WO2019124080A1

WO2019124080A1 - Authentication device and authentication method

Info

Publication number: WO2019124080A1
Application number: PCT/JP2018/044803
Authority: WO
Inventors: 大輔本田; 貴司中野; 信夫山崎; 幸夫玉井; 足立　佳久; 莉絵子小川
Original assignee: シャープ株式会社
Priority date: 2017-12-20
Filing date: 2018-12-05
Publication date: 2019-06-27

Abstract

Provided are an authentication device and authentication method with which it is possible to accurately authenticate an individual. This authentication device is provided with: an image acquisition unit (120) for acquiring an image including an iris area of a subject's eye; a living body detection unit (130) for determining whether or not a pulse wave is detected in the iris area in the image; and an authentication unit (140) for performing iris authentication for the image. The authentication unit (140) refers to the result of determination by the living body detection unit (130) and the result of iris authentication by the iris authentication unit (140) and outputs an authentication result.

Description

Authentication device and authentication method

One aspect of the present disclosure relates to an authentication device that performs iris authentication, and an authentication method.

BACKGROUND Conventionally, an authentication device that authenticates an individual from an image of an iris of a human eye is known. In these authentication devices, for example, in order to prevent false authentication due to so-called "spoofing" using an artificial object such as a artificial iris, a biological detection function is provided to identify whether the iris is an artificial object or a living thing. (See, for example, Patent Document 1).

In the authentication device described in Patent Document 1, an image of the user's face is acquired, and using the acquired image, pulse waves among a plurality of parts of the face area such as the user's lips, cheeks, forehead, and hair Whether or not the face area is a living body is determined by comparing the amplitudes of the waveforms.

Japanese patent publication "Japanese Patent Laid-Open No. 2014-184002 (October 2, 2014 published)"

However, with an authentication device combining personal authentication with the iris of the eye and living body detection with the pulse wave of the face, it is possible to forge the combination of an artificial object such as an iris picture and the face of another person who is a living body. It was difficult to prevent false authentication due to so-called impersonation.

One aspect of the present disclosure is made in view of the above-described circumstances, and an object of the present disclosure is to provide an authentication device capable of performing personal authentication with high accuracy.

In order to solve the above problems, an authentication device according to an aspect of the present disclosure includes an image acquisition unit that acquires an image including an iris region of a subject's eye, and presence or absence of pulse wave detection in the iris region of the image. A biometric detection unit to be determined, and an authentication unit for performing iris authentication of the image, the authentication unit refers to the determination result by the biometric detection unit and the iris authentication result by the authentication unit. It is the structure to output.

Moreover, in order to solve the above-mentioned subject, the authentication device according to an aspect of the present disclosure includes an image acquisition unit that acquires an image including an iris region of an eye of a subject, and analyzing the image. A determination unit that determines whether biological detection can be performed using pupil contraction according to the acquired pupil opening degree, and the determination unit performs biological detection using pupil contraction The first living body detection unit performs living body detection using the presence or absence of pulse wave detection in the iris region of the image when it is determined that it can not be performed, and the judgment unit can perform living body detection using pupil contraction. And a second living body detection unit that performs living body detection using pupil contraction generated by irradiating visible light to the eyes of the subject, and an authentication unit that performs iris authentication of the image. The authentication unit includes the first living body detection Or a detection result by the second biological detection portion, with reference to the iris authentication result by the authentication unit is configured to output the authentication result.

Further, in order to solve the above-described problem, an authentication method according to an aspect of the present disclosure includes an image acquisition step of acquiring an image including an iris region of an eye of a subject, and pulse wave detection in the iris region of the image. A method of outputting an authentication result with reference to a judgment result in the living body detection step and an iris authentication result in the authentication step, including a living body detection step for determining presence / absence and an authentication step for iris authentication of the image. It is.

According to one aspect of the present disclosure, personal identification can be performed with high accuracy.

FIG. 1 is a block diagram showing a schematic configuration of an authentication device 100 according to a first embodiment. Is a diagram showing an oxyhemoglobin (HbO ₂₎ and the wavelength dependence of the absorbance of melanin. It is a figure which shows typically the iris area | region of a subject's eye. (A) is a figure which shows typically the structure of an iris area | region, (b) is a figure which showed the formation aspect of the blood vessel in an iris area | region typically. (A) is a figure showing a rectangular section, (b) shows a frequency power spectrum in a rectangular section area including the crimp wheel 15, (c) is a frequency in a rectangular section area including only skin A power spectrum is shown, and (d) is a figure showing a relation between a pulse wave extraction place and a pulse wave detection accuracy. It is a figure which shows the structure of a human eye typically. It is a flowchart which shows the flow of a pulse wave detection process. It is a flow chart which shows a flow of living body determination processing. FIG. 7 is a block diagram showing a schematic configuration of an authentication device according to a second embodiment. FIG. 7 is a block diagram showing a schematic configuration of an authentication device according to a third embodiment. It is a figure for demonstrating the calculation method of pupil opening degree. It is a figure which shows the relationship between the ratio of a pupil diameter and an iris diameter, and illumination intensity. FIG. 16 is a block diagram showing a schematic configuration of an authentication device 400 according to a fourth embodiment. It is a figure which shows the waveform of an acceleration pulse wave.

Embodiment 1
The first embodiment of the present disclosure will be described in detail below. FIG. 1 is a block diagram showing a schematic configuration of the authentication device 100 according to the first embodiment. As will be described in detail below, the authentication device 100 authenticates the target person by the iris authentication technology using the iris image of the eye of the authentication target person, and detects the pulse wave in the iris region, It is determined whether or not the subject is a living body. The authentication apparatus 100 may be configured to perform authentication using images of the irises of both eyes of the authentication target person, or may be configured to perform authentication using the image of the iris of one eye of the authentication target person. Good.

As shown in FIG. 1, the authentication device 100 includes an image acquisition unit 120, a living body detection unit 130, and an authentication unit 140. The authentication device 100 further includes an imaging unit 110. The image acquisition unit 120, the living body detection unit 130, the authentication unit 140, and the imaging unit 110 are not limited to the configuration integrally provided in the authentication device 100, but may be separate from the authentication device 100. May be In the configuration in which the image acquisition unit 120, the living body detection unit 130, the authentication unit 140, and the imaging unit 110 are separate from the authentication device 100, for example, the configuration may be configured to communicate with the authentication device 100 via wireless communication.

