JP2005165527A - User presumption device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a user presumption device which can perform presumption of a user who has deposited a counterfeit bill when the counterfeit bill is found in bank bills processed by a bank bill processor in the central bank, etc. later, without complicating a main part of a bank bill processor. <P>SOLUTION: The user presumption device 1 stores the user with high probability of having deposited the counterfeit bill to an ATM 3 in a second database 15 for each counterfeit bill discovered so far. When presuming the user who has used the counterfeit bill, the presumption is made by not only matching rate with the counterfeit bill but also possibility of using the counterfeit bill until now by using counterfeit bill management data stored in the second database 15. Thereby, presumption accuracy of a user of the counterfeit bill can be improved. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  This invention discriminate | determines that a banknote processing apparatus is a genuine note, and when the counterfeit ticket was discovered in the central bank etc. later from the banknote received and processed, the counterfeit ticket was thrown into the banknote processing apparatus. The present invention relates to a user estimation device capable of estimating a user.

  2. Description of the Related Art Conventionally, an automated teller machine (ATM) that performs deposit / withdrawal processing on an account designated by a user is installed in a store of a financial institution. A general ATM has a banknote processing device for discriminating the authenticity and denomination of banknotes inserted into the ATM. The banknote handling apparatus uses the basic information registered for each denomination to discriminate the denomination of whether the inserted banknote is a genuine note. This basic information is based on sensor output (banknote data) obtained when a new bill is read by a sensor of the same type as the sensor of the banknote processing apparatus. The banknote processing device collates banknote data read by a sensor for each inserted banknote and basic information, and detects the degree of similarity as the degree of authenticity. And if the degree which is a detected genuine note is larger than the predetermined threshold value, the bill will be identified as a genuine note, and if it is smaller than the threshold value, the bill will be identified as a false note.

  ATM accepts the banknote which the banknote processing apparatus identified as a genuine note, and accommodates it in the cartridge provided in ATM. Conversely, banknotes that have been identified as fake tickets are returned to the user. Then, the ATM calculates the total amount of accepted banknotes and coins as the deposit amount, and deposits the calculated deposit amount into the account designated by the user. This process is a process of notifying the management server that manages each account that the deposit amount calculated for the designated account has been received. The management server updates the balance of the corresponding account based on this notification.

  By the way, a counterfeit note in which a bill processing device is inserted may be misidentified as a genuine note. Although it is technically possible to make the occurrence probability of this misclassification zero, a serious problem arises in terms of ATM operation. Specifically, even if it is a very sophisticated counterfeit note, there is a difference between it and a genuine note, so the occurrence of misclassification that the counterfeit note is genuine by raising the above threshold extremely high Probability can be reduced to zero. However, since ATMs handle banknotes in circulation, if a banknote with a certain amount of wrinkles and dirt (genuine bill) cannot be identified as a genuine note, only a very limited number of banknotes can be handled. It becomes a situation that can not be operated properly. In this way, if the threshold is set too high, the probability of misidentification that it is a fake note will increase even though it is a genuine note, and ATM will not operate properly, so the threshold is set too high. In rare cases, it is possible to misclassify a sophisticated counterfeit ticket as a genuine note. In many cases, a counterfeit ticket misidentified as a genuine note is discovered when a banknote (accepted banknote) stored in a cartridge taken out from an ATM is scrutinized.

In this way, although rarely, the banknote processing device may misclassify a counterfeit note as a genuine note, so when a counterfeit note is found, the user who put the counterfeit note into an ATM is estimated. Technology (Patent Document 1) that prints information such as the date and time received on the periphery of the accepted banknote with a coating agent mixed with a fluorescent agent with a short effective life, the serial number of the banknote, the received date and time, and usage Has been proposed (Patent Document 2) in which a cartridge is provided with a storage unit for storing a person's ID and the like in association with each other.
Japanese Patent No. 31481717 JP 11-328493 A

  However, since the said patent document 1 prints information, such as the date received on the peripheral part for every banknote received as a genuine note, a banknote processing apparatus becomes complicated. Furthermore, this printing takes time, the time required for transaction processing (payment processing) increases, and as a result, the ATM processing speed decreases. At the same time, there is a problem that the running cost is greatly increased by the coating agent used for printing.

