unit is worn by a user. [Figure 4] FIG. 4 is a diagram illustrating an example of a spectrogram. [Diagram 5] FIG. 5 is a block diagram showing an example of a hardware configuration of the wearing user identification device. [Figure 6] FIG. 6 is a flowchart showing an example of a learning process operation related to learning of a classification model in the wearing user identification device. [Figure 7] FIG. 7 is a flowchart showing an example of an identification process operation related to personal identification of a user in the wearing user identification device. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] An embodiment of the present invention will now be described with reference to the drawings.

[0011] FIG. 1 is a block diagram showing an example of a configuration of a wearing user identification system 1 according to an embodiment of the present invention. The wearing user identification system 1 includes a wearing user identification device 10 according to an embodiment of the present invention, a measurement unit 20, and an audio interface unit 30. The wearing user identification device 10 includes a signal generation unit 11, a signal storage unit 12, a feature generation unit 13, a model learning unit 14, a model storage unit 15, an identification execution determination unit 16, and a user identification unit 17. The measurement unit 20 is a sensor that measures the vibration characteristics of a measurement target and a part for mounting the sensor on a living body of a target user, and includes a signal generation unit 21 and a signal reception unit 22. The audio interface unit 30 is an interface between the wearing user identification device 10 and the measurement unit 20, and includes a signal control unit 31 and a signal amplification unit 32. Note that the signal generation unit 11 and the signal control unit 31, and the signal control unit 31 and the signal generation unit 21 may be connected by wire or wirelessly as long as they have a function of transmitting and receiving signals, and the connection form does not matter.

[0012] The signal generating unit 11 of the wearing user identification device 10 generates an acoustic signal based on arbitrarily set parameters. As an example, the acoustic signal is an ultrasonic wave that sweeps from 20 kHz to 40 kHz. However, the acoustic signal may be set in any manner, such as whether or not to sweep, or whether other frequency bands are used.

[0013] The signal control unit 31 of the audio interface unit 30 generates a drive signal based on the acoustic signal generated by the signal generation unit 11 based on the preset parameters, and applies vibration to the target through the signal generation unit 21 of the measurement unit 20. Note that, if a configuration is adopted in which feature amounts contained in teacher data on an individual to be identified are generated in advance and registered in a registration database (not shown), the vibration at this time may contain other frequencies as long as it contains the frequency contained in the vibration used at that time.

[0014] The signal generating unit 21 and the signal receiving unit 22 of the measuring unit 20 are composed of two piezoelectric elements that do not contact each other. The piezoelectric elements can be realized by, for example, piezo elements. One of the piezoelectric elements becomes the signal generating unit 21 that generates vibrations having the same frequency characteristics as the driving signal generated in the signal control unit 31 of the audio interface unit 30. The other piezoelectric element becomes the signal receiving unit 22 that receives the vibrations. The signal receiving unit 22 acquires the vibrations that have propagated inside and on the surface of the object on which the signal receiving unit 22 is installed. Here, when the vibrations provided by the signal generating unit 21 are propagated to the signal receiving unit 22, the living body of the user that is the measurement object functions as a propagation path, and the frequency characteristics of the acquired vibrations change depending on the material and boundary conditions of this propagation path. The signal receiving unit 22 transmits the received vibration signal (hereinafter referred to as a reaction signal) to the signal amplifying unit 32. The signal generating unit 21 and the signal receiving unit 22 may be of any shape and material as long as they are in contact with the living body of the object and have a mechanism that can propagate vibrations.

[0015] Fig. 2 is a plan view showing the configuration of the measurement unit 20, and Fig. 3 is a schematic diagram showing a state in which the measurement unit 20 is worn by a user who is the measurement subject. Note that, although the measurement unit 20 is configured as a band-type sensor here, other implementation methods such as a biological adhesive tape may be used as long as the signal generating unit 21 and the signal receiving unit 22 can be fixedly attached to the user's skin while maintaining a certain distance therebetween.

[0016] The measuring unit 20 is attached to the fixed part 23 so that the two piezoelectric elements serving as the signal generating unit 21 and the signal receiving unit 22 are kept at a certain distance from each other without contacting each other. The fixed part 23 also functions as a reinforcing member that reinforces the strength of the signal generating unit 21 and the signal receiving unit 22 so that the signal generating unit 21 and the signal receiving unit 22 can be used continuously. A band 24 and a square ring 25 are attached to the fixed part 23 at opposing positions across the signal generating unit 21 and the signal receiving unit 22. A hook-and-loop fastener 26 is provided on the back side of the band 24. The measuring unit 20 having such a configuration is worn by the user by adjusting the length of the band 24, wrapping it around the user's wrist, and fastening it with the hook-and-loop fastener 26. At this time, the wearing location does not matter as long as it is consistent for the individual when performing personal authentication.

