TWI463439B - Paper type processing device - Google Patents

Description

Paper sheet processing device

The present invention relates to a paper sheet in which paper sheets, checks, securities, and the like are discriminated by a feature amount such as an optical pattern or a magnetic pattern of a paper sheet, and a recording number indicated on the paper sheet. The processing device and the recognition algorithm update program for updating the recognition algorithm that recognizes the authenticity by the feature amount of the paper sheet.

In the past, there have been paper sheet processing apparatuses for processing paper sheets such as banknotes, checks, and securities. The paper sheet processing apparatus identifies each paper sheet to be a genuine paper coupon or a fake paper coupon (hereinafter referred to as a fake paper coupon). For example, an automatic teller machine (hereinafter referred to as ATM) installed in a store of a financial institution such as a bank detects the optical pattern or the magnetic pattern of the banknote as a feature amount for each banknote processed, and utilizes The detected feature amount is used to identify the true and false of the paper sheet.

Further, ATM is an identification by a recording number even if a banknote (pseudo paper coupon) which cannot be recognized as a fake paper coupon with respect to the feature amount of the optical pattern or the magnetic pattern detected from the banknote is used. , 俾 to identify as a fake paper coupon. Specifically, the ATM is a black paper for a fake paper ticket that memorizes the record number that has been registered for the fake paper ticket, and compares the record number marked on the banknote with the record number registered on the black paper ticket black list. To identify true and false.

Also, the police station and other relevant agencies will record the record number of the fake paper coupons found. The code is notified to the financial institution. The financial institution is a banknote (pseudo-paper coupon) in which the record number is notified, and at this point, when the feature quantity regarding the optical pattern or the magnetic pattern detected from the banknote is not recognized as a fake paper coupon, the The record number is registered on the black list of fake paper coupons.

Further, there is also a device for determining whether or not the banknote is stolen based on the record number marked on the banknote (refer to Patent Document 1); reading the two record numbers marked on the banknote, and determining the character string of the two record numbers, And a device for matching the record number (refer to Patent Document 2); whether the paper sheets of the same record number are present in the plurality of devices at the same time, and the paper ticket of the record number not registered in the black list of the fake paper ticket can be identified. The device (see Patent Document 3) and the like.

(previous technical literature) (Patent Literature)

Patent Document 1: Japanese Patent Laid-Open No. Hei 5-250546

Patent Document 2: Japanese Laid-Open Patent Publication No. 2010-117803

Patent Document 3: Japanese Patent Laid-Open Publication No. 2003-303367

However, as described above, the financial institution performs the pseudo paper coupon which cannot be recognized as a fake paper coupon by the feature amount regarding the optical pattern or the magnetic pattern detected from the banknote at the time of discovery. The record number is additionally registered in the blacklist of the fake paper coupon, and the use is related to the detection from the fake paper coupon. The feature amount of the optical pattern or the magnetic pattern can be appropriately recognized as the development of the recognition algorithm of the pseudo paper coupon. Next, the recognition algorithm to be installed in the ATM is updated and updated to the developed recognition algorithm. Therefore, in the ATM, the record number registered in the blacklist of the fake paper coupon is a recognition calculation that can be appropriately recognized as a fake paper coupon if the pseudo paper coupon that is updated to the record number uses the feature amount of the optical pattern or the magnetic pattern. Law, it becomes unnecessary.

The black list of the fake paper coupons is a record number that does not need to delete the registered fake paper coupons, and the record number only increases cumulatively. In other words, the black list of the fake paper coupons is also registered with a futile record number (which becomes an unnecessary record number), and thus the size thereof is vainly enlarged (to become fat).

Further, the ATM uses the record number of the read banknote to be compared with all the record numbers registered in the black list of the fake paper ticket, thereby performing the identification of the authenticity of the record number, and thus The number of record numbers registered on the black list of the fake paper coupons will increase, and the time required to identify the true and false will also increase. That is, if the ATM registers the unrecognized recording number in the pseudo paper coupon blacklist, the time required for the authenticity identification by the recording number increases, and there is a problem that the processing efficiency is lowered.

