WO2013145166A1 - 生体認証装置、生体認証方法、および生体認証プログラム - Google Patents
生体認証装置、生体認証方法、および生体認証プログラム Download PDFInfo
- Publication number
- WO2013145166A1 WO2013145166A1 PCT/JP2012/058169 JP2012058169W WO2013145166A1 WO 2013145166 A1 WO2013145166 A1 WO 2013145166A1 JP 2012058169 W JP2012058169 W JP 2012058169W WO 2013145166 A1 WO2013145166 A1 WO 2013145166A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- authentication
- biometric
- sensor
- processing unit
- image
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/12—Fingerprints or palmprints
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/98—Detection or correction of errors, e.g. by rescanning the pattern or by human intervention; Evaluation of the quality of the acquired patterns
- G06V10/993—Evaluation of the quality of the acquired pattern
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/70—Multimodal biometrics, e.g. combining information from different biometric modalities
Definitions
- the present invention relates to a biometric authentication device, a biometric authentication method, and a biometric authentication program.
- Biometric authentication is performed by comparing verification data acquired by a biometric sensor with registered data registered in a database.
- biometric authentication has been diversified, and Patent Documents 1 to 3 disclose techniques related to diversified biometric authentication.
- Patent Documents 1 to 3 there is a risk that authentication accuracy may be lowered when a plurality of biometric sensors are used.
- the present invention has been made in view of the above problems, and an object thereof is to provide a biometric authentication device, a biometric authentication method, and a biometric authentication program that can suppress a decrease in authentication accuracy when a plurality of biometric sensors are used. To do.
- a biometric authentication device in the specification includes a first biometric sensor that acquires user biometric information and a second biometric information that is less reproducible than the first biometric sensor.
- a biometric sensor and an authentication processing unit that performs authentication by verification using biometric information acquired by the first and second biosensors, wherein the authentication processing unit is acquired by the first biosensor.
- the biometric information acquired by the second biometric sensor is collated using the biometric information when the biometric information and the registered information are successfully collated.
- the biometric authentication method disclosed in the specification includes a first authentication processing step of performing authentication by comparing biometric information acquired by a first biometric sensor that acquires biometric information of a user with registration information.
- the biometric authentication program disclosed in the specification is a first authentication that performs authentication on a computer by comparing biometric information acquired by a first biometric sensor that acquires biometric information of a user with registered information.
- a second biometric sensor that acquires biometric information of a user with lower reproducibility than the first biometric sensor is acquired using the biometric information when the verification in the processing step and the first authentication processing step is successful.
- biometric authentication device biometric authentication method, and biometric authentication program disclosed in the specification, it is possible to suppress a decrease in authentication accuracy when a plurality of biometric sensors are used.
- (A) is a block diagram for demonstrating the hardware constitutions of the biometrics apparatus which concerns on Example 1
- (b) is a schematic diagram for demonstrating an example of the biosensor which detects a palm vein.
- (A) is a block diagram of each function implement
- (b) is an example of the table showing the registration data registered into the registration database. It is an example of the flowchart performed in the case of the authentication process at the time of using a biometric sensor. It is a figure for demonstrating the example of a production
- FIG. 9 is a block diagram for explaining a hardware configuration of a biometric authentication apparatus according to a second embodiment. It is a block diagram of each function implement
- FIG. 1 is a diagram for explaining an example of a scene to which the following embodiment is applied.
- sensors for entering / exiting management type A sensor
- sensors type B sensor
- Biometric authentication is performed by comparing the biometric information acquired by the biometric sensor with the registered information registered in the registration database.
- type A sensors require high-precision authentication. Therefore, for type A sensors, for example, large-size and high-cost sensors are used, although the authentication accuracy and external light resistance are high.
- a sensor built in each desktop PC, notebook PC, etc. in the office is limited in size. Moreover, since the number of introductions increases, cost cut is required. Therefore, a small and inexpensive sensor is used for the type B sensor. In this case, the authentication accuracy at the sensor in the office is lowered. There is a need to improve the authentication accuracy of such type B sensors. Therefore, in the following embodiments, a biometric authentication device, a biometric authentication method, and a biometric authentication program that can suppress a decrease in authentication accuracy when a plurality of biometric sensors are used will be described.
- FIG. 2A is a block diagram for explaining a hardware configuration of the biometric authentication device 400 according to the first embodiment.
- the biometric authentication device 400 has a configuration in which a terminal 100, a terminal 200, and a storage device 300 are connected via a network.
- a network in this case, a communication network such as an intranet, a public line network, or the Internet can be used.
- the terminal 100 is a device that determines permission to enter the office, for example, and the terminal 100 is a PC terminal or the like disposed in the office.
- the terminal 100 includes a CPU 101, a RAM 102, a storage device 103, a first biological sensor 104, a display device 105, a communication unit 106, and the like. Each of these devices is connected by a bus or the like.
- the terminal 200 includes a CPU 201, a RAM 202, a storage device 203, a second biosensor 204, a display device 205, a communication unit 206, and the like. Each of these devices is connected by a bus or the like.
- CPU Central Processing Unit
- 201 is a central processing unit.
- the CPUs 101 and 201 include one or more cores.
- a RAM (Random Access Memory) 102 is a volatile memory that temporarily stores programs executed by the CPU 101, data processed by the CPU 101, and the like.
- the RAM 202 is a volatile memory that temporarily stores programs executed by the CPU 201, data processed by the CPU 201, and the like.
- Storage devices 103 and 203 are nonvolatile storage devices. As the storage devices 103 and 203, for example, a solid state drive (SSD) such as a ROM (Read Only Memory) or a flash memory, a hard disk driven by a hard disk drive, or the like can be used.
- the display device 105 is a device for displaying the result of each process performed by the terminal 100.
- the display device 205 is a device for displaying the result of each process performed by the biometric authentication device 200.
- the display devices 105 and 205 are, for example, liquid crystal displays.
- Communication units 106 and 206 are interfaces for transmitting and receiving signals to and from other devices.
