JP2021002357A - Personal authentication device, personal authentication method, and personal authentication program - Google Patents

Abstract

To provide a personal authentication device that can prevent impersonation after authentication by quickly performing highly accurate authentication even during insertion with wearing of a single hands-free headset.SOLUTION: A personal authentication device 300 comprises: an acoustic signal observation unit 303 for, when detecting contact with a transmission path, which is a part of a head of an authentication subject, observing a first observation acoustic signal observed by transmitting an audible range to a part of the head and a second observation acoustic signal observed by transmitting an inaudible range with a higher frequency than that of the audible range to a part of the head; a storage unit for storing a first registration acoustic signal with an audible range of the authentication subject and a second registration acoustic signal with an inaudible range with a higher frequency than that of the first registration acoustic signal and not including the audible range; and an identification unit 304 for performing first authentication for authenticating the authentication subject by comparing the first observation acoustic signal and the first registration acoustic signal, and performing second authentication after succeeding in the first authentication by comparing the second observation acoustic signal and the second registration acoustic signal.SELECTED DRAWING: Figure 17

Description

The present invention relates to a personal authentication device or the like that authenticates an individual.

Biometrics-based authentication, which is based on individual differences in living organisms, has a lower risk of leakage or theft than passwords created by users. For this reason, there are an increasing number of cases where personal authentication based on individual differences in living organisms is introduced for the purpose of identifying individuals and confirming their rights and for the purpose of security protection. In general, personal authentication technology based on individual differences in living organisms is known to use fingerprints (Fingerprint), veins (Vein), faces (Face), irises (Iris), voice (Voice), etc. as living organisms. ing. Among these, personal authentication using voice can be performed using an inexpensive device that is generally used, such as a telephone or a microphone, instead of a special device.

Among biometrics, in recent years, personal authentication (otoacoustic authentication) using changes in the acoustic characteristics of the ear (ear canal) has attracted attention. In the case of a method of acquiring biometric information other than otoacoustic authentication and performing personal authentication, it is necessary to have the user perform some action in order to perform authentication. For example, in the case of personal authentication using fingerprints or veins, a user action such as placing a finger on a dedicated scanner is required. Further, in the case of personal authentication using a face or an iris, a user's action such as turning the face to the camera is required. Further, in the case of personal authentication using voice or bone conduction sound, user actions such as uttering a password are required. A user who is forced to perform such an action bears a psychological and physical burden even if it is a short time. Further, it is not preferable to continue such an operation for a long time because it hinders the user's next operation. However, in the case of otoacoustic authentication, it is sufficient to put an acoustic radio wave transmitter / receiver such as a handset or earphone on the ear or insert it into the ear, so even if it takes a long time, the user's psychological and physical The burden is less than other biometric authentication methods. The technique of otoacoustic emission authentication is disclosed in Patent Document 1 and Patent Document 2.

International Publication No. 2014/061578 Japanese Unexamined Patent Publication No. 2005-032056

However, in the otoacoustic emission as disclosed in Patent Document 1 and Patent Document 2, the vibration and fine movement of the earphone / microphone device generated when the earphone / microphone device for transmitting / receiving the authentication acoustic signal is inserted into and removed from the ear is used. There is a problem that the acoustic characteristics of the ear canal are easily changed.

The problem of acoustic feature change will be described. For example, after inserting the earphone / microphone device, if the earphone / microphone device is reattached due to a feeling of strangeness in the ear, it is difficult for a general user to insert the earphone / microphone device at the exact same position as the previous time. It will be inserted at a position slightly deviated from the previous insertion position. Due to this deviation of the insertion position, the acoustic signal that hits the ear canal and is bounced changes, and authentication based on the acoustic signal may not be accurate.

Further, Patent Document 2 has a problem that another authentication means (for example, input of a user ID) is required in order to obtain highly accurate authentication. In this case, if the user ID is forgotten, authentication is impossible. In addition, when permission is required to quickly access highly confidential information at an emergency site (for example, at an accident site or in an ambulance), inputting a user ID via a keyboard or the like can be performed by rain, gusts, etc. at the site. It may not be possible due to the vibration of an earthquake, or it may take time to input the user ID, and the information may not be accessed quickly. Further, a mouse and a keyboard for inputting a user ID are required separately.

Therefore, in view of the above-mentioned problems, an object of the present invention is to provide a personal authentication device or the like that can quickly execute highly accurate authentication and prevent spoofing after authentication.

In view of the above problems, the personal authentication device according to one aspect of the present invention includes a detection means for detecting contact with a propagation path which is a part of the head of the person to be authenticated, and the detection means for detecting the contact. Then, the first observed acoustic signal observed by sending the audible sound range to a part of the head and the inaudible sound range having a frequency higher than the audible sound range were sent to a part of the head and observed. An acoustic signal observing means for observing the second observed acoustic signal, a first registered acoustic signal in the audible sound range of the person to be certified, and an inaudible sound range having a frequency higher than that of the first registered acoustic signal and not including the audible sound range. After the storage means for storing the second registered audio signal is compared with the first observed audio signal and the first registered audio signal and the first authentication to be authenticated by the authentication target person is executed, and the first authentication is successful. The identification means for executing the second authentication for comparing the second observed acoustic signal with the second registered audio signal is provided.

The personal authentication method according to one aspect of the present invention detects contact with a propagation path that is a part of the head of the person to be authenticated, and when the contact is detected, sends an audible sound range to a part of the head. The first observed acoustic signal observed in the above and the second observed acoustic signal observed by sending the inaudible sound range having a frequency higher than the audible sound range to a part of the head are observed, and the audible person to be certified. The first registered audio signal of the sound range and the second registered audio signal which has a frequency higher than that of the first registered audio signal and is an inaudible sound range which does not include the audible sound range are stored, and the first observed sound signal and the first first registered sound signal are stored. The first authentication that compares the registered acoustic signals and authenticates the person to be authenticated is executed, and after the first authentication is successful, the second authentication that compares the second observed acoustic signal with the second registered audio signal is executed. To do.

The personal authentication program according to one aspect of the present invention detects contact with a propagation path that is a part of the head of the person to be authenticated, and when the contact is detected, sends an audible sound range to a part of the head. The first observed acoustic signal observed in the above and the second observed acoustic signal observed by sending the inaudible sound range having a frequency higher than the audible sound range to a part of the head are observed, and the audible person to be certified. The first registered audio signal of the sound range and the second registered audio signal which has a frequency higher than that of the first registered audio signal and is an inaudible sound range which does not include the audible sound range are stored, and the first observed sound signal and the first first registered sound signal are stored. The first authentication that compares the registered acoustic signals and authenticates the person to be authenticated is executed, and after the first authentication is successful, the second authentication that compares the second observed acoustic signal with the second registered audio signal is executed. Make the computer do what you want. The personal authentication program may be stored on a non-temporary computer-readable storage medium.