An object of the present invention is to provide a paper sheet processing apparatus and a recognition algorithm update program, which can ensure the recognition accuracy of the authenticity of the paper sheets, and can suppress the hypertrophy of the fake paper coupon list, and The processing efficiency of the recognition of the true or false number of the record number is reduced.

The paper sheet processing apparatus of the present invention solves the above problems, and in order to achieve the object, the following structure is obtained.

The feature amount detecting unit is for detecting the feature amount of the paper sheet. The feature amount detecting unit detects, for example, an optical pattern (reflected light pattern, transmitted light pattern) or a magnetic pattern of a sheet as a feature amount. The recognition unit recognizes the authenticity of the paper sheet by using the attached recognition algorithm using the feature amount detected by the feature amount detecting unit. The recognition unit calculates, for example, the degree of similarity between the feature amount detected from the paper sheet and the feature amount of the genuine paper ticket stored in advance, or calculates the feature amount detected from the paper sheet and memorizes it in advance. The degree of similarity of the feature amount of the pseudo paper coupon, and the authenticity of the paper sheet is identified based on the calculated similarities.

The record number identification unit is for identifying the record number marked on the paper sheet. The determination unit is a list of the fake paper coupons for determining whether or not the recording number recognized by the recording number identifying unit is registered in the recording number of the pseudo paper coupon registered in the storage unit.

Therefore, even if the paper sheet (pseudo paper coupon) that cannot be recognized as a fake paper coupon is recognized by the recognition algorithm installed from the feature amount of the paper sheet, the record number is a fake paper coupon registered in the fake paper coupon list. It can also be recognized as a fake paper coupon.

In addition, when the recognition algorithm is updated, the recognition algorithm update unit deletes the paper sheet (pseudo paper ticket) that can be recognized as a fake paper ticket by the recognition unit by using the updated recognition algorithm. The record number registered in the list of fake paper coupons. That is, when the feature quantity regarding the optical pattern or the magnetic pattern detected from the paper sheet or the like is used, the recognition and recognition of the true and false are updated. In the algorithm, the record number of the fake paper coupon that can be recognized as a fake paper coupon by the updated recognition algorithm is deleted from the fake paper coupon list. Therefore, it is possible to prevent the recording number that is unnecessary in the fake paper coupon list from continuing to be registered in a vain manner, to suppress the enlargement of the fake paper coupon list, and to reduce the processing efficiency for identifying the true and false by the recording number.

The recognition algorithm update unit acquires, for example, a deletion list of the record number deleted from the pseudo paper coupon list together with the updated recognition algorithm, and deletes the registration of the fake paper coupon list using the deletion list. The required record number is fine.

According to the present invention, it is possible to prevent the registration of the fake paper coupon list from being unnecessarily registered, and it is possible to suppress the enlargement of the fake paper coupon list and to suppress the processing efficiency regarding the identification of the authenticity of the recorded number. Reduced.

Hereinafter, embodiments of the present invention will be described.

Fig. 1 is a schematic diagram showing an automatic teller system of a financial institution such as a bank. The automatic teller system connects a plurality of cash dispensers 1 (hereinafter referred to as ATM1) to the host device 5 via the network 6. The ATM 1 is a store or a convenience store installed in a financial institution. The ATM 1 is equivalent to the paper sheet processing apparatus of the present invention. The main device 5 is installed at the center of the financial institution. Each ATM 1 is communicable with the host device 5 via the network 6.

ATM1 handles various transactions such as cash transactions and payment transactions in response to user input operations. The main device 5 is in accordance with the opening of gold Each account of the organization manages account information about the account. Account information is the address, name, communication number, password number, account balance, transaction history, etc. of the account holder. In addition, the main device 5 also performs a process of notifying the record number of the counterfeit banknote (hereinafter referred to as a "pick paper ticket") to each ATM 1 connected via the network 6.

