JP2009037463A - Personal authentication device and personal authentication method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a personal authentication technique using a nail as biometrics information which is updated, for example, about every several weeks. <P>SOLUTION: An automatic door opening and closing system is provided with: an authentication information database 13 registering a user ID, a password, an image of nail pattern (registered image) and a registration time of a registrant, as personal authentication information for authenticating the existing registered person, in association with each other; a user authentication unit 12 for comparing the user ID and the password of a person to be authenticated with registered information of the authentication information database 13 to authenticate the person, in the authentication; a nail pattern extracting unit 14 for extracting the nail pattern from the image of a nail picked up by an imaging unit 4, after authenticated; a nail pattern predicting unit 15 for reading out the registration time of the nail pattern image of the authenticated user, and for changing the registered image of the authentication information database 13, based on an elapsed time from the registration time to the present time; and a nail pattern authenticating unit 16 for collating the changed registered image with an actual pattern to authenticate the person to be authenticated. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a personal authentication device and a personal authentication method for performing personal authentication using a fingernail (biological body) of a human finger.

  In recent years, there have been many crimes of improperly acquiring personal authentication information such as user IDs and passwords to impersonate users, illegally entering the company, or acquiring money from personal accounts of financial institutions. In addition to the above personal authentication information, personal authentication technology that combines biometric information that is difficult to be stolen has been developed.

  For example, user authentication systems using fingerprints, irises, faces, vein shapes, etc., PC login management systems using fingerprints, faces, etc., automatic door opening / closing systems, etc. have been put into practical use.

  By the way, since most biometric information such as fingerprints and irises does not change throughout their lifetime, there is a problem that they cannot be changed if the information is stolen.

  Therefore, when using biometric information, it is often desirable to use information that changes due to aging, such as a face, while updating it over a certain period.

  However, there is a concern that photographing of the face may be problematic from the viewpoint of privacy protection. Further, although the face shape is biological information that changes over time, since it is exposed, it is biological information that can be easily voyeurized from a distance with a camera or the like and is relatively easily stolen.

  Further, the face is difficult to be invalidated because the period until the change appears is relatively long. Furthermore, since there is usually only one face shape for an individual, it is impossible to take countermeasures such as changing to other finger information such as a fingerprint even in an emergency such as theft.

Therefore, in recent years, a technique using a fingernail for biometric authentication has been proposed (see, for example, Patent Document 1).
JP 2003-256840 A

  In the case of the technique of the above-mentioned patent document 1, as the characteristic amount of the fingernail, the arc shape of the boundary between the nail plate and the upper nail skin near the nail half moon at the base of the nail, the width of the nail plate, the three-dimensional shape of the nail plate, etc. In order to authenticate a portion of the nail that does not change substantially, it is assumed that periodic updates are performed at a cycle of several months to one year.

  However, because it is easy to forge with a body part that is easy to cut or cut like a nail, or forgetting that the registrant has registered, inadvertently cutting the nail, etc. In the biometric authentication to be used, it is desirable to update in a short period of time, for example, from several days to several weeks.

  The present invention has been made to solve such a problem. In consideration of the change in using a fingernail for biometric authentication, for example, personal authentication suitable for renewal in a short period of time such as several days to several weeks. An object of the present invention is to provide a personal authentication apparatus and a personal authentication method capable of providing a technique.

  In order to solve the above problems, the personal authentication device of the present invention provides first personal information that does not change over time as personal authentication information for authenticating an existing registrant individual, 2 personal information, a database in which registration times of the second personal information are registered correspondingly, a reception unit that receives the first personal information input from the person to be authenticated, and the first personal that is received by the reception unit A first authentication unit that compares the information with the first personal information of the registrant registered in the database and authenticates when the first personal information matches, and when authenticated by the first authentication unit, An imaging unit that captures an image including the nail surface of the authenticator, and an image of the nail pattern or an image obtained by normalizing the image from the image including the nail surface captured by the imaging unit. And a registration time of the second personal information corresponding to the first personal information authenticated by the first authentication unit is read from the database, and based on the elapsed time from the registration time to the present time The nail pattern prediction unit for changing the nail pattern image of the registrant registered in the database or the normalized data thereof, the nail pattern image changed by the nail pattern prediction unit or the normalized data and the nail pattern extraction And a second authentication unit that verifies the image of the nail pattern extracted by the unit or the normalized data thereof and authenticates the person to be authenticated as the registrant based on the matching rate.

  In the present invention, personal authentication is performed using a nail pattern such as a scratch or a stripe pattern that appears on the surface of a nail, which is a living body that changes at a relatively high speed. The surface of the nail is a part that tends to change over time in the human body. For this reason, even if the information is stolen, it is invalidated in a relatively early period compared to, for example, a face, and is suitable as biometric information to be used while being updated in a relatively short period of time. In addition, since information corresponding to the number of fingers can be used as in the case of fingerprints and finger veins, information can be replaced relatively easily during theft.

  According to the present invention, it is possible to provide a personal authentication technique suitable for renewal in a short period of time such as several days to several weeks in consideration of the change in using a nail for biometric authentication.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram showing the configuration of the automatic door opening and closing system of the first embodiment.