[0017] The signal amplifier 32 of the audio interface unit 30 amplifies the reaction signal acquired by the signal receiver 22 of the measurement unit 20, and transmits it to the wearing user identification device 10. The signal is amplified by the signal amplifier 32 because vibrations that pass through the measurement object are attenuated, and therefore it is necessary to amplify the signal to a level at which processing is possible.

[0018] In the wearer identification device 10 , the response signal transmitted from the signal amplifier 32 of the audio interface unit 30 is stored in the signal storage unit 12 .

[0019] The feature generating unit 13 extracts the reaction signal stored in the signal storage unit 12 at regular time intervals, and performs, for example, FFT (Fast Fourier Transform) on the extracted reaction signal to generate a spectrogram, which is a feature representing the acoustic frequency characteristics of the living body to be measured. FIG. 4 is a diagram showing an example of this spectrogram. When performing a learning process operation related to learning a classification model, the feature generating unit 13 generates teacher data that is a pair of the generated spectrogram and a user ID corresponding to the user to be measured, and outputs the teacher data to the model learning unit 14. The teacher data may be generated by extracting it from the above-mentioned registration database created in advance. When performing a discrimination process operation related to personal discrimination of a user, the feature generating unit 13 outputs the generated spectrogram to the discrimination execution determination unit 16.

[0020] The model learning unit 14 generates and learns a classification model with the teacher data obtained from the feature generating unit 13 as input and the user ID as output. The model learning unit 14 registers the classification model itself or the parameters of the model obtained by this learning process in the model storage unit 15, which is a model database. As long as it is possible to learn to obtain an optimal output by performing parameter tuning or the like on the teacher data, the type of the classification model and the library used for learning it does not matter. For example, using a generally known machine learning library, an algorithm for generating a classification model such as SVM (Support Vector Machine) or a neural network may learn to obtain an optimal output by performing parameter tuning or the like on the teacher data.

[0021] The identification execution determination unit 16 determines whether or not to execute user identification processing by the user identification unit 17. The identification execution determination unit 16 obtains the stability of the spectrogram within a set fixed time from the spectrogram acquired from the feature amount generation unit 13. For example, to obtain the stability of the spectrogram for 2 seconds, the average dB value for each frequency for 2 seconds is obtained (e.g., average value of 20 kHz=0 [dB], ..., average value of 40 kHz=-5 [dB]). Then, the identification execution determination unit 16 uses these average values ​​to calculate the standard deviation at each frequency of the spectrogram as the stability. The identification execution determination unit 16 determines whether or not the state of the part where the measurement unit 20 is attached to the user is stable depending on whether or not the standard deviation at each frequency is higher than a set arbitrary threshold value. In other words, if the standard deviation at each frequency of the spectrogram is not higher than the threshold value, the identification execution determination unit 16 considers that the state of the part where the measurement unit 20 is attached to the user is stable, and causes the user identification unit 17 to perform processing. Specifically, the identification execution determination unit 16 outputs the spectrogram acquired from the feature amount generation unit 13 to the user identification unit 17. Conversely, when the standard deviation at each frequency of the spectrogram is higher than the threshold value, the identification execution determination unit 16 determines that the user wearing the measurement unit 20 is moving and has low stability, and prevents processing in the user identification unit 17. In other words, the identification execution determination unit 16 does not output a spectrogram to the user identification unit 17. Note that this processing may be limited to only values ​​having a certain value or more in each frequency characteristic (for example, -50 [dB] or more).

[0022] The user identification unit 17 inputs a spectrogram, which is a feature obtained from the identification execution determination unit 16, into a classification model registered in the model memory unit 15, and obtains a numerical value for identifying an individual user as an output of the classification model.

[0023] Here, when SVM is used as the algorithm of the model learning unit 14, a score indicating the similarity of each label (user ID) held by the classification model for the input spectrogram is output as a reference value from the classification model. For example, when the similarity is normalized and expressed between "0" and "1", the score can be output as "1-(similarity)".