Fig. 2 is a block diagram showing the configuration of the main part of the ATM. The ATM 1 includes a main control unit 10, a display/operation unit 11, a banknote processing unit 12, a coin processing unit 13, a card/detail processing unit 14, a passbook processing unit 15, and a biometric information reading unit. 16. The communication unit 17, the CD drive 18, and the memory unit 19. The main control unit 10 is an operation for controlling each part of the ATM1 body.

The display/operation unit 11 has a display provided on the front surface of the main body, a touch panel attached to the screen of the display, and the like. The display/operation unit 11 displays an operation guide screen for the user on the display. Further, the display/operation unit 11 detects an input operation of the user regarding the password number or the transaction content (transaction type, cashier amount, etc.) by detecting the pressing position of the touch panel. Further, the display/operation unit 11 also includes a display device or an input device that is operated by a carrier or the like during maintenance inspection.

The banknote processing unit 12 processes the cashier's banknotes. On the front side of the main body, a banknote deposit opening 12a for storing the banknotes is provided. At the note insertion opening 12a, a door is provided. Further, the banknote dispensing opening 12a is provided with a partition panel that distinguishes the incoming banknotes, the deposited banknotes, and the storage place where the banknotes are returned. The banknote processing unit 12 uses the opening and closing of the door to restrict the discharge of the banknote The banknote of the port 12a is put in or taken out. The detailed configuration of the banknote processing unit 12 is as follows.

The coin processing unit 13 transports coins along a coin transport path formed between a coin banknote opening provided on the front surface of the main body and a coin cassette housed inside the main body. Further, the coin processing unit 13 is a coin identifying unit that recognizes the type of money and the authenticity of each coin carried along the coin conveying path.

The card/detail processing unit 14 is a cash card (hereinafter simply referred to as a card) inserted into a card insertion slot provided on the front surface of the main body, and reads card information recorded on the magnetic strip of the card (financial institution) Number, store number, account number, etc.), or rewrite card information, etc. Further, when the inserted card is an IC card, the information recorded on the IC chip of the card is read or rewritten as necessary. On the IC chip, biometric information (registration information) of the card owner, that is, the user is recorded. Further, the card/detail processing unit 14 is a printing unit having a transaction content printed on the list. The card/detail processing unit 14 discharges the list in which the transaction content is printed to the list discharge port provided on the front side of the main body.

The passbook processing unit 15 is a printing unit that has a passbook inserted into a passbook insertion opening provided on the front surface of the main body, and a transaction history is printed on the passbook. Further, the passbook processing unit 15 is also provided with a page turning mechanism for turning the page of the deposited passbook or a bar code reader for reading the bar code indicating the page number printed on the passbook, and will be recorded. A magnetic head or the like that reads the passbook information (financial institution number, store number, account number, etc.) of the magnetic strip attached to the passbook.

The biological information reading unit 16 has a biological information reading sensor provided on the front surface of the main body. The biological information reading sensor is a finger vein pattern for reading a user. Here, as a biological information reading sensor, a sensor that reads a finger vein pattern is taken as an example, but as a sensing for reading other types of biological information such as palm vein, iris, and omentum. Can also be. The biological information reading unit 16 compares the biometric information (read information) of the user read by the biometric information reading sensor with the biometric information (registration information) of the IC chip recorded on the card. ), and its similarity to verify whether the user and the card owner are the same person.

The communication unit 17 controls communication with the host device 5 of the financial institution via the network 6. The CD drive 18 is a file for reading a program or material recorded on the placed CD. Further, for the CD to be placed, a file relating to a program or data stored in the body is written.