- the first biological sensor 104 and the second biological sensor 204 are sensors that acquire user's biological information.
- the biometric information of the user is not particularly limited, and is information related to any living body such as a fingerprint, a vein, an iris, a voiceprint, and a shape.
- sensors that detect palm veins as images are used as the first biological sensor 104 and the second biological sensor 204.
- FIG. 2B is a schematic diagram for explaining an example of the first biological sensor 104 that detects a palm vein.
- the first biological sensor 104 is a photographing device that photographs the subcutaneous vein of the palm using near infrared rays having high permeability to the human body.
- the first biological sensor 104 is provided with, for example, a CMOS (Complementary Metal Oxide Semiconductor) camera.
- the illumination etc. which irradiate the light containing a near infrared ray may be provided.
- the second biosensor 204 has the same configuration as the first biosensor 104.
- the first biological sensor 104 has higher reproducibility than the second biological sensor 204 when acquiring the biological information of the user. That is, the similarities of the biological information acquired by the first biological sensor 104 a plurality of times are higher than the similarities of the biological information acquired by the second biological sensor 204 a plurality of times.
- the first biological sensor 104 and the second biological sensor 204 are biological image acquisition sensors, the first biological sensor 104 has a device configuration that realizes higher image quality than the second biological sensor 204.
- the image quality of an image that satisfies the following requirements is referred to as high image quality
- the image quality of an image that does not satisfy the following requirements is referred to as low image quality.
- the image quality of an image with less noise may be referred to as high image quality. This is because the smaller the noise, the more stable the authentication feature can be extracted, and the higher the authentication accuracy.
- the noise of the image is determined by the light quantity of the light source and the lens performance. For example, noise is reduced when the amount of light is large.
- an F value of an index representing brightness can be used.
- the image quality of an image with small peripheral dimming may be referred to as high image quality. This is because the edge feature of the captured image can be stably extracted when the peripheral light attenuation is small.
- the authentication part extraction process is stable, the authentication accuracy is improved.
- the clipping process is, for example, a process of clipping a palm region that is an authentication target from an image. Ambient dimming is determined by lens performance, illumination uniformity, and the like.
- the image quality of a high resolution image may be referred to as high image quality. This is because if the resolution of the image is high, fine features can be photographed, and the authentication accuracy increases.
- the resolution of an image is determined by the performance of an image sensor (CMOS or CCD), the performance of a transmission path (for example, USB transfer capability), and the like. Regarding the performance of the transmission path, even if high-resolution data can be acquired by the image sensor, the high-resolution data generally has a large data capacity, so it is difficult to actually use it if the capacity of the transmission path is small. Because of the circumstances.
- the image quality of a high resolution image may be referred to as high image quality.
- the resolution is an index indicating how much information can be distinguished. For example, it is an index for photographing a pair of white and black lines and determining whether a fine line pair can be correctly identified as a line pair. Specifically, when the resolution is low, white and black are mixed and appear gray. The resolution depends on the performance of the lens together with the resolution.
- the MTF (Modulation Transfer Function) of the lens may be used as an index of resolution.
- the image quality of an image with less surface reflection may be referred to as high image quality. This is because if the surface reflection is small, the vein image of the internal feature can be clearly captured and the authentication accuracy is improved.
- the surface reflection component is a reflection component generated at the boundary between the palm and air. Surface reflection can be reduced by mounting a polarizing plate. Moreover, surface reflection can be reduced by the arrangement of the light source and the lighting method.
- the storage device 300 is a nonvolatile storage device.
- a solid state drive such as a ROM or a flash memory, a hard disk driven by a hard disk drive, or the like can be used.
- the storage device 300 stores a biometric authentication program.
- the biometric authentication program may be distributed and stored in the storage devices 103 and 203.
- the biometric authentication program stored in the storage device 300 is expanded in the RAMs 102 and 202 so as to be executable.
- the CPU 101 executes a biometric authentication program expanded in the RAM 102.
- the CPU 201 executes a biometric authentication program developed in the RAM 202. Thereby, each process by the terminal 100 is performed.
- FIG. 3A is a block diagram of each function realized by executing the biometric authentication program.
- the terminal 100 by executing the biometric authentication program, the terminal 100 causes the overall control unit 11, the imaging processing unit 12, the authentication processing unit 13, the template acquisition unit 14, the temporary template generation unit 15, and the image processing. It functions as the unit 16.
- the terminal 200 By executing the biometric authentication program, the terminal 200 functions as the overall control unit 21, the imaging processing unit 22, the authentication processing unit 23, the template acquisition unit 24, and the template cache unit 25.
- the storage device 300 functions as the registration database 30 by executing the biometric authentication program.
- the overall control unit 11 controls the imaging processing unit 12, the authentication processing unit 13, the template acquisition unit 14, the temporary template generation unit 15, and the image processing unit 16.
- the imaging processing unit 12 acquires a palm image from the first biological sensor 104 in accordance with an instruction from the overall control unit 11.
- the authentication processing unit 13 extracts the vein information of the palm as an authentication feature from the palm image acquired by the imaging processing unit 12 in accordance with an instruction from the overall control unit 11, and performs authentication processing.
- the template acquisition unit 14 acquires a registration template (registration data) from the registration database 30 for authentication in the authentication processing unit 13.
- the temporary template generation unit 15 generates a temporary template from the authentication feature extracted by the authentication processing unit 13.
- the image processing unit 16 processes the image acquired by the imaging processing unit 12.
- the overall control unit 21 controls the photographing processing unit 22, the authentication processing unit 23, the template acquisition unit 24, and the template cache unit 25.
- the imaging processing unit 22 acquires a palm image from the second biological sensor 204 in accordance with an instruction from the overall control unit 21.
- the authentication processing unit 23 extracts the vein information of the palm as an authentication feature from the palm image acquired by the imaging processing unit 22 in accordance with an instruction from the overall control unit 21, and performs authentication processing.
