Authentication system

Abstract

To provide an authentication system configured to measure the change in capacitance in face authentication, to determine whether the face authentication is a fraud or not, thereby enabling accurate authentication processing.SOLUTION: An authentication system includes: a face authentication system 3 which executes authentication processing through face authentication of a person who is to enter a predetermined restricted area; a capacitive authentication system 4 which executes authentication processing using a parameter which is different from that of the face authentication system 3; and a releasing unit 5 which releases the restriction for entry when the face authentication in the face authentication system 3 is successful and when the authentication in the capacitive authentication system 4 is successful. The capacitive authentication system 4 includes: a measurement sensor 45 which measures capacitance that is changed as the person approaches the predetermined region; and a determination unit 44 which determines whether the face authentication in the face authentication system 3 is a fraud or not on the basis of a result of the measurement sensor 45.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to an authentication system for entering a restricted area. [Background technology]

[0002] For example, there are known technologies related to authentication when entering an area with entry restrictions, such as an auto-lock in an apartment building. For many years, authentication by key or PIN number has been common, but recently, technologies that release entry restrictions by facial recognition have been on the rise. However, in the case of facial recognition, it is possible to fake the face by using a photograph, etc., so to compensate for this, technologies such as those shown in Patent Documents 1 and 2 below have been developed.

[0003] The technology shown in Patent Document 1 is a status management system that includes an acquisition unit that acquires first information related to a person's face from at least one of an apparatus included in an intercom system installed in an apartment building or an apparatus connected to the intercom system, an identification unit that identifies the person based on the acquired first information, and an output unit that acquires second information related to the status of the identified person and outputs the acquired second information, and it is disclosed that the second information is the person's body surface temperature.

[0004] The technology disclosed in Patent Document 2 is an information processing system that includes an acquisition means for acquiring facial information from an image captured by a first camera and acquiring iris information from an image captured by a second camera, an authentication means for performing an authentication process using the facial information and iris information, and an execution means for executing a predetermined process to allow the subject to enter a facility to be used if the authentication process is successful, and allows the first subject and the second subject to enter the facility on the condition that the authentication process using both the facial information and iris information of the first subject is successful, and the authentication process using at least one of the facial information and iris information of a second subject different from the first subject is successful. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] JP 2023-48276 A [Patent Document 2] Patent No. 7243885 Summary of the Invention [Problem to be solved by the invention]

[0006] The technology shown in Patent Document 1 acquires a person's body surface temperature in addition to face recognition, but due to the effects of global warming in recent years, the outside temperature in summer often exceeds 35° C. Therefore, there is almost no difference between a person's body temperature and the outside temperature, which makes it difficult to acquire a person's body surface temperature accurately.

[0007] The technology disclosed in Patent Document 2 acquires iris information in addition to face authentication, but there is a problem in that it is difficult to acquire iris information, for example, when sunglasses are worn or depending on the light reflection state.

[0008] The present invention has been made to solve the above-mentioned problems, and aims to provide an authentication system that can determine whether or not facial authentication has been camouflaged by measuring changes in capacitance when performing facial authentication, thereby performing accurate authentication processing. [Means for solving the problem]

[0009] The authentication system of the present invention comprises a first authentication system that performs authentication processing by recognizing the face of a person when the person enters a specified area where entry is restricted, a second authentication system that performs authentication processing using parameters different from those of the first authentication system, and a release means for releasing the entry restriction when the face authentication of the first authentication system is successful and the authentication of the second authentication system is successful, and the second authentication system has a measurement sensor that measures a capacitance that changes when the person approaches the specified area, and a determination means for determining whether the face authentication in the first authentication system has been spoofed based on the result of the measurement sensor.

[0010] In this way, the authentication system of the present invention comprises a first authentication system that performs authentication processing by recognizing the face of a person when the person enters a specified area where entry is restricted, a second authentication system that performs authentication processing using parameters different from those of the first authentication system, and a release means for releasing the entry restriction when face authentication of the first authentication system is successful and authentication of the second authentication system is successful.The second authentication system measures the capacitance that changes when the person approaches the specified area and determines whether the face authentication in the first authentication system has not been spoofed based on the measurement result, so that by measuring the change in capacitance, it is possible to detect people regardless of the environment, and it is possible to determine whether face authentication is being performed through spoofing, thereby achieving the effect of achieving high security. [Brief description of the drawings]