(Configuration of image acquisition unit 120)
The image acquisition unit 120 acquires an image including the iris area of the eye of the authentication target person captured by the imaging unit 110. The imaging unit 110 is provided with a filter suitable for observing increase / decrease in blood volume in the artery, that is, change in concentration information of oxygenated hemoglobin. FIG. 2 shows the absorbance wavelength dependence of oxygenated hemoglobin (HbO ₂ ) and melanin. As for the wavelength selection of the filter, it is desirable to select a wavelength at which the sensor sensitivity is high and the absorbance of oxyhemoglobin is relatively high with respect to the absorbance of melanin. The imaging unit 110 is preferably, for example, a camera having sensitivity to a wavelength of 500 to 600 nm.

The imaging unit 110 may be, for example, a general silicon sensor camera provided with an RGB filter. The RGB filter included in the imaging unit 110 has, for example, a configuration in which the G pixel has sensitivity to a wavelength of 500 nm to 600 nm, the R pixel has sensitivity to a wavelength of 600 nm to 700 nm, and the B pixel has sensitivity to a wavelength of 400 nm to 500 nm can do.

The imaging unit 110 continuously captures an image including the iris area of the subject's eye at a predetermined time interval (frame rate) based on the intensity of light transmitted through each of the RGB filters. The frame rate of the imaging unit 110 can be, for example, 30 fps.

The image acquisition unit 120 acquires a plurality of captured images that are continuously captured by the imaging unit 110 and that include the iris region of the subject's eyes. The image acquisition unit 120 supplies the acquired plurality of captured images to the living body detection unit 130.

(Configuration of living body detection unit 130)
The living body detection unit 130 analyzes the plurality of captured images acquired by the image acquisition unit 120 to confirm the presence or absence of pulse wave detection in the iris region of the subject's eye, and determines whether the subject is a living body or not. judge. The living body detection unit 130 provides the authentication unit 140 with an image including the iris area of the eye of the subject, along with the determination result as to whether the subject is a living body. The living body detection unit 130 may not be configured to provide the authentication unit 140 with an image including the iris region of the eye of the subject when it is determined that the subject is not a living body.

The living body detection unit 130 includes a detection area setting unit 131, a pixel value calculation unit 132, a pulse wave detection unit 133, and a living body determination unit 134.

The detection area setting unit 131 sets a detection area for detecting a pulse wave in each of a plurality of captured images including the iris area of the subject's eyes. The detection area setting unit 131 detects the iris area of the eye of the subject in the captured image acquired by the image acquisition unit 120 for each frame. FIG. 3 is a diagram showing an eye including the iris area of the subject detected by the detection area setting unit 131 from the captured image. The iris region is a doughnut-shaped band region whose outer edge is the outer edge of the pupil 5 and whose outer edge is the boundary with the sclera (white eyes). As shown in FIG. 3, the detection area setting unit 131 further detects an area including the crimped ring 15 present at the boundary between the large iris ring and the small iris ring with respect to the iris area detected from the captured image. Set as.

FIG. 4A is a diagram showing the configuration of the iris region of the human eye, and the diagram shown in the lower part is a cross-sectional view taken along line II of the diagram shown in the upper part, showing the iris and the lens 30 in cross section. Is shown. FIG. 4 (b) is a view schematically showing the formation of blood vessels in the iris region. As shown in FIG. 4 (a), the iris 10 of the human eye includes a small iris ring 12 which is a concentric narrow band having a pupil 5 as an inner edge, and a large iris ring 11 which is a band surrounding the outside thereof. I am divided.

The small iris ring 12 has a pupil sphincter, and the large iris ring 11 has a pupil dilated muscle. On the surface of the iris 10 there are elevations, grooves and depressions of tissues called iris fringes, and from the difference of these tissues, radial fringes in the small iris ring 12 and annular fringes in the large iris ring 11 . In addition, at the boundary between the large iris ring 11 and the small iris ring 12, a crimped ring 15 having an irregular ring-like protuberance is present.

The detection area setting unit 131 detects an area including the crimped ring 15 present at the boundary between the large iris ring 11 and the small iris ring 12 based on, for example, the difference in the shape of the shading.

As shown in FIGS. 4 (a) and 4 (b), in the iris of the human eye, the arterial blood vessel 20 runs between the sclera and the choroid to reach the ciliary body, forming the large iris arterial ring 21. The large iris artery ring 21 is present at the junction of the iris 10 and the ciliary body. Subsequently, a large number of blood vessels enter the iris 10 radially from the large iris arterial ring 21. The anastomosing produces a small iris artery ring 25. That is, the small iris arterial ring 25 is a blood vessel region in which blood vessels are concentrated and formed. The small iris arterial ring 25 is present at the boundary between the large iris ring 11 and the small iris ring 12. A very thin capillary runs from the small iris ring 25 to the tip of the iris 10 and leads to a vein (not shown).

As described above, the detection area setting unit 131 sets an area including the crimp ring 15 at the boundary between the large iris ring 11 and the small iris ring 12 in which the small iris artery ring 25 in which blood vessels are concentrated is formed. , It sets as a detection area which detects a pulse wave.

FIG. 5 (a) shows an image including an iris 10 provided with a rectangular section, and FIG. 5 (b) shows a frequency power spectrum in a rectangular section area including the crimp ring 15. c) shows the frequency power spectrum in a rectangular segmented region containing only skin. As shown in FIGS. 5 (a) to 5 (c), the frequency power spectrum in the rectangular section area including the crimp ring 15 has a peak at a frequency of 1.3 Hz.

In addition, the frequency power spectrum in the rectangular section area including only the skin also has a peak at a frequency of 1.3 Hz. From this, it is understood that the peak of the frequency of the pulse wave is 1.3 Hz, and it is understood that the pulse wave can be detected with high accuracy in the region including the crimp ring 15 of the iris 10.