  In addition, the above-mentioned Patent Document 2 requires a high-resolution image sensor and a large-capacity image memory in order to read the serial number from the inserted bill, and the bill processing apparatus becomes large and complicated. is there.

  The object of the present invention is to insert a counterfeit note when a counterfeit note is later discovered in a central bank or the like from banknotes processed by the note processing device without complicating the main body of the note processing device. The object is to provide a user estimation device capable of estimating a user.

  The present invention has the following configuration in order to solve the above problems.

(1) Read banknote data for each inserted banknote, detect the degree that the banknote is a genuine note from the read banknote data, and discriminate that it is a genuine note if the detected degree is greater than a first threshold, From a device with a built-in banknote processing apparatus that has a means for acquiring user data of a user who has inserted a banknote accepted by the banknote processing apparatus while receiving a banknote processing apparatus that accepts a banknote identified as a genuine note Banknote management data acquisition means for acquiring banknote management data including the banknote data and the user data for banknotes accepted by the banknote handling apparatus;
Banknote management data storage means for storing the banknote management data acquired by the banknote management data acquisition means;
When a bill that is later determined to be a fake note is found from the bills received by the bill processing apparatus, the bill data of the fake note is read, and the bill data of the fake note read here and the bill management A degree of coincidence calculating means for calculating the degree of coincidence with the banknote data of the banknote stored in the data storage means;
For each discovered counterfeit ticket, user data stored in the banknote management data storage means, and counterfeit ticket management data storage means for storing counterfeit ticket management data including the calculated degree of coincidence,
Appearance frequency calculating means for calculating the appearance frequency of the user for the fake ticket for each user using the fake ticket management data stored in the fake ticket management data storage means,
Utilization in which a counterfeit note whose current coincidence is calculated is inserted into the banknote processing apparatus using the appearance frequency for each user calculated by the appearance frequency calculating unit and the coincidence calculated by the coincidence degree calculating unit Estimating means for estimating a person.

  In this configuration, banknote data used by the banknote management data acquisition means and the banknotes are entered for each banknote that has been identified and accepted as a genuine note from a device incorporating the banknote processing apparatus. The banknote management data including the user data of the user is acquired, and the banknote management data acquired by the banknote management data storage unit is stored. As an apparatus incorporating a banknote processing apparatus, there exists an apparatus which deposits the amount of money which can be used, for example to an automatic teller machine (ATM) and an IC card used as a prepaid card. The coincidence calculation means reads the banknote data of the counterfeit bills later discovered by the central bank from the banknotes accepted by the banknote processing device, and the banknote data of the counterfeit bills read here and the banknote management data storage means store them. The degree of coincidence with the banknote data of the banknote being processed is calculated. Usually, the counterfeit bill accepted by the banknote handling apparatus is not found immediately after acceptance, but is discovered after a certain period (several days). For this reason, although the state of fake bills (the state of adhesion of wrinkles and dirt) is slightly different at the time of acceptance in the banknote processing device and at the time of discovery, the banknote data stored in the banknote management data storage means From the above, the banknote data when the counterfeit ticket discovered this time is discriminated by extracting the one having a degree of coincidence with the banknote data of the counterfeit ticket that is larger (relatively larger) than a predetermined value. It is included in the reliably extracted banknote data.

  For example, for a bill with bill data whose calculated coincidence is larger than a predetermined value, the fake bill management data storage means is a user who has inserted this bill (a user who has a high probability of having inserted a fake bill). Fake ticket management data including user data and the calculated degree of coincidence is stored. The counterfeit ticket management data storage means stores counterfeit ticket management data in an accumulated manner for each discovered counterfeit ticket. Using the fake ticket management data stored in the fake ticket management data storage means, the appearance frequency calculation means calculates the appearance frequency for each user. This appearance frequency is a value that indicates the frequency with which each user has a high probability of being put in a fake ticket that has been discovered so far. For example, for each user whose user data is stored in the fake ticket management data storage means, the sum of the degrees of coincidence associated with the user data is used as the appearance frequency of the user identified by the user data. calculate. Alternatively, the total number of user data stored in the fake ticket management data storage means may be calculated as the appearance frequency of the user identified by the user data. Usually, a person who uses a counterfeit ticket uses a counterfeit ticket over and over again, so that the appearance frequency calculated by the above method increases.