According to the present invention, it is possible to provide a personal authentication device or the like that can quickly execute highly accurate authentication and prevent spoofing after authentication.

It is a block diagram which shows the structural example of the personal authentication apparatus which concerns on 1st Embodiment of this invention. It is a block diagram which shows the specific configuration example of the personal authentication apparatus of 1st Embodiment. It is a graph which shows ECTF (Ear Canal Transfer Function). It is a flowchart which shows the operation of the user registration in the personal authentication apparatus which concerns on 1st Embodiment. It is a chart figure which shows the transmission timing of the transmitted acoustic signal. It is a graph which shows schematic the waveform of the transmitted acoustic signal. It is a graph which shows schematic the waveform of the observed acoustic signal. It is a graph which shows an example of the waveform of the signal after the synchronous addition processing. It is a graph which shows an example of the impulse response as an acoustic characteristic. It is a graph which expressed the acoustic characteristic by ECTF (Ear Canal Transfer Function). It is a graph which shows the spectral envelope. It is a graph which shows the spectral envelope of a plurality of people. It is a flowchart which shows the operation of the user identification in the personal authentication apparatus which concerns on 1st Embodiment. It is a chart figure which shows the transmission timing of the transmitted acoustic signal. It is a block diagram which shows the structural example of the personal authentication apparatus which concerns on 2nd Embodiment of this invention. It is a flowchart which shows the operation of the user identification in the personal authentication apparatus which concerns on 2nd Embodiment. It is a block diagram which shows the structural example of the personal authentication apparatus which concerns on 3rd Embodiment of this invention. It is a block diagram which shows the structural example of the information processing apparatus applicable in each embodiment.

Generally, the frequency range (audible range) that can be heard by humans is about 20 Hz for low sounds and about 20 kHz for high sounds. Such sounds that can be heard by humans are "audible sounds", sounds that are so high that they cannot be heard by humans are "ultrasonic waves" (hereinafter, also referred to as "infrasounds"), and sounds that are so low that they cannot be heard by humans are "audible". It is called "infrasound". The frequency of sound that humans can hear well is about 2kHz-4kHz.

In this embodiment, audible sound and inaudible sound are used for otoacoustic emission authentication. In general, the user feels uncomfortable that he / she has to listen to the audible sound for authentication for a long period of time or that he / she has to listen to the audible sound at regular time intervals, which he / she does not want. For this reason, there is an opinion that it is preferable to use an inaudible sound for otoacoustic emission because the burden on the user is light (see Patent Document 2).

However, the inventor of the present invention considers the acoustic feature of the audible range to be the inaudible range even while the acoustic feature of the ear canal is changed by inserting and removing the earphone / microphone device transmitting and receiving the acoustic signal into the ear. I found that it was relatively stable (details will be described later). Each embodiment of the present invention has been devised based on this idea.

Hereinafter, embodiments of the present invention will be described with reference to the drawings. In the description of the drawings below, the same or similar parts are designated by the same or similar reference numerals. However, the drawings schematically show the configuration in the embodiment of the present invention. Further, the embodiment of the present invention described below is an example, and can be appropriately changed as long as the essence thereof is the same.

<First Embodiment>
(Personal authentication device)
As shown in FIG. 1, the personal authentication device 100 according to the first embodiment of the present invention has a first acoustic signal generation unit 101, a second acoustic signal generation unit 102, an acoustic signal observation unit 103, and an acoustic characteristic calculation unit 104. , A feature extraction unit 105, an identification unit 106, a first feature amount storage unit 107, and a second feature amount storage unit 108.

The first acoustic signal generation unit 101 generates an acoustic signal (first acoustic signal) in the audible range. The earphone 14 inputs the first acoustic signal and sends it to a propagation path which is a part of the head of the user (authentication target person). Here, a part of the head to which an acoustic signal is sent is a region in the head that is formed to have an outward opening, for example, the ear canal of an auditory organ. In addition, it may be the nasal cavity. It may be a cavity in the head that does not have an opening to the outside. In addition, it may be at least a part of the area where the device that produces the sound effect can be attached or approached.

The second acoustic signal generation unit 102 generates an acoustic signal (second acoustic signal) in an inaudible range (ultrasonic range) having a frequency higher than that in the audible range. The earphone 14 inputs the second acoustic signal and sends it to the propagation path which is a part of the user's head described above. The inaudible range is preferably an inaudible range having a higher frequency than the audible range. This is because the second authentication does not cause stress on the user due to the generation of audible sound.

The acoustic signal observation unit 103 inputs the output signal of the microphone 15 provided on the earphone 14, and after the acoustic signal in the audible sound range transmitted from the first acoustic signal generation unit 101 propagates through a part of the user's head. Observe the first acoustic signal of. Further, the acoustic signal observation unit 103 also observes the second acoustic signal after the acoustic signal in the inaudible sound range transmitted from the second acoustic signal generation unit 102 propagates in a part of the user's head. It is preferable that the second acoustic signal generation unit 102 generates the second acoustic signal after the acoustic signal observation unit 103 observes the first acoustic signal.

More specifically, the part of the head that is used as the propagation path of the first and second acoustic signals may be at least a part of the skull, the brain, the sensory organs, and the cavities in between. .. In addition, the acoustic signal after propagation includes a signal reflected in the propagation path.

Further, in FIG. 1, the first acoustic signal generation unit 101 and the second acoustic signal generation unit 102 are separate units, but they may be the same unit or device.

The acoustic characteristic calculation unit 104 calculates the acoustic characteristic (first acoustic characteristic) of the first acoustic signal propagating in a part of the user's head based on the first acoustic signal observed by the acoustic signal observation unit 103. Further, the acoustic characteristic calculation unit 104 calculates the acoustic characteristic (second acoustic characteristic) of the second acoustic signal propagating in a part of the user's head based on the second acoustic signal observed by the acoustic signal observation unit 103. ..

The feature extraction unit 105 calculates a feature amount (first feature amount) relating to the user who has propagated the acoustic signal from the calculated first acoustic characteristic. Further, the feature extraction unit 105 calculates a feature amount (second feature amount) relating to the user who has propagated the acoustic signal from the calculated second acoustic characteristic.

The first feature amount storage unit 107 stores at least one of the first acoustic characteristic and the first feature amount associated with a predetermined user. The number of users may be singular or plural. The second feature amount storage unit 108 stores at least one of the second acoustic characteristic and the second feature amount associated with a predetermined user. That is, two feature quantities (first and second acoustic characteristics, or first and second feature quantities) associated with one user are two storage units (first feature quantity storage unit 107 and second feature quantity). It is stored in the quantity storage unit 108). Hereinafter, a user whose acoustic characteristics and feature amounts are stored in the first feature amount storage unit 107 and the second feature amount storage unit 108 may be referred to as a registered user. In FIG. 1, the first feature amount storage unit 107 and the second feature amount storage unit 108 are drawn so as to be stored in different storage media, but even if they are stored in one storage medium. Good. Further, it may be set to be connectable to the storage medium via a wireless or wired line. Further, although the user can be identified by using either the acoustic characteristic or the feature amount, an example of performing authentication mainly using the feature amount will be described below.

