JP2012108645A - Egress/ingress management system and personal identification data reading device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an egress/ingress management system and personal identification data reading device in which a person of a personal authentication target can be identified highly accurately.SOLUTION: An egress/ingress management system includes an image capturing device, an authentication management device, a personal identification data reading device having data reading means and data association means, and an egress/ingress management device. The authentication management device produces and outputs personal identification information processing information representing personal authentication processing in egress/ingress of a personal authentication target person on the basis of personal identification information. In the personal identification data reading device, the data reading means reads personal identification information from a personal identification information medium and outputs the read personal identification information to the authentication management device, and the data association means selects a still image meeting a predetermined condition from among a plurality of still images from the image capturing device, combines the selected still image with the personal identification information processing information from the authentication management device and outputs a result of the combination as detection image data. The egress/ingress management device displays contents corresponding to the personal identification information processing information received from the authentication management device, and displays the still image on the basis of the personal identification information processing information of the detection image data received from the data association means.

Description

  The present invention relates to an entry / exit management system for managing entry / exit of a person and a personal identification data reading device used for the entry / exit management system, and more particularly to an individual based on a personal authentication result and a photographed image of a person subject to personal authentication. The present invention relates to a person who can specify the person to be authenticated.

  This seed access control system includes a personal identification data reading device that performs biometric authentication using a fingerprint, iris of the pupil, or the like, or personal authentication using an IC card. In this system, the personal information read by the personal identification data reader is compared with the stored personal information and a personal authentication result signal indicating the success or failure of the authentication is output. Of people (persons subject to personal authentication) are managed. On the other hand, the entry / exit management system also enables the identification of the person to be personally authenticated who performs the personal authentication operation. In other words, in addition to the result of personal authentication, the personal authentication target person is identified by the function of displaying on the PC terminal a photographed image of the personal authentication target person at the authentication time. By having such a function, when the result of personal authentication is successful, it is possible to specify the name and face of the person to be personally authenticated. Further, when the personal authentication result is an authentication failure, the name of the person subject to personal authentication cannot be specified, but the face as a suspicious person can be confirmed from the photographed image of the person subject to personal authentication. As a result, security can be improved (for example, Patent Document 1).

JP 2003-27796 A

  However, since the conventional entrance / exit management system identifies a person to be personally authenticated using a captured image at the authentication time, for example, when the face is not directed toward the camera at the authentication time, the face is detected from the captured image. There was a problem that it was not possible to recognize and identify the person.

  The present invention has been made to solve such problems, and an object of the present invention is to provide an access control system and a personal identification data reader capable of accurately identifying a person to be personally authenticated.

  An entrance / exit management system according to the present invention includes an image photographing device arranged to photograph an entrance / exit of a specific area near an entrance / exit door, a plurality of pieces of personal identification information stored therein, received personal identification information, and the stored personal identification information And a personal authentication result signal indicating success or failure of authentication as a signal for controlling the unlocking and locking of the door of the entrance door, and a person subject to personal authentication based on the received personal identification information Generating and outputting personal identification information processing information representing personal authentication processing at the time of passing through the doorway, and detecting a personal identification information medium installed at the doorway near the doorway and storing personal identification information Then, the personal identification information is read from the personal identification information medium, the data reading means for outputting the personal identification information to the authentication management apparatus, and the image photographing apparatus. A data association unit that selects a still image that meets a predetermined condition from the selected still image and transmits the selected still image as detected image data that combines personal identification information processing information from the authentication management device. The personal identification data reading device and the personal identification information processing information from the authentication management device are displayed, and the content corresponding to the personal identification information processing information is displayed, and from the data association means of the personal identification data reading device And an entry / exit management device that receives the detected image data and displays a still image based on the personal identification information processing information of the detected image data.

  The personal identification data reading device according to the present invention comprises a data reading means for reading the personal identification information from the personal identification information medium when the personal identification information is sensed from the personal identification information medium storing the personal identification information, and a plurality of stationary A captured image storage means for storing an image, and an evaluation value for calculating an evaluation value of the still image by quantifying the face direction, size, brightness, eye position, etc. in each of the stored face images It is attached each time the personal identification information is read by the calculation means, the image selection means for extracting a predetermined still image from a plurality of still images based on the evaluation value calculated by the evaluation value calculation means, and the data reading means. Personal identification information processing information storage means for storing personal identification information processing information having a personal authentication processing number and the personal identification information, and the predetermined information extracted by the image selection means A data association unit including a detection image data generation unit configured to generate and output detection image data combining a still image and the personal identification information processing information stored in the personal identification information processing information storage unit; Features.

  The access control system and the personal identification data reader according to the present invention have the effect of being able to specify the person who is the subject of personal authentication with high accuracy.

The conceptual diagram which shows the entrance / exit management system which concerns on Embodiment 1 of this invention. The figure which shows an example of the traffic event information used for Embodiment 1 of this invention. The block diagram which shows the access control system which concerns on Embodiment 1 of this invention. The figure which shows an example of the value of each item for calculating the score value used for Embodiment 1 of this invention. The figure which shows an example of the display content in the display means 603 shown in FIG. The figure which shows an example of the display content in the display means 603 shown in FIG. The figure which shows an example of the display content in the display means 603 shown in FIG. 4A and 4B are diagrams illustrating data exchange between devices, where FIG. 4A illustrates a case where there is no reading operation by another person during shooting, and FIG. 4B illustrates a case where a reading operation is performed by another person during shooting. The flowchart showing operation | movement of Embodiment 1 of this invention. The figure which shows the other example of arrangement | positioning of the imaging device 400 in Embodiment 1 of this invention. The block diagram which shows the entrance / exit management system which concerns on Embodiment 2 of this invention. The flowchart showing operation | movement of Embodiment 2 of this invention.

