JP6347349B2 - Biological detection device and biological detection method - Google Patents

Description

  The present invention relates to a living body detection apparatus and a living body detection method for determining whether or not a detection target part is a living body.

  Conventionally, various authentication methods have been proposed for security and the like. In particular, the biometric authentication method that performs personal authentication using biometric information such as fingerprints and veins is more convenient than the conventional password or IC card authentication method because the user does not need to store authentication information. In recent years, it has been widely used in institutions that require high security strength, such as bank ATMs and immigration at airports.

  On the other hand, such a biometric authentication method has been pointed out to be vulnerable to “spoofing” using a fake by a third party. For example, in the conventional fingerprint authentication method, since a fingerprint pattern read by a fingerprint sensor is generally only checked against pre-registered fingerprint data, a thin disguised finger simulating another person's fingerprint pattern is created with silicon or the like. Then, by pasting this disguised finger on a third person's human finger and performing collation, it is determined that it matches the registered fingerprint data, which may cause erroneous recognition. In recent years, in order to eliminate the vulnerability due to such “spoofing” behavior, in addition to the biometric authentication method, a biometric detection method for discriminating between a biometric and a fake is used.

  As a living body detection method, various methods such as a method using distortion of the surface of a human body (for example, refer to Non-Patent Document 1) and a method using sweating of a living body (for example, refer to Non-Patent Document 2) have been proposed. Yes. However, these living body detection methods use the characteristics of the surface of the human body, and thus are strongly influenced by the environment and the condition of the human body when performing living body detection. In addition, as a living body detection method, an impedance method (for example, see Non-Patent Document 3) in which detection is performed based on an electrical constant inside the human body has been proposed. In this method, detection is performed by bringing a finger into contact with an electrode during measurement. Therefore, it is affected by the pressure applied to the electrode of the finger and moisture during sweating. Thus, in the impedance method, since the influence of the human body surface appears strongly, the vulnerability to a medium having an electrical constant of the skin becomes a problem.

  On the other hand, the present inventors used a BPF (Band-pass Filter) in which a CSRR (Complementary Split-Ring Resonator) that is strongly influenced by an adjacent medium is arranged. A biometric detection method for vein authentication has been proposed (see, for example, Non-Patent Document 4). In the living body detection method of Non-Patent Document 4, living body detection is performed using a difference in passage characteristics caused by the electrical characteristics of each medium. Since the living body detection method of Non-Patent Document 4 is caused by the change in the electromagnetic field distribution near the filter due to the influence of the medium close to the CSRR, the response of the electromagnetic field to the depth direction of the close medium is obtained. It is possible to obtain. In other words, this living body detection method uses the electrical characteristics of the human body's layered structure (skin, fat, muscles, bones) in addition to the electrical characteristics of the human body surface part, so that breakthroughs by third parties can be avoided. It can be very difficult. Moreover, in this living body detection method, it is also possible to reduce the influence by the condition at the time of the measurement concerned about the impedance method by performing detection using a frequency that does not largely depend on the electrical constant of the human body surface portion.

Antonelli A., Cappelli R., Maio D., Maltoni D., “Fake Finger Detection by Skin Distortion Analysis”, IEEE Trans. Information Forensics and Security, vol.1, no.3, pp.36-373, Sep. 2006. A.Ross, S.C.Dass, A.K.Jain, “Fingerprint warping using ridge curve correspondences”, IEEE Trans. Pattern Anal., Vol.28, no.1, pp.19-30, Jan. 2006. OGMatinsen, S. Clausen, JBNysaether, S. Grimnes, “Utilizing Characteristic Electrical Properties of the Epidermal Skin Layers to Detect Fake Fingers in Biometric Fingerprint Systems-a Pilot Study”, IEEE Trans. Biomed. Eng., Vol. 54, no.5, pp.891-894, Apr. 2004. Koji Serizawa, Satoru Iguchi, Tadahiko Maeda, "Proposal of a biometric detection method for vein authentication using broadband CSRR-BPF", Theory of Science (B), vol.J95-B, no.10, pp.1284-1287 , Oct. 2012.

  In the living body detection method of Non-Patent Document 4, as reference data used when performing living body detection, a plurality of human fingers are brought close to a CSRR-BPF (bandpass filter using a complementary split ring resonator) in advance. The human finger to be detected becomes a living body using the averaged value of the pass characteristic and the pass characteristic when the finger of the person to be detected is brought close to CSRR-BPF. It is determined whether or not. However, since individual persons have individual differences in the tissue structure, shape, size, etc. inside the human body, the reference data used when performing biometric detection has multiple fingers placed close to the CSRR-BPF. In the case of predetermined data such as an average value of the pass characteristics of the human body, depending on the person, it is determined that it is a living body but not a living body, or is not a living body but is determined to be a living body. There is a risk of false detection.