  This automatic door opening and closing system includes first personal information (user ID, password) that does not change over time and second personal information (a nail pattern that is one of human biological information) that changes over time. Is a personal authentication device that authenticates a user entering and exiting an entrance and opens and closes a door.

  As shown in FIG. 1, the automatic door opening / closing system according to the first embodiment includes a control device 1, a card reader 2, an imaging unit 4, a door opening / closing unit 5, a notification unit 6 such as a display and a speaker, and the like.

  The card reader 2 includes a card reading unit and a numeric keypad. The person to be authenticated reads information (user ID) from an IC card 3 such as an electronic employee ID card passed through the card reading unit, and serves as a password designated by the numeric keypad. It is sent to the control device 1 together with the password.

  Note that the user ID and password may be entered with the numeric keypad without using the IC card 3, and both the user ID and password are stored in advance in the IC card 3 and only the information on the IC card 3 is read. It may be.

  As shown in FIG. 2, the imaging unit 4 includes an illumination device 21, a CCD camera 22, a finger installation base 25, and the like. The finger installation table 25 is a table for positioning the finger 23 to be imaged. The finger placement base 25 is placed with the belly of the finger 23 facing down.

  The CCD camera 22 is arranged at a position about several centimeters above the finger setting table 25 and picks up an image of the finger 23 placed on the finger setting table 25. The illuminating device 21 is arranged so as to illuminate from the side of the CCD camera 22 toward the finger setting table 25, and illuminates the finger 23 placed on the finger setting table 25.

  In this example, as shown in FIG. 2, the finger 23 placed on the finger setting base 25 is illuminated from the upper front, and the VGA (640 × 480) pixel size is almost directly above the finger 23 by the CCD camera 22. A color image or monochrome image is taken. In addition, for example, an infrared camera or an ultraviolet camera may be used.

  In this case, illumination corresponding to these cameras is used. Further, the shooting direction of the camera and the direction of illumination are not limited to this example, and illumination may be performed from above, below or from the side of the finger, for example.

  In this embodiment, direct illumination is performed from the illuminating device 21, but transmitted light may be used, reflected illumination through a reflector may be performed, and if there is a certain level of illuminance, illumination is performed. It does not have to be.

  Furthermore, you may acquire the three-dimensional information of a finger | toe with the image image | photographed from multiple directions. In the present embodiment, the pattern on the surface of the nail is acquired. However, for example, the back side of the nail or the nail pattern inside the nail may be acquired by using transmitted light.

The door opening / closing unit 5 is controlled by the control device 1 to open or close a door disposed at the entrance of the building. That is, the door opening / closing unit 5 permits or rejects the next work (entering the building from the entrance) performed by the person to be authenticated according to the result of personal authentication by the control device 1.
In addition, this example is only an example, and when the control target of the control device 1 is not the door opening / closing unit 5 but a computer (hereinafter referred to as “PC”), for example, when the authenticated person logs in to the PC, You may allow or deny login.

  As a result of the personal authentication process performed by the control device 1, the notification unit 6 notifies the person to be authenticated that the user has not been authenticated by voice or display.

  The control device 1 includes the surface portion of the nail of the finger 23 placed on the finger installation base 25 by controlling the illumination device 21 and the CCD camera 22 when imaging the finger 23 to optimize exposure and adjust the focus. Take an image.

  The control device 1 receives the information of the IC card 3 read by the card reader 2 and the image of the finger 23 imaged by the CCD camera 22, performs authentication processing for each, and according to the results of both authentication processing The door opening / closing mechanism 5 is controlled to open and close the entrance door.

  The control device 1 includes an authentication information database 13, a reception unit 11, a user authentication unit 12 as a first authentication unit, a nail pattern extraction unit 14, a nail pattern prediction unit 15, a nail pattern authentication unit 16 as a second authentication unit, a nail A pattern registration unit 17 and a memory 18 are included.

  The authentication information database 13 includes, as personal authentication information for authenticating existing registrants, first personal information (user ID, password) that does not change over time, and second individual that changes over time. The information (the image of the nail pattern or its normalized data) and the registration time of the second personal information are registered in association with each other.

  The authentication information database 13 may hold the nail image itself in addition to the nail pattern image. In addition, for each nail pattern, information of a registered person corresponding to the nail pattern such as a name may be associated and registered.

  The accepting unit 11 accepts information on the IC card 3 of the person to be authenticated and the person to be registered read by the card reader 2. The accepting unit 11 accepts first personal information (user ID, password) input from an authenticated person or a registered person.

  The user authentication unit 12 receives the first personal information (user ID, password) received by the receiving unit 11 and the first personal information (user ID, password) of the registrant registered in the authentication information database 13. In comparison, the person to be authenticated is authenticated with the registrant himself by matching each other's information. The authentication here is primary authentication (provisional authentication).

  When the user authentication unit 12 performs primary authentication, the imaging unit 4 captures an image including the nail surface of the person to be authenticated.