- the template acquisition unit 24 acquires a registration template (registration data) from the registration database 30 for authentication in the authentication processing unit 23.
- the template cache unit 25 temporarily stores the temporary template generated by the temporary template generation unit 15.
- FIG. 3B is an example of a table representing registration templates registered in the registration database 30.
- the registration template includes an authentication feature associated with each user ID.
- This registration template can be registered in advance in the registration database 30 using the first biological sensor 104 or the like. Details of the authentication process will be described below.
- FIG. 4 is an example of a flowchart executed in the authentication process when the first biosensor 104 is used. This authentication process is performed, for example, when entering the office.
- the imaging processing unit 12 acquires a palm image I for authentication from the first biological sensor 104 (step S1).
- the authentication processing unit 13 extracts an authentication feature F from the palm image I (step S2).
- the authentication processing unit 13, an authentication feature F, the template acquisition section 14 collates the registration template T R obtained from the registration database 30, and calculates both the similarity S (step S3).
- the authentication processing unit 13 calculates the similarity between each user's registered template (T R1 to T RN ) and the authentication feature F.
- step S4 the authentication processing unit 13, the similarity S of any user is equal to or threshold TH 0 or more (step S4).
- the authentication process part 13 outputs the signal which concerns on authentication success (step S5). Thereby, for example, a door for entering the office is unlocked.
- the image processing unit 16 generates the image I ′ having a lower image quality than the image I by processing the palm image I that has been successfully authenticated (step S6).
- a sufficient amount of illumination light may not be obtained.
- the second biosensor 204 is built in a notebook PC, it is necessary to reduce the amount of light from the relationship of power consumption. As a result of reducing the amount of light, a lot of image noise is included.
- the size of an image sensor (CCD, CMOS, etc.) used to acquire an image is reduced. As the area of the image sensor decreases, the amount of light received per unit area decreases, resulting in increased noise. In general, small and inexpensive sensors often use small image sensors, and as a result, noise tends to increase.
- the image processing unit 16 reduces the difference between the image quality acquired by the first biosensor 104 and the image quality acquired by the second biosensor 204 to the image acquired by the first biosensor 104. Noise may be added.
- the image processing unit 16 measures the noise magnitude ⁇ that can express the difference between the sensors in advance, and adds the random noise determined based on the corresponding ⁇ to the pixel of the palm image I. Specifically, the image processing unit 16 acquires the image I ′ according to the following formula (1).
- I ′ (x) is a processed image I ′
- I (x) is a palm image I before processing
- I ′ (x) I (x) + N (x) (1)
- FIG. 5 is a diagram for explaining another example of the generation of the low-quality image I ′.
- the sensor area may be small and the irradiation area of the light source may be limited.
- the sensor area is small, it is difficult to irradiate light uniformly on the subject such as a palm.
- peripheral dimming to the image acquired by the first biological sensor 104, the difference between the image quality of the image acquired by the first biological sensor 104 and the image quality of the image acquired by the second biological sensor 204 is reduced. can do.
- FIG. 5 is a diagram for explaining another example of the generation of the low-quality image I ′.
- the image processing unit 16 represents a difference between the distribution of luminance values of the image acquired by the first biological sensor 104 and the distribution of luminance values of the image acquired by the second biological sensor 204. Obtain a dimming conversion curve in advance. Furthermore, the image processing unit 16 generates the low-quality image I ′ by applying the peripheral dimming conversion curve to the image acquired by the first biological sensor 104. Thereby, the difference between the image quality of the image acquired by the first biosensor 104 and the image quality of the image acquired by the second biosensor 204 is reduced. Specifically, the image processing unit 16 acquires the image I ′ according to the following formula (2).
- I ′ (x) is an image I ′ after processing
- I (x) is a palm image I before processing
- I ′ (x) a (x) ⁇ I (x) (2)
- FIG. 6 is a diagram for explaining another example of the generation of the low-quality image I ′.
- the second biological sensor 204 when an inexpensive sensor is used as the second biological sensor 204, the number of lenses, the lens thickness, and the like are restricted. In addition, the optical characteristics are reduced depending on the material of the lens. In this case, the sharpness (frequency characteristic) of the image acquired by the second biological sensor 204 is lowered. This sharpness can be expressed using the MTF value.
- the MTF value is a value expressing in the frequency domain how much the original image is degraded by the lens.
- the image processing unit 16 obtains in advance a “frequency degradation curve” that matches the frequency characteristics between sensors, and changes the frequency characteristics according to the frequency degradation curve. .
- the image processing unit 16 acquires the image I ′ according to the following formula (3).
- F ′ (f) is the Fourier transform of the processed image I ′
- F (f) is the Fourier transform of the palm image I before processing
- a (f) is the frequency. It is a deterioration curve.
- F ′ (f) a (f) ⁇ F (f) (3)
- the image processing unit 16 adds lens distortion to the palm image I acquired by the first biological sensor 104, whereby the image quality of the image acquired by the first biological sensor 104 and the second biological sensor 204 acquire it. It is possible to reduce the difference from the image quality of the processed image.
- FIG. 7 is a diagram for explaining another example of the generation of the low-quality image I ′.
- the image quality of the image acquired by the first biological sensor 104 and the image quality of the image acquired by the second biological sensor 204 are added by adding high-frequency noise to a region where the luminance value of the palm image acquired by the second biological sensor 204 is high. Can be reduced. Since the surface reflection is a reflection in which the incident angle and the reflection angle are equal, it is highly likely that the surface reflection occurs in the “convex region” of the palm with reference to FIG. Since this area has a high average luminance value, occurrence of surface reflection can be predicted based on the luminance value. Since high-frequency noise is generated in a region where surface reflection exists due to surface wrinkles or the like, the image processing unit 16 may perform processing for adding high-frequency noise to the corresponding region.
- the authentication processing unit 13 extracts an authentication feature from the image I ′, and the temporary template generation unit 15 sends the authentication feature to the template cache unit 25 as a temporary template T I ′ (step S7).
