WO2019102686A1

WO2019102686A1 - Biometric authentication device and biometric authentication system

Info

Publication number: WO2019102686A1
Application number: PCT/JP2018/033113
Authority: WO
Inventors: 友輔松田; 洋野々村; 三浦　直人; 長坂　晃朗; 渓一郎中崎; 宮武　孝文
Original assignee: 株式会社日立製作所
Priority date: 2017-11-27
Filing date: 2018-09-06
Publication date: 2019-05-31
Also published as: JP6846330B2; JP2019096168A

Abstract

There is a problem in the state of the art in that brightness varies significantly between when an organism is contacting a terminal and when there is no such contact, resulting in a difference in the quantity of irradiated light and thus in degraded authentication precision. A biometric authentication system disclosed by the present invention comprises a computation part for calculating brightness saturation information relating to a brightness saturation region occurring in a blood vessel image by the quantity of light irradiated on a finger from a light source for irradiation exceeding a prescribed value.

Description

Biometric authentication apparatus and biometric authentication system

The present invention relates to a biometric authentication device and a system for performing biometric authentication.

In recent years, the danger of illegal use of IDs and passwords has been called for. Therefore, biometric authentication technology has attracted attention as a personal authentication technology that is less likely to be abused. Above all, finger vein authentication technology can realize a high degree of security because it uses the blood vessel pattern inside the finger.

The conventional finger vein authentication device improves the reproducibility of a captured vein image by fixing a finger to the authentication device, and realizes high accuracy authentication. However, since it takes time to fix the finger on the device, it is not suitable for the case where authentication of a large number of people is required in a short time. As a solution to this problem, a device that acquires vein images and authenticates them completely without contact has been developed. However, in non-contact authentication, since the presentation position of the finger can not be fixed, it is necessary to control the light amount according to the presentation position of the finger.

In Patent Document 1, an opening formed on the surface of a housing and a plurality of light sources arranged in a grid and arranged on the side of the opening and position information of a hand presented on the opening are acquired. An illumination light source is selected from among a plurality of light sources based on a sensor and position information, and a light amount control unit that controls the light amount of the illumination light source, and a hand irradiated with light from the illumination light source And an imaging unit configured to capture an image of a blood vessel included in the finger portion of the finger.

WO 2016/0842214

The technique disclosed in Patent Document 1 controls the light amount value of the light source based on the luminance value. In this control method, the luminance largely differs between the state in which the living body contacts the terminal and the state in which the living body does not contact, and a difference occurs between the light amount values to be irradiated, so that the authentication accuracy is lowered.

In order to solve the above problems, in the biometric authentication system disclosed in the present invention, luminance saturation information on a luminance saturated region generated in the blood vessel image when the light amount of light irradiated to the finger from the irradiation light source exceeds a predetermined value. And an arithmetic unit for calculating

Even if there is a difference in the position or posture of the finger to be presented, authentication is performed based on the luminance saturation information on the luminance saturation region generated in the blood vessel image when the light amount of the light irradiated to the finger from the irradiation light source exceeds a predetermined value. By doing this, the authentication accuracy can be improved.

The figure which shows the network structural example of a biometrics system. The figure for demonstrating the example of a program stored in memory. The figure which shows the structural example of a biometrics system. The figure for demonstrating the example of the difference of the illumination conditions of the irradiated light of the light source by the fluctuation | variation of the presentation position of a finger | toe. The figure for demonstrating the example of the difference in the lighting conditions in the state which the hand touched the apparatus, and a non-contact state. The figure for demonstrating brightness saturation. The figure for demonstrating that the difference in the magnitude | size of the brightness | luminance saturation area | region in the finger side surface arises by the difference in the irradiation light quantity of a light source. The figure which shows the example of the flowchart of the registration process which a biometrics system implements. The figure which shows the example of the flowchart of the authentication process which a biometric authentication system implements. The figure which shows the example of the flowchart of the process which feedback-controls light quantity. The figure which shows the example of the relationship of a luminance saturation rate and the light quantity value of a light source. The figure which shows the example of the difference of the illumination conditions of the irradiated light of the light source by the difference in the rotation state of a hand. The figure which shows the example of the flowchart of the process which produces | generates a clear finger blood vessel image by synthesize | combining several finger blood vessel images. FIG. 7 is a view showing an example of a finger blood vessel image taken in a posture in which the finger is closed using the imaging unit 11; The figure which shows the example of the image of the hand image | photographed by reflected light. The figure for demonstrating the example of the method of imaging a clear finger blood vessel image, when the finger of a hand is inclined from horizontal posture.

Several embodiments are described below.

In this embodiment, luminance saturation information is extracted from the blood vessel image of the finger to be presented, and the light intensity of the light source is controlled based on the luminance saturation information to clear the finger blood vessel even if the position and posture of the finger to be presented fluctuate. The configuration for capturing an image and performing authentication will be described. The biometric authentication system described in the present embodiment can be used not only when photographing data at the time of authentication but also when photographing data at the time of registration. In addition, although the present embodiment takes an example of capturing and authenticating a blood vessel image of a finger, it is also possible to perform capturing and authentication using a blood vessel image regarding a site other than the finger, such as a palm.

FIG. 1 is a diagram showing an example of a network configuration of a biometric authentication system. The biometric authentication system 501 assumes a configuration in which a server 502 and a biometric information extraction terminal 503 are connected by a network. Among these, the server includes a storage unit 505 that stores biological information, and an operation unit 506 that performs predetermined processing described later. Also, the biometric information extraction terminal 503 is a terminal that extracts biometric information from the user. Further, as shown in FIG. 1, a plurality of terminals may be connected to the server on the network.

Although this embodiment will be described as an example in which personal authentication is performed in the same housing as the biological information extraction apparatus, the authentication processing may be performed by an operation unit in a server or the like installed outside the rat. . That is, the authentication processing may not be performed by the housing itself, and information of the captured blood vessel image may be transmitted to the calculation unit such as a server. This configuration is suitable for applying the present invention to a cloud system in which a blood vessel image transmitted from many biological information extraction terminals is used for authentication on the server side.