  The nail pattern extraction unit 14 extracts a nail pattern image or data obtained by normalizing the image from the image including the nail surface imaged by the imaging unit 4. In other words, the nail pattern extraction unit 14 performs predetermined image processing (such as nail portion detection, pattern cutout processing, normalization processing) on the captured finger image, identifies the nail portion, and extracts the nail portion from the nail portion. Extract the pattern.

  The nail pattern prediction unit 15 includes second personal information (a nail pattern image or normalized data thereof) corresponding to the first personal information (user ID, password) that has been provisionally authenticated (primary authentication) by the user authentication unit 12. The registration time is read from the authentication information database 13 and the image of the registrant's nail pattern registered in the authentication information database 13 or its normalized data is changed based on the elapsed time from the registration time to the present.

  The nail pattern authentication unit 16 collates the nail pattern image changed by the nail pattern prediction unit 15 or the normalized data thereof with the nail pattern image extracted by the nail pattern extraction unit 14 or the normalized data thereof. Whether the user is the registrant is authenticated based on the similarity (matching rate).

  For example, in the comparison between images, the degree of overlap between the images is regarded as a similarity, and if the similarity is 90% or more, for example, the person to be authenticated is authenticated as a registrant. Note that the similarity value of 90% or more in this example is an example, and other values may be used.

  The nail pattern registration unit 17 registers the nail pattern extracted by the nail pattern extraction unit 14 in the authentication information database 13. In addition to registering the nail image itself in the authentication information database 13, for example, the image itself captured by the imaging unit 4, or an image or vector data subjected to normalization processing such as alignment and size, for example May be saved.

  When the nail pattern registration unit 17 is authenticated as a regular employee (registrant) by the nail pattern authentication unit 16, the nail pattern registration unit 17 extracts the nail pattern image extracted by the nail pattern extraction unit 14 or its normalized data and the current time. It functions as a biometric information update unit (nail pattern update unit) that is registered or overwritten as personal authentication information of the corresponding registrant in the authentication information database 13.

  The memory 18 functions as a work area for personal authentication processing (comparison and collation of images, image extraction, and a calculation area for changing the nail pattern image or its normalized data at a change rate according to the period). The memory 18 stores captured images, changed images, and the like.

  The memory 18 stores in advance the change rate for changing the nail pattern or its normalized data according to the period and the movement calculation formula of the x and y coordinates of the image, and the nail pattern prediction unit 15 stores the registration date. By calculating the period from the date to the authentication date and substituting it into the above formula, the predicted image can be generated by moving the pixel position of the registered image.

  The operation of the automatic door opening / closing system according to the first embodiment will be described below with reference to the flowchart of FIG. First, the nail pattern registration operation will be described.

  In the authentication information database 13, a user ID and a personal identification number, which are first personal information of the registered person, are registered in advance, and the user ID is also registered in advance in the IC card 3 possessed by the registered person. It shall be.

  In this case, first, when the registered person places the IC card 3 on the card reader 2, the card reader 2 reads the user ID from the IC card 3. When the ten-key is pressed by the registered person and a password is input, the card reader 2 sends the first personal information (user ID and password) to the control device 1.

  In the control device 1, the reception unit 11 receives the first personal information (user ID and password) and passes it to the user authentication unit 12.

  When the user authentication unit 12 receives the first personal information (user ID and password) from the reception unit 11 (step S101 in FIG. 3), the user authentication unit 12 performs primary authentication processing based on the first personal information (user ID and password). Perform (step S102).

  Specifically, the user authentication unit 12 searches the registration information in the authentication information database 13 one by one using the received first personal information (user ID and password) as a key, and matches the registration information in the authentication information database 13. Check whether or not to do.

  When the received first personal information (user ID and personal identification number) matches the registration information in the authentication information database 13 (Yes in step S103), the user authentication unit 12 causes the imaging unit 4 to start imaging the finger 23. .

  The imaging unit 4 images the finger image of the finger 23 placed on the finger installation base 25 (step S104) and passes it to the nail pattern extraction unit 14. At this time, the imaging unit 4 captures at least one finger image of the registered person's finger 23.

  The nail pattern extraction unit 14 performs image processing on the finger image I (see FIG. 4) passed from the imaging unit 4 and includes an image F including a pattern on the surface of the nail 24 (hereinafter referred to as “nail pattern F”). To get.

  Specifically, the nail pattern extraction unit 14 detects the contour of the nail 24 from the finger image I passed from the imaging unit 4 and extracts the nail pattern F on the surface of the nail 24 (step S105), and the memory 18 is stored. As a method for extracting the nail portion, for example, a method such as template matching is used.

  Subsequently, the nail pattern extraction unit 14 uses a vertical pattern, a horizontal pattern, a diagonally descending pattern, and a diagonally rising pattern from the image of the nail pattern F stored in the memory 18 using the edge extraction filters in the vertical, horizontal, and diagonal directions. Extract patterns and scars in four directions such as patterns.

  That is, the nail pattern extraction unit 14 filters the extracted nail pattern F to extract a plurality of images such as a vertical pattern image R, a horizontal pattern image S, a diagonally falling pattern image T, and a diagonally rising pattern image U. To do.

