JP2011100317A - Personal authentication method and personal authentication system using vein pattern during bending and stretching motion of finger - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a new personal authentication method and personal authentication system using a vein pattern. <P>SOLUTION: The personal authentication system uses a vein pattern during bending and stretching motion of a finger as authentication information. Since joints of a human finger are hinge joints of one degree of freedom, and besides, a tendon is present between the first and second joints to interlock movements of both joints, bending and stretching motion of the finger is very highly reproducible. So, in the personal authentication system, time-series variation in a vein pattern along with the bending and stretching motion is defined as a dynamic feature amount to generate authentication information. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a biometric authentication technique, and more particularly to a finger vein authentication technique having impersonation resistance.

  In recent years, finger vein authentication using finger vein patterns as authentication information has begun to be put into practical use at bank ATMs and the like. Recently, research reports have pointed out the vulnerability of finger vein authentication. For example, Non-Patent Document 1 tried to authenticate an existing system using an artificial finger made using an OHP sheet and an artificial snow agent to which a two-dimensional image of a person's vein was transferred. Report on results that were determined to be closer to the person compared to the other person's finger. This fact suggests that if a person's vein image is taken by some unauthorized means, the person may be impersonated by making an artificial finger.

  On the other hand, Japanese Unexamined Patent Application Publication No. 2007-287080 (Patent Document 1) discloses a finger vein authentication system in which a finger to be authenticated is rotated and imaged, and finger vein information for each angle is registered for the same finger. In the method of Patent Document 1, since a plurality of different vein patterns are used as authentication information for one finger, impersonation by an artificial finger (Non-Patent Document 1) in which only a vein pattern imaged from one direction is transferred is used. It is assumed that it has a certain tolerance. However, if a vein pattern image for every angle is taken for a finger, the 3D structure of the person's vein pattern is analyzed from these multiple images, and a 3D printer or the like is used based on the result. It is not possible to reproduce the three-dimensional pattern of veins inside the artificial finger, and it is possible to impersonate the person by inserting and rotating such an artificial finger into the imaging part of the system of Patent Document 1. there is a possibility.

  The finger vein authentication system inevitably has to hold a template image, and there is always a risk of the template image being leaked. In this regard, the conventional finger vein authentication system cannot be said to have sufficient spoofing resistance.

JP 2007-287080 A

Tsutomu Matsumoto, Biological detection and registration failure in biometrics, IEICE technical report, Vol.104, No.734, pp81-82, (2005).

  The present invention has been made in view of the above-described problems in the prior art, and an object of the present invention is to provide a personal authentication method and a personal authentication system using a vein pattern having spoofing resistance.

  As a result of earnestly examining the personal authentication method using a vein pattern having spoofing resistance, the present inventor has focused on the extremely high reproducibility of finger bending and stretching movements, and the time series of vein patterns during finger bending and stretching movements The present invention has been conceived by conceiving a configuration in which such changes are used as authentication information.

  That is, according to the present invention, there is provided a personal authentication method using finger flexion / extension motion, imaging light transmitted through the side surface of the finger of the person to be authenticated who performs the flexion / extension motion, and performing a plurality of continuous operations on the finger side surface. Obtaining a digital image, extracting a vein pattern for each of the digital images, and storing the vein pattern in association with a bending angle of the finger. A personal authentication method is provided, in which authentication is performed by pattern matching for each bending angle of a predetermined finger. In the present invention, the bending angle of the finger can be determined based on position information corresponding to the extreme value of the curvature of the contour of the finger.