Embodiment 1 FIG.
1 to 9 are diagrams showing an access control system according to Embodiment 1 of the present invention.

  As shown in the schematic diagram of FIG. 1, the access management system includes a personal identification data reading device 100, an authentication management device 300, an image capturing device 400, and an access management device 600.

  The personal identification data reading device 100 is installed at an entrance / exit of a specific area near the entrance / exit door 700 and includes a data reading unit 200 and a data association unit 500. When the data reading means 200 senses the contact or non-contact state of the personal identification information medium in which the personal identification information is stored, the data reading means 200 takes in the personal identification information from the personal identification information medium and outputs it. For example, the data reading means 200 is an IC card authentication device that reads personal information from an IC card that is a personal identification information medium, or biometric authentication (fingerprint authentication, fingerprint information that reads personal information from a fingerprint or iris of a personal identification information medium). Iris authentication) device. Further, when the data reading means 200 senses the personal identification information medium, it transmits a photographing start signal.

  The authentication management apparatus 300 stores a plurality of pieces of personal identification information, receives personal identification information from the data reading means 200 of the personal identification data reading apparatus 100, compares and collates the received personal identification information with the stored personal identification information, A personal authentication result signal indicating the success or failure of the authentication is output. This personal authentication result signal is used as a signal for controlling the unlocking and locking of the entrance door 700. Further, personal identification information processing information such as traffic event information (hereinafter abbreviated as traffic event information) is generated based on the personal identification information from the data reading unit 200, and the data is associated with the data association unit 500 of the personal identification data reading device 100. Send. Therefore, the authentication management apparatus 300 also functions as personal identification information processing information generation means (hereinafter abbreviated as traffic event information generation means) that generates traffic event information.

  The passing event information is information representing personal authentication processing when the person to be authenticated passes on the floor based on the personal identification information from the data reading means 200, and an example is shown in FIG. That is, the passing event information includes an event number represented by a number or symbol specified for each personal authentication process, personal identification information (ID) from the data reading means 200, authentication time, type of entry and exit, authentication, etc. Contains authentication result information indicating the success or failure of. For example, the event number is sequentially assigned as 0001 for the first personal authentication process in the day and 0002 for the next personal authentication process, and is unique for each personal authentication process. ) Are not assigned the same event number (personal authentication processing number). The traffic event information shown in FIG. 2 includes event number “0001” (indicating the first event today), personal identification information (ID) “001111”, authentication time “12:00:00”, type “room entry” ", The authentication result information" success ".

  Returning to FIG. 1, the image capturing apparatus 400 is an apparatus that captures an image of a camera or the like arranged to capture an entrance / exit of a specific area near the entrance door 700. Upon receiving a shooting start signal from the data reading unit 200 of the personal identification data reading device 100, the image shooting device 400 captures still images of the person to be authenticated at predetermined intervals to the personal identification data reading device 100 until the shooting is completed. Send. The image capturing apparatus 400 has a timer inside, and ends the capturing when a predetermined time elapses after receiving the capturing start signal. Note that shooting may be terminated based on a signal from an external device without using the internal timer.

  In this example, a captured image composed of a plurality of still images is transmitted during a period from the start of capturing to the end of capturing, but a captured image composed of a moving image may be transmitted during the same period. A moving image is an aggregate of a plurality of still images. In the present invention, a moving image is included when expressed as a plurality of still images. The period from the start of shooting to the end of shooting corresponds to one traffic event.

  The data association unit 500 of the personal identification data reading device 100 stores (saves) a plurality of still images from the image capturing device 400 and the passing event information from the authentication management device 300, and stores the plurality of stored images. A still image that meets a predetermined condition is selected from the still images, that is, in this example, a still image (optimum image) showing an appropriate face or a face that is easy to recognize is selected, and the selected still image and the stored traffic event It combines with information and transmits as detection image data. Although the details will be described later, the predetermined condition is determined by the value of the evaluation value calculated by digitizing the face direction, size, brightness, eye position, etc. in the face image of the still image. The detected image data may include an evaluation value calculated for each still image used when detecting a still image showing an appropriate face. Further, when the captured image from the image capturing apparatus 400 is a moving image, a function for extracting a still image from the moving image at predetermined intervals may be added to the data association unit 500.

  The entry / exit management device 600 is connected to the personal identification data reading device 100 and the authentication management device 300 via a network, and a PC terminal that receives, records and displays management information for the administrator to grasp and manage the entry / exit of the person to be personally authenticated. It is a device. This management information is the passing event information transmitted from the authentication management device 300 and the detected image data transmitted from the personal identification data reading device 100. Upon receipt of such management information, the entrance / exit management device 600 displays, for example, traffic event information and a captured image associated with this information as shown in FIGS.

  Next, the access control system according to Embodiment 1 of the present invention will be described in more detail with reference to the block diagram of FIG.

  First, the authentication management apparatus 300 will be described. The authentication management apparatus 300 includes data collating means 301, registered information storage means 302, and personal identification information processing information generating means 303 (hereinafter referred to as traffic event information generating means).

