JP2018019125A - Unauthorized use detection system and unauthorized use detection method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To detect borrowing of a portable medium between and owner and others.SOLUTION: An unauthorized use detection system for detecting unauthorized use of a portable medium, owned by each person satisfying prescribed conditions, by a person other than the owner is constituted of read means reading identification information for identifying each portable medium when it is used, acquisition means for acquiring the feature information of a portable medium user who made the reading means read the identification information, management means for managing the feature information and identification information in association, and detection means for detecting the borrowing of a portable medium, based on whether or not different feature information is included in multiple feature information associated with the same identification information, after the portable medium was used multiple times.SELECTED DRAWING: Figure 1

Description

  The present invention relates to an unauthorized use detection system and an unauthorized use detection method for detecting unauthorized use in which a portable medium owned by each person satisfying a predetermined condition is used by someone other than the owner.

  Conventionally, card-type portable media such as railway commuter passes and ski lift tickets have been used in various places. For example, a card such as a magnetic card or an IC card according to the purpose of use can be obtained if a predetermined condition is satisfied by paying a fee in advance in order to use a railway or a lift. With this card, when passing through a gate such as a ticket gate, it is possible to use a railroad or a lift simply by causing a predetermined reading device to read information recorded on the card.

  When each user is obliged to purchase a card, it is necessary to prevent unauthorized use by borrowing the card from the owner who purchased the card without purchasing the card. If the user can confirm that the user is the card owner when using the card, unauthorized use of the card can be prevented.

  Face authentication technology can be used to confirm that the card user is the card owner. For example, Patent Document 1 discloses a technique for confirming an identity by using a face image. By managing the card identification number and the face image of the owner who purchased the card in association with each other, it is possible to execute face authentication processing when using the card and determine whether or not the card is used by the owner.

  In order to perform the face authentication process, it is necessary to acquire a face image suitable for the process. For example, Patent Document 2 discloses a technique for guiding a user by guidance when acquiring a face image. If the user can be guided and moved to an appropriate position, the user can be imaged by the imaging unit, and a face image suitable for face authentication can be acquired.

JP 2004-157602 A JP 2012-133477 A

  However, according to the above-described conventional technology, for example, when purchasing a card, face images of all purchasers must be acquired in advance and registered and managed in a database. When the number of card users is very large as in the case of railways, it is expensive and time-consuming to register and manage face images in a database.

  If a user is guided by guidance when using a card, a face image suitable for face authentication can be acquired. However, it takes time to acquire a face image. Even after acquiring the user's face image, it will be determined whether it is the card purchaser by comparing it with a large number of face images registered in advance in the database, and it will take time until the determination result is obtained. It takes. When a large number of users pass through the ticket gates continuously, such as at stations in the commuting school time zone, face images are acquired and face authentication processing is performed within a short time when each user passes through the ticket gates. Although it is difficult to obtain an accurate determination result, even in such a case, there has been a demand for a method of detecting card lending and borrowing in order to prevent the card from being illegally used repeatedly.

  The present invention has been made to solve the above-described problems caused by the prior art, and a portable medium owned by each person satisfying a predetermined condition is detected by an unauthorized use used by someone other than the owner. It is an object of the present invention to provide an unauthorized use detection system and an unauthorized use detection method.

  In order to solve the above-described problems and achieve the object, the present invention is an unauthorized use detection system that detects unauthorized use in which a portable medium owned by each person satisfying a predetermined condition is used by someone other than the owner. Then, when using the portable medium, reading means for reading identification information for identifying each portable medium from the portable medium, and characteristic information of the portable medium user who causes the reading means to read the identification information. An acquisition means for acquiring, a management means for managing the feature information in association with the identification information, and a plurality of feature information associated with the same identification information after the portable medium is used a plurality of times, And detecting means for detecting unauthorized use of the portable medium based on whether or not different feature information is included.

  Further, the present invention is the above invention, wherein the management means only allows the feature information satisfying a preset predetermined condition from among the feature information obtained from the acquisition means each time a portable medium is used. It is characterized by being managed in association with identification information of the carrying medium.

  The present invention further includes display means for displaying feature information in the above invention, wherein the detection means is the same on the screen of the display means after the portable medium is used a plurality of times. A plurality of feature information associated with the identification information is displayed so as to be comparable.

  Further, the present invention is the above invention, wherein the detecting means calculates an evaluation value from a value indicating a degree of coincidence with other feature information for each of the plurality of feature information, and based on the evaluation value, It is characterized in that it is determined whether or not all characteristic information belongs to the same user.

  Further, in the present invention according to the above-described invention, the detecting unit compares a plurality of feature information, and calculates an evaluation value of each feature information by assigning different points depending on whether the feature information matches or not. Then, based on the evaluation value, it is characterized in that it is determined whether or not all of the plurality of pieces of feature information belong to the same user.

  Moreover, the present invention is characterized in that, in the above-mentioned invention, the detection means displays a list of a plurality of feature information together with an evaluation value of each feature information on the screen of the display means.

  Also, in the present invention according to the above-described invention, the detection unit displays only the feature information whose evaluation value is within a predetermined range among the plurality of feature information associated with the same identification information on the display unit. It is characterized by that.

  Further, the present invention is the above invention, wherein the reading unit and the acquiring unit are operated during a predetermined business hours, and the detection unit is a portable medium at a predetermined timing after the business hours are over. The process of detecting lending / borrowing is started.

  Further, the present invention is the above invention, wherein registration of at least one of identification information of a portable medium and at least one feature information associated with the identification information is accepted, and the portable information is stored during the business hours. It further comprises monitoring means for monitoring the use of a portable medium, and notifying means for notifying when the monitoring means detects the use of a portable medium registered as a monitoring target.

  Moreover, the present invention is characterized in that, in the above-mentioned invention, the acquisition means acquires a face image of a portable medium user as the feature information.

  Further, the present invention is characterized in that, in the above-mentioned invention, the acquisition means acquires an appearance image obtained by capturing an appearance of the portable medium user as the characteristic information.