Further, according to the present invention, there is provided a personal authentication system using the bending / extending motion of a finger, imaging light transmitted through the side surface of the finger of the person to be authenticated who performs the bending / extending motion, and performing a plurality of continuous operations on the side surface of the finger A finger imaging unit that acquires the digital image obtained, and a personal authentication unit to which information on the digital image acquired by the finger imaging unit is input. The personal authentication unit includes a vein pattern extraction unit and a finger bending angle calculation. Unit, an authentication data generation unit, a template storage unit, and a pattern matching unit, wherein the vein pattern extraction unit extracts the vein pattern for each of a plurality of digital images on the side of the finger, The authentication data generation unit generates authentication data by associating the vein pattern with the bending angle of the finger, and at the time of user registration, the authentication data is stored as template information in a template storage unit. And 憶, during authentication, and outputs the authentication data to the pattern matching unit, the pattern matching unit, personal authentication system that performs authentication by pattern matching and the template information and the authentication data is provided. In the present invention, the finger imaging unit includes a light emitting unit for irradiating the side surface of the finger with near-infrared light, and an imaging unit for imaging light transmitted through the side surface of the finger. An imaging space for the finger can be formed between two polarizing filters that are orthogonal to each other, and a sensor for detecting the insertion and removal of the finger is provided, and the finger is inserted into the imaging space. The camera can be configured to automatically start imaging and automatically end imaging when the finger is removed from the imaging space. In the present invention, the finger bending angle calculation unit may be configured to calculate the finger bending angle based on position information corresponding to the extreme value of the curvature of the finger outline.

  As described above, according to the present invention, a novel personal authentication method and personal authentication system using a vein pattern are provided. The personal authentication system of the present invention has impersonation resistance even if a template image leaks from the system.

The figure which shows the personal authentication system of this embodiment. Schematic which shows the structure of the finger imaging part in this embodiment. The conceptual diagram for demonstrating the mechanism which interrupts | blocks light other than the transmitted light of a finger | toe. The figure which shows the continuous image of the finger | toe which bends and extends. The figure which shows the image processing performed about the digital image of a finger in time series. The functional block diagram of the personal authentication system of this embodiment. The conceptual diagram for demonstrating the method of calculating the bending angle (theta) of a finger. The figure which shows the result of an authentication experiment.

  Hereinafter, the present invention will be described with reference to embodiments shown in the drawings, but the present invention is not limited to the embodiments shown in the drawings. In the drawings referred to below, the same reference numerals are used for common elements, and the description thereof is omitted as appropriate.

  FIG. 1 shows a personal authentication system 10 according to an embodiment of the present invention. The personal authentication system 10 of the present embodiment includes a finger imaging unit 12 and a personal authentication unit 14. In the personal authentication system 10, at the time of authentication, the user 16 who is the person to be authenticated is requested to insert his / her finger 18 (preferably, forefinger) into the finger imaging unit 12 and perform bending and stretching movements. The finger imaging unit 12 acquires the side surface of the finger 18 that bends and stretches as a continuous digital image, and transfers the image data to the personal authentication unit 14.

  The personal authentication unit 14 determines whether or not the user 16 is a regular registered user based on the image data transferred from the finger imaging unit 12. The personal authentication unit 14 in the present embodiment is configured as a server device that includes a CPU that functions as a system controller, a RAM that provides an execution space for application software, a ROM that stores programs, and a storage device such as a hard disk. Using MacOS (TM), Windows (R), UNIX (R), LINUX (R) or other suitable operating systems, C, C ++, Visual C ++, Visual Basic, Java By executing an application program described in (registered trademark) or the like, a functional means for performing personal authentication is provided. Moreover, in this embodiment, the authentication result display part 20 for displaying the authentication result derived | led-out by the personal authentication part 14 can also be provided. In the present embodiment, the authentication result derived by the personal authentication unit 14 can be output to a predetermined control means. For example, the authentication result can be directly output to the door lock unlocking control means, and the authentication result and the door lock unlocking can be linked.

  FIG. 2 is a schematic diagram showing the structure of the finger imaging unit 12 in the present embodiment, and FIG. 2A shows a perspective view of the finger imaging unit 12 as viewed from above. As shown in FIG. 2A, the finger imaging unit 12 is configured as a casing 30 having an imaging space inside, and the casing 30 is formed with an opening 32 for inserting the finger 18. ing. When receiving authentication, the user 16 inserts the finger 18 into the housing 30 through the opening 32 with the belly of the finger 18 facing down. Inside the housing 30, the finger 18 can be inserted up to at least the second joint (proximal interphalangeal joint), and the first and second joints of the finger 18 can be bent and bent and extended. It has a space.