  Thereafter, the nail pattern extraction unit 14 normalizes the image size of each nail part. In this example, each image R, S, T, U is normalized to a square of 50 pixels each vertically and horizontally, and the sum of the values in the corresponding range in the image before normalization is used as the value of each pixel.

  The arrangement of these as vectors is referred to as normalized data X. In this case, the normalized data X is represented by a 10,000-dimensional vector. That is, the normalized data X can be said to be a nail pattern F extracted by the nail pattern extraction unit 14.

  That is, the normalized data X is obtained by reducing the images R, S, T, and U to a size of 50 × 50 pixels and aligning them in the vertical direction to form one pattern).

  Note that the processing of the nail image is an example, and other information such as color information may be added to the characteristics of the nail pattern, and the nail pattern F is selected by performing feature selection on the normalized data X. May be dropped from the normalized data X.

  In addition, when a plurality of nail images are taken, a partial space created by performing principal component analysis on the nail images may be used as the nail pattern F.

  Then, the nail pattern extraction unit 14 registers the normalized data X in the authentication information database 13 together with today's time (registration time) in association with the first personal information of the corresponding registrant in the authentication information database 13 (step S106). ).

  The registered person information stored in the authentication information database 13 includes, for example, an employee name, an employee ID, an employee name, authority information, an authentication expiration date, an operation command for the door opening / closing mechanism 5 and the like. .

  In this example, it is assumed that the authentication information database 13 is stored in a hard disk drive device built in the control device 1, but is directly connected to the control device 1, for example, disposed on the Internet. A database on a server that is not available may be used, or the authentication information database 13 may be stored on a portable medium and connected as necessary.

  Further, in this example, the authentication information database 13 is single and all registered users are registered in one and the same database. This is optional so that each authentication information database includes one or more registered users. It may be divided into a plurality of databases.

  For example, as an example, it is assumed that data for one registered person is stored in an IC chip built-in card or the like, and the registered person and the person to be authenticated are authenticated one-on-one.

  A plurality of control devices 1 may use the same authentication information database 13. As an example of this, when the users of the control devices 1 installed in a plurality of rooms are common, it is conceivable to connect them online and use the centrally managed authentication information database 13.

  Next, with reference to the flowcharts of FIGS. 5 and 6, an authentication operation for a person to be authenticated after a certain period of time has elapsed after nail pattern registration will be described. As a precondition, it is assumed that information of at least one registrant has already been registered in the authentication information database 13.

  In this case, since the first personal information receiving process to the nail pattern extracting process (steps S111 to S115) are substantially the same as those in the registration process shown in FIG. 3, the description of each process is omitted. If the first personal information does not match in the primary authentication (No in step S113), the user authentication unit 12 reports an error from the display, the speaker, and the like (step S118).

  In this case, it is determined that the person to be authenticated is one of those registered in the authentication information database 13 by the primary authentication. After the nail pattern extraction process in step S115, the nail pattern prediction unit 15 reads the registration date and time of the nail pattern from the registration information of the primary authentication person corresponding to the authentication information database 13, and changes over time from the time of registration to the present time. The registered nail pattern is predicted (step S116), and the changed pattern (hereinafter referred to as “predicted pattern”) is stored in the memory 18.

  Specifically, the nail pattern prediction unit 15 reads the registration date and time of the nail pattern from the registration information of the corresponding registrant in the authentication information database 13 and calculates the elapsed time up to the current date and time. For example, if the last registration date / time is 13:00 on July 4, and the current date / time is 13:00 on July 9, then 5 days have passed and 120 hours have passed since 5 days × 24 hours. Become.

  After calculating the elapsed time, the nail pattern prediction unit 15 reads the registered image from the authentication information database 13 and performs an operation to grow the registered image by the elapsed time (120 hours). Predict elongation (change).

  As an example, 120 hours are input to a preset growth function, and a changed image is predicted. For example, the vertical pattern of the vertical pattern image R is extended in the Y-axis direction by 120 hours, the horizontal pattern of the horizontal pattern image S is moved in the Y-axis direction by 120 hours, and the diagonal pattern images T and U are The movement and extension are mixed at a predetermined ratio.

  Note that the length of the registrant's nail 24 growing every day may be maintained, and the pattern of the registered image may be extended by 5 days.

  Subsequently, the nail pattern authentication unit 16 compares the predicted pattern stored in the memory 18 with the actual nail pattern F (hereinafter referred to as “actual pattern”) extracted from the pattern read from the finger of the person to be authenticated. Thus, a determination is made according to the matching rate (similarity).

  In this embodiment, the nail pattern authentication unit 16 calculates the Euclidean distance between the actual pattern and the predicted pattern, determines that the distance is equal to or less than a threshold value, and determines that the distance is greater than the threshold value, and determines that the distance does not match. In addition, about the method of comparison collation of a nail pattern, it is not restricted to said method, For example, you may use the inner product value of a real pattern and a prediction pattern as a similarity.

  If the nail pattern authentication unit 16 determines that there is a mismatch as a result of the comparison of the nail patterns (No in step S117), the nail pattern authentication unit 16 notifies an error from the display, the speaker, and the like (step S118), and prompts re-authentication.