  In addition, the present invention is an unauthorized use detection method for detecting unauthorized use in which a portable medium owned by each person satisfying a predetermined condition is used by another person other than the owner by an unauthorized use detection system. A first step of reading identification information for identifying each portable medium from the portable medium, and a second step of acquiring characteristic information of the portable medium user who has read the identification information in the first step. A step, a third step of managing the feature information in association with the identification information, and a plurality of feature information associated with the same identification information after the first to third steps are executed a plurality of times. And a fourth step of detecting lending / borrowing of the portable medium based on whether different feature information is included.

  According to the present invention, when using a portable medium such as an IC card, the identification information such as an identification number for identifying each portable medium is read, and the characteristic information of the portable medium user is acquired, and the identification information and It can be managed in association with feature information. If different pieces of feature information are included in multiple pieces of feature information acquired for the same portable medium, it is assumed that this portable medium has been illegally used by a person other than the owner. Can be detected. Without selling feature information such as facial images of each purchaser when selling portable media, it is possible to detect lending / borrowing after the start of use of the portable media. The medium can be the object of loan detection.

FIG. 1 is a schematic diagram illustrating an outline of an installation configuration of an unauthorized use detection system. FIG. 2 is a block diagram showing a schematic functional configuration of the unauthorized use detection system. FIG. 3 is a schematic diagram illustrating an example of information managed by the information management unit. FIG. 4 is a diagram for explaining the detection process. FIG. 5 is a diagram for explaining an example of detection processing performed after the number of user face images increases. FIG. 6 is a diagram illustrating an example of a notification screen displayed on the operation display unit. FIG. 7 is a diagram illustrating an example of a list display screen displayed on the operation display unit. FIG. 8 is a diagram for explaining a method for calculating the evaluation value of each face image using the two most recently acquired face images. FIG. 9 is a diagram for explaining a method for calculating the evaluation value of each face image using the most recently acquired one face image. FIG. 10 is a schematic diagram illustrating an example of a screen that is displayed when the clerk determines lending / borrowing.

  Hereinafter, an unauthorized use detection system and an unauthorized use detection method according to the present invention will be described in detail with reference to the accompanying drawings. In the unauthorized use detection system, when a portable medium such as a magnetic card or an IC card is used, identification information such as an identification number can be read from each portable medium, and a facial image, etc. Feature information for distinguishing each portable medium user is acquired. After acquiring a plurality of feature information for the same portable medium, the unauthorized use detection system evaluates whether each feature information matches. When different feature information is included in the plurality of feature information, for example, one portable medium is lent and borrowed between the owner of the portable medium and another person, and other than the owner This is detected as being illegally used by a person, and this is notified.

[Example 1]
Issued to receive the provision of goods and services at the store, such as train commuter passes, ski lift tickets, cards used for games at the amusement store, and other portable media that the fraudulent use detection system is subject to loan detection Card, a permit permitting entry into a predetermined area, and the like. If it is possible to distinguish the portable media owned by each person by identification information such as identification number etc. given to each person who satisfies the predetermined condition by confirming the payment in advance or being qualified The type of portable medium is not particularly limited. In the present embodiment, an IC card (hereinafter simply referred to as “card”) used as a commuter pass on a railway will be described as an example of a portable medium. The characteristic information acquired by the unauthorized use detection system is information for determining whether or not the portable medium is used by the same person. Depending on how the card is used, accessories such as face, fingerprint, iris, vein, sunglasses, and clothing (appearance) can be used. In this embodiment, a face image will be described as an example.

  First, the configuration of the unauthorized use detection system will be described with reference to FIG. 1 and FIG. FIG. 1 is a schematic diagram illustrating an outline of an installation configuration of an unauthorized use detection system. FIG. 2 is a block diagram showing a schematic functional configuration of the unauthorized use detection system.

  As illustrated in FIG. 1, the unauthorized use detection system includes a camera 10, a card reader 20, and a detection device 30. The camera 10, the card reader 20, and the detection device 30 are installed at the ticket gate of the station.

  The camera (acquisition means) 10 has a function of capturing the user 100 in order to acquire a moving image and a still image of the user 100 who uses the card 200. The card reader (reading means) 20 has a function of reading an identification number for distinguishing each card from the card 200 possessed by the user 100.

  The card reader 20 is installed on the floor of the ticket gate. The camera 10 is installed on the ceiling so that the user 100 can be imaged by the card reader 20 reading the information on the card 200. However, the installation mode of the camera 10 and the card reader 20 is not limited to this. For example, the camera 10 may be installed at a position close to the floor, and the user 100 may be imaged from below, or the card reader 20 may have the camera 10 built-in.

  The detection device 30 is communicably connected to the camera 10 and the card reader 20, and detects a fraudulent use of the card 200 based on an image acquired by the camera 10 and an identification number acquired by the card reader 20. Have The installation location of the detection device 30 is not particularly limited. For example, the detection device 30 is installed in a room provided beside the ticket gate, a predetermined room where a staff member of the station waits, or the like. Examples of unauthorized use include use of a stolen card 200, use of a picked up card 200, use of a card 200 borrowed from an owner, etc. In the following, the card 200 is used between an owner and another person. An example of lending and borrowing will be described.

  Although FIG. 1 shows the minimum configuration of the unauthorized use detection system, when there are a plurality of ticket gates in the station, such as a west ticket gate and an east ticket gate, a camera 10 and a card reader 20 are installed at each ticket gate. Is installed. At each ticket gate, a camera 10 and a card reader 20 for users who enter the platform of the station and a camera 10 and a card reader 20 for users who enter from the platform of the station are installed.

  All the cameras 10 and the card reader 20 are connected to the detection device 30, and the detection device 30 includes information for identifying each of the camera 10 and the card reader 20, an installation location of the camera 10 and the card reader 20, and a camera. 10 and information indicating whether the card reader 20 is for entry or entry are managed. As a result, the detection device 30 can recognize whether the information acquired from the camera 10 and the card reader 20 is acquired at any ticket gate at any station and whether the information is at the time of entry or at the time of entry. it can.