Further, instead of using the photographed blood vessel image itself for personal authentication, blood vessel image information may be used as an encryption key for encrypting predetermined data. This makes it possible to reduce the risk that the blood vessel image leaks to the outside.

FIG. 2 is a diagram showing a representative program stored in the memory 6 stored in the computing unit 506. As shown in FIG. The light source control program 601 is a program that mainly executes processing for controlling the light amount value of the light source. The finger position / finger posture detection program 602 is a program that executes processing for detecting the position and posture of a finger using a distance image, a near infrared image, and the like. The luminance information calculation program 603 is a program for executing processing for calculating luminance information in a blood vessel image. The luminance saturation information calculation program 604 is a program for executing processing for calculating luminance saturation information and the like in a blood vessel image. The finger vein image normalization program 605 is a program for executing normalization processing and the like for correcting the enlargement ratio and distortion due to position fluctuation, posture fluctuation and the like of the blood vessel image. The finger vein feature extraction program 606 is a program for executing a process of extracting a blood vessel feature from the blood vessel image after the normalization process. The feature matching program 607 is a program for executing processing of matching the extracted blood vessel feature with the blood vessel feature registered in advance in the storage device or the like to calculate a matching score. The registered fruit skin judgment program 608 is a program for judging whether or not to register the extracted blood vessel feature.

FIG. 3 is a diagram showing an example of the configuration of a biometric authentication system. An opening 3 is provided on the surface of the casing of the blood vessel image photographing apparatus 2 so that the hand 1 can be presented above the opening 3 of the blood vessel image photographing apparatus 2 at the time of photographing a blood vessel image of a finger. The distance sensor 4 disposed inside the housing below the opening 3 converts the received light into an electric signal and takes it into the computing unit 506 as distance data between the hand 1 and the distance sensor through the data input unit 50. The CPU 7 calculates the position of the hand 1, the posture of the hand 1, the position of the finger, the posture of the finger, and the like according to a program stored in the memory 6 based on the distance data taken into the calculation unit 506. The lighting control unit 51 of the light source controls the light source array 9 disposed inside the opening 8 based on the calculation result, and selects the irradiation light source from the plurality of point light sources 10 configuring the light source array 9. The imaging unit 11 disposed below the opening 3 receives the light passing through the optical filter 12. The received light is converted into an electrical signal by the imaging unit 11 and taken into the computing unit 506 as a finger blood vessel image through the image input unit 52. The CPU 7 calculates luminance saturation information and the like in the finger blood vessel image according to a program stored in the memory 6 based on the finger blood vessel image taken into the calculation unit 506. The light amount control unit 53 of the light source sets the light amount value of the light source for irradiation based on the calculation result. Each point light source emits light corresponding to the set light amount value to the finger. The finger blood vessel image captured by the calculation unit 506 is stored in the memory 6 once. Then, the CPU 7 uses the program stored in the memory 6 to collate the image stored in the memory with the image stored in advance in the storage device 14 to perform authentication.

The position and orientation of the hand 1 may be calculated by detecting the position and orientation of the hand using the image of the hand captured by the imaging unit 11 or the image of the hand 1 captured by the imaging unit 11 and the hand 1 And distance data between the distance sensors 4 may be used.

Further, as means for notifying the user of the authentication result, the voice of the speaker 15 may be used, or the authentication result may be displayed on the display unit 16.

In addition, the visible light source 17 is provided in the opening portion 3, and when the hand is detected, the hand is detected, the hand is detected, the authentication process, the authentication success, the authentication failure, etc. You may be notified of the status of

Furthermore, the registrant data to be collated may be narrowed down from a large number of registrant data by causing the user ID input unit 18 to input a personal identification number or an ID or reading an IC chip at a stage before authentication. . This narrowing brings about the effect of improving the image search speed and the authentication accuracy. In particular, when it is possible to uniquely identify an authentication target by narrowing down, it is called 1: 1 authentication, and the above-mentioned effect is further improved.

In addition, the opening part 3 uses the raw material which permeate | transmits the irradiated light from each point light source 10. For example, a transparent member such as acrylic or glass is assumed. In addition, a film that allows only near infrared light to pass may be attached to the opening 3 made of these transparent members. In this way, the inside of the apparatus can be made invisible from the user.

Hereinafter, a subject for capturing a clear finger blood vessel image will be described with reference to FIGS. 4 and 5.

FIG. 4 is a diagram showing the difference in the illumination condition of the illumination light of the light source due to the fluctuation of the presentation position of the finger. When performing clear blood vessel imaging corresponding to a finger of a different size for each user, it is necessary to control the light quantity value of the light source according to the size of the finger. However, if the presentation position of the finger is different as shown in FIGS. 4a and 4b, the illumination light source to be selected is different, so the relative distance between the finger and the light source changes in FIGS. 4a and 4b. The light emitted from the light source diffuses according to the distance and the light intensity attenuates. Therefore, when the relative distance between the finger and the light source changes, the illumination condition of the light irradiating the finger fluctuates, and the image is stable and clear. You can not take a good blood vessel image.

FIG. 5: is a figure which shows the difference in the illumination conditions of the irradiated light of the light source in the state which the hand contacted the apparatus, and each non-contact state. As a conventional method of controlling the light quantity value of a light source, a method of equalizing the luminance value of a finger area of a finger blood vessel image photographed for performing sharp blood vessel photographing corresponding to a change in relative distance between the finger and the light source There is. Hereafter, the subject of the method of controlling the light quantity value of the light source using equalization | homogenization of a luminance value is described using FIG. 5 a and FIG. 5 b.