FIG. 5D is a diagram showing the relationship between the extraction place where the pulse wave extraction process has been performed and the detection accuracy of the pulse wave extracted at each extraction place, and the distance between the center of the rectangular section including the pupil 5 is 0. And, it shows the pulse wave detection accuracy distance dependency which is a relationship between the distance of the extraction location from there and the pulse wave detection accuracy.

As an index of detection accuracy of the extracted pulse wave, SNR (signal to noise ratio) is used. Also, SNR here is defined using a value obtained by dividing the area of the peak signal by the area of the noise signal from the power spectrum of the frequency. In order to show the relationship between the pulse wave detection accuracy and the extraction location, a rectangular section partially including the pupil, white eyes, skin, and eyebrows is excluded from the section for pulse wave extraction processing, and includes only the iris. Pulse wave extraction processing is performed using only the classification.

FIG. 6 is an image diagram showing a schematic configuration of human eyes. As shown in FIG. 6, for example, when there are eyelashes in the pulse wave detection area, the pulse wave detection accuracy in the area is lowered. In addition, a part of the iris 10 may be hidden by the upper and lower eyelids. Therefore, it is desirable for the detection area setting unit 131 to set the left and right areas of the iris as the detection area for the pupil 5.

For example, assuming that a horizontal line extending rightward from the pupil is θ0 = 0 degrees, the range from θ2 = -40 degrees to θ1 = 15 degrees in the right area of the iris and θ3 = 165 degrees to θ4 = 220 degrees in the left area of the iris It is desirable to set the detection area from the range.

Further, as shown in FIG. 5D, the pulse wave detection accuracy has dependency on the extraction location in the iris, and shows a maximum value in the region including the crimp ring 15. Further, it was found that the pulse wave detection accuracy was the highest in the region including the crimp ring 15. From these results, the detection area setting unit 131 can accurately detect an iris pulse wave by setting an area including the crimped ring 15 as a detection area.

The shape of the detection area set by the detection area setting unit 131 may be set as long as an area including the crimp ring 15 can be set, and may be a rectangular shape illustrated in FIG. It may be a donut shape including the reduced ring 15 and the vicinity thereof.

The pixel value calculation unit 132 sets the pixel value (tone value of each color for expressing each pixel included in the captured image in the detection area which is the area including the crimp ring 15 set by the detection area setting unit 131. The calculated value of the pixel value of each color in a predetermined detection area is calculated using. Each color for expressing each pixel contained in a captured image is R, G, B, for example. Further, the operation value of the pixel value of each color is a value obtained by performing a predetermined operation on the pixel values of a plurality of pixels included in the detection area in the captured image, and the calculated value of the pixels included in the detection area It is a value reflecting the size of the pixel value.

The pixel value calculation unit 132 may calculate, for example, an average pixel value which is an average of pixel values of respective colors in the detection area as a calculation value of pixel values in the detection area. In addition, the pixel value calculation unit 132 is, for example, a statistic calculated by increasing the weighting of pixel values of pixels close to the crimping wheel 15 and reducing the weighting of pixel values of pixels away from the crimping wheel 15. The value may be calculated as an operation value of the pixel value in the detection area. In the following description, it is assumed that the pixel value calculation unit 132 calculates the average pixel value of each color in the detection area as an operation value of the pixel value in the detection area.

The pixel value calculation unit 132 calculates an average pixel value for a frame corresponding to a predetermined time (for example, 30 seconds) in the captured image in order to obtain a time change of the average pixel value. The pixel value calculation unit 132 outputs the calculated average pixel value of each color to the pulse wave detection unit 133.

The pulse wave detection unit 133 calculates a pulse wave of a living body by detecting a change in the calculated value of each color in the detection area calculated by the pixel value calculation unit 132. The pulse wave is a change in pressure and a change in volume in the peripheral blood vessel associated with the beating of the heart, and the pulse wave can be detected by measuring the time change of the concentration of oxyhemoglobin in blood. The pulse wave detection unit 133 detects the pulse wave of the living body by detecting the temporal change of the average pixel value of each color calculated by the pixel value calculation unit 132.

FIG. 7 is a flowchart illustrating an example of the flow of pulse wave detection processing by the pulse wave detection unit 133. An example of the flow of pulse wave detection processing by the pulse wave detection unit 133 will be described using FIG. 7.

(Step S1)
The pulse wave detection unit 133 first performs independent component analysis on the average pixel value of each color in the detection area calculated by the pixel value calculation unit 132, and the number is equal to the number of colors (ie, R, G, G, B Take out 4) independent components.

(Step S2)
Next, the pulse wave detection unit 133 removes low frequency components and high frequency components from the four independent components taken out, for example, using a digital band pass filter of 0.75 to 4.0 Hz.

(Step S3)
Next, the pulse wave detection unit 133 performs fast Fourier transform on the four independent components from which the low frequency component and the high frequency component have been removed, and calculates the power spectrum of the frequency of each independent component.

(Step S4)
Next, the pulse wave detection unit 133 calculates the peak value at 0.75 to 4.0 Hz of the power spectrum of the calculated frequency of each independent component.

(Step S5)
Then, the pulse wave detection unit 133 detects an independent component having a peak with the largest peak value among the calculated peak values of each independent component as a pulse wave.

The pulse wave detection unit 133 outputs the detected pulse wave to the living body determination unit 134.

The pulse wave detection by the pulse wave detection unit 133 is not limited to the method of detecting an independent component having a peak with the largest peak value among the peak values of the independent components as a pulse wave. The pulse wave detection unit 133 may calculate the SNR of the peak from the power spectrum, and detect the independent component having the peak with the largest SNR as a pulse wave.

In addition, when the fluctuation with time of the average pixel value calculated by the pixel value calculation unit 132 is large, the pulse wave detection unit 133 may first perform trend removal on the average pixel value of each color (IEEE Trans Biomed Eng, 2002 Feb; 49 (2): 172-175). Then, the pulse wave detection unit 133 may perform the independent component analysis on the average pixel value of each color after removing the above-mentioned fluctuation by performing the trend removal.

The living body determination unit 134 determines whether the waveform detected by the pulse wave detection unit 133 is a pulse wave.