  The present invention has been made in view of the above problems, and by using electromagnetic wave response characteristics at a plurality of different positions in the same human body, the occurrence of false detection due to individual differences in the tissue structure inside the human body. It is an object of the present invention to provide a living body detection apparatus and a living body detection method capable of preventing the above-described problem and improving the accuracy of living body detection.

  In order to achieve the above object, the living body detection apparatus according to claim 1 includes a plurality of band-pass filters having split ring resonators designed to have a predetermined center frequency according to a detection target portion of the subject, and electromagnetic waves. Each of the electromagnetic wave response characteristics when the plurality of reference target parts that are different from the detection target part and the detection target part in the subject are brought close to or in contact with the respective band-pass filters. A relative reference value obtained from a measurement value of the electromagnetic wave response characteristic by the detection target part measured by the measurement part, and a measurement value of the respective electromagnetic wave response characteristics by the plurality of reference target parts And based on the evaluation value calculation unit for calculating the relative evaluation value in the subject, and the relative evaluation value calculated by the evaluation value calculation unit, Serial detection target site is characterized by and a determining biological determination unit that determines whether a living body.

  The living body detection device according to claim 2 is characterized in that the electromagnetic wave response characteristic measured by the measurement unit is a transmission characteristic or / and a reflection characteristic.

  The living body detection device according to claim 3, in a three-dimensional manner, for each of the detection target part and the reference target part so that each of the bandpass filters follows the shape of the detection target part and the reference target part. It is characterized by a plurality of arrangements.

  According to a fourth aspect of the present invention, the living body detection device is formed so as to follow the shapes of the respective band-pass filters, the detection target part, and the reference target part.

  The biological detection apparatus according to claim 5, wherein a plurality of electromagnetic wave responses when a detection target portion of a plurality of subjects different from the subject measured by the measurement unit is brought close to or in contact with a band-pass filter, respectively. A storage unit that preliminarily stores an absolute reference value obtained from a measured value of the characteristic, and the evaluation value calculation unit includes the measured value of the electromagnetic wave response characteristic by the detection target site measured by the measurement unit, and the storage unit The absolute evaluation value is calculated using the absolute reference value stored in the living body, and the living body determination unit determines whether the detection target site is a living body based on the relative evaluation value and the absolute evaluation value. It is characterized by determining whether or not.

  The living body detection apparatus according to claim 6 is characterized in that a plurality of the band-pass filters for bringing the detection target part close to or in contact with each other are arranged side by side.

  The living body detection method according to claim 7, wherein an electromagnetic wave is generated in a plurality of band-pass filters having split ring resonators designed at a predetermined center frequency according to a detection target part of a subject, and the detection target part And a measuring step for measuring each electromagnetic wave response characteristic when a plurality of reference target parts that are different from the detection target part in the subject are brought close to or in contact with the respective band pass filters; Using the measured value of the electromagnetic wave response characteristic by the detection target part and the relative reference value obtained from the measured value of the respective electromagnetic wave response characteristic by the plurality of reference target parts, the relative evaluation value in the subject is obtained. An evaluation value calculating step for calculating, and a living body determining step for determining whether or not the detection target part is a living body based on the relative evaluation value. It is characterized.

  According to the living body detection device of claim 1, the detection target region and the plurality of references are included in the plurality of band-pass filters having split ring resonators designed to have a predetermined center frequency according to the detection target region of the subject. Each electromagnetic response characteristic when the target part is brought close to or in contact with each other is measured by the measuring unit, and the measured electromagnetic response characteristic value by the detection target part and each electromagnetic wave response characteristic by the plurality of reference target parts are measured. The relative evaluation value in the same subject is calculated by the evaluation value calculation unit using the relative reference value obtained from the measured values. Then, the living body determination unit determines whether or not the detection target part is a living body based on the relative evaluation value, so that the erroneous detection due to the individual difference of the tissue structure inside the human body does not occur. Accuracy can be improved.

  According to the living body detection device according to claim 2, since the pass characteristic or / and the reflection characteristic is measured as the electromagnetic wave response characteristic by the measurement unit, the measured values of both the pass characteristic and the reflection characteristic are used for the living body determination. In addition, the accuracy of living body detection can be further improved.

  According to the biological detection device of claim 3, a plurality of each band pass filter is arranged three-dimensionally for each detection target part and reference target part so as to follow the shape of the detection target part and the reference target part. Therefore, it is possible to acquire three-dimensionally appropriate electromagnetic wave response characteristics from the detection target part and the reference target part of the subject having a three-dimensional structure. Accuracy can be improved.