FIG. 5 a is a view showing a state in which the finger is in contact with the opening 3. In the contact state, the brightness (brightness) of the finger area of the finger blood vessel image is determined by the amount of light transmitted through the finger (transmitted light 30) among the light irradiated to the finger. Since the transmitted light 30 of the finger is absorbed and attenuated in the living body, it is necessary to irradiate the finger with strong light in order to increase the brightness of the finger region in the finger blood vessel image.

FIG. 5 b is a view showing a state in which the finger is not in contact with the opening 3 (non-contact state). In the non-contact state, in addition to the transmitted light 30 of the finger, the brightness (brightness) of the finger area is affected by the light (reflected light 31) reflected between the opening 3 such as acrylic and the finger and the irradiation light of the light source ) Is decided. The reflected light 31 in the non-contact state increases the brightness of the finger area even with a small amount of irradiation light. Therefore, in the conventional light amount control method for equalizing the luminance value of the finger area, the light amount value to be irradiated is set smaller in the non-contact state than in the contact state, and the light amount values irradiated in the contact state and the non-contact state are different. Will occur.

As described above, when the illumination condition fluctuates due to the difference in the light quantity value irradiated to the finger, it causes the fluctuation of the image quality such as the sharpness of the finger blood vessel image, so the images were taken in the contact state and the non-contact state respectively. When the finger blood vessel images are compared with each other, the authentication accuracy is reduced.

In this embodiment, by controlling the light amount value based on the luminance saturation information of the finger blood vessel image, the fluctuation of the light amount value irradiated to the finger is suppressed, and the fluctuation of the illumination condition is suppressed, thereby stably and clearly. Blood vessel images can be taken.

Hereinafter, the method of controlling the light amount value based on the luminance saturation and the luminance saturation information will be described in detail with reference to FIGS. 6 and 7.

FIG. 6 is a diagram for explaining luminance saturation used for light amount control in the present embodiment. In the apparatus shown in FIG. 1, when imaging a blood vessel present inside a finger by irradiating light to a plurality of fingers, the imaging unit 11 receives a light which is directly irradiated and reflected on the side of the finger, and the finger blood vessel image In the vicinity of the contour of the finger, the brightness is extremely high and the whiteout occurs, and a region (brightness saturation 40) in which the blood vessel information is lost is generated.

FIG. 7 is a view for explaining that a difference in the size of the luminance saturated region on the side surface of the finger is caused due to the difference in the irradiation light amount of the light source. 7a shows the magnitude of the luminance saturation 40 near the finger contour when the light quantity of the light source is large (that is, when light is irradiated in the state of FIG. 5a), and FIG. That is, the magnitude of the luminance saturation 40 in the vicinity of the finger contour in the case where light is irradiated in the state of FIG. As described above, according to the change in the amount of light irradiated to the finger, the amount of light reflected on the side of the finger also changes, and the area of the luminance saturated region near the contour of the finger also changes. Conventionally, the luminance saturated area where the blood vessel image has disappeared is noise in authentication, and it has been an issue how to reduce the luminance saturated area. However, since there is a strong correlation between the light quantity value of the light source irradiated to the finger and the area of the luminance saturated area near the contour of the finger, if the area of the luminance saturated area is made uniform, the finger is irradiated It is possible to suppress the fluctuation of the light amount value of the light source and to suppress the fluctuation of the illumination condition. This correlation will be described later in the description of FIG.

As described above, in the present embodiment, it is possible to suppress a decrease in the degree of coincidence at the time of matching of blood vessel patterns by capturing a finger blood vessel image so as to equalize the area of the luminance saturated region. The reflected light 31 generated between the finger presented in a non-contact state and the acrylic (the opening 3) raises the brightness of the finger area of the finger blood vessel image, but it is sufficient to saturate the brightness near the finger contour of the finger. It is not a strong light. Therefore, by controlling the light amount value of the light source so as to equalize the area of the brightness saturated region 40, a sufficient amount is not influenced by the reflected light generated between the finger and the acrylic (the opening 3). Light can be transmitted to the finger. Therefore, it is possible to stably capture a clear finger blood vessel image in the contact state and the non-contact state.

In the above example, the area of the luminance saturation region 40 is described as the luminance saturation information used for light amount control, but the luminance saturation rate can be used instead of the luminance saturation region. The luminance saturation rate is calculated as a ratio of luminance saturated pixels in each finger area in the finger blood vessel image. When the luminance saturation rate is used, it is possible to control the light amount value of the light source that illuminates each finger so that the luminance saturation rate of each finger becomes uniform. Further, the light quantity value of the light source may be controlled so that the average value of the luminance saturation rates of the plurality of detected finger areas becomes uniform.

FIG. 8 is a diagram showing a flowchart of registration processing in the biometric authentication system. Hereinafter, in order to explain the flowchart of the registration process of FIG. 8, unless otherwise specified, the CPU 7 in the biometric authentication system shown in FIG. 3 executes the process related to each step by reading the program stored in the memory 6. Do. In step 100, the registration process is started. Step 101 is an operation in which the registrant presents a hand at the top of the opening 3. Step 102 is processing for executing the finger position / finger posture detection program of FIG. 2 and detecting a hand using a distance image or the like acquired by the distance sensor 4. Step 103 is processing to determine whether a hand has been detected. If no hand is detected, the process returns to step 102. If a hand is detected, the process proceeds to step 104. Step 104 executes the light source control program of FIG. 2 and the lighting control unit 51 performs initial light source control processing, that is, processing for lighting the point light source 10 for emitting light to the detected finger of the hand with the initial light amount value. . Step 105 is processing in which the distance sensor 4 acquires a distance image and the imaging unit 11 acquires a near infrared image. Step 106 is a process of executing the finger position / finger attitude detection program of FIG. 2 and detecting the position and attitude of the finger using the distance image and the near infrared image. Specifically, it is processing for detecting and acquiring position information and posture information based on the three-dimensional shape of the finger. Step 107 executes the light source control program of FIG. 2, and the lighting control unit 51 controls the light source array 9 according to the finger position and the finger posture detected in step 106 so that the finger blood vessel can be clearly photographed. The point light source 10 to be turned on is selected, and the light amount control unit 53 determines the light amount value of the point light source 10 to be turned on. The details of step 107 will be described later. Step 108 is a normalization process for executing the finger vein image normalization program of FIG. 2 and correcting the enlargement ratio and distortion due to finger position fluctuation and posture fluctuation in the acquired finger blood vessel image. Step 109 is a process of executing the finger vein feature extraction program of FIG. 2 and extracting a blood vessel feature from the finger blood vessel image after the normalization process. Step 110 is a process of executing the registration availability determination program of FIG. 2 and determining whether or not the finger vein feature extracted in step 109 is to be registered. If it is determined in step 109 that registration is performed, registration processing is performed in step 111, and the registration processing is ended in step 113. If it is determined in step 110 that the registration is not performed, the registration time is determined to be out in step 112. If timed out in step 112, the registration process is ended in step 113, and if timed out, the process returns to step 105.