  Then, the estimation unit uses the fake note whose current coincidence is calculated based on the coincidence calculated by the coincidence calculating unit and the appearance frequency calculated by the appearance frequency calculating unit. Person is estimated. For example, it is estimated that a user having a larger product of the calculated degree of coincidence and the appearance frequency is more likely to be a user who has inserted the counterfeit note whose degree of coincidence has been calculated into the banknote processing apparatus.

  As a result, the user who inserted the counterfeit ticket discovered this time is estimated by considering not only the degree of coincidence with the counterfeit ticket discovered but also the degree of appearance of the counterfeit ticket for each user. It is possible to estimate a user who is highly likely to use a counterfeit note even when the counterfeit note state (the state of adhesion of wrinkles or dirt) is slightly different between the time of acceptance and the time of discovery. Moreover, since it is not necessary to provide a high-resolution image sensor for reading the serial number or a large-capacity image memory in the banknote processing apparatus, the banknote processing apparatus is not complicated.

  (2) The banknote management data storage means stores banknote management data for a banknote whose degree of being a detected genuine note is larger than the first threshold and smaller than a predetermined second threshold. The banknote management data is not stored for banknotes larger than the threshold.

  In this configuration, the banknote management data of the banknote is not stored for banknotes whose degree of authenticity is greater than the second threshold. In other words, banknote management data is stored only for banknotes whose authenticity is between the first threshold value and the second threshold value. Accordingly, the storage capacity of the banknote management data storage means can be reduced, and the total number of banknotes for which the coincidence calculation means calculates the coincidence when a counterfeit is found can be reduced, thereby shortening the processing time.

  As described above, according to the present invention, in addition to the degree of coincidence with the discovered counterfeit ticket, the user who entered the counterfeit ticket discovered this time is considered for each user in consideration of the degree of appearance of the counterfeit ticket. Since the estimation is performed, it is possible to estimate a user who has a high possibility of using a fake ticket. Moreover, since it is not necessary to provide a high-resolution image sensor for reading the serial number or a large-capacity image memory in the banknote processing apparatus, the banknote processing apparatus is not complicated.

  Hereinafter, a user estimation device according to an embodiment of the present invention will be described.

  FIG. 1 is a diagram showing a configuration of a financial system to which a user estimation device according to an embodiment of the present invention is applied, and a flow of banknotes. In FIG. 1A, reference numeral 1 denotes a user estimation device according to an embodiment of the present invention, which is installed at the center of a financial institution. The center also has an account number for each account, deposit / withdrawal history, current account balance, account holder (user) name, user address, user contact phone number, cash card PIN, etc. A management server 2 is installed to manage the account information. Reference numeral 3 denotes an automatic teller machine (ATM) installed in each store of a financial institution. The ATM 3 installed in each store is connected to the user estimation device 1 and the management server 2 installed in the center by a dedicated line, and can perform data communication with the user estimation device 1 and the management server 2. Also, as is well known, ATM 3 is provided with a banknote processing device that discriminates the authenticity of banknotes inserted into ATM 3.

  Here, the operation of ATM3 will be briefly described. ATM 3 performs deposit processing and withdrawal processing for an account designated by the user. Since the withdrawal process is the same as that of the known ATM3, the description is omitted here. Further, the deposit processing is different from the known ATM 3 in that the bill management data is transmitted to the user estimation device 1 after the transaction is completed. Specifically, ATM3 receives the input of the account number of the object which performs a money_receiving | payment process, and acquires banknote data with the sensor of a banknote discrimination part for every banknote thrown into ATM3. The banknote data acquired here is an output value (x1 to xn) of a sensor from an n-point magnetic sensor or an infrared sensor determined in advance with respect to the banknotes obtained while transporting banknotes. ATM3 detects the similarity between the acquired banknote data and preset basic information as the degree of authenticity. And if the degree detected here is larger than a predetermined threshold (the first threshold referred to in the present invention), the bill is identified as a genuine note and accepted. On the other hand, if the detected degree is smaller than a predetermined first threshold value, it cannot be identified as a genuine note and is returned to the user. The ATM 3 notifies the management server 2 of the deposit of the total amount of the banknotes that have been identified and accepted for the designated account as a genuine note. The management server 2 updates the account balance managed for the corresponding account based on the notification from the ATM 3. ATM3 stores the received banknote in the cartridge provided in the banknote processing apparatus. The processing so far is the same as that of a known ATM. And ATM3 uses a banknote data (x1-xn) used for a money type and discrimination for every banknote received, insertion date and time, a store number, an ATM number, a customer number, and banknote management data in this deposit process as a user estimate. Transmit to device 1.