FIG. 8 is a flowchart showing an example of the flow of the living body determination processing by the living body determination unit 134. An example of the flow of the living body determination processing by the living body determination unit 134 will be described using FIG. 8.

(Step S11)
The living body determination unit 134 determines whether the waveform detected by the pulse wave detection unit 133 is a pulse wave. When the living body determination unit 134 determines that the waveform detected by the pulse wave detection unit 133 is a pulse wave, the living body determination unit 134 determines that the subject is a living body. When the living body determination unit 134 determines that the subject is a living body, the process proceeds to step S12. When the living body determination unit 134 determines that the subject is a living body, the living body determination unit 134 may output a signal indicating that a living body has been detected to the authentication unit 140.

On the other hand, when the living body determination unit 134 determines that the waveform detected by the pulse wave detection unit 133 is not a pulse wave, it determines that the target person is not a living body, that is, the target person is a forgery. If the living body determination unit 134 determines that the target person is not a living body, the process ends without outputting an image including the iris region of the target person to the authentication unit 140. The biometric determination unit 134 may be configured to perform non-authentication processing such as notifying the user that the target person can not be authenticated, for example, when determining that the target person is not a living body. In addition, when it is determined that the subject is not a living body, the living body determination unit 134 may output a signal indicating that the living body is not detected to the authentication unit 140.

(Step S12)
The living body determination unit 134 outputs an image including the iris area of the subject to the authentication unit 140.

The living body determination unit 134 may determine whether the waveform detected by the pulse wave detection unit 133 is a pulse wave using any known means. For example, the living body determination unit 134 confirms the periodicity of the waveform detected by the pulse wave detection unit 133, and determines that the pulse wave is generated when the waveform cycle is 0.75 to 3.0 Hz.

Further, the living body determination unit 134 determines whether the waveform is a pulse wave depending on whether the peak value of the power spectrum of the frequency after fast Fourier transform by the pulse wave detection unit 133 or the value of the SNR is equal to or greater than a preset threshold. It may be determined whether or not. In this case, even if at least one of the power spectrum peak value of the frequency and the value of the SNR is input to the living body determination unit 134 instead of the waveform detected by the pulse wave detection unit 133 and used for determination. good.

As described above, the living body detection unit 130 detects a setting area including the crimped ring 15 existing at the boundary between the large iris ring 11 and the small iris ring 12 from the iris area of the subject in the image. Determine the presence or absence of pulse wave detection. As described above, since the pulse wave can be detected with high detection accuracy in the area including the crimped ring 15, the presence or absence of pulse wave detection is determined with the area including the crimped ring 15 as a setting area. It is possible to accurately determine whether the subject is a living body.

(Configuration of authentication unit 140)
When the pulse wave of the subject is detected from the image including the iris area by the living body detection unit 130 and the living body is detected, the authentication unit 140 is provided with the image including the iris area of the subject and the iris authentication is performed. The authentication unit 140 may be provided with a signal indicating that the living body detection is successful, together with the image including the iris area of the subject. When the living body detection unit 130 does not detect a pulse wave in the iris region, that is, when the living body detection unit 130 does not detect a living body, the authentication unit 140 determines not to perform iris authentication of the image.

In addition, when the living body detection unit 130 detects a pulse wave in the iris region, that is, when the living body detection is performed, the authentication unit 140 determines to perform iris authentication of the image.

The authentication unit 140 performs personal authentication based on an image including the iris area of the subject. For example, the authentication unit 140 first collates an iris image registered in advance in the authentication DB 150 with an image including an iris area obtained at the time of authentication. Then, the authentication unit 140 determines whether the mismatch rate by iris authentication between the iris image registered in advance as a result of the collation and the image including the iris area obtained at the time of authentication is less than a predetermined threshold. The identity is determined by authentication.

The authentication unit 140 determines that the user is the person if, for example, the mismatch rate by iris authentication between the iris image registered in advance and the image including the iris region obtained at the time of authentication is less than a threshold, It may be determined that the other person is when the value of L is equal to or greater than the threshold.

The authentication unit 140 includes an iris code creation unit 141 and a collation unit 142.

The iris code creation unit 141 obtains an image including the iris area of the subject output from the living body determination unit 134, and creates an iris code used to perform iris authentication based on the image. A well-known method can be used suitably for preparation of the iris code in the iris code preparation part 141. FIG.

The collation unit 142 collates the registered iris code registered in advance in the authentication DB 150 with the iris code created based on the image including the iris area of the subject output by the biometric determination unit 134. A well-known method can be used suitably for collation with a registration iris code and an iris code. For example, the matching unit 142 may score similarity or non-similarity and perform matching based on the score. For example, the Hamming distance HD between the registration iris code and the iris code may be calculated, and the matching may be performed based on the Hamming distance HD.

Specifically, when the hamming distance HD is equal to or less than a predetermined hamming distance threshold HDth, the matching unit 142 determines that the matching degree between the registered iris code and the iris code is within a predetermined range. In this case, the matching unit 142 determines that the iris authentication for the subject H has succeeded. On the other hand, when the matching degree is out of the predetermined range (when the hamming distance HD is larger than the hamming distance threshold HDth), the matching unit 142 determines that the iris authentication has failed. That is, if the Hamming distance HD is less than or equal to the Hamming distance threshold HDth, the degree of coincidence is high, and if the Hamming distance HD is larger than the Hamming distance threshold HDth, the degree of coincidence is low.

The authentication DB 150 may be integrated with the authentication device 100 or may be separate from the authentication device 100. In a configuration in which the authentication DB 150 is separate from the authentication device 100, the authentication DB 150 may communicate with the authentication device 100 via, for example, wireless communication, and supply the registered iris code to the collation unit 142.

The authentication DB 150 may have a configuration in which registration iris codes of a plurality of registrants are stored, and the collation unit 142 collates each of the plurality of registration iris codes with the iris code. It is also good. Then, the matching unit 142 may determine that the authentication is successful for the registrant whose matching degree between the registration iris code and the iris code is within a predetermined range and the hamming distance HD is the smallest.