The identification unit 106 determines that the first authentication of the user (authentication target person) is successful when the observed first acoustic signal satisfies a predetermined condition, and when the second acoustic signal satisfies the predetermined condition. , Judge whether the user (authentication target person) who succeeded in the first authentication is the same person. Judge that the second authentication was successful. For example, the identification unit 106 compares the first feature amount extracted by the feature extraction unit 105 with the first feature amount of the registered user stored in the first feature amount storage unit 107 (first authentication). Based on, it is determined whether the user to be authenticated is a registered user. Further, the identification unit 106 compares the second feature amount extracted by the feature extraction unit 105 with the feature amount of the registered user stored in the second feature amount storage unit 108 (second authentication). Based on this, it is determined whether or not the registered user has been replaced (spoofed). Generally, the first feature amount has higher authentication accuracy, but the user tends to feel unpleasant because it is in the audible range. Therefore, the identification unit 106 performs the first authentication of the user for the first time immediately after the user inserts the earphone / microphone device (contact with the propagation path) based on the first feature amount. This is because, immediately after the earphone / microphone device is inserted, the audible sound for several seconds is accepted without stress by the user as a signal that the authentication process is properly started. Furthermore, sound waves in the audible range are robust against vibration and fine movement during insertion, and the authentication accuracy is high, so it is possible to determine with high reliability whether or not the user to be authenticated is a registered user. is there.

The identification unit 106 performs the second authentication based on the second feature amount. This is because when using the earphone / microphone device, the user may be doing something or listening to another sound, and audible sound authentication can disrupt the user's concentration or stress the user during authentication. This is because it is possible to feel it. Further, if the user succeeds in the initial authentication process, even if the authentication is performed by an inaudible sound (ultrasonic wave), the authentication can be performed with sufficient accuracy to discriminate spoofing or the like.

In addition, the identification unit 106 permits a predetermined access when the result of the first authentication is successful, or when the result of the first authentication and the second authentication immediately after the first authentication is successful, and the first authentication is performed. If the result of the second authentication after the second authentication is successful, the predetermined access permission may be continued. This makes it possible to strengthen the security of the initial permission.

(Configuration example of personal authentication device)
FIG. 2 is a diagram showing a specific configuration example of the personal authentication device according to the present embodiment. The personal authentication device shown in FIG. 2 includes a personal computer (PC) 11, a sound processor 12, a microphone amplifier 13, an earphone 14, and a microphone 15 (the earphone 14 and the microphone 15 are collectively referred to as an earphone microphone device 17). , And a sensor 18. Reference numeral 16 represents a user to be recognized.

The earphone 14 outputs an acoustic signal transmitted by the first acoustic signal generation unit 101 and the second acoustic signal generation unit 102 described above. Further, the microphone 15 receives the acoustic signal and outputs it to the acoustic signal observation unit 103. As shown in FIG. 2, it is desirable that the microphone 15 and the earphone 14 are integrated so that the relative positional relationship does not change. However, this does not apply if the relative positional relationship between the two does not change significantly. The earphone / microphone device 17 is held in a contact state, a pressure contact state, or a substantially fitted state with a piece of an acoustic signal propagation path. In the example shown in FIG. 2, an earphone / microphone device 17 in which these are integrated is given as a configuration example of the earphone 14, the microphone 15, and the sensor 18. The earphone / microphone device 17 is inserted into the entrance of the ear canal. The sensor 18 detects contact with a propagation path that is a part of the head of the person to be authenticated. Generally, the sensor 18 is a sensor capable of determining whether or not the user wears such an earphone / microphone device 17 on the body, for example, a warmth sensor, an optical sensor, or a touch sensor.

In addition to the above example, the earphone / microphone device 17 may be realized by a type of headphone that covers the auricle with a microphone (auricle type microphone integrated earphone / microphone device). Further, the earphone / microphone device 17 may be realized by a telephone provided with a microphone in the receiver portion. In such a case, the acoustic signal transmitted from the earphone 14 located at the entrance of the left ear canal or the like is observed by the microphone 15 located at the entrance of the ear canal of the right ear, or vice versa. May be divided.

The acoustic characteristic calculation unit 104, the feature extraction unit 105, and the identification unit 106 are each realized by a CPU (Central Processing Unit) and a memory that operate according to a personal authentication program provided in the PC 11. A specific configuration example of the PC 11 will be described later (see FIG. 18). Further, the first feature amount storage unit 107 and the second feature amount storage unit 108 are realized by a storage medium such as a hard disk included in the PC 11.

Next, the difference in the robustness of authentication between the audible range and the inaudible range will be described. FIG. 3 is a graph showing ECTF (Ear Canal Transfer Function) when the same person repeatedly inserts and removes the earphone / microphone device 17 into the same ear 10 times. The vertical axis of the graph shows the amplitude (dB: decibel) of the sound wave, and the horizontal axis shows the frequency (Hz: hertz). With reference to FIG. 3, the ECTF graphs overlap at a high rate even if the insertion and removal are repeated up to about 0 kHz <frequency <20 kHz. That is, it has high robustness (reproducibility) even with respect to slight movements of the insertion position of the earphone / microphone device 17. As the graph shows, the reproducibility of ECTF gradually decreases after 20 kHz. Since the treble range of the audible sound is about 20 kHz as described above, in ECTF, the audible sound is not affected by the insertion / removal operation as compared with the inaudible sound, and it can be said that the authentication rate is high. Generally, the authentication success rate of the audible range is about 97%, and the authentication success rate of the inaudible range is about 90%. Therefore, by combining these authentication results of the audible range and the inaudible range, the user is less burdened than the authentication result of only the audible range, and the authentication result with higher accuracy than the authentication result of only the inaudible range can be obtained. Can be done.

(Operation of personal authentication device)
Next, the operation of the personal authentication device 100 according to the first embodiment will be described. The operation of this embodiment is divided into "1. user registration" and "2. user identification".
Hereinafter, the operation of "1. User registration" will be described with reference to the flowchart shown in FIG. It should be noted that the identification unit 106 determines whether the user is registered or identified. The identification unit 106 may, for example, store a feature amount that matches the first feature amount storage unit 107, turn on the power of the earphone / microphone device 17, or insert the user's earphone / microphone device 17. When the acoustic signal observation unit 103 detects a change in the acoustic signal, it is determined that the user registration process is performed. The judgment method is not limited to this.