  According to the living body detection device of claim 4, each bandpass filter is formed so as to follow the shape of the detection target part and the reference target part, so that the detection target of the subject having a three-dimensional structure is provided. Since a three-dimensionally appropriate electromagnetic wave response characteristic can be efficiently acquired from the part and the reference target part, it is possible to improve the accuracy of living body detection by performing multilateral determination.

  According to the living body detection device of claim 5, a plurality of electromagnetic wave responses when the detection target portions of a plurality of subjects different from the subject measured by the measurement unit are brought close to or in contact with the band-pass filter, respectively. The absolute reference value obtained from the characteristic measurement value is stored in advance in the storage unit, and the evaluation value calculation unit is stored in the storage unit with the measurement value of the electromagnetic wave response characteristic by the detection target part measured by the measurement unit. The absolute evaluation value is calculated using the absolute reference value, and the living body determination unit determines whether the detection target part is a living body based on the absolute evaluation value and the relative evaluation value. It is possible to appropriately improve whether or not the detection target part is a living body even when a forgery is used not only for the detection target part but also for the reference target part.

  According to the living body detection device of the sixth aspect, since a plurality of band pass filters for bringing the detection target parts close to or in contact with each other are arranged side by side, when the band pass filters having different predetermined center frequencies are used. By moving the detection target part, a plurality of different electromagnetic wave response characteristics with respect to the same detection target part can be obtained, and information on different depth directions of the same detection target part can be acquired. In addition, when bandpass filters with the same predetermined center frequency are used, by moving the detection target part, it is possible to average the measured values of the electromagnetic wave response characteristics with respect to the detection target part. Accuracy can be improved.

  According to the living body detection method of claim 7, a plurality of band pass filters having split ring resonators designed to have a predetermined center frequency according to a detection target portion of the subject include a detection target portion and a plurality of references. Measure the electromagnetic wave response characteristics when the target parts are in close proximity or contact with each other, and measure the measured electromagnetic wave response characteristics by the detection target parts and the measured values of the respective electromagnetic wave response characteristics by multiple reference target parts In order to calculate a relative evaluation value in the same subject using the relative reference value obtained from the above, and to determine whether the detection target site is a living body based on the relative evaluation value, Since false detection due to individual differences in tissue structure does not occur, the accuracy of living body detection can be improved.

It is a schematic block diagram which shows an example of the biometric apparatus which concerns on embodiment of this invention. It is a graph which shows the epidermis depth of a human body finger. It is a schematic diagram which shows an example of CSRR-BPF, Comprising: (a) has shown the upper surface side, (b) has shown the bottom face side. It is an expansion schematic diagram which shows a part of CSRR-BPF, Comprising: (a) has shown the A part enclosed with the dashed-two dotted line in FIG. 3, (b) is the B part enclosed with the dashed-two dotted line in FIG. Is shown. It is a schematic diagram which shows an example of a some measurement point. It is a schematic diagram showing an example of a model of a medium to be approached, where (a) is a model of the index finger of the human body, and (b) is a model of the camouflaged finger in which a camouflage is pasted on the index finger of the human body. It is a graph which shows an example of passage characteristic | S21 | at the time of making a medium approach CSRR-BPF. It is a flowchart which shows an example of the flow of a process by the biological detection apparatus which concerns on embodiment of this invention. It is a graph which shows the average difference and similarity of the passage characteristic used as an evaluation value for a biological body determination. It is a schematic diagram which shows an example of the state which has arrange | positioned two or more CSRR-BPF three-dimensionally to the object site | part. It is a schematic diagram which shows another example of CSRR-BPF. It is a schematic diagram which shows an example of arrangement | positioning of CSRR-BPF.

  Hereinafter, embodiments of a living body detection apparatus according to the present invention will be described with reference to the drawings. A living body detection apparatus 1 according to the present invention is for determining whether or not a detection target part of a human body as a subject is a living body in order to prevent an “spoofing” act by a forgery. In addition, it is used in combination with biometric authentication devices such as fingerprint authentication and vein authentication. As shown in FIG. 1, the living body detection apparatus 1 according to the present invention includes a plurality of BPFs (Band-pass Filters) in which CSRRs (Complementary Split-Ring Resonators) are arranged. Bandpass filter (hereinafter referred to as CSRR-BPF) 2 and respective electromagnetic wave responses when the detection target part and a plurality of reference target parts at positions different from the detection target part are brought close to each CSRR-BPF 2 A measurement unit 3 that measures the characteristics, and a computer 4 that performs arithmetic processing for biological determination using the electromagnetic wave response characteristics measured by the measurement unit 3 are provided. In addition, since the mechanism for fingerprint authentication or vein authentication used in combination with the living body detection apparatus 1 can use a conventionally well-known thing, it abbreviate | omits about the detailed description. Moreover, although this embodiment shows an example in which three CSRR-BPFs 2 are arranged, the number of CSRR-BPFs 2 is not limited to this, and a detection target part and a reference target for measuring electromagnetic wave response characteristics You may arrange | position suitably according to the number of site | parts.