  As illustrated in FIG. 2, the detection device 30 includes a control unit 40, a communication unit 50, an operation display unit 60, and a storage unit 70. The communication unit 50 has a function of communicating with the camera 10 and the card reader 20 via a network or the like. The operation display unit (display means) 60 is composed of a touch panel type liquid crystal display device, and has a function of inputting various information related to instruction operation and setting operation, and outputting and displaying various information. The control unit 40 has a function of controlling each unit in order to realize each function and operation described below. The storage unit 70 includes a non-volatile storage device such as a semiconductor memory or a hard disk, and is used for storing data such as an image of the user 100 of the card 200, an identification number of the card 200, and setting information related to each process. Is done.

  The control unit 40 includes an information management unit 41, a detection unit 42, a monitoring unit 43, and a notification unit 44. The information management unit (management unit) 41 has a function of managing information such as an image of the user 100 acquired by the camera 10 and an identification number of the card 200 acquired by the card reader 20. The detection unit (detection means) 42 has a function of detecting that the card 200 has been lent or borrowed between the owner of the card 200 and another person using information managed by the information management unit 41. The monitoring unit (monitoring means) 43 has a function of accepting registration for monitoring the card 200 for which the detection unit 42 has detected lending and borrowing and monitoring the use of the registered card 200 at the ticket gate. When the monitoring unit 43 detects the use of the card 200 to be monitored, the notification unit (notification unit) 44 emits sound and light and displays information on the operation display unit 60 to notify the user of this. Have.

  When entering the station platform or leaving the platform, as shown in FIG. 1, when the user 100 holds the card 200 over the card reader 20 at the ticket gate, the card reader 20 reads the identification number of the card 200. The information management unit 41 acquires the identification number of the card 200 read by the card reader 20 via the communication unit 50. Further, the information management unit 41 acquires the moving images continuously recorded by the camera 10 via the communication unit 50 and recognizes the face of the user 100 when the card reader 20 reads the identification number of the card 200. A possible partial area is cut out from the moving image, and the cut out still image is used as a face image. And the information management part 41 preserve | saves and manages in the memory | storage part 70 in the state which linked | related the acquired identification number and face image.

  FIG. 3 is a schematic diagram illustrating an example of information managed by the information management unit 41. As illustrated in FIG. 3, the information management unit 41 includes the face image of the user 100 acquired by the camera 10, the identification number read from the card 200 of the user 100 by the card reader 20, the use date and time of the card 200, and the card 200. Are associated with each other and managed by the storage unit 70. However, the information that the information management unit 41 stores and manages in the storage unit 70 is not limited to the mode illustrated in FIG. For example, only the information at the time of entering or entering may be stored, or when there are a plurality of ticket gates, only the information at a predetermined ticket gate may be stored. Further, the information management unit 41 determines whether or not the face image is an image suitable for face collation processing, and stores information such as a face image and an identification number only when an appropriate face image can be acquired. An aspect may be sufficient. If conditions are set in advance by the detection device 30, the information management unit 41 refers to the setting conditions and stores only information that satisfies the conditions.

  In the unauthorized use detection system, as shown in FIG. 1, the identification number of the card 200, the face image of the user 100, the use date and time, for all the users 100 who use the card 200 and pass through the ticket gate of the station, Information such as the place of use is acquired, and each information is associated and managed as shown in FIG. However, when the user 100 passes the ticket gate, the detection device 30 only acquires information such as an identification number and a face image to confirm that the user 100 is the owner of the card 200. Lending and borrowing detection processing is not performed. For this reason, even if a large number of customers pass through the ticket gate of the station, the flow of users is not hindered. In addition, even when an unauthorized use detection system is newly installed at the ticket gate, the station user can use the ticket gate as usual.

  The detection device 30 starts detection processing for detecting the user 100 who lent and borrowed the card 200 at a predetermined timing, such as after the ticket gate is idle or after a day at the station. The start timing of the detection process can be changed by setting. Specifically, the detection process can be set to start manually by the operator, or the detection process is automatically performed at a predetermined timing such as after the business day ends on weekdays by specifying the date, day of the week, time, etc. It can also be set to start.

  Next, a detection process for detecting the user 100 who lent and borrowed the card 200 will be described. In the following description, it is assumed that the user 100 who uses the card 200 with the identification number “Card001” as a commuter commuter pass enters the A station platform in the morning and enters the A station platform at night for commuting. To do. Moreover, the detection apparatus 30 shall be set so that a detection process may be started after business hours of A station end.

  FIG. 4 is a diagram for explaining the detection process. In FIG. 4, the name of the face image is shown on the upper side of each face image, and the evaluation result compared with the other face images and the evaluation value calculated based on the evaluation result are shown on the lower side. For example, when the user 100 first comes to work from station A using the pre-purchased card 200 and returns home, as illustrated in FIG. A face image of Time Image002 is obtained.

  After the business of station A ends, the detection unit 42 starts a detection process for detecting whether the card 200 is lent or borrowed at a predetermined timing set in advance. In the detection process, a plurality of face images stored in association with the same identification number are mutually evaluated with brute force to calculate an evaluation value. Specifically, when two face images are compared, the score is 100 if they are from the same person, 50 if they are not from the same person, and the score is given to all other face images. The evaluation value of each face image is calculated by calculating and summing the obtained scores. In addition, about the method of determining whether it is the same person by comparing two face images, since the conventional face authentication technique can be utilized, detailed description is abbreviate | omitted.

  When the detection process is started, when the face images managed in association with the identification number “Card001” are only two face images of Image001 and Image002, the detection unit 42 stores the two face images in the storage unit. The data is read from 70 and detection processing is executed. When the detection unit 42 compares the face image of Image001 and the face image of Image002 and obtains the result that they are the same person, as shown in FIG. 4, the score for this evaluation is set to 100 points. ("001-002: 100 points" in the figure). The detection unit 42 similarly obtains scores for all other face images, and uses the total value of the obtained scores as the evaluation value of the face image. In the example of FIG. 4, since there is only one face image to be compared, the evaluation value of the face image of Image001 is 100 points. Similarly, the evaluation value of the face image of Image002 is 100 points based on the comparison result with the face image of Image001.