First, in step S101, the first acoustic signal generation unit 101 generates the first acoustic signal and outputs it from the earphone 14. Next, the second acoustic signal generation unit 102 generates the second acoustic signal and sends it out from the earphone 14. FIG. 5 is a chart schematically showing the transmission timing of the first acoustic signal and the second acoustic signal. The vertical axis of the graph indicates frequency (kHz), and the horizontal axis indicates time (t). First, the first acoustic signal generation unit 101 transmits the first acoustic signal in the audible range a plurality of times at predetermined intervals (n) toward a part of the head of the user to be authenticated. This is because it is preferable to send a plurality of times from the viewpoint of accuracy. In FIG. 5, the first acoustic signal generation unit 101 transmits the first acoustic signal from the minimum frequency as an audible sound (exemplified as 0 kHz in FIG. 5 for simplification of explanation) to 20 kHz five times every 0.2 seconds. To do.

After that, the second acoustic signal generation unit 102, like the first acoustic signal generation unit 101 described above, has an inaudible sound range at predetermined intervals (n) toward a part of the head of the user to be authenticated. The second acoustic signal is transmitted multiple times. For example, in FIG. 5, the second acoustic signal generation unit 102 transmits the second acoustic signal from 20 kHz to 40 kHz five times every 0.2 seconds.

It is preferable that the first acoustic characteristic and the first feature amount are extracted in the audible range, and the second acoustic characteristic and the second feature amount are extracted in the inaudible range, but this is not always the case. There is no limit. This is because the range of frequencies from which highly accurate acoustic signals and features can be extracted may differ depending on the shape of the user's head and the like to be authenticated and the types of acoustic signals and features used. .. Therefore, the designer may set a band within the audible range and the inaudible range in which the acoustic characteristics and the like can be calculated with high accuracy.

FIG. 6 is a graph diagram schematically showing the waveforms of the transmitted acoustic signals (first acoustic signal and second acoustic signal) in the time domain representation. The vertical axis represents the amplitude of the signal value of the measured acoustic signal at time t, and the horizontal axis represents time t. The acoustic signals generated by the first acoustic signal generation unit 101 and the second acoustic signal generation unit 102 may be transmitted from, for example, the entrance of the ear canal toward the ear canal. At this time, as these acoustic signals, an M-sequence signal (maximal length sequence) widely used for the purpose of measuring an impulse response, a TSP (Time Stretched Pulse) signal, or the like may be used.

The transmission interval and the set value of the range (frequency) are examples, and can be changed as appropriate according to the equipment to be used, the application, and the like.

In step S102, the acoustic signal observation unit 103 observes the first acoustic signal after propagating a part of the head of the user to be authenticated. Similarly, the acoustic signal observation unit 103 observes the second acoustic signal after propagating a part of the head of the user to be authenticated.

FIG. 7 is a graph diagram schematically showing the waveforms of the observed acoustic signals (first acoustic signal and second acoustic signal) in the time domain representation. In the graph, the vertical axis represents the signal value (amplitude) of the measured acoustic signal at time t, and the horizontal axis represents time (t). The observed first acoustic signal is affected by noises generated in the body such as heartbeat sounds, motion sounds (joint sounds and muscle sounds), breath sounds, and vocal sounds during propagation in the external auditory canal, and the graph in FIG. Changes from to as shown in the graph of FIG.

Further, the acoustic signal observation unit 103 performs a synchronous addition process to remove noise in the observed acoustic signals (first acoustic signal and second acoustic signal). FIG. 8 shows the waveform of the signal after the synchronous addition processing in 0.2 seconds described above.

In step S103, the acoustic characteristic calculation unit 104 includes the transmitted first acoustic signal (see FIG. 6) and the observed first acoustic signal (including the one after the synchronous addition process) (see FIGS. 7 and 8). The acoustic characteristics (first acoustic characteristics) when the first acoustic signal propagates through a part of the user's head are calculated from those changes. Similarly, the acoustic characteristic calculation unit 104 picks up the transmitted second acoustic signal (see FIG. 6) and the observed second acoustic signal (including the one after the synchronous addition process) (see FIGS. 7 and 8). By comparison, the acoustic characteristics (second acoustic characteristics) when the second acoustic signal propagates through a part of the user's head are calculated from those changes.

The acoustic characteristics (first acoustic characteristic and second acoustic characteristic) are how an acoustic signal propagates and is observed by propagating an object (a part of the user's head in the present invention) when it is transmitted. It is a characteristic that represents. The observed acoustic signal changes depending on the transmitted signal, but in principle, if the signal observed when a sudden and very short acoustic signal called an impulse signal is transmitted is known, an arbitrary transmitted signal can be obtained. Since the observation signal for the impulse signal can be calculated, the signal observed for the impulse signal, that is, the impulse response is a typical acoustic characteristic. A transfer function (TF) obtained by performing a Fourier analysis on an impulse response is also a typical acoustic characteristic. The acoustic properties preferably include information on how the acoustic signal is reflected and / or attenuated in vivo.

Hereinafter, the description will be made on the assumption that the propagation path is the ear canal. FIG. 9 is a graph of ECIR (Ear Canal Impulse Response) showing the acoustic characteristics calculated by the acoustic characteristic calculation unit 104 in the time domain representation. In the graph shown in FIG. 9, the horizontal axis represents the time t, and the vertical axis represents the value h (t) of the external auditory canal impulse response of the observed acoustic signal at the time t. The following equation (t) is used between the signal value x (t) of the transmitted acoustic signal, the signal value y (t) of the acoustic signal observed after propagation, and the value h (t) of the impulse response of the external auditory canal. There is a relationship shown in 1). Τ is a shift coefficient.

The acoustic characteristic may be ECTF (Ear Canal Transfer Function). FIG. 10 is a graph of ECTF showing the acoustic characteristics shown in FIG. 9 in the frequency domain. In the graph, the vertical axis represents the amplitude (dB) and the horizontal axis represents the frequency (kHz). In the graph, it can be seen that a plurality of maximums (mountains) and minimums (valleys) are clearly observed in the audible range as compared with the inaudible range.

In step S104, the feature extraction unit 105 calculates the feature amount (first feature amount) from the first acoustic characteristic calculated by the acoustic characteristic calculation unit 104.

Here, the feature amount (the first feature amount and the second feature amount described later) is a numerical value useful for the required processing (individual authentication in the present invention) extracted by applying some processing to the acoustic signal. Also called acoustic features. For example, MFCC (Mel-Frequency Cepstrum Coefficients), which is widely used in speech recognition, is a typical acoustic signal obtained by subjecting an acoustic signal to processing such as Fourier analysis, logarithmization, mel conversion, and discrete cosine transform. Acoustic features. MFCC expresses vocal tract characteristics while considering human speech perception. The ECIR processed by FFT (Fast Fourier Transform) is ECTF, and the low-order component of cepstrum obtained by subjecting ECTF to the above processing is MFCC.

The impulse response or transfer function calculated as the acoustic characteristic may be used as it is for the first feature amount. That is, the feature extraction unit 105 may use the value of each time of the impulse response as the first acoustic characteristic and the value of each frequency of the transfer function as the feature quantity.