  When the nail pattern authentication unit 16 determines that the nail patterns match each other as a result of the comparison (Yes in step S117), the nail pattern authentication unit 16 controls the door opening / closing mechanism 5 to open the door (step S119). Allow admission (traffic). Then, the claw pattern authentication unit 16 controls the door opening / closing mechanism 5 so as to close the door after a predetermined time has elapsed (after 1-2 seconds or the like), and closes the door (step S119).

  As described above, according to the automatic door opening and closing system of the first embodiment, the user authentication using the IC card 3 and the biometric authentication of the nail pattern can be used for renewal in a short period of time such as several days to several weeks. It is possible to provide suitable personal authentication technology. It is also possible to construct an authentication system that is robust against theft.

  Then, the automatic door opening and closing system of 2nd Embodiment is demonstrated. In addition, the same code | symbol is attached | subjected to the structure same as 1st Embodiment, and the description is abbreviate | omitted.

  As shown in FIG. 7, the control device 1 of the automatic door opening and closing system according to the second embodiment further includes a feature detection unit 19.

  The feature detection unit 19 detects feature information given (printed or coated with infrared ink) to the nail 24 of the registered person. The characteristic information given to the nail 24 is, for example, registered person information such as a marker, ID and name or the encrypted information, an operation command for the door opening / closing mechanism 5 and the like.

  That is, the feature detection unit 19 detects feature information from the nail pattern extracted by the nail pattern extraction unit 14 and passes it to the nail pattern authentication unit 16.

  As a device for giving characteristic information to the nail 24, for example, a printing device or a writing device for printing special information such as characters and symbols on the nail 24 with special ink such as infrared or ultraviolet ink in addition to normal color or monochrome ink. A printing device that prints on the nail 24 with a material that absorbs or reflects a specific light such as a tool or a contrast agent, a laser processing device that cuts the surface of the nail, and the like.

  When a printing apparatus that performs printing using a contrast agent is used, there is an advantage that feature information can be extracted even when the surface is hidden by, for example, application of nail polish.

  In this case, the imaging unit 4 uses an infrared camera to image feature information printed or drawn on the nail 24 with invisible ink (special ink) such as infrared ink.

  The characteristic information includes characters, symbols, marks, and the like applied to the nail 24 with invisible ink that can be imaged by an infrared camera. Further, not only the infrared camera but also the CCD camera used in the first embodiment may be used together, for example, by imaging a marker or normal ink applied to the nail 24 and combining a plurality of information.

  Next, the operation of the automatic door opening and closing system of the second embodiment will be described with reference to the flowcharts of FIGS. First, the nail pattern registration operation will be described with reference to FIG.

  In the authentication information database 13, a user ID and a personal identification number, which are first personal information of the registered person, are registered in advance, and the user ID is also registered in advance in the IC card 3 possessed by the registered person. It shall be.

  In this case, first, when the registered person places the IC card 3 on the card reader 2, the card reader 2 reads the user ID from the IC card 3. When the ten-key is pressed by the registered person and a password is input, the card reader 2 sends the first personal information (user ID and password) to the control device 1.

  In the control device 1, the reception unit 11 receives the first personal information (user ID and password) and passes it to the user authentication unit 12.

  When the user authentication unit 12 receives the first personal information (user ID and password) from the reception unit 11 (step S201 in FIG. 7), the user authentication unit 12 performs primary authentication processing based on the first personal information (user ID and password). It performs (step S202).

  Specifically, the user authentication unit 12 searches the registration information in the authentication information database 13 one by one using the received first personal information (user ID and password) as a key, and matches the registration information in the authentication information database 13. Check whether or not to do.

  When the received first personal information (user ID and password) matches the registration information in the authentication information database 13 (Yes in step S203), the user authentication unit 12 starts imaging the finger 23 on the imaging unit 4 ( Start).

  Here, it is assumed that the registered person has drawn characters such as “Hanako” as characteristic information on the nail 24 with a pen filled with infrared ink. This example is only an example, and any feature information may be drawn.

  Thereafter, when the registered person places the finger 23 on the finger setting table 25, the imaging unit 4 images infrared imaging of the finger 23 placed on the finger setting table 25 (step S204) and passes it to the nail pattern extraction unit 14. . At this time, the imaging unit 4 captures at least one infrared image of the finger 23 of the registered person.

  The nail pattern extraction unit 14 performs image processing on the infrared image of the finger passed from the imaging unit 4 and performs an image F (hereinafter referred to as “nail outline and“ Hanako ”characters) on the surface of the nail 24. "Nail pattern F").

  Specifically, the nail pattern extraction unit 14 detects the outline of the nail 24 from the image of the finger 23 passed from the imaging unit 4, and a portion of the nail pattern F inside (the portion within the dotted line in the figure). Is extracted (step S205) and stored in the memory 18. As a technique for extracting an image of a part of the nail, for example, a technique such as template matching is used.

  Subsequently, the feature detection unit 19 detects “Hanako” which is feature information obtained by removing the contour from the image of the nail pattern F stored in the memory 18 (step S206), and cuts it out.

  That is, the feature detection unit 19 cuts out an image of characters “Hanako” from the image of the nail pattern F stored in the memory 18 and passes it to the nail pattern registration unit 17.