  The CSRR-BPF 2 is designed to have a predetermined center frequency according to the detection target part of the subject (human body), and includes a complementary split ring resonator (CSRR) that is strongly influenced by an adjacent medium. Have. Since the CSRR-BPF 2 is caused by the change in the distribution of the electromagnetic field in the vicinity of the filter due to the influence of the adjacent medium, an electromagnetic field response in the depth direction of the adjacent medium can be obtained.

  Examples of the detection target part include a human finger that performs fingerprint authentication and vein authentication. The layered structure of the human finger is composed of skin, fat, muscle, and bone, and assuming an approximate uniform medium from the volume ratio of each tissue, the epidermis depth is calculated as shown in FIG. Since it is necessary to simulate a blood vessel pattern that is distributed in three dimensions when creating a camouflage that assumes vein authentication, the structure utilizing the characteristics of the whole human finger is preferable. The CSRR-BPF2 is designed at a frequency where the finger thickness and the skin depth are equal. When a human finger is brought close to the CSRR-BPF2, the electrical characteristics of the portion corresponding to the thickness of the human finger greatly affect the characteristics of the CSRR-BPF2. Therefore, in Non-Patent Document 4, the thickness of the human finger is assumed to be about 10 mm, and the CSRR-BPF 2 is designed with 6 GHz, which is the frequency at which the value is the skin depth, as the center frequency.

  On the other hand, in the present embodiment, FIG. 3 shows a CSRR-BPF 2 designed on the assumption of “spoofing” by attaching a thin fake to the tip of a human finger in order to perform biometric detection for fingerprint authentication. FIG. 3 shows the structure of CSRR-BPF 2 designed with a center frequency of 10 GHz, where (a) shows the state seen from the top surface side, and (b) shows the state seen from the bottom surface side. Is shown. 4 is an enlarged view of a part of CSRR-BPF2 in FIG. 3. FIG. 4A shows a portion A surrounded by a two-dot chain line in FIG. 3, and FIG. A portion B surrounded by a two-dot chain line is shown. In biometric detection for fingerprint authentication, unlike the case where vein authentication is assumed, a fake finger is composed of a human finger and a thin film, so that the influence of the human body surface is affected rather than the structure utilizing the characteristics of the entire human finger. Since the structure utilizing the layered structure of the human finger is preferable while receiving relatively large, the epidermis depth is assumed to be about 5 mm, which is half of the approximate thickness of the human finger, and the value is the frequency that becomes the epidermis depth. CSRR-BPF2 is designed with 10 GHz as the center frequency.

In this CSRR-BPF2, for example, glass thermosetting PPO resin R-4276 (relative dielectric constant ε _γ = 3.4, dielectric loss tangent tan δ = 0.005) is used as the substrate in the same manner as the 6 GHz band CSRR-BPF of Non-Patent Document 4. , Substrate thickness: 1.0 mm) can be used. The hatched portions in FIGS. 3 and 4 are copper foils, and the unhatched portions are made of a dielectric. As shown in FIGS. 3 and 4, the dimensions of CSRR-BPF2 are as follows: a1 = 26.0 mm, a2 = 19.5 mm, a3 = 10.5 mm, a4 = 1.3 mm, a5 = 9.4 mm, a6 = 0.7 mm, a7 = 0.2 mm, a8 = 0.4 mm, a9 = 1.4 mm, a10 = 1.5 mm, a11 = 0.8 mm, a12 = 0.7 mm, a13 = 1.2 mm, b1 = 9 .5mm, b2 = 10.2mm, b3 = 0.3mm, b4 = 0.3mm, b5 = 3.0mm, b6 = 0.3mm, b7 = 1.1mm, b8 = 0.2mm, b9 = 2.2mm B10 = 0.2 mm and b11 = 0.2 mm. The ports P1 and P2 provided at both ends of the CSRR-BPF 2 are provided with connectors (not shown), respectively, and are connected to the measurement unit 3 via the coaxial cable 5 as shown in FIG. Yes. 3 and 4 show an example of CSRR-BPF2 with a center frequency of 10 GHz assuming biometric detection for fingerprint authentication, but the structure of CSRR-BPF2 is not limited to this. Rather, it is appropriately designed according to the location of the detection target part for determining whether or not it is a living body. Although not shown in detail, a thin polypropylene film having a thickness of about 0.2 mm may be installed on the CSRR-BPF 2 in order to flatten the CSRR and protect the copper foil against corrosion.