The memory unit 19 is a memory medium having a large capacity such as a hard disk, and stores various programs necessary for the operation control of the ATM 1 main body or a blacklist of recording numbers to be described below (corresponding to the pseudo paper coupon described in the present invention). List). Further, each part of the ATM1 main body is provided with a control unit having a CPU or a non-volatile memory, and is installed with a program necessary for the operation. The main control unit 10 is to install a program for each part of the ATM1 main body. The installer is memorized in the memory unit 19.

Hereinafter, the structure of the banknote processing unit 12 will be described. Fig. 3 is a schematic view showing the internal structure of the banknote processing unit. The banknote processing unit 12 stores the currency type cassettes 21 to 24 and the collection cassette 25 in the internal. The currency type cards 21 to 24 and the recovery cassette 25 are detachable from the main body. The currency type cards 21 and 22 are 10,000 banknotes. The currency type card 匣 23 is for storing one thousand banknotes. The currency type card 24 is for storing two thousand banknotes and 5,000 banknotes. The recycling cassette 25 is a banknote that is judged to be unusable in the transaction. The banknotes (two thousand banknotes and 5,000 banknotes) accommodated in the currency type card 24 and the banknotes accommodated in the collection cassette 25 are not usable as the banknotes. Therefore, it is also possible to have a structure in which the banknote dispensing mechanism is not provided for the currency type card 24 and the recycling cassette 25.

Further, the banknote processing unit 12 is a temporary storage unit 26 that temporarily stores and deposits a banknote or the like. Further, the banknote transport path 27 is provided with a banknote deposit port 12a, currency type cards 21 to 24, a recovery cassette 25, and a temporary storage unit 26. The banknote processing unit 12 is a banknote transport unit (not shown) that carries banknotes along the banknote transport path 27. In addition, the banknote conveyance unit controls a flapper that is provided at a branch point or the like of the banknote conveyance path 27, and controls the conveyance destination of the banknote being conveyed. Further, the banknote processing unit 12 is a banknote recognition unit 30 having a currency type for identifying banknotes transported along the banknote transport path 27, and authenticity. The banknote recognition unit 30 recognizes the authenticity of each banknote conveyed along the banknote conveyance path 27 using a feature amount such as an optical pattern or a magnetic pattern read from the banknote. Further, the banknote identification unit 30 recognizes the record number also marked on the banknote. The banknote identification unit 30 is provided with the feature amount detecting unit, the recognition unit, and the recording number identifying unit described in the present invention.

Fig. 4 is a block diagram showing the configuration of a main portion of the banknote recognition unit. The banknote recognition unit 30 includes a control unit 31, an optical sensor 32, a magnetic sensor 33, a recording number identification unit 34, an identification unit 35, a storage unit 36, and an input/output unit 37.

The optical sensor 32 acquires an image of the banknote while detecting the reflected light of the light irradiated on the banknote or transmitting the light to obtain an optical pattern of the banknote. The portrait is included in the field in which the record number is marked on the banknote, and not the entire banknote. The banknote recognition unit 30 may have a structure in which an optical pattern is acquired by the same optical sensor 32 and an image is acquired, or an optical sensor 32 that performs optical pattern acquisition and an image is separately provided. The configuration of the obtained optical sensor 32 is also possible.

The magnetic sensor 33 is a magnetic pattern obtained by taking a banknote made of magnetic ink used for banknotes.

The record number identifying unit 34 processes the portrait of the banknote (the original image of the field indicated by the record number) acquired by the optical sensor 32, and recognizes the record number indicated on the banknote.

The recognition unit 35 uses the optical pattern of the banknote obtained by the optical sensor 32 and the magnetic pattern of the banknote obtained by the magnetic sensor 33, and identifies the banknote with the installed recognition algorithm. True and false.

The memory unit 36 memorizes a parameter or the like for identifying the authenticity of the banknote. The parameter for identifying the authenticity of the banknote is the optical pattern of the genuine paper coupon, the feature amount of the magnetic pattern, the optical pattern of the fake paper coupon, and the feature amount of the magnetic pattern.

The input/output unit 37 notifies the banknote processing unit 12 of the recognition result of the recognition unit 35.