As described above, the authentication device 100 acquires an image including the iris region of the subject's eye, determines the presence or absence of pulse wave detection in the iris region of the acquired image, and does not detect the pulse wave in the iris region. Make a decision not to perform iris recognition of the image. That is, the authentication unit 140 outputs an authentication result with reference to the determination result by the living body detection unit 130 that determines the presence or absence of pulse wave detection in the iris region and the iris authentication result by the authentication unit 140. As a result, forgery authentication by combination of an artificial object such as an iris picture and the face of another person who is a living body becomes impossible. Therefore, false authentication due to so-called spoofing can be prevented, and human authentication can be performed with high accuracy.

Second Embodiment
Embodiment 2 of the present disclosure will be described below. In addition, about the member which has the same function as the member demonstrated in the said Embodiment 1 for convenience of explanation, the same code | symbol is appended and the description is not repeated.

FIG. 9 is a block diagram showing a schematic configuration of the authentication device 200 according to the second embodiment. As shown in FIG. 9, the authentication device 200 is different from the authentication device 100 of the first embodiment in that the authentication device 200 further includes a reflection removal unit 260.

The reflection removal unit 260 is provided between the image acquisition unit 120 and the living body detection unit 130. The reflection removal unit 260 performs reflection removal processing on the image including the iris area of the subject's eye acquired by the image acquisition unit 120, and the image subjected to the reflection removal processing is transmitted to the living body detection unit 130. Supply.

The reflection removal unit 260 removes the reflection component by the ambient light specularly reflected on the corneal surface from the image including the iris area of the eye of the subject acquired by the image acquisition unit 120. As described above, by performing reflection removal of ambient light from the image by the reflection removal unit 260, it is possible to accurately perform iris authentication based on the image including the iris region. Therefore, the success rate of living body detection and iris recognition in a highly reflective outdoor environment can be improved.

As a reflection removal method by the reflection removal unit 260, a conventionally known method can be used. For example, the reflection removal method using color characteristics disclosed in ““ The Measurement of Highlights in Color Images ”, GUDRUN J. KLINKER, STEVEN A. SHAFER, AND TAKEO KANADE” may be used.

Also, the reflection removal unit 260 may remove the reflection of ambient light from the image using the following method. For example, the reflection removal unit 260 removes at least a part of the luminance value indicating a specularly reflected light component from the luminance value indicated by the image information of the image including the iris area of the eye of the subject acquired by the image acquisition unit 120 Therefore, ambient light may be reflected and removed from the image.

In addition, the reflection removal unit 260 may use the reflection removal method disclosed in “Japanese Patent No. 3955616”. In this case, the imaging unit 110 includes an imaging element and a polarizing element, and changes the angle of the main axis of the polarizing element by rotating the polarizing element.

The reflection removal unit 260 is configured, based on the normal vector of the subject and the line-of-sight vector, for each pixel in the pixel group in which specular reflection occurs in a plurality of images obtained by the imaging device and having different principal axes of the polarizing element. Identify the plane of incidence and the angle of incidence. Then, the reflection removal unit 260 clusters pixels in which both the incident surface and the incident angle are similar to form a pixel set, and the probability between the diffuse reflection component and the specular reflection component in the pixel set. It separates the reflected components on the assumption of dynamic independence. Thereby, the reflection removal unit 260 can remove the specular reflection component from the image.

Further, the reflection removing unit 260 may be one in which one pixel unit in which the image pickup device is associated with a plurality of polarization elements different in principal axis direction is two-dimensionally arrayed. In this case, the reflection removal unit 260 calculates or estimates the minimum value of luminance (minimum luminance value) for each of the pixel units corresponding to the eyeballs included in the image, and based on the minimum luminance value, At least a portion of the specularly reflected light component at the surface may be removed.

In addition, the reflection removal unit 260 may remove at least a part of the specularly reflected light component by performing an independent component analysis (ICA) process.

Third Embodiment
Embodiment 3 of the present disclosure will be described below. In addition, about the member which has the same function as the member demonstrated in the said Embodiment 1 for convenience of explanation, the same code | symbol is appended and the description is not repeated.

FIG. 10 is a block diagram showing a schematic configuration of the authentication device 300 according to the third embodiment. As shown in FIG. 10, the authentication device 300 differs from the authentication device 100 of the first embodiment in that the authentication device 300 further includes a pupil opening degree calculation unit (determination unit) 370 and a second living body detection unit 380. . The first living body detection unit 330 of the third embodiment and the living body detection unit 130 of the first embodiment have the same configuration, and the detailed description thereof will be omitted.

The pupil opening degree calculation unit 370 is provided between the image acquisition unit 120 and the living

body detection units

330 and 380. The pupil opening degree calculating unit 370 calculates the pupil opening degree of the subject by analyzing the image including the iris area of the subject acquired from the image acquiring unit 120. Then, the pupil opening degree calculation unit 370 determines whether or not the living body detection can be performed using the pupil contraction in accordance with the calculated pupil opening degree of the subject. When it is determined that the living body detection can be performed using the pupil contraction, the pupil opening degree calculation unit 370 provides the second living body detection unit 380 with an image including the iris area of the target person. On the other hand, when it is determined that the biological body detection is not to be performed using pupil contraction, the pupil opening degree calculation unit 370 performs the biological body detection using the presence or absence of pulse wave detection in the iris region of the image. The living body detection unit 330 provides an image including the iris area of the subject.

For example, the pupil opening degree calculation unit 370 may calculate the pupil diameter by analyzing an image including the iris region of the subject, and may set the pupil diameter to the pupil opening degree. The pupil opening degree calculation unit 370 also calculates the pupil diameter and the iris diameter by analyzing the image including the iris region of the subject, and uses the ratio of the pupil diameter and the iris diameter as the pupil opening degree. good.

FIG. 11 is a diagram for explaining a method of calculating the pupil opening degree. As shown in FIG. 11, when the pupil opening degree is calculated using the ratio between the pupil diameter D1 and the iris diameter D2, the pupil opening degree is “pupil opening degree = iris inside diameter (pupil diameter D1) / iris outside diameter ( It is calculated using the formula of "iris diameter D2)".