  The registration information storage unit 302 stores a plurality of pieces of personal identification information in advance. The data collating unit 301 receives the personal identification information from the data reading unit 200 of the personal identification data reading apparatus 100, and compares and collates the received personal identification information with the personal identification information stored in the registered information storage unit 302. If the received personal identification information is stored in the registered information storage means 302, a personal authentication result signal indicating that the authentication has succeeded, and if no matching personal identification information has been stored, a personal authentication result signal indicating failure is displayed. Output. This personal authentication result signal is used as a lock control signal for controlling the unlocking and locking of the doorway 700, and the doorway door 700 is unlocked when the personal authentication result signal is a signal indicating a successful authentication.

  The passage event information generation unit 303 receives the personal identification information from the data reading unit 200, adds an event number to the personal identification information and adds a type such as authentication time, entry, and exit, for example, as shown in FIG. Passage event information is generated and transmitted to the data association means 500 of the personal identification data reader 100.

  Next, the data association unit 500 of the personal identification data reading apparatus 100 will be described. The data association unit 500 includes a first personal identification information processing information receiving unit 501 (hereinafter abbreviated as a first traffic event information receiving unit) and a first personal identification information processing information storage unit 502 (hereinafter referred to as a first personal identification information processing unit). A captured event reception unit 503, a captured image storage unit 504, a captured image evaluation value calculation unit 505, an image selection (detection) unit 506, and a detected image data generation unit 507.

  The first traffic event information receiving unit 501 receives the traffic event information from the traffic event information generating unit 303 via the network 500 and outputs it to the first traffic event information storage unit 502. The first traffic event information means 502 stores traffic event information. The captured image acquisition unit 503 receives a still image from the image capturing apparatus 400 and outputs it to the captured image storage unit 504. The captured image storage unit 504 stores a captured still image.

  The photographed image evaluation value calculation unit 505 extracts a still image stored in the photographed image storage unit 504, detects a human face from the still image, and how much the face image is an objectively balanced face image. Evaluate if there is. That is, the ease of recognizing and appropriateness of the face in the still image is calculated as the evaluation value, and the still image captured as the evaluation result and the passing event information stored in the first passing event information storage unit 502 are calculated. Is output to the captured image storage means 504 as evaluation information associated with the event number.

  The photographed image evaluation value calculation means 505 includes a face detection processing unit that detects a human face area in the captured still image as face image data, and an evaluation that indicates the ease of recognition and appropriateness of the face in the detected face image. An evaluation value calculation unit that calculates a value, an evaluation value from the evaluation value calculation unit, a captured still image, and an event number of the traffic event information stored in the first traffic event information storage unit 502, Evaluation information output means for outputting to the captured image storage means 504.

  The face detection processing unit compares and collates a storage unit that stores a large number of face images and target images other than face images, and a face image and target image stored in the storage unit with a captured still image. A discriminator for discriminating a face area from a non-face area and outputting face image data;

  The evaluation value calculation unit obtains values of each item such as the face direction, size, brightness, and eye position (eye detection) from the face image data output from the face detection processing unit, Each of the values is multiplied by a weighting factor and combined (added), and an evaluation value as a score value is output. For example, it has a direction calculation unit, a size calculation unit, a brightness calculation unit, and an eye position calculation unit, a weight coefficient multiplication unit for each calculation unit, and adds the multiplication results from each weight coefficient multiplication unit And an adder that outputs an evaluation value.

  The direction calculation unit calculates an evaluation value for the direction from the positional relationship between the position of the nose and the eyes detected from the face image data by using a circular filter, and the size calculation unit increases the size by the height pixel of the face image data. The brightness calculation unit calculates an evaluation value for brightness based on an average luminance value obtained by averaging the RGB values of each pixel of the face image data, and the eye position calculation unit calculates the evaluation value for the face image data. The evaluation value for the eye position is calculated by performing template matching.

  The calculation of the score value will be further described with reference to FIG. FIG. 4 is a table showing an example of the value of each item for calculating the score value. That is, an evaluation value corresponding to the state is given to the item to be evaluated, and each item is weighted. The items are “face direction”, “face size”, “face brightness”, “eye position (eye detection)”, and weighting coefficients “6”, “6”, “4” in order. "4".

  The evaluation score according to the state of the item “face direction” is a three-level evaluation. If the state is determined to be “front” based on the positional relationship between the position of the nose and the eye, it is determined that the rotation is within 30 degrees. Is set to 3 points, and “other” is set to 1 point. The evaluation score according to the state of the item “face size” is a five-level evaluation, and the height pixel exceeds 200, 200 or less, 150, 150 or less, 100, 100 or less, 50, 50 or less in the order of 4 points, 5 points, Three points, one point, and zero point are set. The evaluation score according to the state of the item “facial brightness” is a five-level evaluation, and the average luminance value obtained by averaging the RGB values of each pixel is more than 240, more than 240, more than 200, less than 200, less than 150, more than 150, less than 100, less than 100. 2 points, 5 points, 3 points, 1 point, and 0 points are set in this order. The evaluation score according to the state of the item “eye detection” is a three-level evaluation. If it is determined as “both eyes” by template matching, it is 5 points, 3 points when it is determined as “one eye”, and 0 when it is determined as “none”. Set with a point.