The banknote identification unit 30 recognizes the currency type and the true and false of the banknote by using the optical pattern read from the banknote and the magnetic pattern, but the main control unit 10 performs the identification of the true and false according to the record number. .

Figure 5 is a diagram showing a blacklist of record numbers. The blacklist of the record number is memorized in the memory unit 19. As shown in Fig. 5, the blacklist of the record number is registered on each of the fake paper coupons with the record number and currency type corresponding to the pseudo paper ticket. The currency type is a combination of the pass code (if Japanese yen: JPY) and the amount defined by ISO4217.

Hereinafter, the operation of the ATM 1 of this embodiment will be described.

Fig. 6 is a flow chart showing the addition processing of the pseudo paper coupon for the blacklist of the record number. Notify the financial institution of the record number and currency type of the fake paper coupons found by relevant agencies such as the police station. The financial institution is a method of recognizing the discovered pseudo paper coupons by all the ATMs 1 connected to the main apparatus 5, and the recording number and the currency type including the pseudo paper coupons are made from the main apparatus 5 via the network 6 to all the ATMs 1. All the requests for additional registration are required.

In the communication unit 17, when the communication unit 17 receives the additional registration request sent from the host device 5 (s1), the ATM 1 reads the record number of one of the fake paper coupons included in the additional registration request received this time. And the currency type (s2), and determine whether the fake paper coupon is not registered in the blacklist of the record number (s3). ATM1 is a pseudo paper coupon that is not registered in the blacklist of the record number, and the record number and currency type of the fake paper ticket read in s2 are registered in the blacklist of the record number (s4). ATM1, if it is not registered in the record For the fake paper coupons of the blacklist, the record number and currency type of the fake paper coupons read in s2 are not registered in the blacklist of the record numbers (the processing regarding s4 is not performed). The ATM 1 determines whether or not the above-described processing after s2 is performed (s5) for the recording number and currency type of all the fake paper coupons included in the additional registration request received this time. When the record number and the currency type of the fake paper ticket that are determined to be unprocessed in s5 are determined, the ATM 1 returns to s2, and the above-described processing is repeated. When the s5 determines that there is no record number and currency type of the unprocessed fake paper ticket, the ATM 1 ends the present process.

In addition, the additional registration request is for the ATM1, and the fake paper coupon has been sent for the record number and currency type, and may not be sent. That is, the relevant agencies such as the police station have recently been notified, and at this point in time, for the ATM1, only the pseudo paper coupons for which the recording number and the currency type are not transmitted can be transmitted, and the recording number and the currency type can be transmitted.

The various transaction processing for the deposit transaction, the payment transaction, and the like at the ATM 1 is well known, and the detailed description is omitted here, but the identification processing of the banknote is simply described.

Fig. 7 is a flow chart showing the recognition process. The recognition processing unit 12 reads the optical pattern and the magnetic pattern of the banknote in accordance with each banknote conveyed along the banknote conveyance path 27 in the banknote recognition unit 30 (s11). The banknote recognition unit 30 reads the optical pattern of the banknote with the optical sensor 32, and reads the magnetic pattern of the banknote with the magnetic sensor 33.

The banknote identification unit 30 recognizes the currency type of the banknote and the authenticity (s12) based on the installed recognition algorithm. In s12, the optical pattern and the magnetic pattern of the banknote read at s11 are respectively registered with beforehand The optical pattern of the genuine paper coupon is compared with the magnetic pattern, and the similarity of these is calculated. Further, when the similarity of the optical pattern exceeds the optical threshold value determined in advance, and the similarity of the magnetic pattern exceeds the magnetic threshold value determined in advance, it is recognized as a genuine paper coupon. Further, s12 is a comparison with an optical pattern and a magnetic pattern of a pseudo paper coupon registered in advance, and the processing is recognized as a genuine paper coupon as a condition that the similarity can be recognized as not being a fake paper coupon. can.