  Such CSRR-BPF2 can be measured in a specific region having a certain extent in the human body, for example, a plurality of positions A1 to A9 such as the finger surface and the palm as shown in FIG. Several are arranged. In FIG. 5, nine positions A1 to A9 are shown as measurement positions. However, it is not necessary to perform measurement at all these positions, and when performing biometric detection for fingerprint authentication of the index finger. For example, the position of the detection target part may be a position A1 above the first joint of the index finger, and any one of the other positions A2 to A9 may be the reference target part. FIG. 5 shows an example of a measurement position for performing living body detection. The measurement position is not limited to the positions A1 to A9, and the electromagnetic wave response characteristics are measured at other positions. You may do it.

  The measurement unit 3 is configured by, for example, a vector network analyzer or the like, and generates a high-frequency electromagnetic wave for measurement, and brings a detection target part and a plurality of reference target parts close to each CSRR-BPF 2. Each electromagnetic wave response characteristic is measured. As the electromagnetic wave response characteristics, in each CSRR-BPF2, pass characteristics S21 (forward transmission) and S12 (reverse transmission) between the ports P1 and P2, reflection characteristics S11 (forward reflection) and S22 (reverse reflection). 4) can be acquired. S21 represents a pass characteristic that is incident on the port P1 of the CSRR-BPF2 and transmitted to the port P2. S12 is a pass characteristic that is incident on the port P2 and transmitted to the port P1. is there. S11 represents the reflection characteristic incident on the port P1 and reflected from the port P1, and S22 represents the reflection characteristic incident on the port P2 and reflected from the port P2.

  FIG. 7 shows an example of the result of calculating the pass characteristic | S21 | when the medium is brought close to the CSRR-BPF 2 designed with the center frequency shown in FIG. 3 as 10 GHz by the FDTD (Finaite Difference Time Domain) method. Is. In FIG. 7, in order to show the effect of the thickness direction of the human finger on the CSRR-BPF 2, a human finger having a layered structure, a skin phantom, and a human finger (hereinafter referred to as a camouflaged finger) to which a fake material is attached. Each of them shows a passing characteristic | S21 | when close to the CSRR-BPF2. Here, a calculation model as shown in FIG. 6 is used as a human finger and a fake finger. The dimensions of this calculation model are based on the dimensions of each calculation model disclosed in Non-Patent Document 3, and the dimensions of the human finger model are x1 = 22.0 mm and y1 = 14.0 mm and z1 = 10.0 mm are set at a position where the distance from the CSRR-BPF2 is 0.2 mm, and the pass characteristic | S21 | is obtained. For human fingers, the dimensions are set based on the volume ratio of each tissue (skin, fat, muscle, bone), 0.8mm for skin, 1.8mm for fat, 1.4mm for muscle, and 2.0mm for bone. Is set. On the other hand, as the camouflage worn on the camouflaged finger, the passing characteristic | S21 | is obtained by using a skin phantom that simulates the electrical constant of the skin and having a sufficiently thin thickness of 0.5 mm. From the results shown in FIG. 7, it can be seen that CSRR-BPF2 affects not only the electrical characteristics of the skin, which is the surface of the human body, but also human tissues other than the skin. In addition, it can be confirmed that the characteristics are greatly changed when the camouflaged finger is brought close to the human finger, and a biometric detection process described later is performed by using a difference in passing characteristics caused by the electrical characteristics of these media. Living body detection can be performed with high accuracy.

  Thus, the measured values of the electromagnetic wave response characteristics measured by the measuring unit 3 are input to the computer 4. The computer 4 uses the measurement value of the electromagnetic wave response characteristic measured by the measurement unit 3 to perform arithmetic processing for biological determination of whether or not the detection target site is a living body. A central processing unit) 6, a hard disk 7, an arithmetic processing unit 8, a RAM (Random Access Memory) 9, a display unit 10, an operation unit 11, and the like. Further, as shown in FIG. 1, these units are mutually connected to a system bus 12, and various data and the like are input / output via the system bus 12, and various processes are executed under the control of the CPU 6. The

  The hard disk 7 stores a processing program and the like for performing a living body determination as to whether or not the detection target site is a living body using the measured value of the electromagnetic wave response characteristic input from the measuring unit 3. In the present embodiment, an example in which a processing program for performing biometric determination is stored in the hard disk 7 is shown, but instead, it is stored in a computer-readable storage medium (not shown). The processing program can be read from the recording medium.

  The arithmetic processing unit 8 performs arithmetic processing for determining whether or not the detection target part is a living body under the control of the CPU 6 based on a processing program stored in the hard disk 7. The RAM 9 temporarily stores a processing program read from the hard disk 7 and is used as a work area for the CPU 6.