  When the evaluation values of all the face images are obtained, the detection unit 42 confirms whether or not the evaluation values of the face images are the same value. If there is a difference between the evaluation values, the detection unit 42 determines that the plurality of evaluated face images include face images of a plurality of different persons instead of those of the same person. That is, it is determined that one card 200 has been used by a plurality of people and the card has been lent or borrowed. On the other hand, if there is no difference between the evaluation values, the detection unit 42 determines that the evaluated plurality of face images belong to the same person and the card 200 has not been lent or borrowed.

  In the example shown in FIG. 4, the evaluation value of the face image of Image001 and the evaluation value of the face image of Image002 are both 100, and there is no difference between the evaluation values. For this reason, the detection unit 42 determines that the two face images belong to the same person and the card 200 has not been lent or borrowed.

  The number of face images managed in the storage unit 70 in association with this identification number increases every time the card 200 with the identification number “Card001” is used at the A station after the next day. The detection unit 42 starts the detection process at a predetermined timing when the business hours have ended even after the next day. If the number of face images has increased compared to the previous detection process, the evaluation value of each face image including the increased face images is recalculated to determine whether lending or borrowing has been performed.

  FIG. 5 is a diagram for explaining an example of detection processing performed after the number of face images of the user 100 increases. The name of the face image is shown on the upper side of each face image, and the evaluation result compared with the other face images and the evaluation value calculated based on the evaluation result are shown on the lower side.

  When the face image managed in association with the identification number “Card001” is the six face images of “Image001” to “Image006” illustrated in FIG. 5 when the detection process is started, the detection unit 42 selects these six face images. Is read from the storage unit 70 and the detection process is executed.

  FIG. 5 shows an example where the card 200 with the identification number “Card001” is lent and borrowed and illegally used by a person other than the owner after the detection process shown in FIG. 4 is executed. Therefore, in addition to the four face images Image001 to Image004 of the same person, two face images Image005 and Image006 of different persons are included.

  The detection unit 42 that has started the detection process compares the six face images with brute force to calculate an evaluation value. As for the face image of Image001, in comparison with the face images of Image002 to Image004, a result that it is the same person is obtained, and the score is 100 points. On the other hand, in the comparison of the image image of Face001, the face image of Image005 and the face image of Image006, the result of being the same person cannot be obtained and the score is 50 points. As a result, the evaluation value of the face image of Image001 is 400 points in total obtained by comparing with the face images of Image002 to Image006. Similarly, the evaluation value of each face image of Image002 to Image004 is 400 points.

  For the face image of Image 005, the result with the same person is not obtained in the comparison with the face images of Image 001 to Image 004, the score is 50 points, and the result with the same person is obtained in the comparison with the face image of Image 006. Becomes 100 points. As a result, the evaluation value of the face image of Image005 is 300 points. Similarly, the evaluation value of the face image of Image006 is 300 points.

  When the evaluation values of the face images of Image001 to Image006 are obtained, the detection unit 42 compares the evaluation values. The evaluation values of the face images of Image001 to Image004 are 400 points, whereas the evaluation values of the two face images of Image005 and Image006 are 300 points, and there is a difference between the evaluation values. The face images of a plurality of different persons are included in the six face images, and it is determined that the card 200 has been lent or borrowed.

  When the detection unit 42 detects that the card 200 has been lent or borrowed, the detection unit 42 inputs the detection result to the notification unit 44. In response to this, the notification unit 44 emits sound or light at the detection device 30 and displays the detection result on the screen of the operation display unit 60 to notify that the card 200 has been lent or borrowed.

  FIG. 6 is a diagram illustrating an example of a notification screen displayed on the operation display unit 60. As shown in FIG. 6, card information 301, user face image 302, usage count 303, usage history 304, and the like are displayed on the screen together with a message 300 indicating that the card 200 has been detected.

  As the card information 301, the identification number of the card 200 that has been detected by the detection unit 42 and other information read from the card 200 when the card reader 20 reads the identification number are displayed. For example, as shown in FIG. 6, in addition to the identification number, information such as the card type, purchase date, purchase price, purchaser's name and age is displayed as the card information 301.

  In addition, as the user face image 302, face images from which different evaluation values are obtained in the detection process, that is, face images of a plurality of persons who lent or borrowed the card 200 are displayed. As shown in FIG. 5, when the face images of Image001 to Image004 and the face images of Image005 and Image006 are different persons, the face image selected from the face images of Image001 to Image004, and selected from Image005 and Image006 A face image is displayed. For example, as shown in FIG. 6, the face image acquired last among a plurality of face images representing the same person, that is, the face image of Image004 and the face image of Image006 that are the latest face images are displayed. As described above, since the face images of a plurality of persons who have detected the loan of the card 200 are displayed on the notification screen of the operation display unit 60, the station attendant is of a person whose displayed face images are different, and the detection unit 42 It can be confirmed that there is no error in the detection result by.

  In addition, information relating to the number of uses of each person's card 200 is displayed as the number of uses 303 on the notification screen. For example, the number of times each person has used the card 200 during the detection process, the total number of times the card 200 has been used, and the ratio of the number of times each person has used the total number of times are displayed. Specifically, as shown in FIG. 5, when the card 200 with the identification number “Card001” used a total of six times is used four times by a person of the face image of Image004, the usage ratio of this person is about Since it is 67% (= 4/6), “4/6 (67%)” is displayed below the face image of Image004 as shown in FIG. From the information displayed as the usage count 303, for example, a person with a high usage count is the owner of the card 200, and a person with a low usage count borrows the card 200 from the owner, impersonating the owner and improperly using the card 200. The possibility of having done it can be examined.