[0011] [Figure 1] 1 is a system configuration diagram of an authentication system according to a first embodiment of the present invention. [Diagram 2] 1 is a block diagram showing a configuration of an electrostatic capacitance authentication system in an authentication system according to a first embodiment of the present invention. [Diagram 3] FIG. 1 is a diagram showing an example of a circuit diagram of a Colpitts oscillator circuit. [Figure 4]FIG. 11 is a diagram showing a state in which an image of a person to be face-authenticated is captured by a camera in an authentication system according to a second embodiment of the present invention. [Diagram 5] FIG. 11 is a system configuration diagram of an authentication system according to another embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0012] (First embodiment of the present invention) The authentication system according to the present embodiment will be described with reference to Figures 1 to 3. The authentication system according to the present embodiment is an authentication system that authenticates the entry of a person into an area where entry is restricted, and in particular, is an authentication system that enhances authentication accuracy and prevents face authentication from being performed by camouflage.

[0013] 1 is a system configuration diagram of an authentication system according to this embodiment. The authentication system 1 according to this embodiment includes a restriction unit 2 for restricting the entry of people, a face authentication system 3 as a first authentication system, a capacitance authentication system 4 as a second system, and a release unit 5 for releasing the restriction state of the restriction unit 2 based on the authentication results of each authentication system.

[0014] Examples of the restriction unit 2 include an autolock in an apartment building, a security gate in a building, a security entrance gate for entering an event venue or facility, and a security gate for entering a residential area. That is, when a person enters, an entry restriction is usually set, and only authenticated people are allowed to enter by lifting the entry restriction. The above-mentioned systems are generally in practical use, so a detailed description of the configuration and functions will be omitted.

[0015] The face authentication system 3 compares the face image of a person captured by a camera with a preregistered face image by image processing, and authenticates the person if the face image matches the preregistered face image. For example, in the case of an autolock of an apartment building, face images of all residents of the apartment building are preregistered, and a camera 11 as an input device installed at the entrance of the apartment building captures the face of a person who is about to enter the apartment building, and judges whether the captured face matches any of the preregistered face images. If there is a matching face image, it is determined that the person is a resident of the apartment building, and the entry restriction (autolock) is released. If there is no matching face image, it is determined that the person is not a resident of the apartment building, and the entry restriction is maintained. The camera 11 as an input device may be a fixed camera such as a security camera.

[0016] The capacitance authentication system 4 measures the change in capacitance when a person approaches the camera 11 to perform face authentication in the face authentication system 3, and performs authentication by determining whether or not the face authentication is faked from the change in capacitance. As described above, in the face authentication system 3, a person who is not the person who can be authenticated but cannot be authenticated may succeed in authentication by using a photograph or image of the person, which is insufficient in terms of security. Therefore, the capacitance authentication system 4 measures the change in capacitance to authenticate that the subject captured by the camera 11 is not a photograph printed on paper or an image on a display, but is indeed a living person, thereby making security sufficient. Note that, for example, when a fixed camera such as a security camera is used as the camera 11 of the face authentication system 3, the capacitance authentication system 4 may measure the change in capacitance when a person approaches the restriction unit 2 or a predetermined area in the vicinity thereof.

[0017] Here, the configuration of the capacitance authentication system 4 will be described in detail. Fig. 2 is a block diagram showing the configuration of the capacitance authentication system in the authentication system according to this embodiment. The capacitance authentication system 4 includes a helical antenna 41 serving as a signal input unit, an oscillation circuit 42 that outputs an AC signal by changing the oscillation frequency according to a change in capacitance 41a input by the helical antenna 41, and a measurement sensor 45 consisting of a frequency counter 43 that counts the frequency of the AC signal output from the oscillation circuit 42, and a determination unit 44 that calculates a change in capacitance based on the count result of the frequency counter 43 and determines whether the subject captured by the camera 11 is a living person.

[0018] As the oscillator circuit 42, for example, a generally known Colpitts oscillator circuit or a Hartley oscillator circuit can be used. Here, a Colpitts oscillator circuit is used, and an example of a circuit diagram of the Colpitts oscillator circuit is shown in Fig. 3. As can be seen from the circuit diagram in Fig. 3, when a person approaches the oscillator circuit 42, the person is present as a dielectric in a space that originally contained only air, so that the dielectric constant changes, and the capacitance of the capacitor Ca connected in parallel to the helical antenna 41 changes. In the oscillator circuit 42, the change in the capacitance of the capacitor Ca causes a change in the oscillation frequency accordingly.