[0024] Furthermore, when Random Forest is used as the algorithm for the model learning unit 14, data is randomly extracted from the teacher data, multiple decision trees are generated, and the number of judgment results for each label of each decision tree for the input data is output. The higher the number of judgment results, the better, so the classification model outputs "(number of judgments) - (number of judgment results)" as a reference value.

[0025] In addition, the classification algorithm may be other classification algorithms such as DNN (Deep Neural Network). In that case, the reference value may be obtained by subtracting the normalized similarity from "1" or by converting the similarity by the inverse of the similarity.

[0026] The user identification unit 17 uses the obtained reference value list to determine the user ID with the smallest reference value (similar). In the determination process, a threshold value may be set for the similarity, and a determination may be made only when the reference value is smaller than the threshold value.

[0027] FIG. 5 is a diagram showing an example of a hardware configuration of the wearing user identification device 10. As shown in FIG. 5, the wearing user identification device 10 is configured by a computer such as a microcomputer or a personal computer, and has a hardware processor 101 such as a CPU (Central Processing Unit). The CPU can execute multiple information processes simultaneously by using a multi-core and multi-threaded one. The processor 101 may also include multiple CPUs. In the wearing user identification device 10, a program memory 102, a data memory 103, a communication interface 104, and an input / output interface 105 are connected to the processor 101 via a bus 106. In FIG. 5, the "interface" is abbreviated as "IF".

[0028] The communication interface 104 may include, for example, one or more wired or wireless communication modules. In the example shown in FIG. 5, two communication modules 1041 and 1042 are shown. The communication module 1041 is a communication module that uses a short-range wireless technology such as Bluetooth (registered trademark), and transmits and receives signals to and from the signal control unit 31 and the signal amplifier unit 32 of the audio interface unit 30. The communication module 1041 can also transmit and receive signals to and from the signal control unit 31 and the signal amplifier unit 32 of a remote audio interface unit 30 via a network NW. The network is composed of an IP network including the Internet and an access network for accessing this IP network. As the access network, for example, a public wired network, a mobile phone network, a wired LAN (Local Area Network), a wireless LAN, a CATV (Cable Television), or the like is used. Therefore, the wearing user identification device 10 can acquire reaction signals from a plurality of measurement units 20 via a plurality of audio interface units 30, and can also identify a plurality of wearing users who are wearing the respective measurement units 20.

[0029] Further, the input unit 107 and the display unit 108 are connected to the input / output interface 105. The input unit 107 and the display unit 108 may be a so-called tablet-type input / display device in which an input detection sheet using an electrostatic method or a pressure method is arranged on a display screen of a display device using, for example, liquid crystal or organic EL (Electro Luminescence). The input unit 107 and the display unit 108 may be configured as independent devices. The input / output interface 105 inputs operation information inputted in the input unit 107 to the processor 101, and causes the display unit 108 to display display information generated by the processor 101.

[0030] The input unit 107 and the display unit 108 do not have to be connected to the input / output interface 105. The input unit 107 and the display unit 108 can transmit and receive information to and from the processor 101 by being provided with a communication unit for connecting to the communication interface 104 directly or via the network NW.

[0031] The input / output interface 105 may also have a function for reading / writing a recording medium such as a semiconductor memory such as a flash memory, or may have a function for connecting to a reader / writer having a function for reading / writing such a recording medium. In this way, a recording medium that is detachable from the wearing user identification device 10 can be used as a model database that holds the classification model. The input / output interface 105 may further have a function for connecting to other devices.

[0032] The program memory 102 is a non-transient tangible computer-readable storage medium that is a combination of a non-volatile memory that can be written and read at any time, such as a hard disk drive (HDD) or a solid state drive (SSD), and a non-volatile memory such as a read only memory (ROM). The program memory 102 stores programs required for the processor 101 to execute various control processes according to an embodiment. That is, each of the processing function units, namely, the signal generating unit 11, the feature generating unit 13, the model learning unit 14, the identification execution determining unit 16, and the user identifying unit 17, can be realized by having the processor 101 read and execute the programs stored in the program memory 102. Note that some or all of these processing function units may be realized in various other formats, including integrated circuits such as an application specific integrated circuit (ASIC) or a field-programmable gate array (FPGA).

[0033] The data memory 103 is a tangible computer-readable storage medium, and is, for example, a combination of the above-mentioned non-volatile memory and a volatile memory such as a RAM (Random Access Memory). The data memory 103 is used to store various data acquired and created in the process of performing various processes. That is, an area for storing various data as appropriate in the process of performing various processes is secured in the data memory 103. As such areas, the data memory 103 may be provided with, for example, a signal storage unit 1031, a model storage unit 1032, a classification result storage unit 1033, and a temporary storage unit 1034.