When the banknote recognition unit 30 recognizes that the feature amount of the banknote is a genuine paper ticket, the record number recognition unit 34 recognizes the image of the banknote read from the optical sensor 32, and is identified by The record number of the banknote (s13, s14). The banknote identification unit 30 notifies the main control unit 10 of the record number recognized at s14. The main control unit 10 determines whether or not the record number notified from the banknote recognition unit 30 is registered in the record number black list (s15), and if it is not registered, it is recognized as a genuine paper coupon (s16). Conversely, if registered, it is recognized as a fake paper coupon (s17).

Further, if it is recognized that the feature amount of the banknote is not a genuine paper coupon at s12, it is recognized as a fake paper coupon at s17 (s17).

In this way, the identification algorithm to be installed is a pseudo paper coupon that is not accurately and accurately recognized by the optical pattern and the magnetic pattern of the banknote, and is used for the fake paper coupon in which the recording number is registered in the blacklist of the recording number. The record number marked on the banknote can also be identified as a fake paper ticket.

Further, since the banknote recognized as the pseudo paper coupon by the collation of the optical pattern and the magnetic pattern is used, since the marked recording number is not recognized, the processing load of the recording number identifying unit 34 can be reduced.

Hereinafter, the operation at the time of recognizing the update of the algorithm will be described. Here, the case where the updated recognition algorithm is recorded on the CD will be described as an example. In the CD, in addition to the updated recognition algorithm, a deletion list indicated in Fig. 8 is recorded. The deletion list is a list in which the currency type and the record number of the fake paper coupon that can be recognized by the recognition algorithm are generated.

Fig. 9 is a flow chart showing the operation at the time of the recognition algorithm update. The program for updating the recognition algorithm is memorized in the memory unit 19. The main control unit 10 reads out the recognition algorithm recorded on the CD (s21) from the CD placed on the CD drive 18, and mounts the recognition algorithm (s22). When the installation of the recognition algorithm is completed, the main control unit 10 confirms whether or not the deletion list is recorded on the CD placed in the CD drive 18 (s23), and if the deletion list is not recorded, the processing ends.

The main control unit 10 reads the deletion list from the CD when the deletion list is recorded on the CD placed in the CD drive 18 (s24). The main control unit 10 reads out one record number and currency type registered in the read deletion list (s25), and determines whether or not the record number and the currency type are registered in the record number black list (s26). . When the matching person is registered in the blacklist of the record number, the main control unit 10 deletes the record number from the blacklist of the record number (s27). The main control unit 10 determines whether or not to execute the processing after s25 for all the record numbers registered in the deletion list (s28). When the main control unit 10 determines that there is an unprocessed record number in s28, the main control unit 10 returns to s25 and repeats the above processing. The main control unit 10 determines that there is no unprocessed record number at s28. Then, the process ends.

Therefore, the pseudo paper coupon for which the authenticity determination can be performed on the updated recognition algorithm is deleted from the blacklist of the record number. Therefore, it is possible to prevent the blacklist of the record number from vainly memorizing the record number of the fake paper ticket. That is, the blacklist of the record number can be prevented from being enlarged. Further, it is possible to suppress the waste of the time required for discriminating the true and false of the banknote by the recording number, and to suppress the decrease in the processing efficiency.

Further, the deletion list of the above example may be displayed as the tenth figure. The deletion list is a version corresponding to the recognition algorithm in addition to the currency type and record number of the fake paper ticket. The version of the recognition algorithm is a version of the pseudo paper coupon that can be identified.

Fig. 11 is a flow chart showing the operation at the time of updating the recognition algorithm in this example. The main control unit 10 determines the version of the installed recognition algorithm (s31) before updating the recognition algorithm. Thereafter, the main control unit 10 reads out the recognition algorithm recorded on the CD (s32) from the CD placed on the CD drive 18, and installs the recognition algorithm (s33). When the installation of the recognition algorithm is completed, the main control unit 10 confirms whether or not the deletion list is recorded on the CD placed in the CD drive 18 (s34), and if the deletion list is not recorded, the processing ends.