  For example, when the direction calculation unit determines that the front side, 5 is output, and 30 (= 5 × 6) is output from the weight coefficient multiplication unit. When the size calculation unit detects 180 as the height pixel, 5 is output, and 30 (= 5 × 6) is output from the weight coefficient multiplication unit. When the brightness calculation unit detects 140 as an average luminance value obtained by averaging the RGB values of each pixel, 1 is output, and 4 (= 1 × 4) is output from the weight coefficient multiplication unit. When the eye position calculation unit determines that both eyes, 5 is output, and 20 (= 5 × 4) is output from the weight coefficient multiplication unit. Then, the adder receiving the output from these weight coefficient multipliers outputs 84 (= 30 + 30 + 4 + 20). 84 from the addition unit is an evaluation value (score value) from the evaluation value calculation unit.

Upon receiving the evaluation value from the evaluation value calculation unit, the evaluation information output unit associates the evaluation value with the captured still image, and outputs it to the captured image storage unit 504 as evaluation information. These operations are performed on a plurality of still images received by the captured image receiving unit 503 during the period from the start of shooting to the end of shooting, that is, the same period as the event number of the traffic event information. An evaluation value is calculated for each of them and is output as evaluation information to the captured image storage unit 504.

  The evaluation information output means associates the evaluation value with the still image by changing the name to a file name in which the evaluation value is added to the file name of the image, or inserting the evaluation value into the user area or meta information area of the image file. It is done by the method of

  The image selection unit 506 selects a still image from a plurality of still images stored in the captured image storage unit 504 based on the evaluation value calculated by the captured image evaluation value calculation unit 504. For example, a still image having evaluation information with the highest evaluation value is selected and output to the detected image data generation unit 507.

  The detected image data generating unit 507 generates detected image data by associating (combining) the passing event information received from the first passing event information storage unit 502 with the still image received from the image selecting unit 506, and generates the detected image data via the network 500. To the access control device 600.

  The association between the still image and the passing event information in the detected image data generation means 507 is the same method as the association between the evaluation value and the still image, for example, the association by changing the file name, that is, the still image This is done by adding the event number “0001” in the traffic event information to the image file name “image 0001.jpg” and changing it to “image 0001-event 0001.jpg”, and transmitting the still image as the file name of the detected image data To do.

  The entry / exit management device 600 includes a second personal identification information processing information receiving unit 601 (hereinafter abbreviated as a second traffic event information receiving unit) and a second personal identification information processing information storage unit 602 (hereinafter referred to as a second personal information processing information storage unit). A display event 603 such as a liquid crystal monitor, a detected image data receiving unit 604, and a detected image data storing unit 605.

  The second traffic event information receiving means 601 receives the traffic event information from the traffic event information generating means 303 via the network 500, and the detected image data receiving means 604 that outputs to the second traffic event information storage means 602 is detected. The detected image data from the image data generating unit 507 is received via the network 500 and output to the detected image data storage unit 605.

The display unit 603 displays the traffic event information stored in the second traffic event information storage unit 602, and the detected image data stored in the detected image data storage unit 605, that is, an appropriate face and an easily recognized face. A still image in which is displayed is displayed along with the traffic event information. For example, when displaying the optimum still image corresponding to the traffic event information including the event number 0001, the display unit 603 can search and display an image using the file name “event 0001” in the detected image data as a key. .

  By displaying the passing event information on the display unit 603, the administrator can grasp and manage the entry / exit of the person to be personally authenticated. In addition, the display unit 603 displays a still image showing an appropriate face corresponding to the traffic event information and a face that can be easily recognized, so that the administrator can specify the person of the person to be personally authenticated with high accuracy.

  Next, an example of display contents on the display unit 603 will be described with reference to FIGS. As shown in each figure, the left side of the display means 603 is a traffic event information display area in which the traffic event information received from the second traffic event information storage means 602 is displayed, and the right side is the stationary image received from the detected image data storage means 605. This is a still image display area in which an image is displayed.

  FIG. 5 shows an example in which the traffic event information is displayed when the first personal authentication process in one day is performed. In the second traffic event information storage unit 602 corresponding to the event number 0001, FIG. Based on the stored traffic event information, the personal identification ID 123456 that is the first person to be authenticated, the authentication time 9:20:30, the type “room entry”, and the authentication result “success” are displayed.

  FIG. 6 shows a display example when the administrator clicks the event number 0001 in the state shown in FIG. 5, and the still image data stored in the detected image data storage means 605 corresponding to the event number 0001. Is displayed. The displayed still image is a face image of a person subject to personal authentication having a personal identification ID 123456 having the highest evaluation value in the traffic event information period.

  FIG. 7 shows an example of the display of the pass event information after the five individual authentication subjects have been authenticated and the case where the administrator clicks the event number 0005. Information is displayed in the traffic event information area based on the traffic event information stored in the second traffic event information storage unit 602 corresponding to the event number each time the person subject to personal authentication performs personal authentication processing. Still image data stored in the detected image data storage unit 605 corresponding to the event number 0005 is displayed in the image display area. The displayed still image is a face image having the highest evaluation value in the traffic event information period for the fifth person subject to personal authentication. Even a face image of a person who has been subject to personal authentication can be displayed in the image display area.

  In this way, the administrator cannot confirm the person name of the person who has been subjected to the personal authentication process corresponding to the event number 0005, but can easily recognize the face image displayed in the image display area. It becomes possible to perform person confirmation by looking at.

  Next, the operation of the first embodiment of the present invention configured as described above will be described with reference to FIGS. FIGS. 8A and 8B show data exchange between devices when the horizontal axis is a time axis, and FIG. 9 is a flowchart.