[0034] The signal storage unit 1031 stores the response signal transmitted from the signal amplifier 32 of the audio interface unit 30. That is, the signal storage unit 12 can be configured in this signal storage unit 1031.

[0035] The model storage unit 1032 stores the classification model learned by the model learning unit 14. That is, the model storage unit 15 can be configured in this model storage unit 1032.

[0036] The identification result storage unit 1033 stores output information obtained when the processor 101 executes the operation of the user identification unit 17 .

[0037] The temporary memory unit 1034 stores data such as spectrograms, teacher data, classification models, reference values, etc. that are acquired or generated when the processor 101 performs operations as the feature generation unit 13, the model learning unit 14, the identification execution determination unit 16, and the user identification unit 17.

[0038] Next, the operation of the wearing user identification device 10 will be described. In this embodiment, before identifying a user, the wearing user identification device 10 first generates a classification model associated with a user ID using a sensor capable of measuring the state of each user to be identified, and stores the generated classification model in the model memory unit 15 as registration data.

[0039] For this purpose, first, the signal generating unit 21 and the signal receiving unit 22 of the measuring unit 20 are attached to the wrist of a user to be identified using a band 24. The form and type of vibration generated by the signal generating unit 21 are not important as long as the vibration has frequency characteristics similar to those of an acoustic signal. In this embodiment, an acoustic signal will be described as an example.

[0040] 6 is a flowchart showing an example of a learning process operation related to learning a classification model in the wearing user identification device 10. This flowchart shows a part of the wearing user identification device 10, specifically, a processing operation in the processor 101 of a computer that functions as the signal generating unit 11, the feature generating unit 13, and the model learning unit 14. After the measurement unit 20 is worn on the wrist of a user to be registered, when an instruction to start learning is given from the input unit 107 via the input / output interface 105, the processor 101 starts the operation shown in this flowchart. In addition, for a remote user to be registered, when a predetermined wearing completion notification transmitted via the network NW from an information processing device such as a smartphone operated by the user is received via the communication interface 104, the processor 101 starts the operation shown in this flowchart.

[0041] First, the processor 101 functions as the signal generating unit 11 and generates an acoustic signal (drive signal) based on arbitrarily set parameters (step S101). The drive signal is, for example, an ultrasonic wave that sweeps from 20 kHz to 40 kHz. However, the settings of the acoustic signal, such as whether or not to sweep, whether or not to use other frequency bands, etc., are not important. The generated drive signal is transmitted to the audio interface unit 30 by the communication interface 104. Alternatively, a signal generating module that generates a drive signal under the control of the processor 101 may be separately prepared, and the generated drive signal may be transmitted to the audio interface unit 30 by the communication interface 104.

[0042] The audio interface unit 30 transmits this drive signal to the signal generating unit 21 of the measurement unit 20. This drive signal causes vibration to be applied to the living body of the user to be registered through the signal generating unit 21. The signal receiving unit 22 of the measurement unit 20 acquires the vibration that is applied to the living body of the user to be registered by the signal generating unit 21 and propagates through the inside and surface of the living body. Here, when the vibration applied from the piezoelectric element of the signal generating unit 21 propagates to the piezoelectric element of the signal receiving unit 22, the living body of the user to be registered functions as a propagation path, and the frequency characteristics of the applied vibration change according to this propagation path. This frequency characteristic is different for each person. The signal receiving unit 22 detects the propagated vibration and transmits a reaction signal indicated by this detected vibration to the audio interface unit 30. The signal amplifying unit 32 of the audio interface unit 30 amplifies the reaction signal transmitted from the signal receiving unit 22 of the measurement unit 20 and transmits it to the wearing user identification device 10.

[0043] The reaction signal transmitted from the audio interface unit 30 is received by the communication interface 104. The processor 101 stores the received reaction signal in the signal storage unit 1031 of the data memory 103 (step S102).

[0044] Next, the processor 101 functions as the feature generator 13 and performs the following processing operations. First, the processor 101 extracts the reaction signals stored in the signal storage unit 1031 for each fixed time interval. The number of signal samples does not matter. The extracted reaction signals are stored in the temporary storage unit 1034 of the data memory 103. Then, the processor 101 generates a spectrogram, which is a feature representing the acoustic frequency characteristics of a living body, from the extracted reaction signals stored in the temporary storage unit 1034 (step S103). The generated spectrogram is stored in the temporary storage unit 1034 of the data memory 103.