The pupil diameter D1 and the iris diameter D2 can be calculated by circle fitting to the pupil outer edge and the iris outer edge, respectively. Also, by using the method of setting the ratio of the pupil diameter to the iris diameter as the pupillary opening degree, it is difficult to keep the size of the iris constant due to the fluctuation of the photographing distance between the subject and the imaging unit 110. Even if there is, the pupil opening degree can be calculated more accurately than when only the pupil diameter is used. The pupil opening degree calculation unit 370 may calculate the pupil diameter not with a circle but with an ellipse (long axis or short axis). In addition, the pupil opening degree calculating unit 370 may be configured to use any conventionally known means as a method for calculating the pupil opening degree.

The pupil opening degree calculating unit 370 compares the calculated pupil opening degree with a predetermined pupil opening degree threshold value set in advance. Then, the pupil opening degree calculation unit 370 provides the first living body detection unit 330 with an image including the iris area of the subject when the pupil opening degree is less than the threshold. Further, the pupil opening degree calculation unit 370 provides the second living body detection unit 380 with an image including the iris area of the subject when the pupil opening degree is equal to or more than the threshold.

FIG. 12 is a diagram showing the relationship between the ratio of the pupil diameter to the iris diameter and the illuminance. As shown in FIG. 12, the ratio of pupil diameter to iris diameter depends on the illuminance. For example, when the ratio of the pupil diameter to the iris diameter is used as the pupil opening degree, the ratio of the pupil diameter to the iris diameter is 0.2 at an illuminance of 4000 lux. Then, the ratio 0.2 of the pupil diameter to the iris diameter in the case of the illuminance 4000 lux can be used as the threshold of the pupil opening degree. For example, when the ratio of the pupil diameter D1 to the iris diameter D2 calculated by the pupil opening degree calculation unit 370 is 0.1 (corresponding to 100,000 lux) with respect to 0.2 which is the threshold value of the pupil opening degree. And the pupil opening degree is less than 0.2 which is a threshold value. As described above, the pupil opening degree is less than the threshold because the subject is in an environment such as an outdoor environment, especially in fine weather, where the intensity of environmental light is high and significant pupil contraction can not be expected even when the visible light source is irradiated to the eye This is a case where an image including the iris region of the eye is captured. Therefore, when the pupil opening degree is less than the threshold, the pupil opening degree calculation unit 370 determines that the living body detection can not be performed using the pupil contraction. Then, the pupil opening degree calculation unit 370 outputs an image including the iris area of the target person to the first living body detection unit that performs living body detection using the presence or absence of pulse wave detection in the iris area of the image.

When the ratio of the pupil diameter to the iris diameter calculated by the pupil opening degree calculation unit 370 is, for example, 0.35, the pupil opening degree is 0.2 or more, which is a threshold. As described above, the pupil opening degree becomes equal to or more than the threshold value when the image including the iris area of the eye of the subject is captured in the illumination environment where the pupil contraction occurs by irradiating the eye with the visible light source. Therefore, when the pupil opening degree is equal to or more than the threshold, the pupil opening degree calculating unit 370 determines that the living body detection can be performed using the pupil contraction. Then, the pupil opening degree calculation unit 370 generates an image including the iris region of the subject with respect to the second living body detection unit that performs living body detection using pupil contraction generated by irradiating the eye of the subject with visible light. Output

The second living body detection unit 380 includes an irradiation unit 381, a pupil contraction detection unit 382, and a living body determination unit 383. The size of the pupil changes in size in response to the intensity of ambient light. The second living body detection unit 380 irradiates light to the pupil and detects contraction of the pupil with respect to the irradiated light to determine whether the target person is a living body.

The irradiation part 381 can be set as the structure which has a light source which irradiates the light of visible light toward a subject's eyeball, and a light adjustment part which adjusts the intensity | strength of the light from the said light source. The irradiating unit 381 irradiates the light of the light source to the eyeball of the subject at the brightness according to the pupil opening calculated by the pupil opening calculating unit 370, for example.

The pupil contraction detection unit 382 detects the pupil contraction of the eye of the subject by referring to the image including the iris area of the subject acquired via the pupil opening degree calculation unit 370.

The living body determination unit 383 determines whether the target person is a living body based on the irradiation timing and intensity of the irradiation light by the irradiation unit 381, and the timing and degree of contraction of the pupil detected by the pupil contraction detection unit 382. Do.

According to these configurations, when the illumination environment in which the pupil contraction is caused by irradiating the eye with a visible light source, that is, the pupil opening degree is equal to or more than the threshold, the second living body detection unit 380 detects the presence or absence of the pupil contraction. Perform biological detection by In addition, in an environment where the intensity of ambient light is high, such as when the sky is fine, and significant pupil contraction can not be expected even when the visible light source is irradiated to the eye, that is, when the pupil opening degree is less than the threshold, The living body detection unit 330 performs living body detection by detecting a pulse wave in the iris region.

In the method of performing living body detection by detecting the pulse wave in the iris region by the first living body detection unit 330, it is necessary to capture images of a plurality of frames for at least one second for pulse wave detection. On the other hand, the method of detecting the living body by using the pupil contraction with respect to visible light by the second living body detection unit 380 only needs to capture images of a plurality of frames for about 0.1 to 0.2 seconds, There is an advantage that living body detection and personal identification can be performed in a shorter time. Therefore, in the third embodiment, the pupil opening degree of the object person is calculated by the function of the pupil opening degree calculating unit 370, and it is more accurate by using it when the living body detection can be performed using pupil contraction. It is possible to perform living body detection.

As described above, when the first living body detection unit 330 or the second living body detection unit 380 that has performed living body detection does not detect a living body, the authentication unit 140 determines not to perform iris authentication of the image.

According to these configurations, the authentication unit 140 refers to the detection result by the first living body detection unit 330 or the second living body detection unit 380 and the iris authentication result by the authentication unit 140, and outputs the authentication result. . As a result, it is possible to prevent false authentication due to so-called impersonation in which an artificial object such as an iris photo and the face of another person who is a living body are combined, and personal authentication can be performed with high accuracy.