FIG. 11 is a graph showing a certain ECTF spectral envelope. In the graph, the horizontal axis indicates frequency (Hz) and the vertical axis indicates sound pressure (both are logarithmic scales). The user features shown in the graph of FIG. 11 form the first maximum around 1.5kHz-1.8kHz, the first minimum around 3kHz-4kHz, and the maximum again around 5kHz-8kHz. The average distance from the outer mouth of the ear canal to the eardrum at the maximum around 5kHz-8kHz is about 2-3 cm. This shape or a numerical value related thereto is the feature amount of the user. FIG. 12 shows another graph showing the spectral envelope. FIG. 12 (a) shows a graph of MFCC at the first and 30th measurements corresponding to male A, FIG. 12 (b) shows male B, and FIG. 12 (c) shows female C. In the graph, the circled parts are the characteristics unique to each person. Since it is a bass (audible range), it is observed from the graph that a high concordance rate is obtained regardless of whether it is the first measurement or the thirtieth measurement.

The feature extraction unit 105 calculates the feature amount (second feature amount) from the second acoustic characteristic calculated by the acoustic characteristic calculation unit 104 by the same method.

In step S105, the identification unit 106 stores the first feature amount acquired from the feature extraction unit 105 in the first feature amount storage unit 107. Further, the identification unit 106 stores the second feature amount acquired from the feature extraction unit 105 in the second feature amount storage unit 108. Specifically, for example, assuming that the feature amount is ECTF, the identification unit 106 stores the 0kHz-20kHz graph of the graph shown in FIG. 10 or a numerical value representing these as the first feature amount in the first feature amount storage unit 107, and stores the graph at 20kHz. A graph of −40 kHz or a numerical value representing these is stored in the second feature quantity storage unit 108 as the second feature quantity.

With the above, the user registration process is completed.

Next, the operation of "2. User identification" will be described with reference to the flowchart shown in FIG. Steps S201 to S207 are the first authentication process using the first feature amount in the audible range, and steps S208 to S215 are the second authentication process using the second feature amount in the inaudible range. ..

First, when the person to be authenticated inserts the earphone / microphone device 17 into the propagation path (for example, the entrance of the ear canal) which is a part of his / her head, the sensor 18 detects the contact with the propagation path. In step S201, the first acoustic signal generation unit 101 generates a signal in the audible range (first acoustic signal) and sends it to the propagation path via the earphone 14 at predetermined intervals.

In step S202, the acoustic signal observation unit 103 receives and observes the acoustic signal after the first acoustic signal propagates through a part of the head of the user to be authenticated via the microphone 15.

FIG. 14 is a chart schematically showing the transmission timing of the first acoustic signal and the second acoustic signal in steps S201 and S202. The vertical axis of the graph shows the frequency (kHz) of the sound wave, and the horizontal axis shows the time (t). First, the first acoustic signal generation unit 101 generates the first acoustic signal in the audible range a plurality of times (FIG. 14) toward a part of the head of the user to be authenticated at predetermined intervals (n). Then 5 times). After that, the second acoustic signal generation unit 102, like the first acoustic signal generation unit 101 described above, is unable to generate at predetermined intervals (n) toward a part of the head of the user to be authenticated. A second acoustic signal in the listening range is generated and transmitted to the propagation path a plurality of times (five times in FIG. 14) via the earphone 14.

In step S203, the acoustic characteristic calculation unit 104 compares the first acoustic signal with the first acoustic signal observed by the acoustic signal observation unit 103, and from the changes thereof, the first acoustic signal is the head of the user. The acoustic characteristic (first acoustic characteristic) of the acoustic signal when propagating a part of the sound signal is calculated.

In step S204, the feature extraction unit 105 calculates the feature amount (first feature amount) from the first acoustic characteristic calculated by the acoustic characteristic calculation unit 104.

In step S205, the identification unit 106 matches whether the first feature amount acquired from the feature extraction unit 105 matches any of the feature amounts stored in the first feature amount storage unit 107, that is, the user who is the authentication target is registered. Determine if you are a completed user.

If the identification unit 106 determines in step S206 that the user is a registered user, the process proceeds to step S207. If they do not match, the process proceeds to step S215.

In step S207, the identification unit 106 permits the registered user to access the predetermined device, assuming that the first authentication process is successful. Access to a predetermined device is, for example, login to a certain system, use of an internal application or database such as a system after login. The applications and databases that can be used may be set differently for each registered user.

In step S208, when a predetermined time, for example, 10 seconds elapses, the process proceeds to step S209, and a second authentication process for continuing the access permission is executed.

First, in step S209, the second acoustic signal generation unit 102 directs the generated inaudible sound range signal (second acoustic signal) toward a part of the head of the user to be authenticated via the earphone 14. It is sent out at predetermined intervals.

In step S210, the acoustic signal observation unit 103 receives and observes the acoustic signal after the second acoustic signal propagates through a part of the head of the user to be authenticated via the microphone 15.

In step S211 the acoustic characteristic calculation unit 104 compares the second acoustic signal with the second acoustic signal observed by the acoustic signal observation unit 103, and from the change thereof, the second acoustic signal is the head of the user. The acoustic characteristic (second acoustic characteristic) of the acoustic signal when propagating a part of the sound signal is calculated.

In step S212, the feature extraction unit 105 calculates the feature amount (second feature amount) from the second acoustic characteristic calculated by the acoustic characteristic calculation unit 104.

In step S213, in the identification unit 106, whether the second feature amount acquired from the feature extraction unit 105 matches any of the feature amounts stored in the second feature amount storage unit 108, that is, the user who is the authentication target is registered. Determine if you are a completed user.

If the identification unit 106 determines in step S214 that the user is a registered user, the process proceeds to step S216. If they do not match, the process proceeds to step S215.

In step S215, the identification unit 106 disallows the access to the predetermined device of the authentication target user and disconnects the access, assuming that the first authentication process or the second process has failed.

In step S216, the identification unit 106 permits the registered user to continue accessing the predetermined device, assuming that the second authentication process is successful. After that, the process proceeds to step S208, and the second authentication process is executed again after the elapse of a predetermined time. Specifically, after the lapse of the waiting time (m) after the transmission of the first second acoustic signal shown in FIG. 14, the second acoustic signal is transmitted again.

The second authentication process is repeated unless the sensor 18 detects the removal of the earphone / microphone device 17 by the user.

On the contrary, when the sensor 18 detects the insertion of the earphone / microphone device 17, the process of "1. User registration" is restarted. A sensor other than the sensor 18 may be used to detect the insertion of the earphone / microphone device 17. For example, the user may be requested to turn on the power of the earphone / microphone device each time the earphone / microphone device 17 is inserted / removed. Alternatively, the insertion of the user's earphone / microphone device 17 may be detected by a change in the acoustic signal observed by the acoustic signal observation unit 103.