- the authentication process part 13 outputs the signal which concerns on authentication failure (step S8).
- FIG. 8 is an example of a flowchart executed in the authentication process when the second biosensor 204 is used.
- This authentication process is performed after successful authentication using the first biosensor 104. For example, it is performed at the time of BIOS authentication at the time of PC activation after entering the office, OS logon authentication, or the like.
- the imaging processing unit 22 acquires a palm image I for authentication from the second biosensor 204 (step S11).
- the authentication processing unit 23 extracts an authentication feature F from the palm image I (step S12).
- the authentication processing unit 23 includes an authentication feature F, the template acquisition section 24 collates the registration template T R obtained from the registration database 30, and calculates both the similarity S (step S13).
- the authentication processing unit 23 determines whether or not the similarity S is equal to or higher than a threshold value TH 2 ( ⁇ TH 1 ) (Step S15).
- the template acquisition unit 24 acquires the temporary template TI ′ from the template cache unit 5 (Step S16).
- the authentication processing unit 23 collates the authentication feature F with the temporary template T I ′ and calculates a similarity S ′ between them (step S17).
- the authentication processing unit 23 determines whether or not the similarity S ′ is greater than or equal to the threshold value TH 3 (step S18). If “Yes” is determined in step S18, or if “Yes” is determined in step S14, the authentication processing unit 23 outputs a signal related to the authentication success (step S19). If it is determined “No” in step S18 or “No” in step S15, the authentication processing unit 23 outputs a signal related to the authentication failure (step S20). After execution of step S19 or after execution of step S20, execution of the flowchart of FIG. 8 ends.
- the biometric information regarding the authentication success acquired by the first biosensor 104 with high reproducibility is obtained.
- highly reliable biological information can be used.
- the second biological sensor 204 is not required to have high reproducibility.
- an inexpensive apparatus can be used as the second biological sensor 204, the cost can be suppressed.
- a registration template may be required for each biosensor. This is because in the case of a low image quality sensor, it is greatly affected by slight posture fluctuations of a subject (such as a palm), and thus has a large influence on the authentication accuracy.
- a temporary template based on the biometric information related to the authentication success can be used, it is not necessary to individually create a registration template for each biosensor. Thereby, the amount of registered templates can be reduced.
- the cause of the authentication failure can be estimated as follows. First, after registering the registration template T R in the registration database 30 may over time until the actual authentication process. In this case, (1) fluctuation of the living body itself, (2) fluctuation of posture, fluctuation unique to collation such as noise, and the like cause authentication failure. As a factor of (1), the influence of the surface flaw etc. are mentioned, for example. On the other hand, the factor (1) is not easily included in the high-quality image collected on that day. For this reason, the cause of the authentication failure can be estimated by comparing the high-quality image acquired by the first biosensor 104 on the day (or a very close time interval) with the image acquired by the second biosensor 204. .
- FIG. 9 is a diagram for explaining an example of a processing flow for estimating the cause of the authentication failure.
- the imaging processing unit 22 acquires the palm image I for authentication from the second biometric sensor 204 (step S21).
- the authentication processing unit 23 an authentication feature F extracted from the palm image I, with the registered template T R the template acquisition section 24 acquires from the registration database 30 collates, calculates the similarity S 0 of both ( Step S22).
- the authentication processing unit 23 determines whether or not the similarity S 0 is greater than or equal to a threshold value TH 0 (step S23). If it is determined as “Yes” in step S23, the authentication processing unit 23 outputs a signal related to the authentication success (step S24). When it is determined as “No” in step S23, the template acquisition unit 24 acquires the temporary template TI ′ from the template cache unit 25 (step S25).
- the authentication processing unit 23 includes an authentication feature F, collates and temporary template T I', calculates both the similarity S T (step S26).
- the authentication processing unit 23, the similarity S T is equal to or more than the threshold value TH T (step S27).
- the authentication process part 23 will estimate that the fluctuation
- the authentication feature F and the temporary template T I ′ are collated. If the similarity in this case is high, it is estimated that the biometric feature has changed. This is because the similarity between the temporary template T I ′ and the authentication feature F is high, while the similarity between the previously registered template and the authentication feature F is low. At this time, an accurate determination can be made by performing image processing for absorbing the difference in image quality between sensors.
- the degree of influence of surface reflection varies from individual to individual. This is because surface reflection is greatly affected by the incident angle and the reflection angle of light, but the shape of the palm differs from person to person. Therefore, there are users who are strongly affected by surface reflection and users who are not so affected. The above effect can be obtained by applying the surface reflection addition process only to a specific user who is strongly influenced by the surface reflection.
- the presence or absence of application may be determined according to the history of authentication results. In other words, the authentication failure rate for a predetermined period or number of times is checked, and when the authentication failure rate exceeds a predetermined value, the image quality of the image at the time of successful authentication by the first biometric sensor 104 is reduced to make a temporary template. May be generated.
- the present invention is not limited thereto.
- a combination of the similarity S R between the authentication feature F and the registered template T R extracted from the image I obtained by the second biometric sensor 204, a similarity S T with the authentication feature F and temporary template T I' A fusion score (total similarity) may be used.
- the total similarity S TOTAL may be calculated according to the following equation (4).
- S TOTAL (1.0-w) S R + wS T (4)
- “w” is a coefficient representing the weight of each similarity. If you just registered in the registration template T R, differences between the registered template T R one o'clock template T I'is small. On the other hand, over time, differences between the registered template T R one o'clock template T I'increases. Therefore, “w” may be reduced immediately after registration, and “w” may be increased as time passes. In the example of FIG. 11, an upper limit (0.5) is set for “w”.
- the number of biosensors having a low reproducibility is not necessarily one.
- a plurality of types of low-quality biosensors may be provided.
- a biosensor 204a and a biosensor 204b are provided.
- a plurality of factors are mixed as image quality deterioration factors.