  Furthermore, the finger imaging unit 12 captures a plurality of light emission means 34 for irradiating the side surface of the finger 18 with near-infrared light, and near-infrared light transmitted through the side surface of the finger 18. The image pickup means 36 for acquiring continuous digital images is included. The imaging means 36 is positioned so that the side surface of the finger 18 can be imaged, and acquires a continuous digital image by imaging the side surface of the finger 18 in time series. In the present embodiment, the imaging means 36 can be constituted by a CCD video camera. In the present embodiment, the finger imaging unit 12 is provided with a sensor for detecting the insertion / removal of the finger, and imaging is automatically started when the finger is inserted into the imaging space, and the finger is removed from the imaging space. It is preferable to configure so that the imaging is automatically terminated upon the occasion. It is also possible to provide a predetermined time limit for the imaging time and output an error when a timeout occurs.

  The finger imaging unit 12 according to the present embodiment acquires a vein pattern image using a phenomenon in which a vein portion appears dark when hemoglobin in blood absorbs near-infrared light, as in the conventional vein authentication technique. In the finger imaging unit 12, it is preferable to provide a band filter 37 on the imaging unit 36 side in order to block light having a wavelength other than the near infrared region. Here, in the conventional finger vein authentication device, the path of near-infrared light is completely blocked with a finger, and only light that passes through the finger is imaged. On the other hand, in the present embodiment, as shown in FIG. 2B, since it is necessary to continuously capture an image of the finger 18 bending and stretching, the movement range of the finger 18 surrounded by a broken line is shown. Although it is necessary to always irradiate near-infrared light, direct light that does not pass through the finger 18 as it is enters the imaging means 36 and causes whiteout, so that a clear image cannot be captured. In this regard, the finger imaging unit 12 in the present embodiment solves this problem by forming an imaging space for the finger 18 between two polarizing filters whose polarization directions are orthogonal.

  FIG. 3 is a conceptual diagram for explaining a mechanism for blocking light other than the transmitted light of the finger 18 in the finger imaging unit 12. In FIG. 3, for convenience of explanation, the band filter 37 is omitted. In the finger imaging unit 12, a polarizing filter 38 is provided on the light emitting means 34 (not shown), and only polarized light B out of the near infrared light A emitted from the light emitting means 34 enters the imaging space. On the other hand, since the polarization filter 39 having a polarization direction orthogonal to the polarization direction of the polarization filter 38 is provided on the imaging means 36 side, the polarization B 1 that does not transmit the finger 18 is blocked by the polarization filter 39. On the other hand, the polarized light B <b> 2 that has entered the finger 18 is scattered within the finger 18 and becomes scattered light C and is emitted from the finger 18. As a result, a predetermined polarization component (polarized light D) included in the scattered light C passes through the polarization filter 39 and enters the imaging means 36 to form an image. By adopting the above-described mechanism, it becomes possible to block light other than the transmitted light of the finger 18 in accordance with the bending / extending movement of the finger 18, and the change in the vein pattern accompanying the bending / extending movement of the finger 18 as a clear image. Can be acquired.

  The personal authentication method of the present invention uses, as authentication information, a vein pattern that is acquired by the above-described finger imaging unit and changes with the bending and stretching motion of the finger. It is known that the reproducibility of finger flexion and extension is extremely high because the joints of human fingers are hinge joints with one degree of freedom and the movements of the first and second joints are linked. Yes. The present inventor has focused on this point, and has come up with the idea that a vein pattern that changes as the finger bends and stretches is regarded as a dynamic feature amount and is used as authentication information.