In addition, the identification unit 106 may use one-to-one authentication or one-to-N authentication. Here, N is an integer of 1 or more.

When one-to-one authentication is used, the identification unit 106 sets a pair of the feature amount of the user to be authenticated (the feature amount obtained by the feature extraction unit 105) and the feature amount of the registered user. Compare with 1. At this time, the administrator of the personal authentication device may give the identification unit 106 a designation as to which registered user is to be compared with the user ID or the like in advance. When one-to-one authentication is used, the identification unit 106 calculates, for example, the distance between the feature amount of the user to be authenticated and the feature amount of the designated registered user, and the distance is greater than the threshold value. If they are small, it may be determined that they are the same person. On the other hand, the user identification unit 106 may determine that they are different human beings when the calculated distance is larger than the threshold value.

When 1-to-N authentication is used, the identification unit 106 compares the user to be authenticated with N registered users. The identification unit 106 calculates the distance between the feature amount of each of the N registered users with respect to the feature amount of the user to be authenticated, sets an appropriate threshold value, and sets the distance most within the threshold value. It is determined that the closest registered user is the user to be authenticated. Further, the identification unit 106 can also be used in combination with one-to-one authentication and one-to-N authentication. In that case, the user identification unit 106 may perform 1-to-N authentication to extract the registered user having the shortest distance, and then perform 1-to-1 authentication with the extracted registered user as a comparison target. Further, the scale of the calculated distance may be, but is not limited to, the Euclidean distance, the cosine distance, and the like.

Further, in the above description, an example in which the first feature amount storage unit 107 and the second feature amount storage unit 108 store the feature amounts obtained from a plurality of people in advance has been described, but a statistical model is stored instead of the feature amount. You may. The statistical model may be, for example, an average value and a variance value obtained by acquiring the feature quantity a plurality of times for each user, or a relational expression calculated by using them. Further, the statistical model may be a GMM (Gaussian Mixture Model), an SVM (Support Vector Machine), a model using a neural network, or the like.

(Modified example of the first embodiment)
In step S207 of FIG. 13, when the first authentication process is successful, the identification unit 106 permits the registered user to access the predetermined device. However, the identification unit 106 may allow the registered user to access the predetermined device when the first authentication process and the first second authentication process are successful. Specifically, if YES in step S214, the identification unit 106 may allow the registered user to access the predetermined device (step S207). By changing the design in this way, it is possible to make the access permission stricter and, by extension, enhance the security of authentication.

(Effect of the first embodiment)
As described above, according to the first embodiment of the present invention, a personal authentication device or the like capable of quickly performing highly accurate authentication even at the time of insertion and preventing spoofing after authentication by attaching the earphone / microphone device alone. Can be provided. The reason for this is that the identification unit 106 executes the first authentication of the authentication target person when the observed first acoustic signal satisfies a predetermined condition, and after the first authentication, the second acoustic signal is predetermined. This is because, when the condition is satisfied, the authentication target person who succeeds in the first authentication executes the second authentication for determining whether the person is the same person. In the present embodiment, the sensor unit 301 detects the insertion, and using this as a trigger, the propagation result of the first acoustic signal including the audible sound range is obtained, and then the second sound range including the inaudible sound range having a frequency higher than the audible sound range is included. The propagation result of the acoustic signal is acquired, and based on these, the identification unit 106 determines whether the user is a registered user (first authentication), and misplaces the user determined to be the registered user in the first authentication. This is because it is determined (second authentication) whether or not (spoofing) has occurred.

<Second embodiment>
In the authentication device 100 according to the first embodiment, the second authentication process is performed at predetermined time intervals. However, when it comes to the second authentication process for a long time, it requires power to continue to output ultrasonic waves at short time intervals, and it may be a burden depending on the constitution of the user who continues the authentication for a long time. is there. Therefore, in the second embodiment, the number of times of the second authentication process is gradually reduced for the user who has succeeded in the second authentication process more than a certain number of times, that is, the user who has a track record of not performing the spoofing act. The personal authentication device 200 to be used will be described.

(Personal authentication device)
As shown in FIG. 15, the personal authentication device 200 according to the second embodiment of the present invention has a first acoustic signal generation unit 101, a second acoustic signal generation unit 102, an acoustic signal observation unit 103, and an acoustic characteristic calculation unit 104. , A feature extraction unit 105, an identification unit 106, a first feature amount storage unit 107 and a second feature amount storage unit 108, a detection unit 109, a counter unit 110, and an adjustment unit 111.

The detection unit 109 detects the insertion / removal of the user's earphone / microphone device 17 (see FIG. 2) into the ear canal. The detection unit 109 is, for example, a warmth sensor, an optical sensor, or a touch sensor. The detection unit 109 may detect insertion / removal based on a change in the acoustic signal observed by the acoustic signal observation unit 103. The judgment method is not limited to this.

When the detection unit 109 detects the insertion of the earphone / microphone device 17 (contact with the propagation path), the detection unit 109 outputs the detection result to the first acoustic signal generation unit 101. This is to notify the timing of transmitting the first acoustic signal, which is the start of the first authentication. On the contrary, when the detection unit 109 detects the removal of the earphone / microphone device 17 (non-contact with the propagation path), the detection unit 109 returns the counter value of the counter unit 110 to the initial value (for example, 0), and the first acoustic signal generation unit 101 and The detection result is output to the second acoustic signal generation unit 102. This is to stop the transmission of the acoustic signal and stop the first and second authentication.

The counter unit 110 counts the number of times the user to be authenticated is identified as having succeeded in the first authentication, the number of times the second acoustic signal is transmitted, and the like. The counter unit 110 may count the number of times the first acoustic signal is transmitted. The counter unit 110 may count the number of times of the first authentication process and the second authentication process.

As an example, when counting the authentication processes, the count may be counted individually such as the number of times of the first authentication process "1 time" and the number of times of the second authentication process "45 times", or the first time is the first time. It may be counted as the number of authentications, and the nth second authentication process may be calculated as the (n-1) th time.

The adjusting unit 111 adjusts the waiting time interval of the second authentication process according to the number of successful times of the second authentication process. For example, if the number of successful authentications is p or more, the waiting time is delayed by (initial value + q) seconds, and the waiting time is lengthened according to the magnitude of the absolute value of the number of successful authentications. As a specific example, when the number of successful authentications is 10 or more, the waiting time is set to (initial value +5) seconds, and when the number of successful authentications is 20 or more, the waiting time is set to (initial value +10) seconds. The adjusting unit 111 controls the second acoustic signal generation unit 102 so as to transmit the second acoustic signal at the adjusted waiting time interval.

Other devices are the same as in the first embodiment.

(Operation of personal authentication device)
Next, the operation of the personal authentication device 200 according to the second embodiment will be described. The operation of the personal authentication device 200 is divided into "1. user registration" and "2. user identification". The operation of "1. User registration" is the same as that of the first embodiment (FIG. 4). Hereinafter, the operation of "2. User identification" will be described with reference to the flowchart shown in FIG.