  In addition, the usage history of each person's card 200 is displayed as the usage history 304 on the notification screen. For example, the date and time when each person used the card 200, the station and the ticket gate, and information indicating whether the person was entering or exiting are displayed. From the information displayed as the usage history 304, for example, the card 200 used for commuting mainly on weekday mornings and evenings is used by different persons on weekends. The possibility of borrowing the card 200 from the unauthorized use can be examined.

  In addition, as shown in FIG. 6, a check box 305 for registering the card 200 that has detected the loan as a monitoring target is displayed on the notification screen. When the check box 305 is checked, the monitoring unit 43 of the detection device 30 displays the identification number of the card 200 displayed on the notification screen, the face image of the person of Image004 who borrowed and borrowed the card 200, and the person of Image006. The face image is registered in the storage unit 70 as information related to the monitoring target.

  When the monitoring target is registered, monitoring by the monitoring unit 43 is started from the next day of registration. Monitoring by the monitoring unit 43 is performed in real time when the card 200 is used. The monitoring unit 43 acquires a face image of each user who passes the ticket gate using the camera 10 and immediately executes face collation processing. The acquired face image is the face image of Image 004 registered as a monitoring target or the image 006. It is determined whether or not it matches the face image. Also, the monitoring unit 43 immediately reads the identification number of each user's card 200 with the card reader 20 and determines whether or not it matches the identification number registered as the monitoring target. At least one of the match between the face image acquired by the camera 10 and the face image registered as the monitoring target and the match between the identification number of the card 200 read by the card reader 20 and the identification number registered as the monitoring target are established. Then, the monitoring unit 43 determines that the monitoring target has been detected, and inputs the determination result to the notification unit 44. Receiving this, the alerting | reporting part 44 alert | reports having detected the monitoring object. The notification unit 44 emits sound and light from the detection device 30 and displays the detected identification number of the card 200, the detected face image of the user, and the like on the screen of the operation display unit 60. Upon receiving the notification, the staff in the station can take a predetermined response, for example, by checking the information displayed on the operation display unit 60 and asking the detected user 100 about the use of the card 200.

  For monitoring by the monitoring unit 43, information used for detecting a monitoring target can be selected from the identification number of the card 200, the owner of the card 200, and the user who borrowed the card 200 and used it illegally. . For example, when the person of the face image of Image004 shown in FIG. 6 is the owner of the card 200, the face image of Image004 is not used for monitoring, and the identification number of the card 200 and the card 200 are used. A setting is made so that the borrowed and used face image Image006 of the unauthorized user is used for detection of the monitoring target. Thereby, when the same person as the person who detected the unauthorized use of the card 200 borrows the card 200 again and uses it illegally, the detection device 30 detects this unauthorized use when passing the ticket gate of the station and notifies the station staff. be able to. The station staff who receives the notification can take a predetermined response to the unauthorized user.

  A list display button 306 and a confirmation button 307 are displayed on the notification screen shown in FIG. When the confirmation button 307 is pressed, the detection result display shown in FIG. 6 is terminated. Since the detection unit 42 performs detection processing for all identification numbers managed in the storage unit 70, if a loan or borrow is detected by another card 200 having a different identification number, the detection result is subsequently displayed. A notification screen for notification is displayed.

  When the list display button 306 is pressed on the notification screen shown in FIG. 6, all face images and related information stored in association with the identification number of the card 200 displayed on the notification screen are displayed on the screen of the operation display unit 60. Listed.

  FIG. 7 is a diagram illustrating an example of a list display screen displayed on the operation display unit 60. An item name 311 of a data number (“Data No.” in the figure), a face image, a date, a place, and an evaluation value is displayed on the upper left portion of the screen. Under the item name 311, information such as a face image associated with the identification number of the card 200 that has detected the loan is displayed in a list. When the operation display unit 60 is operated to select an item name 311 portion, the information displayed in a list can be rearranged. For example, each time an operation for selecting an item name of “evaluation value” is performed, the information displayed in a list is rearranged and displayed in ascending order of evaluation value or descending evaluation value.

  On the right side of the screen, card information 312, an evaluation value range 313, a check box 314 relating to monitoring settings, and a return button 315 are displayed. The card information 312 corresponds to the card information 301 on the notification screen shown in FIG. 6, and the check box 314 corresponds to the check box 305 on the notification screen, each showing the same information as the notification screen. When the return button 315 is pressed, the list display screen returns to the notification screen shown in FIG.

  The evaluation value range 313 on the list display screen is used to narrow down the information displayed in the list. For example, by operating the operation display unit 60 and changing the evaluation value range 313 to “300 points or less” on the screen shown in FIG. Only the data of 005 and 006 will be displayed. By changing the evaluation value range, only data relating to the owner of the card 200 can be displayed in a list, or only data relating to an unauthorized user of the card 200 can be displayed in a list.

  4 and 5 show examples in which evaluation values are calculated by comparing and evaluating each face image and all other face images in a brute force manner. The card 200 used as a commuter or commuter commuter pass is usually used by the owner of the card 200 on a daily basis. Therefore, an unauthorized user who borrows the card 200 is detected by a brute force comparison and evaluation. can do. On the other hand, depending on the usage mode of the card 200, for example, when half of the total usage count of the card 200 is illegal use, or when the usage by the owner and the usage by the illegal user are repeated in order, the brute force In the calculation method, the evaluation value of the owner and the evaluation value of the unauthorized user may have the same score. In the unauthorized use detection system, the evaluation value calculation method can be changed according to the usage mode of the card 200 so that the lending / borrowing of the card 200 can be detected even in such a case.

  Specifically, in addition to the brute force evaluation value calculation method described above, a method of calculating the evaluation value of each face image by comparison evaluation with the two most recent face images acquired before this face image, or the latest It is possible to select a calculation method by comparison evaluation with one face image. Hereinafter, these two evaluation value calculation methods will be described in detail.