- the degradation of the lens MTF exists as a degradation factor in the biosensor 204a, and that there are two degradation factors of the biosensor 204b, that is, a decrease in lens MTF and noise. This may occur when the biosensor 204a and the biosensor 204b share a lens but have different image sensors.
- the processing may be made more efficient by sequentially applying the processing. Specifically, first, a temporary template T1 to which the MTF deterioration process is applied may be generated, and the temporary template T1 may be used for authentication by the biometric sensor 204a. Subsequently, a temporary template T2 obtained by applying noise addition processing to the temporary template T1 may be generated, and the temporary template T2 may be used for authentication by the biometric sensor 204b. By doing in this way, processing can be made more efficient than generating individual sensor images.
- the mode of performing the authentication process at each terminal has been described.
- the authentication process may be performed by an authentication server.
- an authentication server For example, a case where a high image quality sensor is used at the time of entry and a low image quality sensor is used for identity confirmation after entering the country corresponds. In such a case, authentication processing is performed collectively by the authentication server.
- FIG. 12 is a block diagram for explaining a hardware configuration of the biometric authentication device 400a according to the second embodiment.
- biometric authentication device 400a has a configuration in which terminal 100a, terminal 200a, storage device 300, and server 500 are connected via a network.
- the terminal 100a is an apparatus that performs authentication at the time of entry, for example.
- the terminal 200a is a terminal that is used, for example, for confirmation of identity after entering the country, and the server 500 is an authentication server that collectively performs authentication processing.
- the terminal 100a includes a CPU 101, a RAM 102, a storage device 103, a first biological sensor 104, a display device 105, a communication unit 106, and the like. Each of these devices is connected by a bus or the like.
- the terminal 200a includes a CPU 201, a RAM 202, a storage device 203, a second biosensor 204, a display device 205, a communication unit 206, and the like. Each of these devices is connected by a bus or the like.
- the server 500 includes a CPU 501, a RAM 502, a storage device 503, a communication unit 504, and the like. Each of these devices is connected by a bus or the like.
- the CPU 501 is a central processing unit.
- the RAM 502 is a volatile memory that temporarily stores programs executed by the CPU 501, data processed by the CPU 501, and the like.
- the storage device 503 is a nonvolatile storage device.
- the communication unit 504 is an interface for transmitting and receiving signals to and from other devices.
- the storage device 300 is a nonvolatile storage device and stores a biometric authentication program. The biometric authentication program may be distributed and stored in the storage devices 103, 203, and 503.
- the biometric authentication program stored in the storage device 300 is expanded in the RAMs 102, 202, and 502 so as to be executable.
- the CPU 101 executes a biometric authentication program expanded in the RAM 102.
- the CPU 201 executes a biometric authentication program developed in the RAM 202.
- the CPU 501 executes a biometric authentication program expanded in the RAM 502. Thereby, each process by the biometric authentication device 400a is executed.
- FIG. 13 is a block diagram of each function realized by executing the biometric authentication program.
- terminal 100 a functions as overall control unit 11, imaging processing unit 12, authentication processing unit 13, temporary template generation unit 15, and image processing unit 16.
- the terminal 200a functions as the overall control unit 21, the imaging processing unit 22, and the authentication processing unit 23.
- the storage device 300 functions as the registration database 30 by executing the biometric authentication program.
- the server 500 functions as the overall control unit 31, the authentication processing unit 32, the template acquisition unit 33, and the template cache unit 34 by executing the biometric authentication program.
- the authentication processing unit 13 extracts an authentication feature from the palm image acquired by the biometric sensor 104, and sends the authentication feature to the server 500 without performing a matching process.
- the authentication processing unit 23 extracts an authentication feature from the palm image acquired by the second biometric sensor 204 and sends the authentication feature to the server 500 without performing a matching process.
- the overall control unit 31 controls the authentication processing unit 32, the template acquisition unit 33, and the template cache unit 34.
- the authentication processing unit 32 performs authentication processing using the authentication feature sent from the authentication processing unit 13 and the authentication processing unit 23 in accordance with an instruction from the overall control unit 31.
- the template acquisition unit 33 acquires a template from the registration database 30 for authentication in the authentication processing unit 32.
- the template cache unit 34 temporarily stores the template generated by the temporary template generation unit 15.
- FIG. 14 is an example of a flowchart executed in the authentication process when the biosensor 104 is used. This authentication process is performed when entering Japan.
- the imaging processing unit 12 acquires a palm image I for authentication from the biometric sensor 104 (step S31).
- the authentication processing unit 13 extracts an authentication feature F from the palm image I (step S32).
- the authentication processing unit 13 sends the authentication feature F to the server 500 (step S33).
- the authentication processing unit 32 includes an authentication feature F, the template acquisition section 33 collates the registration template T R obtained from the registration database 30, and calculates both the similarity S (step S34).
- the authentication processing unit 32 the similarity S is equal to or threshold TH 0 or more (step S35). If it is determined as “Yes” in step S35, the authentication processing unit 32 outputs a result related to the authentication success (step S36). When it determines with "No” in step S35, the authentication process part 32 outputs the result which concerns on authentication failure (step S37).
- the authentication processing unit 13 receives the authentication result from the authentication processing unit 32 (step S38).
- the authentication processing unit 13 determines whether or not the received authentication result is successful (step S39). When it determines with "Yes” in step S39, the authentication process part 13 outputs the signal which concerns on authentication success (step S40).
- the image processing unit 16 processes the palm image I to generate an image I ′ having a lower image quality than the image I (step S41).
- the authentication processing unit 13 extracts an authentication feature from the image I ′, and the temporary template generation unit 15 sends the authentication feature to the server 500 as a temporary template T I ′ (step S42).
- the authentication process part 13 outputs the signal which concerns on authentication failure (step S43).
- FIG. 15 is an example of a flowchart executed in the authentication process when the second biosensor 204 is used.
- This authentication process is performed after the authentication process using the first biosensor 104 is successful. For example, it is performed at the time of identity verification after entering Japan.