First, when the person to be authenticated inserts the earphone / microphone device 17 into the propagation path (for example, the entrance of the ear canal) which is a part of his / her head, the detection unit 109 constituting the sensor detects the contact with the propagation path. Hereinafter, the operations of steps S301 to S316 are the same as the operations of the first embodiment (see steps S201 to S216 in the flowchart of FIG. 13).

In step S317, the counter unit 110 counts, for example, the number of successful first authentication processes and the second authentication process in the identification unit 106. As the count, the number of times of the first authentication process and the number of times of the second authentication process may be individually counted, or these may be added up and counted. The counter unit 110 may count the number of times the acoustic signal is transmitted from the first acoustic signal generation unit 101 and the second acoustic signal generation unit 102. When the detection unit 109 notifies that the earphone / microphone device 17 has been removed, the counter unit 110 returns the counter value to the initial value (for example, 0).

In step S318, the adjusting unit 111 determines whether the number of successful authentications counted by the counter unit 110 has reached a predetermined number or more. When the number exceeds the predetermined number, the process proceeds to step S319. If it is not more than a predetermined number of times, the process proceeds to step S308.

In step S319, the coordinating unit 111 determines the waiting time interval for the second authentication process according to the number of successful authentications. The adjusting unit 111 may include a mathematical formula or a decision table for determining the waiting time interval. The adjusting unit 111 controls the second acoustic signal generation unit 102 so as to transmit the second acoustic signal at the determined waiting time interval. As a result, the second acoustic signal generation unit 102 transmits the second acoustic signal generated at the determined waiting time interval.

(Modified example of the second embodiment)
In the second embodiment, when the removal of the earphone / microphone device 17 is detected, the counter value of the counter unit 110 is returned to the initial value, and the process of "1. User registration" is restarted (that is, that is). The first acoustic signal is transmitted again). However, in addition to this, the first acoustic signal generation unit 101 may transmit the first acoustic signal as long as it does not burden the user. For example, when the number of successes of the second authentication process of the counter unit 110 exceeds a predetermined number A, or when the number of successes becomes a multiple of this A, the first acoustic signal generation unit 101 is the first. The acoustic signal may be transmitted.

(Effect of the second embodiment)
As described above, according to the second embodiment of the present invention, the same effect as that of the first embodiment, that is, by mounting the earphone / microphone device alone, quickly performs highly accurate authentication even at the time of insertion and authenticates. It is possible to provide a personal authentication device or the like that can prevent subsequent spoofing. Further, the waiting time interval of the authentication process is adjusted according to the number of successful authentications, and the adjusting unit 111 determines the transmission interval of the second acoustic signal according to the number of successful authentications counted by the counter unit 110. As a result, the number of times of the second authentication process can be gradually reduced for a user who has a track record of not performing spoofing. As a result, the power for ultrasonic output can be reduced, and the possibility of imposing a burden on the authentication point in the user's body can be reduced.

<Third embodiment>
As shown in FIG. 17, the personal authentication device 300 according to the third embodiment of the present invention includes a sensor unit 301, an acoustic signal generation unit 302, an acoustic signal observation unit 303, and an identification unit 304.

The sensor unit 301 detects contact with a propagation path that is a part of the head of the person to be authenticated.

When the sensor unit 301 detects contact, the acoustic signal generation unit 302 generates a first acoustic signal including an audible sound range, and further, a second inaudible sound range having a frequency higher than the audible sound range and not including the audible sound range. Generates an acoustic signal and sends it toward the propagation path.

The acoustic signal observation unit 303 observes the acoustic signal after the first acoustic signal propagates through a part of the head, and observes the acoustic signal after the second acoustic signal propagates through a part of the head.

The identification unit 304 executes the first authentication of the person to be authenticated when the observed first acoustic signal satisfies a predetermined condition, and after the first authentication, the second acoustic signal satisfies the predetermined condition. In addition, the second authentication for determining whether the authentication target person who succeeded in the first authentication is the same person is executed.

According to the third embodiment of the present invention, it is possible to provide a personal authentication device or the like that can quickly execute highly accurate authentication even at the time of insertion and prevent spoofing after authentication by attaching the earphone / microphone device alone. .. The reason for this is that the identification unit 304 executes the first authentication of the person to be authenticated when the observed first acoustic signal satisfies a predetermined condition, and after the first authentication, the second acoustic signal is predetermined. This is because, when the condition is satisfied, the authentication target person who succeeds in the first authentication executes the second authentication for determining whether the person is the same person. In the present embodiment, the sensor unit 301 detects the insertion, and using this as a trigger, the propagation result of the first acoustic signal including the audible sound range is obtained, and then the second sound range including the inaudible sound range having a frequency higher than the audible sound range is included. The propagation result of the acoustic signal is acquired, and based on these, the identification unit 106 determines whether the user is a registered user (first authentication), and misplaces the user determined to be the registered user in the first authentication. This is because it is determined (second authentication) whether or not (spoofing) has occurred.

In each of the above-described embodiments of the present invention, the personal authentication devices 100, 200, and 300 may be equipped with a personal authentication program inside the earphone / microphone device 17 and authenticated by a single device. As shown in FIG. 2, the earphone / microphone device 17 and the PC 11 may be connected, and the overall operation may be controlled on the PC 11 side. That is, the application program for personal authentication on the PC11 side may be used under the following configuration.

(Information processing device)
In each of the above-described embodiments of the present invention, a part or all of each component in the personal authentication device shown in FIGS. 1, 2, etc. is an optional combination of the information processing device 500 (PC11) and the program as shown in FIG. It is also possible to realize by using the combination of. The information processing device 500 includes the following configuration as an example.

-CPU (Central Processing Unit) 501
-ROM (Read Only Memory) 502
-RAM (Random Access Memory) 503
-Program 504a loaded into RAM 503
Storage device 505 that stores program 504b and other data.
Drive device 507 that reads and writes the recording medium 506.
-Communication interface 508 that connects to the communication network 509
-I / O interface 510 for inputting / outputting data
-Bus 511 connecting each component
Each component of the personal authentication device according to each embodiment of the present application is realized by the CPU 501 acquiring and executing the program 504b that realizes these functions. The program 504b that realizes the functions of each component of the personal authentication device (for example, the acoustic signal generation unit, the acoustic signal observation unit, and the identification unit) is stored in, for example, in the storage device 505 or the RAM 503 in advance, and is stored as needed. Read by CPU 501. The program 504b may be supplied to the CPU 501 via the communication network 509, or may be stored in the recording medium 506 in advance, and the drive device 507 may read the program and supply the program to the CPU 501.

The input / output interface 510 is connected to the sensor 18 and the sound processor 12 shown in FIG. The storage device 505 includes a feature amount storage unit (first feature amount storage unit and second feature amount storage unit) 504a shown in FIGS. 1 and 15.