  FIG. 8 is a diagram for explaining a method for calculating the evaluation value of each face image using the two most recently acquired face images. FIG. 8 shows an example in which evaluation values are obtained using the two most recent face images for the six face images of Image001 to Image006 shown in FIG. The face images of Image001 to Image006 are acquired in this order. Also, the name of the face image is shown above each face image, and the evaluation result compared with the other face images and the evaluation value calculated based on the evaluation result are shown below.

  For the face image of Image001 and the face image of Image002, the evaluation value is not calculated because there are no two face images acquired before these face images. The evaluation value of each face image of Image003 to Image006 is calculated based on the comparison result with the two most recent face images shown in FIG.

  Specifically, as shown in FIG. 8, for the face image of Image003, the result of being the same person is obtained by comparison with the face images of Image001 and Image002, and thus the evaluation value is 200 points. Similarly, the evaluation value of the face image of Image004 compared with the face images of Image002 and Image003 is 200 points.

  On the other hand, as for the face image of Image005, a result that it is not the same person is obtained by comparing with the face images of Image003 and Image004, so the evaluation value is 100 points. As for the face image of Image006, a result that it is not the same person is obtained by comparison with the face image of Image004, and a result that it is the same person is obtained by comparison with the face image of Image005. It becomes.

  As a result, as shown in FIG. 8, there is a difference between the evaluation value of each face image of Image003 and Image004, the evaluation value of the face image of Image005, and the evaluation value of the face image of Image006. In 42, a plurality of persons are included in the face image, and it can be determined that the card 200 has been lent or borrowed.

  FIG. 9 is a diagram for explaining a method for calculating the evaluation value of each face image using the most recently acquired one face image. FIG. 9 shows an example in which evaluation values are obtained using the latest one face image for the six face images of Image001 to Image006 shown in FIG. The face images of Image001 to Image006 are acquired in this order. Also, the name of the face image is shown above each face image, and the evaluation result compared with the other face images and the evaluation value calculated based on the evaluation result are shown below.

  For the face image of Image001, the evaluation value is not calculated because there is no face image acquired earlier. The evaluation value of each face image of Image002 to Image006 is calculated based on the comparison result with the most recent face image shown in FIG.

  Specifically, as shown in FIG. 9, for the face image of Image002, a result that it is the same person is obtained by comparison with the face image of Image001, so the evaluation value is 100 points. Similarly, the evaluation value of each face image of Image003 and Image004 is 100 points by comparison with the most recent face image.

  On the other hand, for the face image of Image005, a result that it is not the same person is obtained by comparison with the face image of Image004, so the evaluation value is 50 points. As for the face image of Image006, a result that it is the same person is obtained by comparison with the face image of Image005, so the evaluation value is 100 points.

  As a result, as shown in FIG. 9, there is a difference between the evaluation values of the face images of Image002 to Image004 and Image006 and the evaluation value of the face image of Image005. It is possible to determine that the card 200 has been lent or borrowed.

  In the present embodiment, for convenience of explanation, the case where the use of a commuter commuter pass is started has been described as an example. However, the unauthorized use detection system sets an acquisition period of identification information and feature information to be processed in detection processing. It can be done. Specifically, for example, when the detection process is started manually, if the operation display unit 60 is operated and a period is specified by date or time, the unauthorized use detection system acquires the identification information and feature information acquired within the specified period. The detection process is executed for the processing target. Further, for example, if a period is set in advance together with the start timing of the detection process, the unauthorized use detection system automatically executes the detection process using the identification information and feature information acquired within the set period as processing targets. For example, if a period is specified such as from Monday to Friday and from 1st to the end of the month, the unauthorized use detection system executes the above-described detection process based on information acquired within this period.

  Further, in the present embodiment, when the evaluation value is calculated, an example is given in which different points are given depending on whether or not the two face images belong to the same person. In this regard, whether or not the face images are of the same person, that is, whether or not the feature information matches, is determined by a numerical value such as a correlation coefficient indicating the matching degree of the feature information satisfying a predetermined threshold value. Needless to say, it is done based on whether or not. The evaluation value calculation method includes not only a method of calculating an evaluation value by assigning points based on the degree of coincidence of feature information, but also a method of calculating an evaluation value by directly using a numerical value indicating the degree of coincidence. May be. Specifically, in a mode in which a numerical value indicating the degree of coincidence is obtained between the characteristic information for which the evaluation value is to be calculated and each of a plurality of other characteristic information, and the total value of the obtained numerical values is used as the evaluation value. There may be. When the feature information of a plurality of different persons is included in the plurality of feature information, the evaluation value calculation method is particularly limited if an evaluation value different from the other feature information can be obtained with at least one feature information. Not.

  In addition, regarding the determination of whether there is a difference in the evaluation value of feature information, the evaluation value varies depending on the type of feature information used, the accuracy of feature information acquisition, the evaluation value calculation method, etc. In the detection system, when there is a significant difference between a plurality of evaluation values, the detection unit 42 determines that there is a difference in the evaluation values. Specifically, a determination condition such as a numerical range in which the detection unit 42 determines that there is a difference in evaluation value can be changed by setting.

[Example 2]
In the present embodiment, another usage method of the unauthorized use detection system shown in the first embodiment will be described. In the present embodiment, description of functions and operations of the unauthorized use detection system described in the first embodiment will be omitted, and differences from the first embodiment will be described.

  In the unauthorized use detection system, the execution timing of the detection process can be changed according to the usage mode of the card 200. For example, it can be set to execute the detection process every time the card 200 is used.

  For example, in order to detect lending / borrowing for a card 200 used as a lift ticket at a ski resort, a camera 10, a card reader 20 and a detection device 30 are installed at the lift platform, and each time the card 200 is used. Set to execute the process. Each time the card 200 is used at the lift platform, the information management unit 41 acquires the face image of the user 100 and the identification number of the card 200. When the face image and the identification number are obtained, as described with reference to FIG. 9, the detection unit 42 performs a comparative evaluation of the card 200 having the same identification number with the most recently acquired face image. When the detection unit 42 detects lending / borrowing of the card 200, the notification unit 44 executes notification processing, and displays the identification number and face image of the card 200 on the screen of the operation display unit 60 as shown in FIG. The staff at the lift platform can check the information displayed on the operation display unit 60 and take a predetermined response to the unauthorized user.