  Banknotes accepted at ATM3 of each store, such as banknotes accepted at each ATM3 of store A and banknotes accepted at each ATM3 of store B, are sent to the central bank as shown in FIG. And whether or not a fake ticket is included is confirmed. At this time, if it is confirmed that the counterfeit ticket is included, the counterfeit ticket is temporarily returned to the financial institution, and the financial institution is required to identify the user who has inserted the counterfeit ticket into the ATM 3.

  In addition, ATM3 is conventionally provided with the communication part as a structure for performing data communication between the management servers 2, and performs data communication with the user estimation apparatus 1 in this existing communication part.

  FIG. 2 is a diagram showing the configuration of the user estimation device according to the embodiment of the present invention. The user estimation device 1, when a banknote that is later determined to be a fake by a central bank or the like is found from among the control unit 11 that controls the operation of the user estimation device 1 and the banknote received by the ATM 3, In the banknote data reading unit 12 that reads the banknote data of the fake ticket, the communication unit 13 that performs data communication with the ATM 3 installed in each store, and the banknotes accepted by the ATM3, that is, the banknote processing device provided in the ATM3 A first database 14 for storing banknote management data for bills identified as genuine, a second database 15 for storing counterfeit management data for counterfeits discovered at a central bank, and the like; An operation unit 16 that performs an input operation on the user estimation device 1, and a display unit 17 that displays input contents input through the operation unit 16, processing results in the user estimation device 1, and the like. , And a. The banknote data reading unit 12 applies a sensor of the same type as a sensor that reads banknote data from a banknote when identifying the authenticity of a banknote inserted with an ATM3 banknote processing apparatus installed in each store. . This sensor is a relatively inexpensive magnetic sensor or infrared sensor. In other words, the sensor provided in the ATM 3 banknote processing apparatus is also a relatively inexpensive magnetic sensor or infrared sensor.

  The 1st database 14 memorizes bill management data acquired from ATM3 installed in each store. As shown in FIG. 3, the banknote management data associates a denomination, banknote data, and the like for each insertion banknote, store number, ATM number, customer number (user data), and accepted banknotes. The insertion date and time is the date and time when it was inserted into the ATM 3, the store number is a number that identifies the store where the ATM 3 into which the bill was inserted is installed, the ATM number is a number that identifies the ATM 3 into which the bill was inserted, The customer number is the account number that is the subject of the deposit process. Moreover, banknote data are the data which the banknote processing apparatus provided in ATM3 detected with the sensor in order to discriminate authenticity about the banknote. This banknote data is a sensor output (x1 to xn) from a magnetic sensor or an infrared sensor at a predetermined n-point position with respect to the banknote obtained while transporting the banknote.

  The second database 15 stores fake ticket management data for each found fake ticket. As shown in FIG. 4, the fake ticket management data is data in which a fake ticket number, discovery date / time, denomination, customer number, coincidence, insertion date / time, store number, and the like are associated with each other. The counterfeit ticket number is a number for identifying the counterfeit ticket that has been discovered, the date and time of discovery is the date and time that the counterfeit ticket was discovered, and the denomination is the denomination of the counterfeit ticket. Further, among banknotes for which banknote management data is stored in the first database 14, a customer number, a match for each banknote whose degree of coincidence between the banknote data and the banknote data of a fake note is larger than a predetermined value. , Date and time, and store number are related.

  Registration of fake ticket management data in the second database 15 is performed according to the flowchart of FIG. 6 to be described later every time a fake ticket is found, and is used to estimate the user of the found fake ticket.

  In addition, the counterfeit ticket management data as shown in FIG. 4 is not stored, but the counterfeit ticket management data storage means stores the counterfeit ticket number, discovery date, denomination, and counterfeit bill banknote data, and the customer number. The degree of coincidence, the input date and time, the store number, etc. may be calculated each time.

  The banknote management data of banknotes that have been scrutinized by a central bank or the like and confirmed not to be counterfeit are deleted from the first database 14. In other words, the banknote management data of the banknote accepted by the ATM 3 is stored in the first database 14 only for a period (several days) until the banknote is examined by a central bank or the like. On the other hand, the fake ticket management data stored in the second database 15 is not deleted unless there are special circumstances.