  First, a case where the person A subject to personal authentication performs a reading operation and another person does not perform the reading operation during the photographing will be described with reference to FIGS. In this case, the branch of step S11 and step S16 in FIG. 9 is “No”. When the person A subject to personal authentication performs a reading operation on the data reading unit 200 of the personal identification data reading device 100 at 12:00:00 using a personal identification information medium such as an IC card, the data reading unit 200 receives the personal identification information. The personal identification information is read from the medium and transmitted to the data verification unit 301 of the authentication management apparatus 300 (steps 001 and 002 in FIG. 8A and steps S1 and S2 in FIG. 9).

  In response to this, the data verification unit 301 of the authentication management apparatus 300 performs a personal authentication process for the personal authentication target person A. That is, the personal identification information from the data reading unit 200 and the personal identification information stored in the registration information storage unit 302 are compared and collated, and the personal identification information whose received personal identification information matches the registration information storage unit 302 is obtained. If it is stored, a success of authentication is output to the entrance door 700 and a personal authentication result signal indicating failure if no matching personal identification information is stored is output to the entrance door 700 (steps 003 in FIG. 8A, steps in FIG. 9). S3, S4).

  When a personal authentication result signal indicating “authentication success” is output from the data collating means 301, the lock of the doorway 700 is unlocked (step S5 in FIG. 9). When the personal authentication result signal indicating “authentication failure” is output, the door door 700 is not unlocked.

  Further, the passing event information generating means 303 that has received the personal identification information from the data reading means 200 adds an event number to the personal identification information and adds a type such as authentication time, entry, exit, etc. 2) and is transmitted to the data association means 500 of the personal identification data reader 100, and is also transmitted to the access control apparatus 600 via the network 500 (step 004 in FIG. 8A, step in FIG. 9) S6).

  In the entrance / exit management device 600, the second traffic event information receiving means 601 receives the traffic event information via the network 500, stores it in the second traffic event information storage means 602, and displays it as shown in FIG. The contents of the passing event information are displayed on the means 602 (005 in FIG. 8A, step S7 in FIG. 9).

  On the other hand, when the data reading unit 200 of the personal identification data reading device 100 senses the personal identification information medium, it transmits a shooting start signal to the image shooting device 400 (step 006 in FIG. 8A). The image capturing signal 400 that has received the image capturing start signal starts image capturing at the entrance / exit of a specific area in the vicinity of the entrance / exit door 700 at predetermined intervals, and the still images A1 to A1 captured by the captured image receiving means 503 of the personal identification data reader 100 A4 is transmitted (step 007 in FIG. 8A, step S8 in FIG. 9).

  Then, after a predetermined time (10 seconds in the example of FIG. 8A) after the transmission of the shooting start signal from the data reading means 200, the image shooting apparatus 400 causes the timer held therein to pass the predetermined time. Sense and end shooting.

  The captured image receiving unit 503 sends the received still images A1 to A4 to the captured image storage unit 504, and the captured image evaluation value calculation unit 505 performs face detection processing on the image extracted from the captured image storage unit 504, and evaluates it. The value is calculated (008 in FIG. 8A, steps S9 and S10 in FIG. 9).

  When the shooting time has elapsed (step S12 in FIG. 9), the image selection unit 506 selects the still image with the highest evaluation value from the still images A1 to A4 (009 in FIG. 8A, step S13 in FIG. 9). ).

  The detected image data generating unit 507 generates detected image data by combining the still image selected by the image selecting unit 506 and the passing event information generated by the passing event information generating unit 303 of the authentication management apparatus 300. The data is transmitted to the management apparatus 600 (010 in FIG. 8A, step S14 in FIG. 9). The detected image data is stored in the detected image data storage unit 605 (step S15 in FIG. 9). Then, the display unit 603 displays the detected image data stored in the detected image data storage unit 605, that is, a still image showing an appropriate face and a face that can be easily recognized, together with the passing event information.

  Next, with reference to FIGS. 8B and 9, a case where the person A subject to personal authentication performs a reading operation and another person performs the reading operation during the shooting will be described. In this case, the branch of step S11 and step S16 in FIG. 9 is “Yes”. The processing from 001 to 008 in FIG. 8B is the same as that in FIG.

  If another individual authentication target person B performs a reading operation while the person authentication target person A is photographed, the personal identification data reading apparatus 100 stops the process of detecting a person's face even during the process. The optimum image is selected based on the evaluation value from the still images that have been shot before the reading operation is performed and the evaluation value is calculated. Then, detected image data is generated from the selected still image and traffic event information, and transmitted to the access management apparatus 600. For B, a new image is taken for a predetermined time from the reading operation (10 seconds in the example of FIG. 8B).

  Specifically, when another person subject to personal authentication performs a reading operation, the data reading unit 200 of the personal identification data reading apparatus 100 causes the captured image evaluation value calculating unit 505 to stop the face detection process. And an internal timer is reset to the image capturing apparatus 400, and a reset signal is newly transmitted to perform image capturing for a predetermined time. The image capturing apparatus 400 that has received the reset signal captures the person B to be personally authenticated for a predetermined time from when the reset signal is received (009 in FIG. 8B). By doing so, it is possible to perform photographing for a predetermined time even for the person B to be personally authenticated who has performed the reading operation in the middle.

  When the personal authentication target person B performs a reading operation, the image selection unit 506 selects the still image with the highest evaluation value from the still images that have undergone the face detection process until the personal authentication target person B performs the reading operation. (010 in FIG. 8B, step S13 in FIG. 9). In FIG. 8B, one of the images A1 and A2 subjected to face detection processing is selected.