[0019] The oscillation frequency f of the AC signal output from the oscillator circuit 42 φ teeth, f φ = 1 / 2π(√LC) (1) 2. The determination unit 44 calculates the capacitance C1 from (1) above based on the frequency value counted by the frequency counter 43 and the value of the inductance L, which is a known value. If the value of the capacitance C1 exceeds a preset threshold, it is determined that the person performing face authentication with the camera 11 is a living person and that spoofed face authentication is not being performed. If the capacitance C1 is equal to or less than the threshold, it is presumed that the person performing face authentication with the camera 11 is not a living person but, for example, a paper with a photograph printed on it or an image displayed on a display, and it is determined that spoofing is being performed in the face authentication.

[0020] The threshold value is preset and is set with a certain margin in consideration of the value of capacitance C1, which changes when a person approaches camera 11 and captures an image of his / her face. For example, if a person is large and has a large face and the surrounding neck and shoulders, the value of capacitance C1 will inevitably be large, and if a person is small, the value of capacitance C1 will inevitably be small. The threshold value is set with a margin large enough to accommodate such differences in physique.

[0021] Returning to FIG. 1, when the result of the face authentication system 3 is successful and the result of the capacitance authentication system 4 is not faked, the release unit 5 transmits to the restriction unit 2 a control signal for releasing the entry restriction set by the restriction unit 2. When either the face authentication system 3 or the capacitance authentication system 4 is not authenticated, the release unit 5 transmits to the restriction unit 2 a control signal for not releasing the entry restriction, or does nothing to maintain the entry restriction. The restriction unit 2 that receives the control signal for releasing the entry restriction performs drive control such as unlocking or opening a gate so that the authenticated person can enter a desired area. Note that, in the case of a device that normally restricts entry and releases the entry restriction only when an authenticated person enters, such as an autolock in an apartment building or a security gate in a building, the restriction unit 2 detects that a person has entered a predetermined area and immediately returns to the entry restriction state.

[0022] In this way, the authentication system according to this embodiment includes a face authentication system 3 that performs authentication processing by recognizing the face of a person when the person enters a specified area where entry is restricted, and a capacitance authentication system 4 that performs authentication processing using capacitance, which is a parameter different from that of the face authentication system 3. The capacitance authentication system 4 measures the capacitance that changes when a person approaches the camera 11, which is an input device for releasing the entry restriction, and determines whether or not the face authentication in the face authentication system 3 has been spoofed based on the measurement result. If the release unit 5 determines that the face authentication in the face authentication system 3 has been successful and that there has been no spoofing in the face authentication, it releases the entry restriction. Therefore, by measuring the change in capacitance, it is possible to detect people regardless of the environment, and it is possible to determine whether or not face authentication is being performed through spoofing, thereby achieving high security.

[0023] In addition, the measurement sensor 45 for detecting changes in capacitance is equipped with an oscillator circuit 42 whose frequency changes in accordance with changes in capacitance in the measurement area, and a frequency counter 43 that counts the frequency of the signal oscillated by the oscillator circuit 42, thereby making it possible to realize a very simple and low-cost measurement sensor 45.

[0024] (Second embodiment of the present invention) The authentication system according to this embodiment will be described with reference to Fig. 4. The authentication system according to this embodiment compares the capacitance calculated from parameters obtained by face authentication in the face authentication system 3 with the capacitance measured in the electrostatic capacitance authentication system 4, thereby mutually enhancing the authentication accuracy of each. Note that in this embodiment, descriptions that overlap with those in the first embodiment will be omitted.

[0025] The capacitance C of a capacitor is C = ε S / d (2) where ε is the dielectric constant, S is the area, and d is the distance.

[0026] From equation (2), the capacitance C2 between the camera 11 and the person can be calculated by obtaining the distance d between the lens of the camera 11 and the person whose face is to be recognized, and the area S of the biometric part centered on the face. FIG. 4 is a diagram showing a state in which a person whose face is to be recognized is captured by a camera. The camera 11 is equipped with an autofocus function, and the distance d between the lens of the camera 11 and the person whose face is to be recognized can be obtained from the focus signal. In addition, by analyzing the image captured by the camera 11, the area of ​​the face can be obtained, and depending on the analysis algorithm, the entire area constituting the person (including the neck, shoulders, chest, arms, etc.) can be obtained, and the area can be obtained. That is, since the information of ε (using the dielectric constant of air), the area S, and the distance d in equation (2) can be obtained, it becomes possible to calculate the capacitance C2.