  Hereinafter, the operation of the user estimation device 1 of this embodiment will be described. FIG. 5 is a flowchart showing the operation of the user estimation device of this embodiment.

  The user estimation device 1 of this embodiment waits for some input operation to be performed in the operation unit 16 or to receive banknote management data sent from the ATM 3 in the communication unit 13 (s1, s2). .

  The user estimation apparatus 1 discriminates that the banknote processing apparatus applied to ATM3 is a genuine note by receiving the banknote management data sent from ATM3 by the communication part 13, and manages the banknote of the accepted banknote. Get the data. The user estimation device 1 registers the banknote management data acquired from the ATM 3 in the first database 14 (s6).

  Thus, since the banknote management data of the banknote which each ATM3 received is transmitted to the user estimation apparatus 1, the banknote management data of the banknote received by each ATM3 in the user estimation apparatus 1 can be managed by the first database 14. .

  Here, the banknote management data is stored in the first database 14 for all banknotes received by each ATM 3, but the banknote management data for banknotes that can be clearly determined to be genuine is the first database 14. May not be stored. For example, a second threshold value larger than a threshold value (first threshold value referred to in the present invention) for distinguishing that the ATM3 banknote processing apparatus is a genuine note is set, and the similarity to the basic information is set to the second threshold value. For banknotes larger than the threshold value, the banknote data may not be stored in the first database 14. In this way, the storage capacity of the first database 14 can be suppressed. Moreover, although it may be determined by the user estimation apparatus 1 whether it is banknote data memorize | stored in the 1st database 14, this judgment is made by ATM3 and it is not necessary to make it memorize | store in the 1st database 14. About banknote data of a banknote, it is preferable to set it as the structure which is not transmitted to the user estimation apparatus 1 from ATM3. The reason is that the amount of data communicated between the ATM 3 and the user estimation device 1 is reduced, and the time required for data communication between both devices can be shortened.

  The second threshold value is set to a value at which the probability of misidentification that a counterfeit note is a genuine note is substantially zero. In addition, when authenticity of the inserted bill is discriminated by the second threshold value, the probability of misidentification that it is a fake note is increased despite the fact that it is a genuine note, and ATM3 is appropriately Although it cannot be operated, such a problem does not occur because the threshold for discriminating the authenticity of the inserted bill is smaller than the second threshold.

  Further, when any input operation is performed on the operation unit 16, the user estimation device 1 determines whether or not the fake ticket for which the current input operation is found is a use estimation request for requesting the user's estimation. (S3). If it is determined that the request is a user estimation request, a user estimation process is executed (s4). In addition, the process requested this time (other process) for determining that the request is not a user estimation request but another request is executed (s5).

  Here, the user estimation process concerning s4 will be described. FIG. 6 is a flowchart showing this user estimation processing. The administrator inputs a user estimation request at the operation unit 16 when a counterfeit note is found at the central bank after the ATM3 accepts the bill and the found counterfeit note is returned to the financial institution. Perform the operation.

  The user estimation device 1 receives the found counterfeit note (s11), and the bill data reading unit 12 reads the bill data of the counterfeit note (s12). The bill data of the fake bill read in s12 is also an output value of the sensor at n points predetermined for the bill. Here, it is assumed that the banknote data of the fake bill read in s12 is y1 to yn. Moreover, the banknote processing apparatus of ATM3 and the sensor of the banknote data reading part 12 are the same kind.

  The user estimation device 1 accepts input of the found information that has been found about the counterfeit ticket discovered this time (s13). The found information is, for example, information such as the store where the ATM 3 in which the counterfeit ticket discovered this time is inserted, the date and time when the ATM 3 is inserted, and the like is input by the administrator through the operation unit 16. At this time, the administrator inputs only those based on obvious facts, and does not input those inferred. If there is no clear information that can be input, this step is skipped. The user estimation device 1 determines a target banknote from banknotes whose banknote data is stored in the first database 14 based on the found information received in s13 (s14). For example, when the store in which the ATM 3 in which the counterfeit note is inserted is input as the clarification information, the processing according to s14 sets all banknotes whose store numbers are the store as target banknotes and other banknotes as non-targets. Banknotes. In addition, when the insertion date and time is input as the clarification information, the banknotes inserted on that day are set as target banknotes, and other banknotes are set as non-targets. Furthermore, when no finding information is input, all banknotes stored in the first database 14 are set as target banknotes.