The registration propriety determination of step 110 will be described. In order to realize highly accurate authentication, it is desirable that the captured finger vein image be clear. Therefore, whether the registration is performed is performed using the average luminance of the captured finger vein image, luminance saturation information, blood vessel contrast, and the like. In addition, in order to capture a clear finger vein image, there are desirable hand presentation positions and presentation postures. Therefore, whether or not the positions and postures of the hands and fingers detected in step 106 are at desired positions and postures can be used for registration availability determination.

In the registration process of step 111, as information to be registered, not only finger vein characteristics, but also luminance saturation information of a photographed finger vein image, luminance information, and irradiation light amount value of a light source can be registered together.

FIG. 9 is a flowchart of processing performed by the biometric authentication system. Hereinafter, although the flowchart of the authentication process in FIG. 9 will be described, unless otherwise specified, the CPU 7 in the biometric system shown in FIG. 3 executes the process related to each step by reading the program stored in the memory 6 Do. Step 201 is an operation in which the certifier presents a hand at the top of the opening 3. Step 202 is processing for executing the finger position / finger posture detection program of FIG. 2 and detecting a hand using a distance image or the like acquired by the distance sensor 4. Step 203 is processing to determine whether a hand has been detected. Here, if a hand is not detected, the process returns to step 202, and if a hand is detected, the process proceeds to step 204. Step 204 executes the light source control program of FIG. 2 and the lighting control unit 51 performs initial light source control processing, that is, processing for lighting the point light source 10 for emitting light to the detected finger of the hand with the initial light amount value. . Step 205 is processing in which the distance sensor 4 acquires a distance image and the imaging unit 11 acquires a near infrared image. Step 206 is a process of executing the finger position / finger posture detection program of FIG. 2 and detecting the position and posture of the finger using the distance image and the near infrared image. Specifically, it is processing for detecting and acquiring position information and posture information based on the three-dimensional shape of the finger. Step 207 executes the light source control program of FIG. 2, and the lighting control unit 51 controls the light source array 9 according to the finger position and the finger posture detected in step 206 so that the finger blood vessel can be clearly photographed. The point light source 10 to be turned on is selected, and the light amount control unit 53 determines the light amount value of the point light source 10 to be turned on. The details of step 207 will be described later. Step 208 is a normalization process for executing the finger vein image normalization program of FIG. 2 and correcting the enlargement ratio and distortion due to finger position fluctuation and posture fluctuation in the acquired finger blood vessel image. Step 209 is a process of executing the finger vein feature extraction program of FIG. 2 and extracting a blood vessel feature from the finger blood vessel image after the normalization process. In step 210, the feature matching program shown in FIG. 2 is executed, and the extracted blood vessel feature is compared with the blood vessel feature registered in the storage device 14 to calculate a matching score. Step 211 is processing to compare the calculated matching score with a predetermined threshold TH1. Here, if the verification score is larger than the predetermined threshold value TH1, the process proceeds to step 212 to perform a predetermined post-authentication success process, and proceeds to step 214 to end the authentication flow. On the other hand, if the verification score is equal to or less than TH1, the process proceeds to step 213, and authentication timeout determination is performed. If the time-out time has not elapsed, the process returns to step 205 and the authentication process is repeated. On the other hand, when the time-out time has elapsed, the process proceeds to step 214 and the authentication flow is ended.

Next, step 207 will be described. In step 207, in order to suppress the illumination variation due to the irradiation light of the light source and to capture a clear finger blood vessel image, the luminance saturation information such as the luminance saturation rate is made uniform, and the light quantity value of the light source is controlled. At this time, the target luminance saturation information may not be a fixed value, but may be a predetermined range. In this case, the corresponding light amount value is not a value but a predetermined range based on the range of the luminance saturation information. The details of light source control will be described later in steps 307 to 309 in FIG. Further, as to the method of determining the target light amount value, the optimum method differs depending on the difference in the authentication method. For example, in the above-mentioned 1: 1 authentication method, the luminance saturation information of each finger of the finger blood vessel image captured at the time of registration in the registration process of step 111 in the flowchart of FIG. By performing the light amount control so that the luminance saturation information matches the luminance saturation information at the time of registration, the fluctuation of the illumination conditions at the time of registration and at the time of authentication can be suppressed, and highly accurate authentication can be realized.

On the other hand, when the user does not specify the user before authentication (1: N authentication method) without performing the ID input etc., the statistical information of the luminance saturation ratio for each finger calculated from a large amount of database in advance is the basic luminance The light intensity is controlled so that the basic luminance saturation information of each finger registered and the luminance saturation information of each finger of the certifier match when the user authenticates by registering in the DB as saturation information, and finger blood vessel image Taking a picture is effective. At this time, the statistical information may be an average value or a median value of the luminance saturation rates, and other statistical quantities may be used as appropriate.