  The nail pattern registration unit 17 registers the passed image “Hanako” in the authentication information database 13 in association with the first personal information of the corresponding registrant in the authentication information database 13 together with today's current time (registration time). (Step S207).

  The registered person information stored in the authentication information database 13 includes, for example, a user ID such as an employee number, an employee name, authority information, an expiration date of authentication, and an operation command for the door opening / closing mechanism 5. . The registered person information need not be all of them, but only some of them are necessary.

  In this example, it is assumed that the authentication information database 13 is stored in a hard disk drive device built in the control device 1, but is directly connected to the control device 1, for example, disposed on the Internet. The authentication information database 13 may be stored in a portable medium, and may be connected and used as necessary.

  Further, in this example, the authentication information database 13 is single and all registered users are registered in one and the same database. This is optional so that each authentication information database includes one or more registered users. It may be divided into a plurality of databases.

  For example, as an example, it is assumed that data for one registered person is stored in an IC chip built-in card or the like, and the registered person and the person to be authenticated are authenticated one-on-one.

  A plurality of control devices 1 may use the same authentication information database 13. As an example of this, when the users of the control devices 1 installed in a plurality of rooms are common, it is conceivable to connect them online and use the centrally managed authentication information database 13.

  In this example, the user ID is used as the first personal information given to the registered person in the authentication information database 13. However, for example, the name of the employee may be used. Not all information is needed.

  Next, referring to FIGS. 8 and 9, the authentication operation for the person to be authenticated when a certain period has elapsed after the nail pattern is registered in the authentication information database 13 will be described. As a precondition, it is assumed that information of at least one registrant has already been registered in the authentication information database 13.

  In this case, since the first personal information receiving process to the nail pattern extracting process (steps S211 to S215) are substantially the same as those in the registration process shown in FIG. 7, the description of each process is omitted. When the first personal information does not match in the primary authentication (No in step S213), the user authentication unit 12 reports an error from the display, the speaker, and the like (step S219).

  It has already been determined in step S212 that the person to be authenticated is one of those registered in the authentication information database 13 in the primary authentication process (authentication using the user ID and password).

  Further, in the nail pattern extraction process in step S215, the nail pattern F (the actual pattern 51 in FIG. 9), which is an image of the nail 24 including the image of the characters “Hanako”, from the actually captured infrared image of the finger 23. Extracted and stored in the memory 18.

  The feature detection unit 19 detects “Hanako”, which is feature information obtained by removing the outline of the nail 24 from the actual pattern 51 stored in the memory 18 (step S216).

  That is, the feature detection unit 19 detects an image of characters “Hanako” from the image of the nail pattern F stored in the memory 18 as feature information and passes it to the nail pattern authentication unit 16.

  On the other hand, the nail pattern prediction unit 15 reads the registration date and time of the nail pattern from the registration information of the primary authentication person corresponding to the authentication information database 13, and predicts the registered nail pattern that changes over time from the time of registration to the present time. (Step S217), the changed predicted pattern 42 is stored in the memory 18.

  Specifically, the nail pattern prediction unit 15 reads the registration date and time of the nail pattern from the registration information of the corresponding registrant in the authentication information database 13 and calculates the elapsed time up to the current date and time. For example, if the last registration date / time is 13:00 on July 4, and the current date / time is 13:00 on July 9, then 5 days have passed and 120 hours have passed since 5 days × 24 hours. Become.

  After calculating the elapsed time, the nail pattern prediction unit 15 reads a registered image 41 (in this case, a nail image including “Hanako”) from the authentication information database 13 as shown in FIG. Calculation is made to move the position of “Hanako” by the elapsed time (120 hours), and the change of the registered image for 5 days is predicted.

  As an example, 120 hours are input into a preset transfer function, and a changed image is predicted. For example, since the image “Hanako” is an image that is only applied on the nail 24, the image is moved in the vertical direction, that is, the Y-axis direction for 120 hours according to the extension of the nail 24.

  In addition, as a method for predicting the position of “Hanako” that changes, the average length of the nail 24 that extends for each elapsed time (changes by one pixel in the Y-axis direction per day, etc.) is stored. A change in the position of the characters “Hanako” may be predicted from the elapsed time.

  Subsequently, the nail pattern authentication unit 16 compares and collates the predicted pattern 42 stored in the memory 18 with the actual pattern 51 that has been read from the finger 23 of the person to be authenticated and extracted the pattern (step S218). According to (similarity), determination of coincidence or non-coincidence (judgment of changed nail pattern is performed)

  In this embodiment, the nail pattern authentication unit 16 calculates the Euclidean distance between the actual pattern 51 and the predicted pattern 42, determines that the distance is equal to or less than a threshold value, and determines that the distance is greater than the threshold value. . In addition, about the method of comparison collation of a nail pattern, it is not restricted to said method, For example, you may use the inner product value of a real pattern and a prediction pattern as a similarity.

  If the nail pattern authentication unit 16 determines that they do not match as a result of the comparison of the nail patterns (No in step S218), the nail pattern authentication unit 16 reports an error from the display, the speaker, etc. (step S219), and prompts re-authentication.