  In addition, in the RAM 9, a plurality of electromagnetic wave response characteristics measured when a plurality of human body detection target parts different from the subject (human body) that performs biological determination are brought close to the CSRR-BPF 2 are measured in advance by the measurement unit 3. The obtained measurement values of the plurality of electromagnetic wave response characteristics are stored in advance by the arithmetic processing unit 8 as absolute reference values, which are reference data for use in biological determination, for example, averaged values. The number of human bodies necessary for generating the absolute reference value is not particularly limited, but it is preferable to use five or more human bodies in order to increase the reliability of the absolute reference value.

  The display unit 10 is composed of, for example, a liquid crystal display or the like, and displays a biological determination result or the like. Note that the living body determination result may be notified by voice instead of the display unit 10 or together with the display unit 10. The operation unit 11 includes a mouse, a keyboard, and the like, and is used by an operator to input various data and operation commands. The display unit 10 and the operation unit 11 may be integrally configured as a touch panel.

  Hereinafter, the flow of the living body detection process of the detection target part by the living body detection apparatus 1 will be described with reference to the flowcharts of FIGS. 1 and 8. Here, the detection target part is a position A1 which is a part above the first joint of the index finger shown in FIG. 5, and the reference target part is a part A2 which is a part above the first joint of the middle finger and the first part of the ring finger. A case where the position is A3 above the joint will be described as an example.

  First, in the living body detection apparatus 1, as shown in FIG. 8, each subject to be detected detects electromagnetic wave response characteristics when the detection target part A1 and the reference target parts A2 and A3 are brought close to the respective CSRR-BPF2. Is measured by the measurement unit (vector network analyzer) 3 (S101). Here, the vector network analyzer 3 generates an electromagnetic wave having a frequency of 5 GHz to 12 GHz, and 5 GHz as each electromagnetic wave response characteristic when the detection target part A1 and the reference target parts A2 and A3 are brought close to the respective CSRR-BPF2. To pass characteristic | S21 | at 12 GHz. Then, these pass characteristics acquired by the vector network analyzer 3 are input to the computer 4.

The evaluation value calculating unit 81 of the arithmetic processing unit 8, those measurements of pass characteristics by the reference target region A2, A3 obtained by a vector network analyzer 3 were averaged valued relative reference value | S21 H _| is calculated as ( S102). Thus, the relative reference value is calculated from the same human body. The relative reference value is not limited to the average value of the pass characteristics of the reference target part, and other statistical values such as the median value of the pass characteristic of the reference target part may be used as the relative reference value. good.

Next, the evaluation value calculation unit 81 calculates a relative evaluation value using the passage characteristic | S21 _O | of the detection target part acquired by the vector network analyzer 3 and the relative reference value | S21 _H | (S103). ). Here, two relative evaluation values, that is, an average difference in pass characteristics and a similarity, are calculated as relative evaluation values. The average difference of the pass characteristics is obtained by comparing the relative reference value | S21 _H | calculated in S102 from 5 GHz to 12 GHz with the pass characteristics | S21 _O | of the detection target part acquired in S101 from 5 GHz to 12 GHz. calculate. The similarity is calculated using the following mathematical formula (1). However, N to M in Expression (1) are ranges of frequencies to be used. The frequency used is the same as the average difference. Note that the formula for calculating the similarity is not limited to this, and other formulas for calculating similarities may be used. Moreover, you may make it weight numerical formula (1) for every use frequency.

  FIG. 9 shows an example of the result of calculating the average difference and the similarity by performing biometric detection on 16 subjects (20's). FIG. 9 is a two-dimensional plane graph in which the horizontal axis represents the average difference in pass characteristics and the vertical axis represents the degree of similarity, corresponding to the average difference and the degree of similarity, which are relative evaluation values calculated by the evaluation value calculation unit 81. A plot is made at a position on the two-dimensional plane. The hollow plot in FIG. 9 shows a human finger whose detection target part is a living body, and the black-painted plot shows a fake finger with a fake wearing the detection target part. As shown in FIG. 9, by using the average difference and the similarity value, which are relative evaluation values within the same human body, a human body finger that is a living body and a camouflaged finger with a camouflaged part are well distinguished. I understand that.

  The living body determination unit 82 determines whether or not the detection target part is a living body based on the average difference and the similarity value, which are relative evaluation values within the same human body (S104). In the biometric determination unit 82, for example, threshold values are respectively set for the average difference and the similarity of the passage characteristics, and it is determined whether or not the average difference and the similarity are included in each threshold. It is determined whether it is a living body by doing. That is, when the average difference and the similarity calculated by the evaluation value calculation unit 81 are both within the threshold value, it is determined that the detection target part is a living body, and in other cases, the detection target part is It is determined that it is not a living body. In this way, the living body determination unit 82 uses the average difference and the similarity value, which are relative evaluation values within the same human body, to reduce the influence due to individual differences, and wears a human body finger that is a living body and a camouflaged part. A fake finger can be detected with high accuracy. The biometric determination unit 82 may perform biometric determination using only one of the average difference and similarity of the pass characteristics as a relative evaluation value. However, in order to improve detection accuracy, the average of the pass characteristics is used. It is preferable to use two relative evaluation values such as difference and similarity.