  In addition, said s13, s14 is a process for reducing the processing load in the user estimation apparatus 1 by reducing the total number of the banknotes which calculate the coincidence with the banknote data of the fake note mentioned later. When these two processes are not provided, all banknotes stored in the first database 14 may be set as target banknotes.

  The user estimation device 1 calculates, for each target banknote determined in s14, the degree of coincidence between the banknote data and the banknote data of the counterfeit ticket discovered this time (banknote data read in s12), and the coincidence calculated here. A bill having a degree greater than a predetermined value is extracted (s15 to s18). The calculation of the coincidence in s15 is, for example,

Calculated by [Equation 1] takes the concept of the inner product of vectors, and obtains the correlation between counterfeit bill data and bill data stored in the first database 14 as a cosine value (cos Θ). The closer the value calculated in [Equation 1] is to 1, the greater the degree of coincidence. The user estimation device 1 extracts banknotes having a calculated degree of coincidence of, for example, 0.85 or more.

  After completing the processes of s15 to s18, the user estimation device 1 determines that the banknote extracted in s17 is likely to be a banknote discovered this time, and is found this time based on the banknote extracted in s17. The counterfeit ticket management data of the counterfeit ticket is generated (s19) and registered in the second database 15 (s20). The fake ticket management data generated in s19 includes the fake ticket number assigned to the fake ticket discovered this time, the discovery date and time, the denomination, and the customer number registered in the first database 14 for each bill extracted in s17. , The input date and time, the store number, and further the data associated with the degree of coincidence calculated in s15. In this way, the user estimation device 1 generates counterfeit ticket management data for the counterfeit ticket discovered this time and stores it in the second database 15. The second database 15 also stores fake ticket management data for fake tickets discovered in the past.

  Next, the user estimation device 1 calculates the appearance frequency in the fake note management data stored in the second database 15 for each user who has inserted the banknote extracted in s17 into the ATM 3 (s21). The calculation of the appearance frequency in s21 is performed for each user who has inserted the banknote extracted in s17 into the ATM 3, and the fake ticket management data stored in the second database 15 is stored in the second database 15. Calculate the total sum of coincidences. For example, as shown in FIG. 4, since the customer whose customer number is 12345678 is included in both the fake ticket numbers AAAA and BBBB, 0.87 + 0.88 = 1.75 is calculated as the appearance frequency. For the user included only in the fake ticket management data of the fake ticket discovered this time, the degree of coincidence calculated in s15 is calculated as the appearance frequency.

  Moreover, when calculating the appearance frequency, the appearance frequency for each user may be calculated from counterfeit ticket management data regarding a counterfeit ticket of the same denomination as the counterfeit ticket discovered this time.

  As described above, the user estimation device 1 calculates the possibility of being a user in all previously discovered counterfeits (including the counterfeits discovered this time) as the appearance frequency for the counterfeits discovered this time. ing. In general, since counterfeit bills are repeatedly used, the number of registrations in the second database 15 is large, and the appearance frequency calculated in s21 is large.

  The user estimation device 1 calculates the possibility of use of the counterfeit bill discovered this time for each user who has inserted the banknote extracted in s17 into the ATM 3 (s22). In s22, the usability is calculated based on the degree of coincidence calculated in s15 and the appearance frequency calculated in s21. Specifically, the product of the degree of coincidence calculated in s15 and the appearance frequency calculated in s21 is calculated as availability. And the user estimation apparatus 1 performs the estimation result output process which displays the estimation result which listed the user in order with the high usability calculated by s22 on the display part 17 (s23). FIG. 7A shows an example of a display screen on the display unit 17 in s23.

  As described above, the user estimation device 1 of this embodiment does not estimate the user who inserted the counterfeit note into the ATM 3 based only on the degree of coincidence with the bill data of the found counterfeit note, but falsely identifies the user in the past. Since the person who used the fake ticket discovered in consideration of the possibility of using the ticket is estimated, it is possible to estimate the user who is highly likely to use the fake ticket. Thereby, even if it does not apply the banknote processing apparatus which provided the high-resolution image sensor and the large-capacity image memory to ATM3, the estimation precision of the user who used the fake note can fully be secured, ATM3 Can be sufficiently prevented.