[0045] Then, processor 101 assigns a user ID, which is a unique identifier, to the generated spectrogram, and generates teacher data that pairs the spectrogram with the user ID (step S104). The generated teacher data is stored in temporary storage unit 1034 of data memory 103. Furthermore, processor 101 may extract registered data created in advance from model storage unit 15 configured in model storage unit 1032 of data memory 103, and use the extracted data to generate teacher data.

[0046] Next, the processor 101 functions as the model learning unit 14 and performs the following processing operations. First, the processor 101 generates and learns a classification model that receives as input the spectrogram in the training data, and outputs the user ID in the training data as a label and a reference value that is the difference from the input (step S105).

[0047] Then, the processor 101 registers the classification models and classification models obtained by this learning process, or the model parameters, in the model storage unit 15 configured in the model storage unit 1032 of the data memory 103 (step S106).

[0048] When learning for one user to be registered is completed in this manner, the processor 101 stops generating the drive signal and ends transmission of the drive signal to the audio interface unit 30 by the communication interface 104 (step S107). Then, the learning process operation shown in this flowchart is terminated.

[0049] Thereafter, learning can be carried out in a similar manner for other users.

[0050] Next, the operation of the wearing user identification device 10 when identifying an individual user to be identified will be described. The wearing user identification device 10 inputs a spectrogram, which is a feature amount acquired from a user to be identified who is wearing the measurement unit 20, to a classification model registered in the model storage unit 15 or the like, and identifies the individual user. Specific processing of these will be described below.

[0051] FIG. 7 is a flowchart showing an example of an identification processing operation related to personal identification of a user in the wearing user identification device 10. This flowchart shows the processing operation in a part of the wearing user identification device 10, specifically, the processor 101 of a computer that functions as the signal generating unit 11, the feature generating unit 13, the identification execution determining unit 16, and the user identification unit 17. After the measurement unit 20 is worn on the wrist of the user to be identified, when an instruction to start personal identification is given from the input unit 107 via the input / output interface 105, the processor 101 starts the operation shown in this flowchart. In addition, for a remote user to be identified, when a predetermined identification start notification transmitted via the network NW from an information processing device such as a smartphone operated by the user is received via the communication interface 104, the processor 101 starts the operation shown in this flowchart.

[0052] First, the processor 101 functions as the signal generating unit 11 and generates a drive signal based on arbitrarily set parameters (step S201). The generated drive signal is transmitted to the audio interface unit 30 via the communication interface 104. Alternatively, a signal generating module that generates a drive signal under the control of the processor 101 may be separately prepared, and the generated drive signal may be transmitted to the audio interface unit 30 via the communication interface 104.

[0053] The audio interface unit 30 transmits this drive signal to the signal generating unit 21 of the measurement unit 20. This drive signal causes vibration to be applied to the living body of the user to be registered through the signal generating unit 21. The vibration at this time may contain other frequencies as long as it contains the frequency contained in the vibration used to generate the feature amount contained in the registration data related to the learned user (hereinafter referred to as the registered user) registered in the model storage unit 15. The signal receiving unit 22 of the measurement unit 20 acquires the vibration that is applied to the living body of the user to be identified by the signal generating unit 21 and propagates through the inside and surface of the living body. Here, when the vibration applied from the piezoelectric element of the signal generating unit 21 propagates to the piezoelectric element of the signal receiving unit 22, the living body of the user to be identified functions as a propagation path, and the frequency characteristics of the applied vibration change depending on the propagation path. The signal receiving unit 22 detects the propagated vibration and transmits a reaction signal indicated by the detected vibration to the audio interface unit 30. The signal amplifier 32 of the audio interface unit 30 amplifies the response signal transmitted from the signal receiver 22 of the measurement unit 20 and transmits it to the wearing user identification device 10 .

[0054] The reaction signal transmitted from the audio interface unit 30 is received by the communication interface 104. The processor 101 stores the received reaction signal in the signal storage unit 1031 of the data memory 103 (step S202).