Further, in the present embodiment, the case where the living body determination unit 82 performs the living body determination of the detection target part using the average difference and the similarity value that are relative evaluation values within the same human body has been described. The stored absolute reference value | S21 _H1 | may be further used to calculate an absolute evaluation value and perform biometric determination. In this case, in S103, the evaluation value calculation unit 81 calculates both the relative evaluation value and the absolute evaluation value. The evaluation value calculation unit 81 calculates an absolute evaluation value using the passage characteristic | S21 _O | of the detection target part acquired by the vector network analyzer 3 and the absolute reference value | S21 _H1 |. Here, as the absolute evaluation value, two absolute evaluation values, that is, an average difference in pass characteristics and a similarity, are calculated as in the case of the relative evaluation value. The average difference of the pass characteristics is calculated in advance and stored in the RAM 9 in the absolute reference value | S21 _H1 | from 5 GHz to 12 GHz, and the pass characteristics from 5 GHz to 12 GHz of the detection target part acquired in S101 | S21 _O | Is calculated by comparing. Similarity (similarity) is also calculated using the following mathematical formula (2), similarly to the relative evaluation value. It should be noted that the formula for calculating the degree of similarity for the absolute evaluation value is not limited to this, and other conventionally known formulas for calculating the degree of similarity may be used. Moreover, you may make it weight numerical formula (2) for every use frequency.

  In S104, the living body determination unit 82 determines whether the detection target part is a living body based on, for example, the average difference and the similarity value that are relative evaluation values, and determines that the living body is a living body. Whether or not the detection target part is a living body is further determined based on the average difference and the similarity value, which are absolute evaluation values. The biometric determination unit 82 sets threshold values for the average difference and similarity of pass characteristics, which are absolute evaluation values, and whether or not the average difference and similarity are included in each threshold value. It is determined whether it is a living body by determining. As a result, it is possible to appropriately determine whether or not the detection target part is a living body even when a forgery is used not only for the detection target part but also for the reference target part. Further improvement can be achieved.

  In this embodiment, the case where the passing characteristic | S21 | is used as the electromagnetic wave response characteristic for the living body detection is described as an example. However, the reflection characteristic | S11 | of the detection target part and the reference target part by the vector network analyzer 3 Alternatively, other transmission characteristics | S12 | and reflection characteristics | S22 | may be used. Moreover, you may make it perform biometric determination using two or more of these electromagnetic wave response characteristics.

  Moreover, in this embodiment, although the example which has arrange | positioned one CSRR-BPF2 in each of the detection target part A1 and the reference target parts A2 and A3 is shown, as shown in FIG. 10, the detection target part A1 and the reference target part A plurality of CSRR-BPF2s may be arranged three-dimensionally along the shapes of A2 and A3. FIG. 10 shows an example in which three CSRR-BPFs 2 are arranged. In this case, a total of six ports can be used for each of the target parts A1 to A3, so that 6 × 6 = 36 electromagnetic waves. Response characteristics can be obtained. Therefore, the accuracy of living body detection can be further improved by performing living body detection using the plurality of electromagnetic wave response characteristics.

  Moreover, as shown in FIG. 11, you may make it form CSRR-BPF2 in the three-dimensional shape which follows the shape of detection object site | part A1 and reference | standard object site | part A2, A3. Also in this case, 5 × 5 = 25 electromagnetic wave response characteristics can be obtained by providing a plurality of ports P1 to P5 (five in FIG. 11) in the CSRR-BPF2. Therefore, the accuracy of living body detection can be further improved by performing living body detection using a number of electromagnetic wave response characteristics as in the case of FIG. In addition, since the CSRR-BPF 2 is formed in a three-dimensional shape that follows the shapes of the detection target part A1 and the reference target parts A2, A3, it is possible to reduce the occurrence of errors due to pressing pressure and the like. Biological detection can be performed with high accuracy.