  If the nail pattern authentication unit 16 determines that the nail patterns match each other as a result of the comparison (Yes in step S218), the nail pattern authentication unit 16 controls the door opening / closing mechanism 5 to open the door (step S220). Allow admission (traffic). Then, the claw pattern authentication unit 16 controls the door opening / closing mechanism 5 so as to close the door after a predetermined time has elapsed (after 1-2 seconds or the like), and closes the door (step S221).

  Here, with reference to FIG. 9, the processes of the nail pattern prediction unit 15 and the nail pattern authentication unit 16 will be described in accordance with the change of the image.

  In this case, first, the nail pattern prediction unit 15 reads the registration time of the nail pattern from the registered information of the already registered tentatively registered by the primary authentication registered in the authentication information database 13, and is the difference from the current time. Calculate elapsed time.

  Next, the nail pattern prediction unit 15 reads the registered image 41 of the registered person and changes it by the calculated elapsed time (step S301 in FIG. 9), and the position of the feature information “Hanako” changes (moves). A predicted pattern 42 of the nail image thus generated is generated.

  On the other hand, the nail pattern authentication unit 16 reads the actual pattern 51 stored in the memory 18 and obtains an error between the position of the predicted pattern 42 in the Y-axis direction and the position of the actual pattern 51 in the Y-axis direction. That is, the mutual displacement is measured (step S302).

  After the measurement, the nail pattern authentication unit 16 compares each other's error with the first threshold value Th2 preset in the memory 18, and determines whether the error is less than the first threshold value Th2 or exceeds the first threshold value Th2. .

  As a result of the comparison, if the error is equal to or less than the first threshold Th2, the nail pattern authentication unit 16 assumes that the position change state is correct with respect to the elapsed time.

  Next, the nail pattern authentication unit 16 superimposes the feature image 52 of the “Hanako” portion of the actual pattern 51 extracted from each image and the feature image 43 of the “Hanako” portion extracted from the predicted pattern 42. (Step S304), the coincidence rate (similarity) between the images is calculated (Step S305).

  As a result of this calculation, when the matching rate (similarity) is equal to or higher than the second threshold Th3 preset in the memory 18 and less than the third threshold Th4, the nail pattern authentication unit 16 determines that the images match each other. (Step S306). Note that the third threshold value Th4 may change (decrease) in value according to the elapsed time.

  This is because the feature information given to the nail surface (coating, printing, scratching, etc.) is intended to detect counterfeits that do not generate wear by utilizing wear and the like in daily life. In other cases, the nail pattern authentication unit 16 determines that there is a mismatch.

  In this example, the wear state and the movement amount of the actual pattern 51 (feature image 52) with the passage of time are detected. However, only one of them is used, that is, the nail 24 is scratched as feature information. If so, other measures may be taken, such as examining the progress of natural healing.

  Further, the content of the feature image 52 may be verified using, for example, an OCR or a barcode scanner. Further, the nail pattern F itself or a part thereof may be handled as the feature image 52, and these may be arbitrarily combined.

  When updating the registration information in the authentication information database 13, it is not necessary to add new feature information to the nail 24 each time authentication is performed, and the registration information in the authentication information database 13 is used with the feature information already provided. It may be overwritten and updated and used continuously.

  In addition, when updating the registration information, a registered user who matches the registered person is specified from the authentication information database 13 instead of adding a new one, and the registered information of the registered person is used as the registered person information. It may be rewritten.

  Examples of a method for designating a registered person include designation by key input and the like, and designation of a registered person as an authenticated person.

  In addition, when updating registration information, update confirmation may be performed with respect to a registered person or an administrator of the system.

  Further, when the authentication is successful as a result of the comparison of the nail patterns, the registered information of the registered person in the authentication information database 13 that is determined to be coincident with the information of the person to be authenticated is updated with the nail pattern F read this time. .

  In this case, imaging by the imaging unit 4, nail pattern extraction by the nail pattern extraction unit 14, feature information detection by the feature detection unit 19, etc. are not necessarily performed.

  As described above, according to the automatic door opening and closing system of the second embodiment, the user authentication by the IC card 3 and the change prediction of the special information (character “Hanako”) drawn with invisible ink on the nail are used in combination. For example, it is possible to provide a personal authentication technique suitable for renewal in a short period of time such as several days to several weeks. It is also possible to construct an authentication system that is robust against theft.

  In addition, this invention is not limited only to the said embodiment, You may deform | transform a component in the range which does not deviate from the summary in an implementation stage. Moreover, various inventions can be configured by appropriately combining a plurality of components disclosed in the embodiment.

  For example, some components may be deleted from all the components shown in the embodiment. Furthermore, constituent elements over different embodiments may be appropriately combined.

  In the above embodiment, the card reader 2 is connected to the control device 1, and the user ID stored in advance in the IC card 3 and the password entered from the numeric keypad are used as the first personal information. For example, as shown in FIG. 10, the fingerprint reading unit 7 may be connected to the control device 1 to authenticate the first personal information with the fingerprint of the person to be authenticated.