  For the person A subject to personal authentication, the selected image and the traffic event information are combined and transmitted to the entrance / exit management device 600 as detected image data (011 in FIG. 8B, steps S14 and S15 in FIG. 9). . On the other hand, for the person B to be personally authenticated, since another reading operation is performed by B (step S16-Yes in FIG. 9), as described above, the personal identification information is read and authentication processing is performed, and the passing event information and Detected image data is generated from the captured still image and transmitted to the access management apparatus 600 (012 to 018 in FIG. 8B, steps S2 to S15 in FIG. 9). In the example shown in FIG. 8B, since there is no reading operation by another person for personal authentication for a predetermined time (10 seconds), any one of the images B1 to B4 taken for 10 seconds is used. An optimal image is selected.

  As described above, according to the first embodiment of the present invention, a still image showing an appropriate face is selected from a plurality of stored still images, and the selected still image and traffic event information are displayed. As a result, it is possible to specify the person of the person subject to personal authentication with high accuracy.

  In addition, the image selected based on the result of the face detection process is stored in association with the traffic event information representing the personal authentication process (event) at the time of passing the floor of the person to be authenticated, not the time when the personal authentication is performed. Thus, the administrator can display and search for an image associated with each event, and the operation burden on the administrator can be reduced.

  For example, it is assumed that personal authentication at a certain time is an authentication failure. In the conventional entrance / exit management system, since the image taken at the personal authentication time is displayed, if the face in the image taken at that time is unclear, the administrator Images must be searched from the database, and if the displayed image is in the form of a moving image, the movie must always be watched carefully to find the moment when the face can be recognized.

  On the other hand, in the first embodiment of the present invention, the search can be performed based on the passing event information without depending on the personal authentication time. That is, since the optimal image for each traffic event is stored in association with each traffic event, as described above, the most optimal image for the traffic event can be obtained by simply clicking the event number in the displayed traffic event information. This makes it possible to display various images, and the administrator does not need to search for appropriate images one by one from the database, thereby reducing the operation burden.

  In addition, if there is a request for reading by another person for personal authentication during the shooting / face detection process of the person for personal authentication, the previous face detection process for the person for personal authentication is stopped and the image is selected. Since the processing such as personal authentication / photographing for another person subject to personal authentication is immediately performed, it is possible to prevent another person subject to personal authentication not being photographed. For this reason, even when a plurality of passers-bys enter and exit at short time intervals, it is possible to shoot all passers-by without omissions.

  Note that the camera type of the image capturing apparatus 400 is not particularly limited, such as an analog camera, a digital camera, or a network camera. Further, the image capturing apparatus 400 is not necessarily separate from the personal identification data reading apparatus 100, and may be integrated. The image capturing apparatus 400 may be connected to the network 700 and transmit the captured image to the personal identification data reading apparatus 100 via the network.

  In FIG. 8A, the image capturing apparatus 400 captures four still images in 10 seconds. However, the present invention is not limited to this, and both the capturing time and the number of captured images can be set arbitrarily.

  Note that the image capturing apparatus 400 may start capturing an image before a personal authentication operation is performed by a person subject to personal authentication. By doing so, the frequency | count of imaging | photography of a person for personal authentication increases, the probability that an image with a higher evaluation value will be increased, and an administrator can specify a person for personal authentication more easily. Alternatively, the image capturing apparatus 400 may have a motion detect function for detecting a motion in the video, and the capturing may be started when the motion is detected. By doing so, shooting is performed only when the person subject to personal authentication is within the shooting range, so that the person who is subject to personal authentication is photographed before the personal authentication operation while suppressing the accumulation of useless images in which no one is shown. can do.

  1, only one authentication management device 300, image photographing device 400, and entry / exit management device 600 are shown for the sake of simplicity, but each device is appropriately selected. A plurality may be arranged to manage entry / exit.

  FIG. 10 shows a configuration in which two image capturing apparatuses 400 are arranged. By arranging two image capturing apparatuses 400a and 400b as shown in FIG. 10, it is possible to capture the person to be personally authenticated for as long as possible within the capturing setting time. For example, even when the person subject to personal authentication passes the door 700 that has been unlocked by performing personal authentication processing, the person subject to personal authentication after the entrance door 700 can be photographed by the image photographing device 400b.

  Note that the recording medium to be read by the personal identification data reader 100 of FIG. 1 is not limited to a non-contact type IC card, but may be a contact type magnetic stripe card. Alternatively, it may be requested to input a password at the time of reading.

  It should be noted that the authentication management device 300 does not unlock the entrance door 700 immediately after the personal authentication is successful, and does not allow the person to be personally authenticated to pass through until the photographing time of the image photographing device 400 has elapsed. Also good. By doing so, an image of the person to be personally authenticated can always be taken during the set shooting time, and the probability of obtaining an image with a high evaluation value and easy to recognize the face is increased. Can more easily identify the person to be personally authenticated.

  2 may include door numbers, individual camera numbers, and the like in the traffic event information.

  Note that the captured image evaluation value calculation unit 505 may perform the evaluation based on the size of the face detected from the still image without calculating the score value as described above, or the brightness of the face based on the luminance value. Evaluation may be performed based on this. Alternatively, the number of detected faces when face detection is performed may be calculated as an evaluation value, and the image selection unit 506 may select a captured image based on the number of faces. In that case, if the number of faces is plural, the evaluation value becomes large, and the number of faces can be easily obtained by counting the number of areas identified as faces in the image.