[0055] Next, the processor 101 functions as the feature generator 13 and extracts the reaction signals stored in the signal storage unit 1031 at regular time intervals. The number of signal samples does not matter. The extracted reaction signals are stored in the temporary storage unit 1034 of the data memory 103. Then, the processor 101 performs, for example, FFT on the extracted reaction signals stored in the temporary storage unit 1034 to generate a spectrogram, which is a feature representing the acoustic frequency characteristics of a living body (step S203). The generated spectrogram is stored in the temporary storage unit 1034 of the data memory 103.

[0056] Next, the processor 101 functions as the discrimination execution determination unit 16 and performs the following processing operations. First, processor 101 calculates the stability of a spectrogram within a set fixed time from the spectrogram stored in temporary storage unit 1034 as follows. First, processor 101 determines whether spectrograms for a fixed time, for example, two seconds, have been generated (step S204). If it is determined that spectrograms for two seconds have not yet been generated (NO in step S204), processor 101 transitions to the process of step S201 and repeats the above-mentioned processing operations.

[0057] On the other hand, when it is determined that a spectrogram for two seconds has been generated (YES in step S204), the processor 101 calculates the stability (step S205). For example, the processor 101 calculates an average dB value for each frequency for two seconds from the spectrogram for two seconds stored in the temporary storage unit 1034, and uses this average value to calculate the standard deviation at each frequency of the spectrogram as the stability.

[0058] Then, processor 101 determines whether the state of the part where measuring unit 20 is attached to the user is stable based on whether the standard deviation at each frequency of these spectrograms is equal to or less than a preset arbitrary threshold (step S206).

[0059] If the standard deviation at each frequency of the spectrogram is higher than the threshold, the processor 101 determines that the user wearing the measuring unit 20 is moving and not stable (NO in step S206). In this case, the processor 101 deletes the oldest spectrogram within a certain period of time stored in the temporary storage unit 1034 (step S207). After that, the processor 101 transitions to the process of step S201 and repeats the above-mentioned processing operations.

[0060] On the other hand, when the standard deviation at each frequency of the spectrogram is equal to or less than the threshold, the processor 101 determines that the state of the part of the user where the measuring unit 20 is attached is stable (YES in step S206). In this case, the processor 101 functions as the user identifying unit 17 and performs the following processing operations. First, the processor 101 performs personal identification of the user (step S208). That is, the processor 101 inputs one of the spectrograms, which are the features generated in the above step S203 and stored in the temporary storage unit 1034, for example, the latest spectrogram, into the classification model registered in the model storage unit 15 configured in the model storage unit 1032 of the data memory 103, and acquires a list of reference values ​​of the classification model. The acquired list of reference values ​​is stored in the temporary storage unit 1034 of the data memory 103. Next, the processor 101 specifies the smallest reference value from the list of reference values ​​stored in the temporary storage unit 1034. The processor 101 determines in the model storage unit 15 that a registered user associated with the same feature as the specified reference value is a similar user. Then, the processor 101 stores the user ID of the determined registered user in the identification result storage unit 1033 of the data memory 103 as the personal identification result of the user to be identified. In this determination process, a threshold value may be set for the similarity, and a similar user may be determined only when the specified reference value is smaller than this threshold value.

[0061] Then, the processor 101 outputs the user ID, which is the personal identification result stored in the identification result storage unit 1033 (step S209). For example, the processor 101 displays the user ID on the display unit 108 via the input / output interface 105. The processor 101 can also provide the user ID to an application or the like that requires personal authentication of the user.

[0062] When personal identification of one user to be identified is thus completed, the processor 101 stops generating the drive signal and terminates transmission of the drive signal by the communication interface 104 to the audio interface unit 30 (step S210). Then, the identification processing operation shown in this flowchart is terminated.

[0063] The wearing user identification device 10 according to the embodiment described above receives a response signal, which is a measurement signal corresponding to the vibration characteristics of the user's body part measured by the measurement unit 20, from the measurement unit 20, which is a sensor attached to the body part of the user to be identified, by the feature generation unit 13, generates a feature representing the vibration characteristics from the response signal, and determines whether the state of the attachment part of the measurement unit 20 to the user is stable based on the magnitude of the fluctuation of the feature generated by the feature generation unit 13 by the identification execution determination unit 16 as a determination unit. When it is determined that the state of the attachment part is stable, the user identification unit 17 performs personal identification of the user based on the feature generated by the feature generation unit 13. In this way, by performing personal identification of the user only in a stable state where there is no significant change in the vibration characteristic value, it is possible to reduce the influence of disturbances such as moving the attachment part or giving an external stimulus to the attachment part, and to reduce erroneous determination in personal identification. That is, the wearing user identification device 10 according to the embodiment can reduce erroneous determination when active acoustic sensing is used for personal identification of the user.