In the present embodiment, when the pupil opening degree is less than the threshold value, the pupil opening degree calculation unit 370 determines that biological detection can not be performed using pupil contraction, but the present invention is limited thereto. is not. For example, the pupil opening degree calculation unit 370 is configured to be able to refer to the output of an illuminance meter that measures the illuminance of the surrounding environment, and when the illuminance of the surrounding environment measured by the illuminance meter is equal to or more than a predetermined threshold, An image including the iris area of the subject may be output to the first living body detection unit. The threshold of the illuminance may be, for example, 10,000 lux, and when the illuminance measured by the illuminance meter is 10,000 lux or more, the configuration may be such that the living body detection is performed using the first biological detection unit.

Embodiment 4
The fourth embodiment of the present disclosure will be described below. In addition, about the member which has the same function as the member demonstrated in the said Embodiment 1 for convenience of explanation, the same code | symbol is appended and the description is not repeated.

FIG. 13 is a block diagram showing a schematic configuration of the authentication device 400 according to the fourth embodiment. As shown in FIG. 13, the authentication device 400 differs in configuration from the authentication device 100 of the first embodiment in that the health information management unit 490 is further provided.

The health information management unit 490 receives the pulse wave of the subject detected by the pulse wave detection unit 133 of the living body detection unit 130. The health information management unit 490 manages and stores the pulse wave of the subject detected by the pulse wave detection unit 133 as health information in association with the subject.

In addition, the health information management unit 490 can output health information of the subject in response to a request from the external device, for example, to an external device (not shown) connected communicably with the authentication device 400. It is also good.

In addition, even if the health information management unit 490 detects that the pulse wave of the subject is a pulse wave indicating a predetermined health condition, it can output the health information of the subject to the external device. Good.

The health information management unit 490 may be able to manage and store, for example, information including a pulse wave and health information calculated or estimated from the pulse wave as health information of the subject person, and output the information to an external device. . The health information of the subject may be, for example, pulse, blood pressure, stress information and the like.

The health information management unit 490 may calculate the pulse rate, the blood pressure, the stress state, and the like from the pulse wave of the subject detected by the pulse wave detection unit 133 using any method known in the related art. For example, the health information management unit 490 can detect the pulse rate as the biological information by counting the number of peaks of the pulse wave.

For example, the following method may be used to calculate the blood pressure from the pulse wave.

FIG. 14 is a diagram showing a waveform of an acceleration pulse wave obtained by differentiating the pulse wave twice in time. In FIG. 14, a to e indicate amplitudes, x / x (x is a to e) indicate ratios of respective amplitudes, and Tx-x indicates time intervals between peaks. As shown in FIG. 14, in order to calculate the blood pressure from the pulse wave, the health information management unit 490 first differentiates the pulse wave twice to obtain feature amounts a, b, c, d, e, f, b. / A, c / a, d / a, e / a, Ta-a, Ta-b, Ta-c, Ta-d, and Ta-e are derived. It is known that these feature quantities change according to the pulse wave velocity, and the pulse wave velocity and the blood pressure value have a correlation.

The health information management unit 490 has an estimation formula created by performing, for example, multiple regression analysis using these feature quantities and blood pressure value data acquired in advance. Then, the health information management unit 490 can obtain the estimated value of the blood pressure by applying the feature quantity obtained from the pulse wave of the subject detected by the pulse wave detection unit 133 to the estimation formula.

For example, the following method is available as a method of calculating stress information from pulse waves.

The health information management unit 490 detects the peak of the pulse wave of the subject detected by the pulse wave detection unit 133, performs fast Fourier transform (FFT) on the time interval between the peaks, and calculates the power spectrum of the frequency. The integrated value of the power spectrum of 0.04-0.15 Hz is LF, and the integrated value of the power spectrum of 0.15-0.4 Hz is HF, and the stress level value LF / HF can be calculated. In addition, the frequencies of LF and HF may have a width around the above (see: http://hclab.sakura.ne.jp/stress_novice_LFHF.html).

In addition, the health information management unit 490 may have a function of controlling whether or not authentication by the authentication unit 140 is based on the health information of the target person. For example, the health information management unit 490 records a predetermined threshold or an allowable range set in advance for the health information of the subject.

The health information management unit 490 compares the health information of the target person obtained from the pulse wave of the target person detected by the pulse wave detection unit 133 with a predetermined threshold or tolerance set in advance. The health information management unit 490 outputs an authentication OK even when authentication is enabled by the authentication unit 140 when the health information of the subject is out of a predetermined threshold or tolerance set in advance. May have a function to prevent

In addition, the health information management unit 490 has a function of permitting the authentication unit 140 to output an authentication OK only when the values such as the target patient's pulse, blood pressure, and an index indicating a stress state are within an appropriate range. It may be

For example, when the authentication device 400 is used as a travel start lock of a car, the function of the health information management unit 490 can be suitably used. The health information management unit 490 is a case where the driver is authenticated by the authentication unit 140 when the blood pressure value calculated from the pulse wave of the subject is out of the appropriate range, that is, an abnormal state. Can also be used in applications such as not releasing the vehicle's travel start lock.

[Modification 1]
Modification 1 of the present disclosure will be described below.

In the first to fourth embodiments, the imaging unit 110 continuously generates an image including the iris area of the subject's eye at a predetermined time interval (frame rate) based on the intensity of light transmitted through each of the RGB filters. Has been described. In order to acquire a pulse wave from an image including the iris region of the subject's eye, the imaging unit 110 preferably includes a pixel having sensitivity to the G wavelength (500 to 600 nm). On the other hand, in order to obtain iris information for authentication, it is desirable to employ an imaging unit having pixels having sensitivity in the IR region (750 to 950 nm).

Therefore, the imaging unit 110 is provided with RGBIR filters, and based on the intensities of light transmitted through the R, G, B, and IR filters, an image including the iris area of the subject's eyes is continuously formed at predetermined time intervals. The configuration may be such that the image is taken in a targeted manner.

According to this configuration, acquisition of a pulse wave and acquisition of iris information can be performed with high sensitivity from an image including the iris region of the subject's eyes, and personal authentication can be performed with high accuracy.

[Modification 2]
A second modification of the present disclosure will be described below.

In the above-mentioned modification 1, the imaging part 110 demonstrated the structure provided with the filter of RGBIR. The imaging unit 110 picks up an image including the iris area of the subject's eye using two sensors, an RGB sensor provided with a filter of RGGB and an IR sensor provided with a filter of IR. It may be a configuration.