  When the evaluation is performed based on the number of detected faces, the captured image evaluation value calculation unit 505 notifies the image selection unit 506 of the number of detected faces as an evaluation result. By doing so, the administrator can easily search and find traffic by multiple persons. For example, in cases where unauthorized traffic is considered but there is no problem with personal authentication processing, it is possible that a person who does not have a personal identification ID can enter the room, but the number of faces detected as described above is evaluated. As a result, an image with a plurality of faces detected is displayed preferentially, and the administrator can easily search for an image with a high evaluation value without having to check each image one by one and easily pass a plurality of people. It becomes possible to discover.

  Further, the above score value may be calculated by including the number of detected faces in one of the items. By doing so, the evaluation value can be calculated by adjusting the number of detected faces and the ease of grasping the faces.

  Further, the images may be narrowed down to images having a plurality of detected faces from the captured images, and the image selecting unit 506 may select an image having a high score value. In this case, the photographed image evaluation value calculation unit 505 performs face detection processing on the received image and stores it in the photographed image storage unit 504 if there are a plurality of detected faces. If it is not plural, it is not stored, or is stored in another storage area of the photographed image storage means 504.

  When the photographing time has elapsed, the photographed image evaluation value calculation unit 505 receives the image (an image in which a plurality of faces are detected) stored in the photographed image storage unit 504, performs face detection processing again, and calculates a score value. . Then, the image selection unit 506 is notified to select the image having the highest score value among the images for which the score values have been calculated.

  By doing so, the image to be displayed has a plurality of faces and the faces are most objectively balanced, so the administrator should check each image one by one It is possible to easily find the traffic of a plurality of people and easily grasp the plurality of faces.

  Note that the photographed image evaluation value calculation unit 505 may store, in the photographed image storage unit 504, a moving image before and after a time when an image with the highest evaluation value is photographed when the image photographing apparatus 400 is in a moving image format. Good. By doing so, the administrator can perform person confirmation by combining a still image and a moving image having the highest evaluation value, and can easily identify a suspicious person.

  Note that the captured image evaluation value calculation unit 505 may store only the image with the highest evaluation value without storing all the evaluated images in the captured image storage unit 504. Further, an image having a relatively high evaluation value may be stored, for example, a certain threshold value may be set, and only an image having an evaluation value exceeding the threshold value may be stored. By doing so, it is not necessary to store an image with a small evaluation value and a face that is not sufficiently detected, and the data capacity stored in the captured image storage unit 504 can be reduced.

  Note that the image selection unit 506 does not necessarily need to select one image, and a plurality of images may be selected. A plurality of images may be selected in descending order of evaluation value, or a fixed threshold value is provided, and an image having an evaluation value exceeding the threshold value among the still images for which the evaluation value is calculated by the captured image evaluation value calculation unit 505 You may choose only. By doing so, the administrator can specify the face of the person subject to personal authentication from the plurality of images, and can more easily determine the suspicious person.

  In the case where the pass event information is displayed on the display means 603, when the authentication fails, the content “failure” in the authentication item of the pass event information may be displayed conspicuously or an error sound may be sounded. . By doing so, it becomes easier for the administrator to find the abnormality.

Embodiment 2. FIG.
FIG. 11 shows a configuration diagram of an access management system according to Embodiment 2 of the present invention. The present embodiment is different from the first embodiment in that the photographed image storage unit 504 includes a first photographed image storage area 504-1 and a second photographed image storage area 504-2. Parts corresponding to those of the first embodiment are denoted by the same reference numerals as those in FIG.

  In the first captured image storage area 504-1, one of the captured images received by the captured image acquisition unit 503 from the image capturing apparatus 400 is stored. For example, the first photographed image photographed at approximately the same time as the time when the personal authentication is performed is stored. In the following description, it is assumed that the first captured image storage area 504-1 stores the first captured image. However, the present invention is not limited to this, and any of the images captured during the capturing time is captured. It may be an image.

  The second captured image storage area 504-2 stores captured images other than the images stored in the first captured image storage area 504-2 among the images captured during the capturing time by the image capturing apparatus 400. Is done.

  FIG. 12 is a flowchart showing the operation of the access control system according to Embodiment 2 of the present invention. The processing from steps S1 to S9 is the same as that in FIG.

  If the received photographed image is the first one (step S10-Yes), the image is stored in the first image storage area (step S11). If the received photographed image is not the first one (step S10-No), that is, if it is the second and subsequent photographed images, they are stored in the second image storage area (step S12).

  The captured image evaluation value calculation unit 505 receives the image from the first captured image storage area 504-1 for the first captured image and performs face detection processing. For the second and subsequent captured images, the image is received from the second captured image storage area 504-2 and face detection processing is performed (step S13).

  Then, there is no other reading operation during the shooting time (step S14-No), the shooting time elapses (step S15-Yes), and face detection succeeds in any one of the shot images (step S16-Yes). ), The image selection means 506 selects the image with the highest evaluation (step S17).

  When face detection has not succeeded in any of the photographed images (step S16—No), the photographed image evaluation value calculation unit 505 notifies the image selection unit 506 of a face detection result indicating that face detection has failed. The image selection unit 506 selects the first image from the first captured image storage area 504-1 (step S18).

  The detected image data generating unit 507 generates detected image data by associating the first image received from the image selecting unit 506 with the traffic event information (step S19). At this time, the result that the face cannot be detected is associated. As the association method, the same method as in the first embodiment can be used. The associated detected image data is stored in the detected image data storage unit 605 (step S20) and displayed on the display unit 603 as necessary.