[0064] The reaction signal received from the measuring unit 20 is a vibration signal that detects vibrations propagated inside the user's body, which is the part where the measuring unit 20 is attached, and the feature generated by the feature generating unit 13 can be a spectrogram that represents the frequency characteristics of the vibration signal, which is generated by performing, for example, FFT (Fast Fourier Transform) on the reaction signal. In this way, a spectrogram can be generated as the feature.

[0065] Then, the discrimination execution determination unit 16 calculates the standard deviation at each frequency of the spectrogram using the average dB value for each frequency for a certain period of time from the spectrogram for a certain period of time, and if the standard deviation at each frequency is equal to or less than a set threshold, it determines that the condition of the attachment part is stable. In this way, the stability can be easily calculated, and a stable state in which no significant change occurs in the vibration characteristic value can be determined based on the stability.

[0066] Moreover, the wearing user identification device 10 according to one embodiment further includes a model storage unit 15 which is a database in which feature amounts for each of a plurality of users to be registered are registered in advance, and the user identification unit 17 identifies, from among the plurality of users to be registered registered in the model storage unit 15, a user having a feature amount corresponding to the feature amount generated by the feature amount generating unit 13 from the reaction signal received from the measurement unit 20, as a user to be identified. In this way, by registering the feature amount for each of a plurality of users to be registered in the model storage unit 15, it is possible to easily identify the user to be identified based on the feature amount.

[0067] Here, the model storage unit 15 inputs a feature amount, and stores a model that outputs a value based on the difference between at least one feature amount of a user to be registered and the input feature amount in association with a user ID that is an identifier uniquely assigned to the user to be registered. This model is learned based on a spectrogram that is a feature amount generated by the feature amount generating unit 13 from a reaction signal received from the measurement unit 20 for each of a plurality of users to be registered. The user identification unit 17 inputs a spectrogram that is a feature amount generated for the user to be identified into the model stored in the model storage unit 15, and determines the user ID output in association with a value that indicates the highest relevance to the spectrogram of the user among the values ​​output from the model as the user ID of the user to be identified, thereby identifying the user to be identified. Thus, the spectrogram that is the feature amount of the registered user can be used to appropriately identify the user to be identified.

[0068] Moreover, the wearing user identification system 1 according to one embodiment includes the wearing user identification device 10 according to one embodiment, and a measurement unit 20 that generates a first vibration to be applied to a wearing part of the user's body by a piezoelectric element and acquires, as a measurement signal, a vibration signal corresponding to a second vibration that has propagated inside the body out of the first vibration applied to the user's body. Thus, by having each user to be identified wear the measurement unit 20, it becomes possible to individually identify each user.

[0069] [Other embodiments] In the embodiment described above, the user's personal identification based on the feature amount is performed only in a stable state where the vibration characteristic value does not change significantly. However, this is not limited to the case of the user's personal identification, and the model learning unit 14 during learning may also perform learning based on the feature amount only in a stable state. This makes it possible to use only stable learning data for the data used for learning.

[0070] In addition, the audio interface unit 30 is disposed between the wearing user identification device 10 and the measurement unit 20, but the audio interface unit 30 may be incorporated into either the wearing user identification device 10 or the measurement unit 20.

[0071] In the above embodiment, the processing function unit of the wearing user identification device 10 is described as being configured by one computer, but it may be configured by multiple computers by arbitrary division. For example, the model learning unit 14 and the model storage unit 15 may be configured in a computer or server device other than the computer that configures the wearing user identification device 10 and that can communicate via the network NW by the communication interface 104.

[0072] The method described in the above embodiment can be stored as a program (software means) that can be executed by a calculator (computer) on a recording medium such as a magnetic disk (floppy disk, hard disk, etc.), an optical disk (CD-ROM, DVD, MO, etc.), or a semiconductor memory (ROM, RAM, flash memory, etc.), and can also be distributed by transmitting it via a communication medium. The program stored on the medium side also includes a setting program that configures the software means (including not only execution programs but also tables and data structures) that the calculator executes within the computer. The computer that realizes this device reads the program recorded on the recording medium, and in some cases, configures the software means by the setting program, and executes the above-mentioned processing by controlling the operation of the software means. The recording medium referred to in this specification is not limited to a recording medium for distribution, but also includes a storage medium such as a magnetic disk or semiconductor memory provided inside the computer or in a device connected via a network.