  If you are wearing a hat, sunglasses, goggles, etc. or your face is difficult to acquire, such as ski resort customers, it is difficult to obtain a face image suitable for face matching. May be reduced. In order to cope with such a case, in the unauthorized use detection system, instead of the detection unit 42, a staff member can set to determine unauthorized use by lending and borrowing. In addition, in order to cope with the case where it is difficult to acquire a face image suitable for face matching, the unauthorized use detection system acquires an appearance image that captures the appearance of the user 100 in addition to the face image as feature information. It can also be set.

  When the user 100 causes the card reader 20 to read the identification number of the card 200 in order to use the lift, the information management unit 41 uses a still image obtained by capturing the whole body of the user 100 with the camera 10 as an appearance image. get. Further, the information management unit 41 cuts out the face portion of the user 100 from the appearance still image and uses it as a face image.

  Each time the identification number of the card 200 is acquired by the card reader 20, the detection unit 42 includes an appearance image obtained by capturing the whole body of the user 100, a face image obtained by cutting a face portion from the appearance image, an identification number of the card 200, and the like. Information is displayed on the screen of the operation display unit 60. The attendant checks this screen and determines whether or not the card 200 is lent or borrowed.

  FIG. 10 is a schematic diagram illustrating an example of a screen that is displayed when the clerk determines lending / borrowing. As shown in FIG. 10, card information 401 including the type and identification number of the card 200 read by the card reader 20 and the date and time 407 when the identification number was read by the card reader 20 are displayed at the top of the screen.

  In the center of the screen, the face image 405 and appearance image 406 of the user 100 acquired by the camera 10 when the card reader 20 reads the identification number of the card 200 are displayed. On the left side of the screen, a use date and time 402 when the same card 200 was used last time, and a face image 403 and an appearance image 404 acquired at the previous use are displayed. That is, it corresponds so that the face image 403 and appearance image 404 of the user 100 who last used the card 200 shown in the card information 401 can be compared with the face image 405 and appearance image 406 of the user 100 currently used. The images are displayed side by side.

  The staff at the lift platform compares the face image 403 of the previous user 100 with the face image 405 of the current user 100, and the appearance image 404 of the previous user 100 and the appearance of the current user 100. The image 406 is compared, and it is determined whether or not two persons using the same card 200 are the same person. Even when a part of the face is covered with a hat, goggles or the like and the facial features are difficult to understand, it is possible to determine whether or not they are the same person from the difference in clothes by comparing the appearance images 404 and 406.

  When the clerk determines unauthorized use, the detection device 30 may be set to perform face matching processing and appearance matching processing in order to assist the clerk's determination. When setting to perform face matching processing is performed, the detection unit 42 executes face matching processing for determining whether or not the two face images 403 and 405 belong to the same person. The face matching process is performed using a conventional face authentication technique, similar to the above-described evaluation value calculation.

  As shown in FIG. 10, a processing result 408 is displayed on the right side of the screen side by side with the face images 403 and 405 that are processing targets of the face matching process. When the detection unit 42 that collates the two face images 403 and 405 determines that the face images are the same person, information indicating that they are the same person is displayed as the processing result 408. On the other hand, if it is determined that the face images are not the same person, information indicating that they are different persons is displayed as the processing result 408. In addition, when a part of the face is hidden or other features such as eyes, nose, and mouth cannot be extracted from the face image, and the detection unit 42 determines that the face matching process cannot be performed, As shown in FIG.

  When setting to execute the appearance matching process, the detection unit 42 executes the appearance matching process for determining whether or not the two appearance images 404 and 406 belong to the same person. Appearance verification processing determines whether two appearance images are of the same person using appearance features such as clothes, clothing colors, hats and other ornaments, ornament colors, and belongings. However, since it is performed using the prior art, detailed description is omitted.

  As shown in FIG. 10, a processing result 409 is displayed on the right side of the screen side by side with the appearance images 404 and 406 that have been subjected to appearance matching processing. When the detection unit 42 that collates the two appearance images 404 and 406 determines that the appearance images are the same person, information indicating that they are the same person is displayed as the processing result 409. On the other hand, when it is determined that they are not appearance images of the same person, information indicating that they are different persons is displayed as shown in FIG. When the detection unit 42 determines that the appearance matching process cannot be executed, information indicating that the determination is impossible is displayed as the processing result 409.

  As a result of the collation processing by the detection unit 42, when a result that it is not the same person is obtained by at least one of the face collation processing and the appearance collation processing, the notification unit 44 emits sound or light. Also, as shown in FIG. 10, information 410 is displayed on the screen of the operation display unit 60 to call attention to a staff member who determines unauthorized use. Thus, the detection unit 42 can assist the clerk's judgment by notifying that the card 200 may be lent or borrowed.

  As described above, according to the unauthorized use detection system and the unauthorized use detection method according to this embodiment, after the use of a portable medium such as a card is started, the portable medium is used when the portable medium is used. It is possible to acquire and manage characteristic information for distinguishing each user, such as a face image of the user who has performed. Then, after a plurality of pieces of characteristic information of users who use the same portable medium are obtained, it is determined whether or not the portable medium is used by the same person by comparing and evaluating each piece of characteristic information at a predetermined timing. can do. When the same portable medium is used by a plurality of different persons such as lending and borrowing, the unauthorized use detection system can detect this as unauthorized use and notify it. In addition, if the identification information of the portable medium that has detected unauthorized use and the feature information of the user of this portable medium are registered as monitoring targets, the unauthorized use detection system monitors the use of the portable medium next time. When used, this can be detected and notified.