[0027] The determination unit 44 compares the value of the capacitance C1 measured by the measurement sensor 45 consisting of the oscillator circuit 42 and the frequency counter 43 described above with the value of the capacitance C2 obtained from the above-mentioned formula (2), and determines that the face authentication has not been spoofed if they match or the difference between them is equal to or less than a predetermined value. In other words, by setting the determination standard of the capacitance C1 based on the information measured by the measurement sensor 45 to the capacitance C2 calculated based on the information captured by the camera 11 at the time of the measurement, the determination accuracy is improved, and the electrostatic capacitance authentication system 4 can perform authentication more accurately.

[0028] Note that the distance d obtained above is the distance between the lens of the camera 11 and the person undergoing face recognition because it is the focus signal of the camera 11, but it is better to use the distance between the measurement sensor 45 and the person for the capacitance C2 obtained as the judgment criterion. Therefore, it is desirable to arrange the oscillation circuit 42 of the measurement sensor 45 adjacent to or as close to the lens of the camera 11 as possible.

[0029] In addition to obtaining information on distance d and area S from the image capture information of camera 11 as described above, the distance d and area S may also be obtained by, for example, installing multiple distance sensors and measuring the area in which the face of the person undergoing face authentication is determined to be present and the distance thereto (specifically, for areas in which a face is present, the value of the distance sensor becomes smaller, making use of this fact).

[0030] In this way, when the difference between the capacitance C2 calculated based on the distance between the person undergoing face authentication using the input device, camera 11, and the measurement sensor 45, and the area of ​​the person in the measurement area when the measurement sensor 45 measures the capacitance C1, and the value of the capacitance C1 measured by the measurement sensor 45, is within a predetermined threshold range, it is determined that the face authentication in the face authentication system 3 has not been spoofed, so that accurate authentication can be achieved using the capacitance based on C=ε·S / d and the measurement results of the measurement sensor 45.

[0031] In addition, the distance between the person undergoing face authentication and measurement sensor 45 (the distance from measurement sensor 45 based on the lens position of camera 11) is obtained based on the focus signal from camera 11 that captures the image of the person's face in face authentication system 3, and the area of ​​the human body in the measurement region when measurement sensor 45 measures capacitance C2 is obtained based on the image processing results when face authentication is performed in face authentication system 3. Therefore, parameters for calculating capacitance C=ε·S / d can be easily obtained from face authentication system 3, and high security can be achieved with a simple configuration.

[0032] Other embodiments of the present invention Other embodiments will be described with reference to Fig. 5. In the first embodiment, the determination criterion for capacitance C1 is a preset threshold range, and in the second embodiment, the determination criterion for capacitance C1 is a preset threshold range for the difference with capacitance C2, but in this embodiment, the threshold is determined according to the characteristics of the person performing face authentication.

[0033] Fig. 5 is a system configuration diagram of an authentication system according to this embodiment. What is different from Fig. 1 in the first embodiment in Fig. 5 is that it includes a feature extraction unit 6 that extracts feature information of a person who performs face authentication based on image information captured by a camera 11 of a face authentication system 3, and a threshold setting unit 7 that sets a reference value (threshold) that serves as a criterion for capacitance C1 based on the extracted feature information, and a determination unit 44 compares capacitance C1 with the threshold set by the threshold setting unit 7 to determine whether camouflage is being performed in face authentication.

[0034] When performing image analysis on a person undergoing face authentication, it is relatively easy to extract the facial region using existing algorithms, but for example, when wearing a large scarf or knit cap in winter, it may be difficult to accurately extract body parts such as the neck, shoulders, chest, and arms through image analysis. That is, body parts such as the neck, shoulders, chest, and arms affect the measurement of capacitance C1 of measurement sensor 45 due to differences in physique, so when calculating capacitance C2 from imaging information from camera 11, it is desirable to calculate capacitance C2 taking these body parts into account (i.e., area S including the above body parts).