- the imaging processing unit 22 acquires a palm image I for authentication from the second biosensor 204 (step S51).
- the authentication processing unit 23 extracts an authentication feature F from the palm image I (step S52).
- the authentication processing unit 23 sends the authentication feature F to the server 500 (step S53).
- the authentication processing unit 23 receives an authentication result from the server 500 (step S54).
- step S55 determines whether or not the received authentication result is successful. If it is determined as “Yes” in step S25, the authentication processing unit 23 outputs a signal related to the authentication success (step S56). If it is determined “No” in step S55, the authentication processing unit 23 outputs a signal related to the authentication failure (step S57).
- the flowchart of FIG. 15 ends after execution of step S56 and after execution of step S57.
- FIG. 16 is an example of a flowchart executed by the server 500 after execution of step S53 of FIG.
- authentication processing unit 32 receives authentication feature F from authentication processing unit 23 (step S61).
- the authentication processing unit 32 includes an authentication feature F
- the template acquisition section 33 collates the registration template T R obtained from the registration database 30, and calculates both the similarity S (step S62).
- the authentication processing unit 32 determines whether or not the similarity S is equal to or higher than a threshold value TH 2 ( ⁇ TH 1 ) (Step S64). If it is determined as “Yes” in step S64, the template acquisition unit 33 acquires the temporary template TI ′ from the template cache unit 34 (step S65).
- the authentication processing unit 32 collates the authentication feature F with the temporary template T I ′ and calculates a similarity S ′ between them (step S66).
- the authentication processing unit 32 determines whether or not the similarity S ′ is greater than or equal to the threshold value TH 3 (step S67). If “Yes” is determined in step S67, or if “Yes” is determined in step S64, the authentication processing unit 32 returns a signal related to the authentication success to the terminal 200a (step S68). If it is determined “No” in step S64, or if “No” is determined in step S67, the authentication processing unit 32 returns a signal related to the authentication failure to the terminal 200a (step S69). After execution of step S19 or after execution of step S20, execution of the flowchart of FIG.
- the biometric information regarding the authentication success acquired by the first biosensor 104 with high reproducibility is obtained.
- highly reliable biological information can be used.
- the second biological sensor 204 is not required to have high reproducibility.
- an inexpensive apparatus can be used as the second biological sensor 204, the cost can be suppressed.
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Human Computer Interaction (AREA)
- Quality & Reliability (AREA)
- Collating Specific Patterns (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
Description
図4は、第1生体センサ104を用いた場合の認証処理の際に実行されるフローチャートの一例である。この認証処理は、例えば、オフィスへの入室の際に実施される。撮影処理部12は、第1生体センサ104から認証用の手のひら画像Iを取得する(ステップS1)。次に、認証処理部13は、手のひら画像Iから認証特徴Fを抽出する(ステップS2)。次に、認証処理部13は、認証特徴Fと、テンプレート取得部14が登録データベース30から取得した登録テンプレートTRとを照合し、両者の類似度Sを算出する(ステップS3)。なお、認証処理部13は、各ユーザの登録テンプレート(TR1~TRN)と認証特徴Fとの類似度を算出する。
I´(x)=I(x)+N(x) (1)
I´(x)=a(x)・I(x) (2)
F´(f)=a(f)・F(f) (3)
上記各例では、全てのユーザに対して一時テンプレートを生成する構成について説明したが、それに限られない。例えば、一時テンプレートの生成を個人単位で有効または無効としてもよい。例えば、図10を参照して、周辺減光は手の大きい人で影響が大きい一方、手の小さい人にはあまり影響がない。したがって、手のひらが小さいユーザに対しては一時テンプレートの生成を抑制することによって画像処理によるCPU負荷や消費電力、記憶容量を削減する効果が得られる。
STOTAL=(1.0-w)SR+wST (4)
図14は、生体センサ104を用いた場合の認証処理の際に実行されるフローチャートの一例である。この認証処理は、入国時などに実施される。撮影処理部12は、生体センサ104から認証用の手のひら画像Iを取得する(ステップS31)。次に、認証処理部13は、手のひら画像Iから認証特徴Fを抽出する(ステップS32)。次に、認証処理部13は、認証特徴Fをサーバ500に送る(ステップS33)。
12 撮影処理部
13 認証処理部
14 テンプレート取得部
15 一時テンプレート生成部
16 画像加工部
21 全体制御部
22 撮影処理部
23 認証処理部
24 テンプレート取得部
25 テンプレートキャッシュ部
30 登録データベース
100 端末
104 第1生体センサ
200 端末
204 第2生体センサ
Claims (7)
- ユーザの生体情報を取得する第1生体センサと、
前記第1生体センサよりも低い再現性でユーザの生体情報を取得する第2生体センサと、
前記第1生体センサおよび前記第2生体センサが取得した生体情報を用いた照合によって認証を行う認証処理部と、を備え、
前記認証処理部は、前記第1生体センサが取得した生体情報と登録情報との照合が成功した際の当該生体情報を利用して、前記第2生体センサが取得した生体情報を照合することを特徴とする生体認証装置。 - 前記第1生体センサが取得した生体情報と登録情報との照合が成功した際の当該生体情報を加工する加工部を備え、
前記認証処理部は、前記加工部による加工によって得られた加工生体情報と、前記第2生体センサが取得した生体情報との照合によって認証を行うことを特徴とする請求項1記載の生体認証装置。 - 前記認証処理部は、前記第2生体センサが取得した生体情報と前記登録情報との照合による認証が失敗した場合に、前記加工生体情報と、前記第2生体センサが取得した生体情報との照合によって認証を行うことを特徴とする請求項2記載の生体認証装置。