FIG. 10 is a flowchart of processing for feedback control of the light amount. Hereinafter, a mechanism for feedback control of the light quantity of the point light source 10 of the light source array 9 based on the information of the position and posture of the finger and the luminance saturation information in the finger blood vessel image will be described using this figure. As long as the CPU 7 in the biometric authentication system shown in FIG. 3 reads the program stored in the memory 6, the process related to each step is executed.

Step 301 is an operation in which the certifier presents the hand 1 to the blood vessel imaging device 2. Step 302 is processing of executing the finger position / finger posture detection program of FIG. 2 and detecting the hand presented using the distance image or the like acquired by the distance sensor 4. Step 303 is processing to determine whether a hand has been detected. Here, if a hand is not detected, the process returns to step 302, and if a hand is detected, the process proceeds to step 304. Step 304 is a process of executing the light source control program of FIG. 2, the lighting control unit 51 performing initial light source control, and lighting the point light source 10 for irradiating light to the detected finger of the hand with the initial light amount value. The initial light amount value may be a light amount value set in advance, or may be changed according to the size of the detected hand and finger. Further, the relative distance between the hand and the light source detected in step 303, the relative distance between the hand and the opening 3, and the like may be determined according to the position of the hand.

Step 305 is processing in which the distance sensor 4 acquires a distance image and the imaging unit 11 acquires a near infrared image. Step 306 is a process of executing the finger position / finger posture detection program of FIG. 2 and performing processing of detecting the position and posture of the hand and the finger in a three-dimensional space using the distance image acquired in step 305. Step 307 executes the brightness saturation information calculation program of FIG. 2 and uses the finger blood vessel image for use in controlling the light quantity of the point light source 10 to be lit from the finger blood vessel area of the distance image and near infrared image acquired in step 305 It is a process of calculating luminance saturation information and the like in the inside. The details of step 307 will be described later. Step 308 is processing to execute the light source control program of FIG. 2 and the lighting control unit 51 to determine the point light source 10 for lighting the light source array 9 based on the calculation result of the position and posture of the finger in step 306. Step 309 is processing in which the light amount control unit 53 performs light amount control of each point light source 10 to be lit. Specifically, based on the luminance saturation ratio in the finger blood vessel image of each finger calculated in step 307, the light amount value of the point light source 10 to be lighted for clearly photographing the blood vessel of each finger is determined. The point light source 10 is turned on. Details of the method of determining the light amount value in step 309 will be described later with reference to FIG. After the point light source 10 is turned on, the process returns to step 305, and the control of the light source array 9 is repeated. In this way, each point light source 10 is turned on with the light amount value determined in step 309, and the process returns to 305 again to repeat the processing, thereby changing the position and posture of the hand and changing the illumination condition. The light amount value can be feedback controlled using

Next, a method of calculating luminance saturation information such as the luminance saturation rate in the finger blood vessel image in step 307 will be described. When calculating the luminance saturation rate, it can be calculated as the ratio of the luminance saturated area over the entire finger area. Alternatively, the luminance saturation rate may be calculated as the rate of luminance saturation occurring near the finger contour. Here, if the luminance saturation rate is calculated as the ratio of luminance saturation occurring near the first joint to the second joint of the finger, it is possible to calculate the luminance saturation rate excluding the vicinity of the fingertip where the luminance saturation is likely to occur and the vicinity of the finger root. And is preferable.

FIG. 11 is a graph showing the relationship between the brightness saturation information and the light intensity value of the light source, for explaining the control method of the light intensity value of the light source based on the brightness saturation information in step 309. There is a correlation between the luminance saturation rate and the light amount value of the light source, and linear approximation or logarithmic approximation can be performed. When the relationship between the luminance saturation rate and the light amount value of the light source is linearly approximated, the slope of the approximate straight line can be calculated from the light amount value and the luminance saturation rate in a certain imaging frame. Using this, the light amount value corresponding to the target luminance saturation rate is calculated, and is set as the light amount value at the time of finger blood vessel imaging of the next frame. The target luminance saturation rate may be a predetermined range instead of a value. In this case, the corresponding light quantity value is also a range rather than a value based on the range of the luminance saturation rate. In addition, the control of the light amount value of the light source in step 309 may control the light amount value of the light source together with the luminance information such as the average luminance as well as the luminance saturation information. For example, the upper limit value of the average luminance may be set in advance, and the light amount value of the light source may be controlled to be closer to the target luminance saturation information within the range not exceeding the upper limit of the average luminance. Further, it is also possible to control the light quantity value of the light source so that the value of the evaluation index is adjusted to a preset target value using one evaluation index combining the luminance saturation information and the luminance information. As information used for light amount control of a light source in combination with luminance saturation information, an index representing the definition of blood vessels such as blood vessel contrast in a finger blood vessel image may be used in addition to luminance information.

In Example 1, a sharp finger blood vessel image was taken by controlling the light amount of the light source based on the luminance saturation information. On the other hand, in the present embodiment, in connection with step 207 of the flow of authentication processing in the first embodiment and steps 307 and 309 of the flow of light source control, it is clearer when presented with the hand rotated. A configuration for realizing finger blood vessel imaging will be described. The configuration of the biometric authentication system is the same as that in the first embodiment shown in FIG.

FIG. 12 shows the difference in the illumination condition by the illumination light from the light source in the state where the hand 1 is not rotated (non-rotational state) and in the state where the hand 1 is rotated. In the non-rotational state of FIG. 12a, the light emitted from the light source (optical axis direction) is parallel to the direction from the wrist side toward the fingertip (main direction of the hand 1). There is little light reflected from the side of the finger. On the other hand, in the rotation state of FIG. 12 b, the light axis direction of the light source and the main direction of the hand 1 are orthogonal to each other, so that the light reflected by the light source emits more light than the non-rotation state. .