  Next, the processing executed by the personal authentication unit 14 in this embodiment will be described in order. First, a plurality of digital image data including the state of bending / extending movement of the finger 18 of the person to be authenticated, acquired by the finger imaging unit 12 described above, is transferred to the personal authentication unit 14. FIG. 4 illustrates a continuous image of the finger 18 transferred from the finger imaging unit 12. In FIG. 4, six frames of continuous images are shown for convenience of explanation, but it is actually preferable to acquire images of several tens of frames or more. In the personal authentication unit 14, first, image processing for extracting the vein pattern of the finger 18 is performed on each digital image acquired from the finger imaging unit 12. This point will be described below with reference to FIG.

  FIG. 5 shows, in time series, image processing performed on the digital image of the finger 18 acquired from the finger imaging unit 12. FIG. 5A shows an original image before image processing. Since the background other than the finger 18 to be noticed is reflected in the original image, the process of extracting only the area of the image of the finger 18 from the original image is first executed in the following procedure. First, binarization processing focusing on the brightness of the finger is performed. As shown in FIG. 5A, since the brightness of the finger 18 is within a predetermined range, an appropriate threshold (upper limit value and lower limit) corresponding to the brightness of the finger 18 in terms of luminance value. Value), and a pixel having a luminance value within the threshold is converted to white (luminance value 255), and a pixel having a luminance value in the other range is converted to black (luminance value 0). As shown in b), an image in which the part of the finger 18 is white is generated. Next, noise removal is performed on the image shown in FIG. 5B, and AND processing of the image after noise removal and the original image shown in FIG. As shown in FIG. 5, an image is generated in which only the portion of the finger 18 is trimmed on a black background.

  Next, in order to extract the vein pattern, the binarization process is performed on the image shown in FIG. That is, in the image shown in FIG. 5C, paying attention to the luminance of the pixel, the portion corresponding to the vein is stored as a black pixel and the other portion is stored as a white pixel. As a result, the image shown in FIG. 5D is generated. Finally, noise is removed from the image shown in FIG. 5D to obtain the vein pattern image shown in FIG. The vein pattern extraction procedure in the present embodiment has been described above. Next, an algorithm for personal authentication in the personal authentication system 10 of the present embodiment will be described below with reference to FIG.

  FIG. 6 shows a functional block diagram of the personal authentication system 10 of the present embodiment. The personal authentication unit 14 includes a vein pattern extraction unit 40, a finger bending angle calculation unit 42, an authentication data generation unit 44, a template storage unit 46, and a pattern matching unit 48. When the user uses the system, the finger imaging unit 12 images a finger bending / extension motion in order to perform user registration. When a series of digital images accompanying finger flexion / extension motion is transferred to the personal authentication unit 14, the vein pattern extraction unit 40 generates a finger vein pattern image from each image, and the finger bending angle calculation unit 42 performs each image. The “bending angle θ of the finger” is calculated. The method of obtaining “finger bending angle θ” will be described later.

  For each image, the authentication data generation unit 44 generates authentication data by associating the extracted vein pattern image with the “finger bending angle θ” at that time, and uses this as template information to register user registration information (name / user In the template storage unit 46 constituted by a storage device such as a hard disk. User registration is completed by the above procedure.

  Next, a personal authentication procedure executed by the personal authentication system 10 of the present embodiment will be described with reference to FIG. During authentication, when the person to be authenticated inserts his / her finger into the finger imaging unit 12 and performs a bending / extending motion, the outputs from the vein pattern extraction unit 40 and the finger bending angle calculation unit 42 are obtained based on the above-described user registration. Then, the authentication data generation unit 44 generates authentication data and outputs it to the pattern matching unit 48.

  The pattern matching unit 48 selects an arbitrary finger bending angle θ, and extracts only the vein pattern image corresponding to the angle θ from the authentication data input from the authentication data generation unit 44. Here, for convenience of explanation, a case where a vein pattern image corresponding to seven angles of finger bending angles θ = 15 °, 20 °, 25 °, 30 °, 35 °, 40 °, and 45 ° is extracted. Let's take an example.