  In addition, the user list display based on the availability shown in FIG. 7A may be switched between the user list display based on the degree of coincidence calculated in s15 illustrated in FIG. 7B. Alternatively, as shown in FIG. 7C, a list of users based on availability and a list of users based on the degree of coincidence may be displayed side by side.

  Further, as other processing related to s5, for example, there is a list output of appearance frequency and a fake ticket data output for each user. For the appearance frequency list output, the appearance frequency is calculated for all the users whose customer numbers are registered in the second database 15, and a list in which the users are arranged in descending order of the appearance frequency calculated here is shown in FIG. This is a process of displaying on the display unit 17 as shown in FIG. Further, as shown in FIG. 8 (B), the user's counterfeit ticket data is output for each user whose customer number is registered in the second database 15 and registered for the user. This is a process for displaying data on the display unit 17.

  In the above embodiment, the ATM 3 is described as an example of a device incorporating the banknote processing device, but there are devices other than ATM 3 for depositing an available amount into an IC card used as a prepaid card.

It is a figure which shows the structure of the financial system to which the user estimation apparatus which is embodiment of this invention is applied, and the flow of a banknote. It is a figure which shows the structure of the user estimation apparatus which is embodiment of this invention. It is a figure which shows the banknote management data memorize | stored in the 1st database of the user estimation apparatus which is embodiment of this invention. It is a figure which shows the fake ticket management data memorize | stored in the 2nd database of the user estimation apparatus which is embodiment of this invention. It is a flowchart which shows operation | movement of the user estimation apparatus which is embodiment of this invention. It is a flowchart which shows the user estimation process in the user estimation apparatus which is embodiment of this invention. It is a figure which shows the example of a display screen of the user estimation result in the display part of the user estimation apparatus which is embodiment of this invention. It is a figure which shows the example of a display screen in the display part of the user estimation apparatus which is embodiment of this invention.

1-User estimation device 3-ATM
11-control unit 12-banknote data reading unit 13-communication unit 14-first database 15-second database 16-operation unit 17-display unit

Claims (4)

The banknote data is read for each inserted banknote, the degree to which the banknote is a genuine note is detected from the read banknote data, and if the detected degree is greater than the first threshold, it is identified as a genuine note. The banknote processing apparatus includes a banknote processing apparatus that accepts a banknote that has been identified as having a banknote processing apparatus that includes means for acquiring user data of a user who has inserted the banknote received by the banknote processing apparatus. Banknote management data acquisition means for acquiring banknote management data including the banknote data and the user data,
Banknote management data storage means for storing the banknote management data acquired by the banknote management data acquisition means;
When a bill that is later determined to be a fake note is found from the bills received by the bill processing apparatus, the bill data of the fake note is read, and the bill data of the fake note read here and the bill management A degree of coincidence calculating means for calculating the degree of coincidence with the banknote data of the banknote stored in the data storage means;
For each discovered counterfeit ticket, user data stored in the banknote management data storage means, and counterfeit ticket management data storage means for storing counterfeit ticket management data including the calculated degree of coincidence,
Appearance frequency calculating means for calculating the appearance frequency of the user for the fake ticket for each user using the fake ticket management data stored in the fake ticket management data storage means,
Utilizing the appearance frequency of each user calculated by the appearance frequency calculating means and the degree of coincidence calculated by the coincidence degree calculating means using the fake bill whose current coincidence has been calculated is inserted into the banknote processing apparatus A user estimation device comprising: estimation means for estimating a person.   The banknote management data storage means stores banknote management data for banknotes having a detected degree of authenticity greater than the first threshold value and smaller than a predetermined second threshold value, and greater than the second threshold value. The user estimation apparatus according to claim 1, wherein banknote management data is not stored for banknotes.   The fake ticket management data storage means is configured to calculate the user data stored in the banknote management data storage means for the banknote whose degree of coincidence calculated by the coincidence degree calculation means is greater than a predetermined value, and the calculated The user estimation device according to claim 1 or 2, which stores fake ticket management data including a degree of coincidence.   The estimation unit calculates the present coincidence for a user with a larger product of the appearance frequency for each user calculated by the appearance frequency calculation unit and the degree of coincidence calculated by the coincidence degree calculation unit. The user estimation device according to any one of claims 1 to 3, wherein the user is estimated to be highly likely to be a user who has entered the bill processing device.

Images