- 前記第1生体センサおよび前記第2生体センサは、生体画像を取得するセンサであり、
前記加工部は、前記第1生体センサが取得した生体画像の画質を低下させる処理を行うことを特徴とする請求項2または3記載の生体認証装置。 - 前記認証処理部は、前記第1生体センサが取得した生体情報と登録情報との照合が成功した際の当該生体情報と前記第2生体センサが取得した生体情報との照合に、前記第2生体センサが取得した生体情報と前記登録情報との照合結果を反映させることを特徴とする請求項1~4のいずれか一項に記載の生体認証装置。
- ユーザの生体情報を取得する第1生体センサが取得した生体情報と、登録情報との照合によって認証を行う第1認証処理ステップと、
前記第1認証処理ステップにおける照合が成功した場合の当該生体情報を利用して、前記第1生体センサよりも低い再現性でユーザの生体情報を取得する第2生体センサが取得した生体情報を照合することによって認証を行う第2認証処理ステップと、を含むことを特徴とする生体認証方法。 - コンピュータに、
ユーザの生体情報を取得する第1生体センサが取得した生体情報と、登録情報との照合によって認証を行う第1認証処理ステップと、
前記第1認証処理ステップにおける照合が成功した場合の当該生体情報を利用して、前記第1生体センサよりも低い再現性でユーザの生体情報を取得する第2生体センサが取得した生体情報を照合することによって認証を行う第2認証処理ステップと、を実行させることを特徴とする生体認証プログラム。
Priority Applications (6)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
EP12873357.3A EP2833320B1 (en) | 2012-03-28 | 2012-03-28 | Biometric authentication device, biometric authentication method, and biometric authentication program |
JP2014507134A JP5850138B2 (ja) | 2012-03-28 | 2012-03-28 | 生体認証装置、生体認証方法、および生体認証プログラム |
CN201280071769.1A CN104221052B (zh) | 2012-03-28 | 2012-03-28 | 生物体认证装置、生物体认证方法、以及生物体认证程序 |
KR1020147026224A KR101620774B1 (ko) | 2012-03-28 | 2012-03-28 | 생체 인증 장치, 생체 인증 방법 및 기억 매체 |
PCT/JP2012/058169 WO2013145166A1 (ja) | 2012-03-28 | 2012-03-28 | 生体認証装置、生体認証方法、および生体認証プログラム |
US14/494,944 US9208391B2 (en) | 2012-03-28 | 2014-09-24 | Biometric authentication device, biometric authentication method, and computer readable, non-transitory medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/JP2012/058169 WO2013145166A1 (ja) | 2012-03-28 | 2012-03-28 | 生体認証装置、生体認証方法、および生体認証プログラム |
Related Child Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US14/494,944 Continuation US9208391B2 (en) | 2012-03-28 | 2014-09-24 | Biometric authentication device, biometric authentication method, and computer readable, non-transitory medium |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2013145166A1 true WO2013145166A1 (ja) | 2013-10-03 |
Family
ID=49258532
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2012/058169 WO2013145166A1 (ja) | 2012-03-28 | 2012-03-28 | 生体認証装置、生体認証方法、および生体認証プログラム |
Country Status (6)
Country | Link |
---|---|
US (1) | US9208391B2 (ja) |
EP (1) | EP2833320B1 (ja) |
JP (1) | JP5850138B2 (ja) |
KR (1) | KR101620774B1 (ja) |
CN (1) | CN104221052B (ja) |
WO (1) | WO2013145166A1 (ja) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2015100210A1 (en) * | 2013-12-28 | 2015-07-02 | Intel Corporation | Extending user authentication across a trust group of smart devices |
Families Citing this family (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9197637B2 (en) * | 2011-07-08 | 2015-11-24 | Research Foundation Of The City University Of New York | Method of comparing private data without revealing the data |
CH707230B1 (de) * | 2012-11-20 | 2016-02-29 | Frank Türen Ag | Türsystem mit berührungsloser Zutrittskontrolle und berührungsloser Türbedienung. |
JP6531395B2 (ja) * | 2015-01-08 | 2019-06-19 | 富士通株式会社 | 三次元形状測定装置、三次元形状測定方法、及び、プログラム |
US10740335B1 (en) | 2016-01-15 | 2020-08-11 | Accenture Global Solutions Limited | Biometric data combination engine |
JP6717068B2 (ja) * | 2016-06-13 | 2020-07-01 | コニカミノルタ株式会社 | 情報処理端末、情報処理システム、プログラム、および制御方法 |
US11275820B2 (en) * | 2019-03-08 | 2022-03-15 | Master Lock Company Llc | Locking device biometric access |
US11495041B2 (en) * | 2019-03-29 | 2022-11-08 | Jumio Corporation | Biometric identification using composite hand images |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2001014462A (ja) | 1999-07-02 | 2001-01-19 | Mitsubishi Electric Corp | 指紋照合システム |
JP2001084371A (ja) * | 1999-09-14 | 2001-03-30 | Fujitsu Ltd | 生体情報を用いた個人認証システム |
JP2004021491A (ja) * | 2002-06-14 | 2004-01-22 | Canon Inc | 生体パターン認証装置、生体パターン認証方法及びこの方法を記述したプログラム |
JP2006107340A (ja) | 2004-10-08 | 2006-04-20 | Fujitsu Ltd | 生体情報認証装置及び方法,並びに生体情報認証プログラム及び生体情報認証プログラムを記録したコンピュータ読取可能な記録媒体 |
JP2009169796A (ja) | 2008-01-18 | 2009-07-30 | Hitachi Omron Terminal Solutions Corp | システム管理装置、およびセキュリティシステム |
Family Cites Families (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR100342159B1 (ko) | 1996-08-25 | 2002-06-27 | 센사르 인코포레이티드 | 홍채영상 포착장치 및 홍채영상 포착방법 |
JP3877959B2 (ja) | 2000-12-28 | 2007-02-07 | 花王株式会社 | 肌色測定装置および肌画像処理装置 |
US20080177994A1 (en) * | 2003-01-12 | 2008-07-24 | Yaron Mayer | System and method for improving the efficiency, comfort, and/or reliability in Operating Systems, such as for example Windows |
JP2006202212A (ja) * | 2005-01-24 | 2006-08-03 | Konica Minolta Business Technologies Inc | 個人認証装置、情報処理装置及び個人認証システム |