  Next, the pattern matching unit 48 reads out the template information stored in the template storage unit 46 for each registered user, and from the template information, the finger bending angle θ = 15 °, which is the same as the previously selected angle, 20 Seven vein pattern images stored in correspondence with °, 25 °, 30 °, 35 °, 40 °, and 45 ° are extracted.

  Subsequently, the pattern matching unit 48 performs pattern matching on the vein pattern image extracted from the authentication data using the vein pattern image extracted from the template information as a template. This pattern matching is performed every finger bending angle θ = 15 °, 20 °, 25 °, 30 °, 35 °, 40 °, 45 °, and the matching result (similarity) for each angle Find the average value.

  The pattern matching unit 48 determines whether the average value of the matching results (similarity) obtained in the above-described procedure is equal to or greater than a predetermined threshold value, and derives an average value equal to or greater than the threshold value for any template information. In this case, a registered user corresponding to the template information is specified, and it is determined that the person to be authenticated is the registered user. On the other hand, if the average value of the matching results (similarity) does not exceed a predetermined threshold for any of the template information, the pattern matching unit 48 determines that the authentication of the person to be authenticated has failed. The above-described determination result can be output from the pattern matching unit 48 to the authentication result display unit 20 as shown in FIG. Note that the above-described threshold value can be appropriately determined in view of the required accuracy of authentication and the like.

  Here, a method in which the finger bending angle calculation unit 42 calculates “finger bending angle θ” for each image will be described. The “bending angle θ of the finger” in the present embodiment is calculated according to the same procedure as the method disclosed in Japanese Patent Application No. 2008-91920 (Japanese Patent Laid-Open No. 2009-245231) filed earlier by the present applicant. Can do. The procedure will be outlined below with reference to FIG. The digital image transferred from the finger imaging unit 12 is subjected to a binarization process for binarizing the background pixel to black (luminance value 0) and the finger portion to white (luminance value 255), and then appropriately using a Laplacian filter or the like. The edge pixel which becomes the outline of the finger is extracted. As a result, the outline of the side surface of the finger is acquired as a thin line E having a width of one pixel as shown in FIG. In FIG. 7, for convenience of explanation, black and white are shown inverted.

  Next, with respect to the contour line E shown in FIG. 7A, the pixels constituting the contour line E are scanned from the point “START” shown in the figure to the point “END” shown in the figure. Then calculate the curvature sequentially. Here, the curvature means an amount representing the degree of curve bending, and in the case of two dimensions, a value given by the reciprocal (1 / R) of the circumscribed circle radius R of a triangle formed by any three points on the curve. And the greater the degree of bending, the greater.

  The curvature calculated by the above-described procedure is shown in FIG. 7B by reference numerals 1 and 6 related to the second joint of the finger, reference numerals 2 and 5 related to the first joint of the finger, and the nail. The extreme value (maximum value / minimum value) is always shown in the portion of reference numeral 3 corresponding to the root of the symbol and the portion of reference numeral 4 corresponding to the tip of the finger. Here, since the curvature is always the minimum value in the portion 4 corresponding to the tip of the finger, the position where the calculated curvature shows the minimum value is specified as the tip of the finger, and from the position of the tip of the finger The positions of reference numerals 1 to 6 are specified a priori.

  The “finger bending angle” in the present embodiment is obtained from the coordinates of the pixels indicated by the reference numerals 1 to 6 specified in the above-described procedure. That is, the vertical bisector L1 passing through the midpoint M of the line segment connecting the pixel O at the position 1 and the pixel P at the position 6 related to the second joint of the finger is used as a reference line. The angle θ formed by the line segments L2 and L1 connecting the point M and the pixel Q located at the position 4 corresponding to the tip of the finger is defined as the “finger bending angle” in this embodiment.

  Finally, the superiority of the personal authentication method and personal authentication system of the present invention will be described below from the viewpoint of impersonation resistance. Here, suppose a case where a continuous template image (a continuous image of a vein pattern that changes with the bending and stretching motion of a finger) flows out from the personal authentication system of the present invention. A third party who acquires these template images can produce an artificial finger to which the template images are transferred. However, in order to impersonate the personal authentication system of the present invention, it is necessary to continuously present finger vein patterns of different bending angles to the system in time series.