  When the detected image data associated with the result that the face could not be detected is stored in the detected image data storage unit 605, that is, when the result that the face cannot be detected is notified to the access management apparatus 600, the access management is performed. The apparatus 600 may display that the face cannot be detected by the display unit 603 and an event number corresponding to the passing event. By doing so, the administrator can easily search and recognize that the face was not detected without checking each image one by one. In particular, illegal traffic by a suspicious person is often effective when searching for such illegal traffic because the user often intentionally does not face his face or covers his face.

  As described above, according to the second embodiment of the present invention, in addition to having the effects of the first embodiment, the captured image can be displayed even when the captured image evaluation value calculating unit 505 cannot detect the face. By saving and displaying in association with the traffic event information, the administrator can check the status at the time of entering and exiting with an image.

  Also, as described above, by displaying the result of the face detection failure and the corresponding event number on the display device 603, the administrator can easily search for traffic where face detection could not be performed. It becomes possible.

DESCRIPTION OF SYMBOLS 100 Personal identification data reader 200 Data reader 300 Authentication management apparatus 301 Data collation means 302 Registration information storage means 303 Passing event information generation means 400 Image capturing apparatus 500 Data association means 501 Passing event information receiving means 502 Passing event information storage means 503 Photographed image acquisition means 504 Photographed image storage means 504-1 First photographed image storage area 504-2 Second photographed image storage area 505 Photographed image evaluation value calculation means 506 Image selection means 507 Detected image data generation means 600 Access control Device 601 Passing event information receiving means 602 Passing event information storing means 603 Display means 604 Detected image data receiving means 605 Detected image data storing means

Claims (9)

An image capturing device arranged to image the entrance / exit of a specific area near the entrance / exit door;
A plurality of personal identification information is stored, the received personal identification information is compared with the stored personal identification information, collated, and a personal authentication result signal indicating the success or failure of the authentication is displayed. An authentication management device that outputs and outputs personal identification information processing information representing personal authentication processing at the time of entrance / exit passage of the person to be personally authenticated based on the received personal identification information, and outputs the signal as a control signal;
Data reading that is installed at an entrance near the entrance door and senses a personal identification information medium storing personal identification information, reads the personal identification information from the personal identification information medium, and outputs the personal identification information to the authentication management device And a detected image obtained by selecting a still image that meets a predetermined condition from among a plurality of still images from the image photographing device, and combining the selected still image and personal identification information processing information from the authentication management device A personal identification data reader having data association means for transmitting as data;
The personal identification information processing information is received from the authentication management device, the contents corresponding to the personal identification information processing information are displayed, and the detected image data is received from the data association means of the personal identification data reading device. An entry / exit management system comprising: an entry / exit management device that displays a still image based on personal identification information processing information of detected image data.   2. The predetermined condition is determined by a value of an evaluation value calculated by digitizing a face direction, size, brightness, eye position, etc. in a face image of a still image. The entry / exit management system described.   3. The personal identification information processing information includes a personal authentication process number assigned for each personal authentication process, personal identification information, authentication time, and authentication result information. The entry / exit management system described. When there is a reading operation by another person for personal authentication during shooting of the person for personal authentication,
The personal identification data reading device selects a still image that meets a predetermined condition from among still images that have been photographed and evaluated by a reading operation performed by the other individual authentication target person. The access management system according to any one of claims 1 to 3. The data association means includes
Photographed image storage means for storing the still image received from the image photographing device;
A photographed image evaluation value calculating means for detecting a human face from the still image and calculating the evaluation value;
Image selecting means for selecting a still image from the plurality of still images based on the calculated evaluation value;
First personal identification information processing information storage means for storing personal identification information processing information received from the authentication management device;
Detection image data generation means for generating detection image data by associating the still image selected by the image selection means with the personal identification information processing information stored in the first personal identification information processing information storage means. The access management system according to any one of claims 1 to 4, further comprising:   The image selection means has a threshold for the evaluation value, and the still image having an evaluation value exceeding the threshold among the still images for which the evaluation value is calculated by the captured image evaluation value calculation means. 6. The access control system according to claim 5, wherein an image is selected. The photographed image evaluation value calculating means detects a human face from the still image and calculates the number of faces as an evaluation value.
The entrance / exit management system according to claim 5 or 6, wherein the image selection unit selects the captured image based on the number of faces calculated by the captured image evaluation value calculation unit.   The said access management apparatus displays that face detection was not possible, when the said picked-up image evaluation value calculation means cannot detect a face from the said still image. Access control system. Data reading means for reading the personal identification information from the personal identification information medium when sensing the personal identification information from the personal identification information medium in which the personal identification information is stored;
Captured image storage means for storing multiple still images, and calculating the still image evaluation value by quantifying the face direction, size, brightness, eye position, etc. in each stored still image Evaluation value calculating means, image selecting means for extracting a predetermined still image from a plurality of still images based on the evaluation value calculated by the evaluation value calculating means, and each time the personal identification information is read by the data reading means Personal identification information processing information storage means for storing personal identification information processing information having a personal authentication processing number attached to the personal identification information, and the predetermined still image extracted by the image selection means and the individual A data association unit having a detection image data generation unit that generates and outputs detection image data combined with the personal identification information processing information stored in the identification information processing information storage unit; Personal identification data reader, characterized by comprising.

Images