JP4929804B2 (ja) * | 2006-04-10 | 2012-05-09 | 富士通株式会社 | 認証方法、認証装置および認証プログラム |
US8953849B2 (en) * | 2007-04-19 | 2015-02-10 | Eyelock, Inc. | Method and system for biometric recognition |
CN102460507B (zh) * | 2009-06-17 | 2014-12-03 | 富士通株式会社 | 生物体认证装置、生物体认证方法以及生物体认证用计算机程序 |
JP5422326B2 (ja) * | 2009-09-30 | 2014-02-19 | 株式会社日立製作所 | 生体認証装置 |
JP5810581B2 (ja) * | 2011-03-29 | 2015-11-11 | 富士通株式会社 | 生体情報処理装置、生体情報処理方法、および生体情報処理プログラム |
-
2012
- 2012-03-28 WO PCT/JP2012/058169 patent/WO2013145166A1/ja active Application Filing
- 2012-03-28 KR KR1020147026224A patent/KR101620774B1/ko active IP Right Grant
- 2012-03-28 CN CN201280071769.1A patent/CN104221052B/zh active Active
- 2012-03-28 JP JP2014507134A patent/JP5850138B2/ja not_active Expired - Fee Related
- 2012-03-28 EP EP12873357.3A patent/EP2833320B1/en active Active
-
2014
- 2014-09-24 US US14/494,944 patent/US9208391B2/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2001014462A (ja) | 1999-07-02 | 2001-01-19 | Mitsubishi Electric Corp | 指紋照合システム |
JP2001084371A (ja) * | 1999-09-14 | 2001-03-30 | Fujitsu Ltd | 生体情報を用いた個人認証システム |
JP2004021491A (ja) * | 2002-06-14 | 2004-01-22 | Canon Inc | 生体パターン認証装置、生体パターン認証方法及びこの方法を記述したプログラム |
JP2006107340A (ja) | 2004-10-08 | 2006-04-20 | Fujitsu Ltd | 生体情報認証装置及び方法,並びに生体情報認証プログラム及び生体情報認証プログラムを記録したコンピュータ読取可能な記録媒体 |
JP2009169796A (ja) | 2008-01-18 | 2009-07-30 | Hitachi Omron Terminal Solutions Corp | システム管理装置、およびセキュリティシステム |
Non-Patent Citations (1)
Title |
---|
See also references of EP2833320A4 |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2015100210A1 (en) * | 2013-12-28 | 2015-07-02 | Intel Corporation | Extending user authentication across a trust group of smart devices |
US9684778B2 (en) | 2013-12-28 | 2017-06-20 | Intel Corporation | Extending user authentication across a trust group of smart devices |
Also Published As
Publication number | Publication date |
---|---|
EP2833320A1 (en) | 2015-02-04 |
EP2833320A4 (en) | 2015-06-24 |
KR20140138185A (ko) | 2014-12-03 |
US9208391B2 (en) | 2015-12-08 |
JPWO2013145166A1 (ja) | 2015-08-03 |
US20150010217A1 (en) | 2015-01-08 |
CN104221052B (zh) | 2017-05-10 |
KR101620774B1 (ko) | 2016-05-12 |
JP5850138B2 (ja) | 2016-02-03 |
EP2833320B1 (en) | 2018-04-18 |
CN104221052A (zh) | 2014-12-17 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP5850138B2 (ja) | 生体認証装置、生体認証方法、および生体認証プログラム | |
TWI706264B (zh) | 證件的驗證方法、裝置、電子設備和儲存介質 | |
US10049361B2 (en) | Dynamic authentication technology | |
US10268910B1 (en) | Authentication based on heartbeat detection and facial recognition in video data | |
US10135817B2 (en) | Enhancing authentication and source of proof through a dynamically updatable biometrics database | |
US11348370B2 (en) | Iris authentication device, iris authentication method, and recording medium | |
JP6028453B2 (ja) | 画像処理装置、画像処理方法および画像処理プログラム | |
US20110206244A1 (en) | Systems and methods for enhanced biometric security | |
CN110348195A (zh) | 数据凭证授权方法、装置、计算机设备和存储介质 | |
KR101622065B1 (ko) | 생체 인증 장치, 생체 인증 방법 및 기억 매체 | |
US20230334136A1 (en) | Adhoc enrollment process | |
US20220044014A1 (en) | Iris authentication device, iris authentication method and recording medium | |
JP2006085289A (ja) | 顔認証システムおよび顔認証方法 | |
JP6432634B2 (ja) | 認証装置、認証方法、及びプログラム | |
KR102570071B1 (ko) | 라이브니스 검사 방법 및 장치, 얼굴 인증 방법 및 장치 | |
Chhabra et al. | Low quality iris detection in smart phone: a survey | |
US20240046709A1 (en) | System and method for liveness verification | |
WO2023084667A1 (ja) | 認証装置、エンジン生成装置、認証方法、エンジン生成方法、及び、記録媒体 | |
Lunerti et al. | Selfies for Mobile Biometrics: Sample Quality in Unconstrained Environments | |
Singh et al. | Biometric Methods of Face Recognition: A Mirror Review | |
KR20210050650A (ko) | 모바일 기기의 페이스 인증을 통한 잠금 해제 방법 | |
KR20210001270A (ko) | 블러 추정 방법 및 장치 | |
Wenhan et al. | Cloud-based face recognition algorithm | |
Nandhini et al. | Multimodal biometric analysis against spoofing attacks using image quality assessment |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 12873357 Country of ref document: EP Kind code of ref document: A1 |
|
ENP | Entry into the national phase |
Ref document number: 2014507134 Country of ref document: JP Kind code of ref document: A |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2012873357 Country of ref document: EP |
|
ENP | Entry into the national phase |
Ref document number: 20147026224 Country of ref document: KR Kind code of ref document: A |
|
NENP | Non-entry into the national phase |
Ref country code: DE |