Here, even if it is possible to create a plurality of corresponding artificial fingers for all of the acquired continuous template images, it is physically impossible to present them to the system in a short time in order. It is. Moreover, even if it is assumed that this is possible, such impersonation can be easily eliminated by defining the imaging period from the time when the finger is inserted into the imaging space to the next time it is removed. . In other words, in order to impersonate this system, it is necessary to create a three-dimensional movable artificial finger that has the same vein pattern and skeleton as the person's finger and can bend and stretch like the person's finger. However, since information about the skeleton of the person's finger cannot be acquired from the template image, it is practically impossible to produce such an artificial finger. Therefore, even if the template image (authentication information) leaks from the system, the personal authentication system of the present invention has high resistance against impersonation.

  As described above, the present invention has been described with the embodiment. However, the present invention is not limited to the above-described embodiment. For example, as an alternative to the above-described polarizing filter in the finger imaging unit, a black solution (light-absorbing It is possible to employ a transparent container in which liquid is stored to such an extent that the side surface of the finger can be dipped lightly. By irradiating near-infrared light from the bottom surface of the transparent container and bending and stretching with the side surface of the finger in contact with the bottom surface of the transparent container, light other than the transmitted light of the finger is automatically generated according to the movement of the finger. Can be blocked. In addition, it is included in the scope of the present invention as long as the effects and effects of the present invention are exhibited within the scope of embodiments that can be considered by those skilled in the art.

  Hereinafter, the present invention will be described more specifically with reference to examples. However, the present invention is not limited to the examples described below.

  The personal authentication system 10 described above with reference to FIG. 1 was constructed, and an evaluation experiment of this system was performed with the cooperation of four subjects (A, B, C, D). First, for each of the four subjects (A, B, C, D), a continuous image was acquired in which the index finger's flexion and extension motion was imaged from the side. From each of the acquired continuous images, calculate the `` finger bending angle θ '', and for each image when the angle θ is 15 °, 20 °, 25 °, 30 °, 35 °, 40 °, 45 °, A vein pattern image (registered image) was generated, and each was associated with an angle θ to serve as template information for each person.

  The binarization process for extracting the vein pattern described above with reference to FIG. 5D was performed according to the following procedure. That is, a 13 × 13 pixel area centered on the target pixel X is defined, and a pixel constituting the outermost periphery of the area is defined as a peripheral pixel Y. The region is defined such that the peripheral pixel Y is separated from the target pixel X by a distance corresponding to about half of the vein thickness on the image (therefore, the target pixel X is located in the vein portion). In this case, the luminance value of the target pixel X should be smaller than the average value of the luminance values of the peripheral pixels Y). Therefore, when “the luminance value of the pixel of interest <the average of the luminance values of the surrounding pixels”, the pixel of interest X is determined as a vein part and stored as a black pixel. It was judged that it was not part and was saved as a white pixel.

  Further, the noise processing described above with reference to FIG. 5 was performed according to the following procedure. That is, a 9 × 9 pixel area centered on the target pixel is defined, the number of black pixels and white pixels in the area is counted, and if the number of black pixels is less than 70%, the target pixel is The process of saving as pixels was repeated for all pixels in the image.

  Next, in the same procedure as described above, for each of the four subjects (A, B, C, D), a continuous image of the index finger that flexes and stretches is acquired again, and from each of the acquired continuous images, “ For each image when the finger bending angle θ is 15 °, 20 °, 25 °, 30 °, 35 °, 40 °, 45 °, a vein pattern image (input image) is generated, and each is set to an angle θ Corresponding information was input for each person.

  Then, using the template information of each person, pattern matching was performed with the input information of the person and each person (three persons) other than the person, and the similarity was calculated for each combination. Pattern matching was performed twice for each combination. In this example, the similarity was calculated using the following formula 1.