As in the first embodiment, by equalizing the luminance saturation rate of the finger area in the finger blood vessel image and controlling the light amount of the light source, it is possible to bring the illumination conditions in the rotated state and the non-rotated state closer to each other. However, as shown in FIG. 12b, in the luminance saturation region generated on the side surface of the finger when the hand 1 rotates, the area of the luminance saturation region on the finger side closer to the light source becomes larger among the left and right side surfaces of the finger. The area of the luminance saturation region on the side of the finger is smaller. Even if the luminance saturation rate of the finger area is equalized in the rotated state and in the non-rotated state, the deviation of the area of the luminance saturated area on the left and right sides of the finger is small in the non-rotational state, while in the rotational state The bias of the area of the luminance saturation region on the left and right sides becomes large. As described above, when the illumination condition changes due to the difference in the area of the luminance saturation region, the degree of coincidence when the finger blood vessel images photographed in the rotation state and the non-rotation state are compared with each other is reduced, and the authentication accuracy is deteriorated. .

Therefore, in the present embodiment, in order to reduce the deviation of the area of the luminance saturated region on the left and right sides of the finger, the luminance saturation rate of each of the left side and right side is calculated for each finger, and rotation of finger blood vessel image is performed. In the state, the light amount control of the light source is performed so that the luminance saturation rate of each of the finger side surfaces becomes a uniform value with the luminance saturation rate in the non-rotational state. The luminance saturation ratio of the left side surface and the right side surface of the finger can be calculated as the ratio of the luminance saturated region over the entire finger region, as in the first embodiment. Further, the luminance saturation rate may be calculated as the ratio of the luminance saturation region generated near the finger contour. The luminance saturation rate can also be calculated as the ratio of luminance saturation occurring in the region from the first joint to the second joint of the finger excluding the vicinity of the fingertip and the vicinity of the base of the finger.

In the case where the luminance saturation of the left and right sides of the finger is largely uneven, and the luminance saturation of both the left and right sides of the finger can not be made uniform with the luminance saturation in the non-rotational state by light amount control. By equalizing only the luminance saturation rate of the finger side surface having the large luminance saturation rate among the left side surface and the right side surface, it is also possible to suppress the fluctuation of the illumination condition.

In the first embodiment and the second embodiment, acquisition of a clear finger blood vessel image is realized by controlling the light amount of the light source based on the luminance saturation information of the finger blood vessel image according to the position and posture variation of the finger to be presented. The method was described. In this embodiment, when a clear finger blood vessel image can not be obtained by one shooting, a clear blood vessel image can be obtained by combining the finger blood vessel images taken at a plurality of timings with different light amounts of the light source. Describes how to achieve successful authentication. The configuration of the biometric authentication system is the same as that of the first embodiment shown in FIG. 1 as in the second embodiment, so the description will be omitted.

First, a case in which it is difficult to obtain a clear finger blood vessel image in one shooting will be described. As described in the first embodiment, when the irradiation light amount of the light source is increased, the light transmitted through the finger is increased, and a sharp finger blood vessel image is obtained. May increase. In addition, as described in the second embodiment, a luminance saturation region in a finger region of a finger blood vessel image is likely to be generated with a small irradiation light amount, such as a state where the hand is rotated. When the amount of light emitted from the light source is controlled so that the luminance saturation of each of the left and right sides of the finger is equalized with the luminance saturation in the non-rotational state in a state in which the hand rotates easily and luminance saturation easily occurs. The amount of light passing through the finger may decrease, and the finger blood vessel image may become unclear.

Therefore, in the present embodiment, a finger blood vessel image (saturation condition uniform image) captured so as to make the luminance saturation uniform, and a blood vessel contrast large in a state where a sufficient light amount is transmitted through the finger even if the luminance saturation region is large. A finger blood vessel image (high contrast image) is synthesized to obtain a clear finger blood vessel image. By combining the saturated condition uniform image and the high contrast image captured at a plurality of different timings, it is possible to generate a finger blood vessel image having a uniform luminance saturation rate and a large blood vessel contrast. Even when the hand position and posture change are large and a clear finger blood vessel image can not be obtained in one shooting, a clear finger blood vessel image can be obtained, so that more accurate authentication is possible.

FIG. 13 is a flowchart of processing for generating a clear finger blood vessel image by combining a plurality of finger blood vessel images acquired at different timings. Hereinafter, a mechanism for combining two finger blood vessel images and generating a sharp finger blood vessel image will be described using this figure, but unless otherwise specified, the CPU 7 in the biometric system shown in FIG. By reading the program stored in the memory 6, the process related to each step is executed.