[0073] In short, this invention is not limited to the above-mentioned embodiment, and various modifications can be made in the implementation stage without departing from the gist of the invention. Moreover, each embodiment may be implemented in combination as appropriate as possible, and in that case, the combined effects can be obtained. Furthermore, the above-mentioned embodiment includes inventions at various stages, and various inventions can be extracted by appropriate combinations of the disclosed constituent elements. [Explanation of symbols]

[0074] 1... Wearer identification system 10... Wearable user identification device 11...Signal generation section 12...Signal storage section 13...Feature generation unit 14…Model Learning Section 15…Model memory section 16...Identification execution judgment unit 17...User identification section 20…Measuring section 21...Signal generating section 22...Signal receiving section 23…Fixed part 24…Band 25…Square ring 26…Velcro 30...Audio interface section 31...Signal control section 32...Signal amplifier 101…Processor 102…Program memory 103...Data memory 1031...signal storage unit 1032…Model memory section 1033... Classification result storage unit 1034...Temporary storage 104...Communication interface 1041, 1042...Communication module 105... Input / Output Interface 106…Bus 107...Input section 108...Display section NW…Network

Claims

1. a feature generating unit that receives a measurement signal corresponding to a vibration characteristic of the user's body part measured by a sensor attached to the body part of the user to be identified, and generates a feature representing the vibration characteristic from the measurement signal; a determination unit that determines whether or not a state of the attachment site of the sensor on the user is stable based on a magnitude of variation in the feature amount generated by the feature amount generation unit; an identification unit that performs personal identification of the user based on the feature amount generated by the feature amount generation unit when the determination unit determines that the state of the attachment part is stable; Equipped with A wearing user identification device, wherein the measurement signal received from the sensor is a vibration signal that detects vibrations that are applied to the user's body, which is the wearing part of the sensor, and propagated inside the body.

2. The wearing user identification device as described in Claim 1, wherein the feature generated by the feature generation unit is a spectrogram representing the frequency characteristics of the vibration signal.

3. 3. The wearing user identification device according to claim 2, wherein the determination unit calculates a standard deviation at each frequency of the spectrogram using an average dB value for each frequency over a certain period of time from the spectrogram, and determines that the condition of the wearing part is stable if the standard deviation at each frequency is equal to or less than a set threshold value.

4. A database in which the feature amount for each of a plurality of users to be registered is registered in advance, the identification unit identifies, from among the plurality of users to be registered registered in the database, a user having a feature amount corresponding to the feature amount generated by the feature amount generation unit from the measurement signal received from the sensor, as the user to be identified; 4. A user identification device according to claim 1.

5. the database stores a model that inputs the feature amount, and outputs a value based on a difference between the feature amount of at least one of the users to be registered and the input feature amount in association with an identifier that is uniquely assigned to the user to be registered; the model is trained based on features generated by the feature generator from the measurement signal received from the sensor for each of the plurality of users to be registered; the identification unit inputs the feature amount generated for the user to be identified into the model, and identifies the user by determining, as the identifier of the user to be identified, an identifier output in association with a value that indicates the highest relevance to the feature amount of the user to be identified among values ​​output from the model.

5. The wearer identification device according to claim 4.

6. A user identification device according to any one of claims 1 to 5, the sensor generates a first vibration to be applied to the wearing part of the body of the user by a piezoelectric element, and acquires, as the measurement signal, a vibration signal corresponding to a second vibration that is part of the first vibration applied to the body of the user and that has propagated inside the body; A wearing user identification system comprising:

7. A method for identifying a user wearing a sensor in a user identification device that includes a processor and identifies a user who is an identification target and wears a sensor on a body part, comprising: generating, by the processor, a feature quantity representing the vibration characteristics from a measurement signal corresponding to the vibration characteristics of the body part of the user measured by the sensor; determining whether or not a state of the attachment site of the sensor on the user is stable based on a magnitude of the variation of the generated feature amount by the processor; When the processor determines that the condition of the attachment site is stable, the processor performs personal identification of the user based on the generated feature amount; The measurement signal received from the sensor is a vibration signal obtained by detecting vibrations that are applied to the user's body, which is the mounting portion of the sensor, and propagated through the inside of the body. A method for identifying the wearer.

8. A wearer-identifying program that causes a processor to function as each of the units of the wearer-identifying device according to any one of claims 1 to 5.

Images