  FIG. 8 shows the experimental results. Table (a) shows the result of pattern matching between the input information of subjects A to D and the template information of the person. Table (b) shows the result of pattern matching between the input information of subject A and the template information of subjects B, C, and D (three persons). Table (c) shows the input of subject B. The results of pattern matching between the information and template information of subjects A, C, and D (3 persons) are shown. Table (d) shows the input information of subject C and subjects A, B, and D (3 persons). (E) shows the result of pattern matching between the template information of the subject D and the pattern matching between the input information of the subject D and the template information of the subjects A, B, and C (three persons). Results are shown. In Tables (a) to (e), the degree of similarity calculated for each combination is shown for each “finger bending angle θ” (15 ° to 45 °), and the average value is shown in the bottom row.

  As shown in the above table (a), when pattern matching was performed between the input information of the principal and the template information of the principal, the degree of similarity exceeding 0.50 was shown for any combination. On the other hand, as shown in Tables (b) to (e), when pattern matching is performed between the input information of the person and the template information of the other person, the similarity is less than 0.50 for any combination. A significant difference was observed between the two. The above results showed the practicality of the personal authentication system of the present invention.

  As described above, according to the present invention, a novel personal authentication method and personal authentication system using a vein pattern are provided. The present invention is expected to contribute to improvement of impersonation resistance in finger vein authentication technology.

DESCRIPTION OF SYMBOLS 10 ... Personal authentication system, 12 ... Finger imaging part, 14 ... Personal authentication part, 16 ... User, 18 ... Finger, 20 ... Authentication result display part, 30 ... Case, 32 ... Opening part, 34 ... Light emission means, 36 ... Imaging means 37... Band filter 38. Polarization filter 39. Polarization filter 40. Vein pattern extraction unit 42. Finger bending angle calculation unit 44. Authentication data generation unit 46. Template storage unit 48. Matching part

Claims (6)

A personal authentication method using finger flexion and extension,
Imaging light transmitted through the side surface of the finger of the subject performing bending and stretching, and obtaining a plurality of continuous digital images for the side surface of the finger;
Extracting a vein pattern for each digital image;
Storing the vein pattern in association with the bending angle of the finger,
A personal authentication method, wherein authentication is performed by pattern matching the vein pattern of the person to be authenticated for each predetermined finger bending angle.   The personal authentication method according to claim 1, wherein the bending angle of the finger is obtained based on position information corresponding to an extreme value of curvature of the contour of the finger. A personal authentication system that uses the bending and stretching movements of fingers,
A finger imaging unit that images light transmitted through the side surface of the finger of the person to be authenticated performing bending and stretching movements, and acquires a plurality of continuous digital images for the side surface of the finger
Including a personal authentication unit to which information of a digital image acquired by the finger imaging unit is input,
The personal authentication unit
A vein pattern extraction unit, a finger bending angle calculation unit, an authentication data generation unit, a template storage unit, and a pattern matching unit;
The vein pattern extraction unit extracts the vein pattern for each of a plurality of digital images on the side surface of the finger,
The authentication data generation unit generates authentication data by associating the vein pattern with the bending angle of the finger, and stores the authentication data as template information in a template storage unit at the time of user registration, and the authentication data at the time of authentication. Is output to the pattern matching unit,
The pattern matching unit performs authentication by pattern matching a vein pattern included in the authentication data and a vein pattern included in the template information for each predetermined finger bending angle.
Personal authentication system.   The finger imaging unit includes a light emitting unit for irradiating the side surface of the finger with near-infrared light, and an imaging unit for imaging light transmitted through the side surface of the finger. The personal authentication system according to claim 3, wherein an imaging space of the finger is formed between two polarizing filters.   5. The personal authentication system according to claim 3, wherein the finger bending angle calculation unit calculates the finger bending angle based on position information corresponding to an extreme value of a curvature of the finger outline. The finger imaging unit includes a sensor for detecting the insertion / removal of a finger, automatically starts imaging when the finger is inserted into the imaging space, and automatically when the finger is removed from the imaging space. The personal authentication system according to claim 3, wherein the imaging is terminated.