Step 401 is an operation in which the certifier presents the hand 1 to the blood vessel imaging device 2. Step 402 is processing for executing the finger position / finger posture detection program of FIG. 2 and detecting a hand presented using a distance image or the like acquired by the distance sensor 4. Step 403 is processing to determine whether a hand has been detected. If a hand is not detected here, the process returns to step 402, and if a hand is detected, the process proceeds to step 404. Step 404 is a process of executing the light source control program of FIG. 2, the lighting control unit 51 performing initial light source control, and lighting the point light source 10 for irradiating light to the detected finger of the hand with the initial light amount value. The details of step 404 are omitted because the same method as step 304 of the first embodiment can be used. Step 405 is processing for acquiring a distance image by the distance sensor 4 and acquiring a near infrared image by the imaging unit 11. Step 406 is processing for executing the finger position / finger posture detection program of FIG. 2 and performing position and posture detection processing in a three-dimensional space of a hand and a finger using the distance image acquired in step 405. Step 407 executes the brightness saturation information calculation program of FIG. 2 and uses the finger blood vessel image for use in controlling the light quantity of the point light source 10 to be lit from the finger blood vessel area of the distance image and near infrared image acquired in step 405 It is a process of calculating luminance saturation information and the like in the inside. Step 408 is a process of executing the light source control program of FIG. 2 and determining the point light source 10 where the light source array 9 is turned on based on the calculation result of step 406. Step 409 is processing to determine whether luminance saturation information such as luminance saturation of the finger area is within the target range (whether the luminance saturation condition is satisfied) in the photographed finger blood vessel image. If the brightness saturation condition is not satisfied, the light source control program of FIG. 2 is executed in step 410, and the light quantity control is feedback controlled so that the brightness saturation becomes the target value by the process of light quantity control 1 performed by the light quantity control unit 53. And return to step 405. If the brightness saturation condition is satisfied, the process proceeds to step 411. Step 411 is a process of executing the light source control program of FIG. 2 and controlling the light amount so that the light amount control unit 53 sets the luminance information such as the average luminance of the finger region in the finger blood vessel image to a target value. Step 412 is processing of acquiring a distance image by the distance sensor 4 and acquiring a near infrared image by the imaging unit 11. Step 413 is processing for executing the finger position / finger posture detection program of FIG. 2 and performing position and posture detection processing in a three-dimensional space of hands and fingers using the distance image acquired in step 412. Step 414 executes the brightness information calculation program shown in FIG. 2 from among finger blood vessel regions of the distance image and the near infrared image acquired in step 412, for use in the light quantity control of the point light source 10 to be lit. Is processing for calculating luminance information such as average luminance in Step 415 is processing to determine whether luminance information such as average luminance of the finger area is within the target range (whether the luminance condition is satisfied) in the photographed finger blood vessel image. If the brightness condition is not satisfied, the process returns to the process of the light amount control 2 of step 411. If the brightness condition is satisfied, the process proceeds to step 416. Step 416 is processing of synthesizing a finger blood vessel image satisfying the brightness saturation condition of step 409 and a finger blood vessel image satisfying the brightness condition of step 415 to generate a sharp finger blood vessel image. Details of step 416 will be described later.

In step 415, the average luminance is described as an example of the index for obtaining and determining the finger blood vessel image for combination, but the blood vessel contrast value of the finger blood vessel image may be used as an index for determining the finger blood vessel image. The blood vessel contrast value can be calculated, for example, by dividing the finger area of the finger blood vessel image into a blood vessel area and a background area, and calculating the difference between the brightness of the blood vessel area and the background area. In this case, blood vessel contrast values are calculated instead of the luminance information of the finger area in

steps

407 and 414.

Next, the synthesis processing of the finger blood vessel image in step 416 will be described. The composition process generates a clear finger blood vessel image with uniform brightness saturation information such as brightness saturation and a large contrast between the blood vessel area and the background area. Specifically, based on the finger blood vessel image satisfying the brightness saturation condition at step 409, contrast enhancement of the blood vessel region is performed based on the finger blood vessel image satisfying the brightness condition at step 415 in the finger region where the brightness saturation does not occur. By doing this, a clear finger blood vessel image can be generated.

In the present embodiment, the method of combining the saturated condition uniform image satisfying the brightness saturation condition of step 409 and the high contrast image satisfying the brightness condition of step 415 has been described, but the saturated condition uniform image and the high contrast image are registered and It may be used for authentication. Specifically, at the time of authentication, the saturated condition uniform image and the high contrast image are respectively compared with the registered image, and a result with a high degree of coincidence can be adopted to realize highly accurate authentication. Further, the saturated condition uniform image and the high contrast image may be registered data respectively.

In the present embodiment, the presented finger is related to step 205 of the flow of authentication processing in the first embodiment, step 305 of the light source control flow, and step 405 of the finger blood vessel image synthesis flow in the third embodiment. In the closed posture, the finger area can not be specified in the finger blood vessel image, and the light amount of the light source can not be controlled. Light is emitted from the light source arranged below the opening to the finger on the palm side and reflected. A method of accurately detecting a finger area using an image obtained by photographing the above-described light will be described.

The configuration of the biometric authentication system is the same as that of the first embodiment shown in FIG. 1 as in the first and second embodiments, and therefore the description thereof is omitted.

FIG. 14 is a view showing a finger blood vessel image taken using the imaging unit 11 in a posture in which the finger is closed. When photographing in a posture in which the fingers are closed, the luminance saturated regions occurring near the contours of the fingers overlap, so the boundary line between adjacent fingers can not be accurately determined. However, as described in the first embodiment, in order to capture a sharp finger blood vessel image by controlling the light amount of the light source using the brightness saturation information, the brightness saturation such as the finger area of each finger and the brightness saturation rate We need to ask for information. In particular, since the luminance saturated region is likely to be generated near the contour line of the finger, if the finger boundary line can not be detected, it may cause failure and instability of the light amount control of the light source.

FIG. 15 is a view showing an image (reflected light image) obtained by irradiating the finger on the palm side with light from the light source 19 disposed below the opening 3 and capturing the reflected light. In the reflected light image, it is possible to obtain an accurate finger area because the border between the adjacent finger and finger is clear even in the posture in which the finger is closed. In this embodiment, the luminance saturation information of the finger blood vessel image is accurately calculated using the finger area accurately determined from the reflected light image, and the light amount control of the light source is performed based on the determined luminance saturation information. It is possible to take pictures of various finger blood vessel images.

The finger area in the finger blood vessel image can be determined by converting the coordinates of the finger area detected from the reflected light image into the finger blood vessel image coordinate system. Hereinafter, a method of converting the finger area detected from the coordinate system of the reflected light image into the coordinate system of the finger blood vessel image will be described. When the reflected light image and the finger blood vessel image are captured at different timings using the imaging unit 11, the coordinates of the finger area detected from the reflected light image can be used as the coordinates of the finger area in the finger blood vessel image. When using the ToF distance camera 4, the finger area is obtained from the reflected light image captured by the distance camera 4, and the coordinate system of the imaging unit 11 is obtained from the coordinate system of the distance camera 4 using stereo calibration or the like. The finger area in the finger blood vessel image can be obtained by performing the conversion into

In this embodiment, when the presented finger of the hand is inclined from the horizontal position, the method of performing the lighting control of the light source and the light amount control more accurately using the luminance saturation information and photographing a clear finger vein is described. Do.