There are various modifications in the method of realizing each device. For example, the personal authentication device may be realized by any combination of an information processing device and a program that are separate for each component. Further, a plurality of components included in the personal authentication device may be realized by any combination of one information processing device 500 and a program.

In addition, some or all of the components of the personal authentication device are realized by other general-purpose or dedicated circuits, processors, etc., or a combination thereof. These may be composed of a single chip or may be composed of a plurality of chips connected via a bus.

A part or all of each component of the personal authentication device may be realized by a combination of the above-mentioned circuit or the like and a program.

When a part or all of each component of the personal authentication device is realized by a plurality of information processing devices and circuits, the plurality of information processing devices and circuits may be centrally arranged or distributed. You may. For example, the information processing device, the circuit, and the like may be realized as a form in which each of the client and server system, the cloud computing system, and the like is connected via a communication network.

Although the present invention has been described above with reference to the present embodiment, the present invention is not limited to the above embodiment. Various changes that can be understood by those skilled in the art can be made within the scope of the present invention in terms of the structure and details of the present invention.

The present invention is applicable, for example, to sites that require urgency and access to confidential information, such as emergency personnel, police officers, the Self-Defense Forces, and other systems that require quick and highly accurate authentication. Furthermore, it can also be applied to a personal authentication device or the like that authenticates an individual using an audio device. It is also applicable to personalization systems, content rights management systems, communication control systems, etc. that use such personal authentication devices.

11 PC
12 Sound processor 13 Microphone amplifier 14 Earphone 15 Microphone 16 User 17 Earphone microphone device 18 Sensor 101 1st acoustic signal generator 102 2nd acoustic signal generator 103 Acoustic signal observation unit 104 Acoustic characteristic calculation unit 105 Feature extraction unit 106 Identification unit 107 1st feature quantity storage unit 108 2nd feature quantity storage unit 109 Detection unit 110 Counter unit 111 Adjustment unit 100 Personal authentication device 200 Personal authentication device 300 Personal authentication device 301 Sensor unit 302 Acoustic signal generation unit 303 Acoustic signal observation unit 304 Identification unit 500 Information processing device 501 CPU
503 RAM
504b Program 505 Storage device 506 Recording medium 507 Drive device 508 Communication interface 509 Communication network 510 Input / output interface 511 Bus

Claims (10)

A detection means that detects contact with the propagation path that is part of the head of the person to be authenticated,
When the detection means detects the contact, the first observed acoustic signal observed by sending the audible sound range to a part of the head and the inaudible sound range having a frequency higher than the audible sound range are one of the heads. An acoustic signal observation means for observing the second observation acoustic signal sent to the unit and observed,
A storage means for storing the first registered acoustic signal in the audible range of the person to be authenticated and the second registered acoustic signal which is an inaudible range having a frequency higher than that of the first registered acoustic signal and not including the audible range.
The first observed acoustic signal is compared with the first registered acoustic signal, the first authentication to be authenticated by the person to be authenticated is executed, and after the first authentication is successful, the second observed acoustic signal and the second registration are performed. An identification means that performs a second authentication that compares acoustic signals,
A personal authentication device equipped with. The personal authentication device according to claim 1, wherein the identification means performs the second authentication when the first authentication succeeds, and does not execute the second authentication when the first authentication fails. Further provided with an acoustic characteristic calculation means for calculating the first acoustic characteristic from the observed first observed acoustic signal and calculating the second acoustic characteristic from the observed second observed acoustic signal.
The identification means executes the first authentication based on information including at least one of the first acoustic characteristic and the first feature amount extracted from the first acoustic characteristic, and executes the first authentication, the second acoustic characteristic and the second acoustic characteristic. The personal authentication device according to claim 1 or 2, wherein the second authentication is executed based on information including at least one of the second feature amounts extracted from the characteristics. The invention according to any one of claims 1 to 3, further comprising a second acoustic signal generating means for transmitting an acoustic signal in the inaudible range after the acoustic signal observing means observes the first observed acoustic signal. Personal authentication device. The identification means permits a predetermined access when the result of the first authentication is successful, or when the result of the first authentication and the second authentication immediately after the first authentication is successful. The personal authentication device according to any one of claims 1 to 4, which continues to permit predetermined access when the result of the second authentication after the second authentication is successful. A storage means for storing at least one of the first and second acoustic characteristics and the first and second feature quantities associated with one or more persons in advance is provided.
The identification means compares the calculated first and second acoustic characteristics with the first and second acoustic characteristics stored in the storage means, or extracts the first and second characteristics. Any one of claims 1 to 5 that compares the amount with the first and second feature quantities stored in the storage means to identify whether the person is an authentication target person who has succeeded in the first authentication. The personal authentication device described in. Further provided with a counter means for counting at least one of the number of times the authentication target person is identified as the authentication target person who succeeded in the first authentication and the number of times the acoustic signal in the inaudible range is transmitted.
The personal authentication device according to any one of claims 1 to 6, wherein when the detection means detects that the contact has become non-contact, the counter value of the counter means is returned to an initial value. When the number of times to be counted by the counter means is acquired and it is determined that the number of times to be acquired exceeds a predetermined number, after the acoustic signal observing means observes the first observed acoustic signal, the inaudible sound range The personal authentication device according to claim 7, further comprising adjusting means for adjusting the time interval until the acoustic signal is transmitted. Detects contact with the propagation path that is part of the head of the person to be authenticated,
When the contact is detected, the first observed acoustic signal observed by sending the audible sound range to a part of the head and the inaudible sound range having a frequency higher than the audible sound range are sent to the part of the head. Observe the second observation acoustic signal observed in
The first registered audio signal in the audible range of the person to be certified and the second registered audio signal in the inaudible range having a frequency higher than that of the first registered acoustic signal and not including the audible range are stored.
The first observed acoustic signal is compared with the first registered acoustic signal, the first authentication to be authenticated by the person to be authenticated is executed, and after the first authentication is successful, the second observed acoustic signal and the second registration are performed. A personal authentication method that performs a second authentication that compares acoustic signals. Detects contact with the propagation path that is part of the head of the person to be authenticated,
When the contact is detected, the first observed acoustic signal observed by sending the audible sound range to a part of the head and the inaudible sound range having a frequency higher than the audible sound range are sent to the part of the head. Observe the second observation acoustic signal observed in
The first registered audio signal in the audible range of the person to be certified and the second registered audio signal in the inaudible range having a frequency higher than that of the first registered acoustic signal and not including the audible range are stored.
The first observed acoustic signal is compared with the first registered acoustic signal, the first authentication to be authenticated by the person to be authenticated is executed, and after the first authentication is successful, the second observed acoustic signal and the second registration are performed. A personal authentication program that enables a computer to perform a second authentication that compares acoustic signals.

Images