  For example, even if it is a place such as a ticket gate of a station where many users pass continuously, it is only necessary to acquire feature information of each user such as a face image when the user passes. The ticket gate is not congested due to processing time or false detection. If detection processing is executed at a predetermined timing such as at the end of business hours at a station, and the user who has made unauthorized use is identified and registered as a monitoring target, this will be detected when the user makes unauthorized use again. And a predetermined response can be taken.

  Also, for example, when playing a game using a card at a game store, the card identification information and characteristic information such as a card user's face image and appearance image are acquired when the card is used. Can detect unauthorized use by a person other than the owner. Even if the card is sold without requesting personal information such as name, such as ski lift tickets, game cards for games, etc., unauthorized use of the card can be detected. A usage detection system can be used.

  Further, in the unauthorized use detection system, different persons can be detected based on the feature information, so that each person can be identified and the number of people can be specified based on the feature information acquired within a predetermined period. For example, the unauthorized use detection system specifies the number of card users based on the characteristic information of the card users acquired during the business day after the business hours of each business day at the amusement store. The game store can manage the number of people specified by the unauthorized use detection system as the number of games on each business day. The first setting that identifies the number of people who have used the same card separately by counting the number of people who have used the same card for the number of people specified by the unauthorized use detection system, the same card It is possible to select any one of the second setting for specifying the number of persons by counting a plurality of persons using the number of persons as one, the third setting for specifying both the number of persons according to the first setting and the number of persons according to the second setting. . For example, the game store can select the third setting and manage the number of people specified by the first setting separately from the number of games and the number of people specified by the second setting separately from the actual number of games.

  In addition, in the unauthorized use detection system, a detection process for detecting unauthorized use can be executed by specifying a period and targeting identification information and feature information acquired within the period. For example, since the period for detection processing can be changed according to the number of users of portable media, usage mode, etc., the unauthorized use detection system is used in accordance with operations that have been performed in various places. be able to.

  In addition, in the unauthorized use detection system, the feature information of each user can be displayed on the screen of the unauthorized use detection system so that the feature information of each user such as a face image obtained by capturing the user's face and an appearance image obtained by capturing the appearance can be compared. . As a result, for example, even when it is difficult to determine whether or not a plurality of users using the same portable medium are the same person, an attendant or the like visually checks the information displayed on the screen. Can be judged.

  As described above, the unauthorized use detection system and the unauthorized use detection method according to the present invention are for detecting unauthorized use in which a portable medium owned by each person satisfying a predetermined condition is used by someone other than the owner. Useful for.

DESCRIPTION OF SYMBOLS 10 Camera 20 Card reader 30 Detection apparatus 40 Control part 41 Information management part 42 Detection part 43 Monitoring part 44 Notification part 50 Communication part 60 Operation display part 70 Storage part

Claims (12)

An unauthorized use detection system that detects unauthorized use in which a portable medium owned by each person satisfying a predetermined condition is used by someone other than the owner,
Reading means for reading identification information for identifying each portable medium from the portable medium when using the portable medium;
Obtaining means for obtaining characteristic information of a portable medium user who has caused the reading means to read the identification information;
Management means for managing the feature information in association with the identification information;
After the portable medium is used a plurality of times, unauthorized use of the portable medium is determined based on whether or not different feature information is included in a plurality of feature information associated with the same identification information. An unauthorized use detection system comprising a detection means for detecting.   The management unit manages only feature information satisfying a predetermined condition set in advance from feature information obtained from the acquisition unit each time a portable medium is used, in association with identification information of the portable medium. The unauthorized use detection system according to claim 1. It further comprises display means for displaying feature information,
The said detection means displays a plurality of feature information associated with the same identification information on the screen of the display means so that they can be compared after the portable medium is used a plurality of times. The unauthorized use detection system according to 1 or 2.   For each of the plurality of feature information, the detection means calculates an evaluation value from a value indicating the degree of coincidence with the other feature information, and based on the evaluation value, all of the plurality of feature information belongs to the same user. 4. The unauthorized use detection system according to claim 3, wherein it is determined whether or not.   The detection means compares a plurality of pieces of feature information, calculates an evaluation value of each piece of feature information by assigning different points depending on whether the pieces of feature information match or does not match, and based on the evaluation values, 4. The unauthorized use detection system according to claim 3, wherein it is determined whether or not all the feature information of the same belongs to the same user.   6. The unauthorized use detection system according to claim 4, wherein the detection means displays a list of a plurality of feature information together with an evaluation value of each feature information on the screen of the display means.   The detection means displays only the feature information whose evaluation value is within a predetermined range among a plurality of feature information associated with the same identification information on the display means. The unauthorized use detection system according to any one of the above. The reading means and the obtaining means operate for a predetermined business hours,
The said detection means starts the process which detects the borrowing and borrowing of a portable medium at the predetermined timing preset after the said business hours end, The one of Claims 1-7 characterized by the above-mentioned. Unauthorized use detection system. Monitoring means for accepting registration of at least one of identification information of the portable medium and at least one feature information associated with the identification information, and monitoring use of the portable medium during the business hours; ,
The unauthorized use detection system according to claim 8, further comprising a notification unit that notifies the use of the portable medium registered as a monitoring target when the monitoring unit detects the use of the portable medium.   The unauthorized use detection system according to claim 1, wherein the acquisition unit acquires a facial image of a portable medium user as the feature information.   The unauthorized use detection system according to claim 1, wherein the acquisition unit acquires an appearance image obtained by capturing an appearance of the portable medium user as the feature information. An unauthorized use detection method in which an unauthorized use detection system detects unauthorized use in which a portable medium owned by each person satisfying a predetermined condition is used by someone other than the owner,
A first step of reading identification information for identifying each portable medium from the portable medium when using the portable medium;
A second step of acquiring characteristic information of the portable medium user who has read the identification information in the first step;
A third step of managing the feature information in association with the identification information;
After the first to third steps are executed a plurality of times, based on whether or not different feature information is included in the plurality of feature information associated with the same identification information, And a fourth step of detecting lending / borrowing.

Images