FIG. 16 is a view for explaining a light source control method when the hand is presented to the blood vessel image photographing apparatus 2 in a state of being inclined from the horizontal posture. FIG. 16 shows the hand viewed from the side, and FIG. 16 a presents the hand horizontally, and FIG. 16 b presents the hand with the fingertip raised above the finger root side. When the height of the light source to be lit is determined according to the position and height of the fingertip detected by the distance sensor 4, since the position and height of the fingertip are the same in FIGS. 16 a and 16 b, the light sources of the same height Lights up. However, as the broken line from the light source in FIG. 16b shows, when the finger is inclined from the horizontal posture, the light irradiated near the fingertip is not irradiated on the back side of the finger and does not transmit inside the finger, so the veins I can not shoot. Furthermore, since the fingertips face upward, many reflected components at the opening 3 of the light indicated by the broken line from the light source in FIG. When the amount of light irradiated to the belly side of the finger is too large, it is easy to generate luminance saturation of the finger area, and when the light amount control of the light source is performed based on the luminance saturation information, a sufficient amount of light It does not transmit, and an unclear finger vein image is taken. Therefore, in the present embodiment, when the hand is presented in a posture in which the fingertip is higher than the finger root side, the light source at a position higher than the height of the light source lit in the horizontal posture is turned on, and the amount of light not irradiated on the finger And control the light amount of the light source based on the luminance saturation information to capture a clear finger vein image. By turning on the light source at a position higher than the horizontal posture of FIG. 16a in the state of FIG. 16b in which the hand is inclined from horizontal, the amount of light not irradiated to the finger is suppressed, and Among them, by suppressing the amount of light irradiated to the ventral side of the finger, it becomes possible to capture a clear finger vein image by light source control based on the luminance saturation information.

1: Hand 2: Blood vessel image capturing device 3: Opening 4: Distance sensor 6: Memory 7: CPU 8: Opening 9: Light source array 10: Point light source 11: Imaging unit 12: Optical filter 13: Interface 14: Storage device 15: speaker 16: display unit 17: visible light source 18: user ID input unit 19: light source 30: transmitted light 31: reflected light 40: luminance saturation 50: data input unit 51: lighting control unit 52: image input unit 53: light amount Control unit 506: Arithmetic unit

Claims

In a biometric authentication system for performing authentication by comparing biometric information for registration, which is biometric information registered in advance, with biometric information for authentication in which biometric information is extracted by predetermined processing from a blood vessel image of a living body acquired at the time of authentication.
It has an operation unit for calculating luminance saturation information on a luminance saturation region generated in the blood vessel image when the light amount of light emitted to the finger from the irradiation light source exceeds a predetermined value, and performs authentication based on the luminance saturation information A biometric authentication system characterized by
The biometric authentication system according to claim 1, wherein the calculation unit calculates the luminance saturation information based on an area of the luminance saturation region existing near a contour line of a finger included in the blood vessel image.
A biometric authentication system characterized by
In the biometric authentication system according to claim 2,
The luminance saturation information is a luminance saturation rate which is a ratio of an area of the luminance saturation area to an area of a finger area of the blood vessel image.
A biometric authentication system characterized by
In the biometric authentication system according to claim 3,
The calculation unit calculates the luminance saturation information of the left and right sides of the finger included in the blood vessel image.
A biometric authentication system characterized by
In the biometric authentication system according to claim 4,
The calculation unit combines the blood vessel image and a blood vessel image obtained by imaging light emitted from a light source controlled using the luminance saturation information.
A biometric authentication system characterized by
In the biometric authentication system according to claim 5,
An opening formed in the surface of the housing;
A plurality of light sources arranged in a light source array part protruding from the opening;
A sensor for acquiring position information and posture information of a finger presented on the opening;
A lighting control unit which selects an irradiation light source to be discarded to the finger from among the plurality of light sources based on the position information and posture information, and controls lighting of the selected irradiation light source;
An imaging unit configured to capture an image of a blood vessel present inside a finger irradiated with the light of the irradiation light source;
A light amount control unit configured to control a light amount of light emitted from the irradiation light source based on the luminance saturation information;
A biometric authentication system comprising:
In the biometric authentication system according to claim 6,
The light amount control unit controls the light amount of the irradiation light source such that the brightness saturation information falls within a predetermined range.
A biometric authentication system characterized by
In the biometric authentication system according to claim 7,
The light of a light source disposed below the opening is irradiated to the palm side of the finger, and the finger is detected using an image obtained by photographing the reflected light.
A biometric authentication system characterized by
In the biometric authentication system according to claim 8,
The posture information includes the positional relationship between the fingertip and the finger root,
The lighting control unit selects an irradiation light source for irradiating the finger from among the plurality of light sources based on the positional relationship between the fingertip and the finger root, and controls lighting of the selected irradiation light source.
A biometric authentication system characterized by
In the biometric authentication system according to claim 9,
It further comprises an input unit for receiving predetermined information from the user,
The light amount control unit controls the light amount of the irradiation light source based on luminance saturation information linked to the information input by the input unit.
A biometric authentication system characterized by
In the biometric authentication system according to claim 10,
The arithmetic unit calculates statistical information of the luminance saturation information as basic luminance saturation information from the plurality of blood vessel images captured in advance.
The light amount control unit controls the light amount of the irradiation light source based on the basic luminance saturation information.
A biometric authentication system characterized by
A biometric authentication method for performing authentication by comparing biometric information for registration, which is biometric information registered in advance, with biometric information for authentication in which biometric information is extracted from a blood vessel image of a living body acquired at the time of authentication by predetermined processing.
Arithmetic processing for calculating luminance saturation information on a luminance saturation region generated in the blood vessel image when the light amount of the light emitted from the irradiation light source to the finger exceeds a predetermined value;
An authentication process for performing authentication based on the luminance saturation information;
Biometric authentication method characterized by