[Modification 3]
A third modification of the present disclosure will be described below.

In the first to fourth embodiments, the authentication device 100 acquires an image including the iris area of the subject's eye, determines the presence or absence of pulse wave detection in the iris area of the acquired image, and detects the pulse wave in the iris area. If not, it is determined that the iris authentication of the image is not performed. However, the configuration of the present disclosure is not limited to this, and living body detection by the living body detection unit 130 and iris authentication by the authentication unit 140 may be processed in parallel.

The image including the iris region of the subject's eye acquired by the image acquisition unit 120 is sent to the living body detection unit 130 that determines presence / absence of pulse wave detection in the iris region of the image and the authentication unit 140 that performs iris authentication of the image. Supplied.

If the result of iris authentication by the authentication unit 140 is determined to be another person, for example, before the determination by the living body detection unit 130 is finished, the authentication unit 140 immediately outputs the authentication result indicating the authentication NG. . In addition, when the result of iris authentication by the authentication unit 140 is determined to be the person prior to the determination by the living body detection unit 130 ending, the authentication unit 140 waits for the determination result of the living body detection unit 130, Only when the determination result by the living body detection unit 130 is a living body, an authentication result indicating authentication OK is output.

As described above, by performing the living body detection by the living body detection unit 130 and the iris authentication by the authentication unit 140 in parallel processing, the processing time can be shortened compared to sequentially performing the living body detection and the iris authentication. be able to.

In the configurations of the first to fourth embodiments as well as in the configuration of the third modification, the authentication unit 140 refers to the determination result by the living body detection unit 130 and the iris authentication result by the authentication unit 140 to obtain an authentication result. Since the output is performed, false authentication due to so-called impersonation can be prevented, and personal authentication can be performed with high accuracy.

[Example of software implementation]
The control blocks (in particular, the image acquisition unit 120, the living

body detection units

130, 330, 380, and the authentication unit 140) of the

authentication device

100, 200, 300, 400 are logic circuits (hardware) formed in an integrated circuit (IC chip) or the like. And may be realized by software.

In the latter case, the

authentication device

100, 200, 300, 400 includes a computer that executes instructions of a program that is software that implements each function. The computer includes, for example, at least one processor (control device) and at least one computer readable storage medium storing the program. Then, in the computer, the processor reads the program from the recording medium and executes the program to achieve the object of the present disclosure. For example, a CPU (Central Processing Unit) can be used as the processor. As the above-mentioned recording medium, a tape, a disk, a card, a semiconductor memory, a programmable logic circuit or the like can be used besides “a non-temporary tangible medium”, for example, a ROM (Read Only Memory). In addition, a RAM (Random Access Memory) or the like for developing the program may be further provided. The program may be supplied to the computer via any transmission medium (communication network, broadcast wave, etc.) capable of transmitting the program. Note that one aspect of the present disclosure may also be realized in the form of a data signal embedded in a carrier wave in which the program is embodied by electronic transmission.

The authentication device according to each aspect of the present disclosure may be realized by a computer, and in this case, the computer is realized as each component (software element) included in the authentication device to realize the authentication device by the computer. A control program of an authentication device and a computer readable recording medium recording the same also fall within the scope of the present disclosure.

The present disclosure is not limited to the above-described embodiments, and various modifications are possible within the scope of the claims, and embodiments obtained by appropriately combining the technical means disclosed in the different embodiments. Is also included in the technical scope of the present disclosure. Furthermore, new technical features can be formed by combining the technical means disclosed in each embodiment.

(Cross-reference to related applications)
This application claims the benefit of priority to Japanese Patent Application filed on Dec. 20, 2017: Japanese Patent Application No. 2017-243979, the entire contents of which are hereby incorporated by reference. Included in this book.

10 iris 15 crimp ring 110

imaging unit

100, 200, 300, 400 authentication unit 120 image acquisition unit 130 living body detection unit 131 detection area setting unit 132 pixel value calculation unit 133 pulse wave detection unit 134 living body determination unit 140 authentication unit 142 verification Part 260 Reflection removal part 330 First living body detection part 370 Pupil opening degree calculation part (determination part)
380 Second vital sign detection unit 490 Health information management unit

Claims

An image acquisition unit that acquires an image including an iris area of the subject's eyes;
A biological detection unit that determines presence or absence of pulse wave detection in an iris region of the image;
An authentication unit that performs iris authentication on the image;
An authentication unit that outputs an authentication result with reference to a determination result by the living body detection unit and an iris authentication result by the authentication unit.
The living body detection unit detects a setting area including a crimped ring existing at a boundary between a large iris ring and a small iris ring from the iris area of the image to determine presence / absence of pulse wave detection in the setting area. The authentication device according to claim 1, characterized in that
The authentication apparatus according to claim 1, further comprising a reflection removal unit that removes a reflection component due to ambient light specularly reflected on the cornea surface from the image including the iris region.
The authentication apparatus according to any one of claims 1 to 3, wherein the living body detection unit includes a health information management unit that manages health information of the subject including the detected pulse wave.
An image acquisition unit that acquires an image including an iris area of the subject's eyes;
A determination unit that acquires a pupillary opening degree by analyzing the image, and determines whether biological detection can be performed using pupil contraction according to the acquired pupillary opening degree;
A first living body detection unit that performs living body detection using presence or absence of pulse wave detection in an iris region of the image, when the determination unit determines that living body detection can not be performed using pupil contraction;
The second living body performs living body detection using pupil contraction generated by irradiating visible light to the eyes of the target person when it is determined that the determination unit can perform living body detection using pupil contraction. A detection unit,
An authentication unit that performs iris authentication on the image;
An authentication apparatus characterized in that the authentication unit refers to the detection result by the first living body detection unit or the second living body detection unit and the iris authentication result by the authentication unit, and outputs an authentication result. .
An image acquisition step of acquiring an image including an iris area of the subject's eyes;
A biological detection step of determining presence or absence of pulse wave detection in an iris region of the image;
And d) authenticating the iris image of the image.
An authentication method comprising: outputting an authentication result with reference to a determination result in the living body detection step and an iris authentication step in the authentication step.