  In this case, as hardware, the imaging unit 4 and the fingerprint reading unit 7 are combined to form a lighting device 21, a CCD camera 22, and a fingerprint sensor 26 as shown in FIG. The reading and the imaging of the image including the nail 24 of the finger 23 are completed at once.

Specific applications for this application include passports for stay-type amusement parks, application to temporary employees who enter and leave the company for several days to several weeks, issuance of limited time tickets for transportation, keys for opening and closing safe doors, etc. It is valid.
In the above embodiment, the case where the characteristic information is characters has been described. However, different information such as characters and symbols, characters and marks, or the like may be combined.

It is a figure showing composition of an automatic door opening and closing system of a 1st embodiment. It is a figure which shows the imaging unit of the automatic door opening and closing system of FIG. It is a flowchart which shows the personal authentication process of the automatic door opening and closing system of FIG. It is a figure which shows the mode of the image processing of the automatic door opening and closing system of FIG. It is a flowchart which shows the authentication process of the image of the nail | claw which changes with time in an automatic door opening and closing system. It is a figure which shows the structure of the automatic door opening and closing system of 2nd Embodiment. It is a flowchart which shows the registration process of the biometric information in the automatic door opening / closing system of FIG. It is a flowchart which shows the personal authentication process of the automatic door opening and closing system of FIG. It is a figure which shows the mode of the image processing of the automatic door opening and closing system of FIG. It is a figure which shows the structure of the automatic door opening / closing system of an application example. It is a figure which shows the imaging unit of the automatic door opening / closing system of the application example.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 ... Control apparatus, 2 ... Card reader, 3 ... IC card, 4 ... Imaging unit, 5 ... Door opening / closing unit, 5 ... Door opening / closing mechanism, 6 ... Notification unit, 7 ... Fingerprint reading unit, 11 ... Reception unit, 12 ... User authentication unit, 13 ... authentication information database, 14 ... nail pattern extraction unit, 15 ... nail pattern prediction unit, 16 ... nail pattern authentication unit, 17 ... nail pattern registration unit, 18 ... memory, 19 ... feature detection unit, 21 ... Illumination device, 22 ... CCD camera, 23 ... finger, 24 ... nail, 25 ... finger mount, 26 ... fingerprint sensor.

Claims (6)

As personal authentication information for authenticating existing registrants, the first personal information that does not change over time, the second personal information that changes over time, and the registration time of the second personal information correspond. A registered database;
A reception unit for receiving first personal information input from the person to be authenticated;
A first authentication unit that compares the first personal information received by the reception unit with the first personal information of a registrant registered in the database, and authenticates by matching the first personal information;
When authenticated by the first authentication unit, an imaging unit that captures an image including the nail surface of the person to be authenticated;
A nail pattern extraction unit that extracts an image of the nail pattern or data obtained by normalizing the image from an image including the nail surface imaged by the imaging unit;
The registration time of the second personal information corresponding to the first personal information authenticated by the first authentication unit is read from the database, and the registrant registered in the database based on the elapsed time from the registration time to the present time. A nail pattern prediction unit that changes the image of the nail pattern or its normalized data;
The nail pattern image changed by the nail pattern prediction unit or the normalized data thereof is collated with the nail pattern image extracted by the nail pattern extraction unit or the normalized data, and the person to be authenticated is determined according to the matching rate. And a second authentication unit for authenticating the user as the registrant himself.   When the person to be authenticated is authenticated as the registrant by the second authentication unit, the nail pattern image extracted by the nail pattern extraction unit or its normalized data and the current time are registered in the database. The personal authentication apparatus according to claim 1, further comprising a biometric information update unit for overwriting and registering as personal authentication information of a person.   The personal authentication device according to claim 1, wherein the nail pattern is a pattern of the nail itself or characteristic information given to the nail at the time of registration.   4. The personal authentication apparatus according to claim 3, wherein the feature information includes any one of characters, symbols, and marks applied to the nails with invisible ink that can be imaged by an infrared camera.   The personal authentication device according to claim 1, wherein the person to be authenticated is allowed or rejected the next operation according to a result of authentication by the second authentication unit. As personal authentication information for authenticating existing registrants, the first personal information that does not change over time, the second personal information that changes over time, and the registration time of the second personal information are associated with each other. Registering with the database;
A reception unit receiving first personal information input from the person to be authenticated;
A first authentication unit comparing the first personal information received by the reception unit with the first personal information of a registrant registered in the database, and authenticating by matching the first personal information; ,
An imaging unit that captures an image including the authenticated nail surface of the person to be authenticated;
The nail pattern extraction unit extracting the image of the nail pattern or data obtained by normalizing the image from the image including the nail surface imaged by the imaging unit;
The registration time of the second personal information corresponding to the first personal information authenticated by the first authentication unit is read from the database, and the nail pattern prediction unit is registered in the database based on the elapsed time from the registration time to the present time. Changing the registrant's nail pattern image or its normalized data;
The second authentication unit collates the nail pattern image changed by the nail pattern prediction unit or its normalized data with the nail pattern image extracted by the nail pattern extraction unit or its normalized data, and the matching rate between them. And authenticating the person to be authenticated by the personal authentication method.

Images