[0035] However, as described above, it may be difficult to accurately determine these body parts using only the image information captured by the camera 11. In other words, there is a possibility that the calculation of the area S may vary depending on the accuracy of the image analysis. Therefore, in the present embodiment, in the face authentication system 3, face images are registered, and the capacitance C0 measured by the measurement sensor 45 for each person is measured and registered in advance.

[0036] When the face authentication of the face authentication system 3 is successful, the feature extraction unit 6 extracts the capacitance C0 registered in advance corresponding to the person performing the face authentication. At the same time, the capacitance authentication system 4 calculates the values ​​of the capacitances C1 and C2, and at this time, the threshold setting unit 7 corrects the value of C2 based on the value of C0. That is, when the accuracy of the image analysis is poor and the reliability of the value of C2 (the value of the area S) is low, the threshold is adjusted with a weighting coefficient that places emphasis on C0, and when the accuracy of the image analysis is good and the reliability of the value of C2 (the value of the area S) is high, the threshold is adjusted with a weighting coefficient that places emphasis on C2. Then, the determination unit 44 compares the threshold value adjusted with the weighting coefficient with the capacitance C1, and if the difference is within a predetermined range, it determines that no camouflage has been performed in the face authentication. The subsequent processing is the same as that in the first and second embodiments.

[0037] As for the accuracy of image analysis, the administrator may grasp the accuracy (whether the algorithm is highly accurate or low accurate) that is fixed in advance according to the image analysis algorithm, and set the weighting coefficient to a fixed value accordingly. Also, for example, when image analysis is performed by AI, if the AI ​​can calculate the success rate of its own image analysis process, the capacitance C2 may be adjusted by a weighting coefficient corresponding to this success rate. In other words, if the success rate is high, the weighting coefficient is set to place emphasis on capacitance C2, and if the success rate is low, the weighting coefficient is set to place emphasis on capacitance C0.

[0038] In this manner, in the authentication system according to this embodiment, the features of the person captured by the camera 11 are extracted based on the results of face recognition in the face recognition system 3 or the results of image processing when face recognition is performed in the face recognition system 3, and the threshold value that serves as the standard for when the judgment unit 44 performs judgment processing is adjusted by a weighting coefficient based on the extracted feature information, so that judgment criteria according to the features of the person to be authenticated can be set, and accurate authentication processing according to the accuracy of the image analysis processing can be achieved. [Explanation of symbols]

[0039] 1 Authentication System 2 Restrictions 3. Face Recognition System 4. Capacitive authentication system 5 Release section 6 Feature Extraction 7 Threshold setting section 41 Helical Antenna 42 Oscillator Circuit 43 Frequency Counter 44 Judgment section 45 Measurement Sensors

Claims

1. a first authentication system that performs an authentication process by performing face authentication on a person when the person enters a predetermined area where entry is restricted; A second authentication system that executes an authentication process using parameters different from those of the first authentication system; a release means for releasing the entry restriction when face authentication in the first authentication system is successful and authentication in the second authentication system is successful, The second authentication system, a measurement sensor that measures a capacitance that changes when the person approaches the predetermined area; and a determination means for determining whether the face authentication in the first authentication system is not spoofed based on the result of the measurement sensor.

2. 2. The authentication system according to claim 1, The measurement sensor is an oscillator circuit whose frequency changes with changes in capacitance within a measurement region; and a frequency counter that counts the frequency of the signal oscillated by the oscillator circuit.

3. 3. The authentication system according to claim 1, An authentication system characterized in that the determination means determines that the face authentication in the first authentication system is not spoofed when the difference between the distance between a person performing an input operation on an input device for releasing the entry restriction and the measurement sensor, and the electrical capacitance calculated based on the area of ​​the person in the measurement area when the measurement sensor measures the capacitance, and the capacitance value measured by the measurement sensor is within a predetermined threshold range.

4. 4. The authentication system according to claim 3, An authentication system characterized in that the determination means obtains the distance between the person performing an input operation on the input device and the measurement sensor based on a focus signal from a camera that images the person's face in the first authentication system, and obtains the area of ​​the person in the measurement area when the measurement sensor measures the capacitance based on the image processing results when performing the face authentication in the first authentication system.

5. 5. The authentication system according to claim 4, a feature extraction means for extracting feature information of the person captured by the camera based on a result of face authentication in the first authentication system or a result of image processing when face authentication is performed in the first authentication system; and and a threshold setting means for setting the predetermined threshold as a standard when the determination means performs a determination process based on the feature information extracted by the feature extraction means.

Images