  Further, as shown in FIG. 12, a plurality of CSRR-BPFs 2 that make the detection target part A1 approach or contact each other may be arranged side by side. These CSRR-BPFs 2 may be designed to have different center frequencies or may be designed to have the same center frequency. For example, by moving (sliding) the detection target part A1 in the direction indicated by the arrow in FIG. 12 with respect to the plurality of CSRR-BPFs 2 arranged side by side, the CSRR-BPFs 2 having different center frequencies are obtained. When used, a plurality of different electromagnetic wave response characteristics with respect to the same detection target part A1 can be obtained, and information in different depth directions of the same detection target part A1 can be acquired. Further, when the CSRR-BPF 2 having the same center frequency is used, a plurality of measured values of electromagnetic wave response characteristics for the detection target part A1 can be obtained by moving the detection target part A1, and the average value thereof Etc. can be used to improve measurement accuracy.

  The configuration in which a plurality of CSRR-BPFs 2 are arranged may be applied not only to the detection target part A1 but also to the reference target parts A2 and A3. In addition, instead of moving the detection target part A1 and the reference target parts A2 and A3 and bringing them close to or in contact with the respective CSRR-BPF2, the detection target part A1 and the reference target parts A2 and A3 are fixed. -By moving the BPF 2 as a whole using a conventionally known moving mechanism or the like, the electromagnetic wave response characteristics are obtained by automatically approaching or contacting the detection target part A1 and the reference target parts A2 and A3, respectively. You may comprise. Further, the way of arranging the CSRR-BPF2 is not limited to the linear arrangement as shown in FIG. 12, and the detection target part A1 and the reference target parts A2, A3 can be easily arranged with respect to each CSRR-BPF2. It suffices if they are arranged side by side so as to be close to or in contact with each other.

  The embodiment of the present invention is not limited to the above-described embodiment, and can be appropriately changed without departing from the scope of the idea of the present invention.

  The biometric detection apparatus and biometric detection method according to the present invention can be effectively used as a highly reliable information security technique by being used in combination with personal biometric authentication techniques such as fingerprint authentication and vein authentication, for example.

DESCRIPTION OF SYMBOLS 1 Living body detection apparatus 2, 2a The band pass filter (CSRR-BPF) which has a split ring resonator
3 Measurement unit (vector network analyzer)
81 Evaluation Value Calculation Unit 82 Biological Determination Unit 9 RAM (Storage Unit)

Claims (7)

A plurality of bandpass filters having split ring resonators designed at a predetermined center frequency according to the detection target part of the subject;
Each electromagnetic wave response when an electromagnetic wave is generated and a plurality of reference target parts which are different from the detection target part and the detection target part in the subject are brought close to or in contact with the respective band pass filters A measuring section for measuring the characteristics;
Using the measured value of the electromagnetic wave response characteristic by the detection target part measured by the measurement unit and the relative reference value obtained from the measured value of the respective electromagnetic wave response characteristic by the plurality of reference target parts, An evaluation value calculation unit for calculating a relative evaluation value in the sample;
A living body detection apparatus comprising: a living body determination section that determines whether or not the detection target part is a living body based on the relative evaluation value calculated by the evaluation value calculation section.   The living body detection apparatus according to claim 1, wherein the electromagnetic wave response characteristic measured by the measurement unit is a transmission characteristic or / and a reflection characteristic.   The plurality of band-pass filters are arranged three-dimensionally for each of the detection target part and the reference target part so as to conform to the shapes of the detection target part and the reference target part. Item 3. The living body detection device according to item 1 or 2.   3. The living body detection device according to claim 1, wherein each of the band pass filters is formed so as to conform to the shapes of the detection target part and the reference target part. Absolute reference values obtained from measured values of a plurality of electromagnetic wave response characteristics when the detection target portions of a plurality of subjects different from the subject measured by the measurement unit are brought close to or in contact with the band-pass filter, respectively. A storage unit for storing in advance;
The evaluation value calculation unit calculates an absolute evaluation value by using the measurement value of the electromagnetic wave response characteristic by the detection target part measured by the measurement unit and the absolute reference value stored in the storage unit And
5. The living body according to claim 1, wherein the living body determination unit determines whether or not the detection target part is a living body based on the relative evaluation value and the absolute evaluation value. Detection device.   The living body detection device according to claim 1, wherein a plurality of the band pass filters that bring the detection target portion close to or in contact with each other are arranged side by side. Electromagnetic waves are generated in a plurality of band-pass filters having split ring resonators designed to have a predetermined center frequency according to the detection target part of the subject, and the detection target part and the detection target part in the subject Measuring steps for measuring respective electromagnetic wave response characteristics when a plurality of reference target portions at different positions are brought close to or in contact with the respective band pass filters;
Using the measured value of the electromagnetic wave response characteristic by the detection target part and the relative reference value obtained from the measured value of the respective electromagnetic wave response characteristic by the plurality of reference target parts, a relative evaluation value in the subject An evaluation value calculating step for calculating
A living body determination step of determining whether or not the detection target part is a living body based on the relative evaluation value.

Images