When the deletion list is recorded on the CD placed in the CD drive 18, the main control unit 10 reads the deletion list from the CD (s35). The main control unit 10 is registered in the record number of the deletion list read in s35, and the recognition algorithm that extracts all the versions before the update cannot be used as a fake paper coupon, and the recognition algorithm of the updated version can be used. As a pseudo The record number of the paper coupon and the currency type corresponding to the record number (s36).

In s36, the corresponding version X is a comprehensive extraction of the record number and currency type that satisfy the following conditions.

Condition: The version before the update <X≦ updated version

For example, if the version of the recognition algorithm before the update is "3" and the version of the updated recognition algorithm is "4", the record number and currency type corresponding to the version "4" are extracted. In addition, if the version of the recognition algorithm before the update is "2" and the version of the updated recognition algorithm is "4", the record number and currency type corresponding to the version "3" or "4" are extracted. . Here, the version of the recognition algorithm is expressed as an integer.

The main control unit 10 reads out one record number and currency type extracted in s36 (s37), and determines whether or not the record number and the currency type are registered in the record number black list (s38). When the matching person is registered in the blacklist of the recording number, the main control unit 10 deletes the recording number from the blacklist of the recording number (s39). The main control unit 10 determines whether or not to execute the processing after s37 for all the record numbers extracted in s36 (s40). When the main control unit 10 determines in S40 that there is an unprocessed record number, the main control unit 10 returns to s37 and repeats the above processing. When the main control unit 10 determines in S40 that there is no unprocessed record number, the main control unit 10 ends the present process.

According to this configuration, the same effects as those of the above-described example can be obtained, and the time required for the process of deleting the record number of the unnecessary paper ticket from the black list of the record number can be shortened.

Also, in the above example, the updated recognition algorithm and deletion will be performed. The list is recorded on the CD and placed in the CD drive 18 so as to be readable at the ATM 1. However, the main device 5 may be configured to be associated with each ATM 1 for the updated recognition algorithm and deletion list.

Further, the present invention is not limited to banknotes, and can be applied to a sheet processing apparatus that processes other types of paper sheets such as checks and securities.

1‧‧‧Automatic Teller Machine (ATM)

10‧‧‧Main Control Department

11‧‧‧Operation Department

12‧‧‧Banknote Processing Department

13‧‧‧Coin Processing Department

14‧‧‧ Schedule Processing Department

15‧‧‧ Passbook Processing Department

16‧‧‧Biological Information Reading Department

17‧‧‧Communication Department

18‧‧‧CD drive

19‧‧‧ Memory Department

30‧‧‧ banknote identification unit

31‧‧‧Control Department

32‧‧‧ Optical Sensor

33‧‧‧Magnetic sensor

34‧‧‧ Record Number Identification Department

35‧‧‧ Identification Department

36‧‧‧Memory Department

37‧‧‧Input and output

Fig. 1 is a schematic diagram showing an automatic teller system of a financial institution such as a bank.

Fig. 2 is a block diagram showing the configuration of the main part of the ATM.

Fig. 3 is a schematic view showing the internal structure of the banknote processing unit.

Fig. 4 is a block diagram showing the configuration of a main portion of the banknote recognition unit.

Figure 5 is a diagram showing a blacklist of record numbers.

Fig. 6 is a flow chart showing the addition processing of the pseudo paper coupon for the blacklist of the record number.

Fig. 7 is a flow chart showing the banknote recognition processing.

Figure 8 is a diagram showing the deletion of the list.

Fig. 9 is a flow chart showing the operation at the time of the recognition algorithm update.

Fig. 10 is a diagram showing a deletion list of ATMs in other examples.

Fig. 11 is a flow chart showing the operation at the time of updating the recognition algorithm of the ATM of another example.

Claims (4)

A paper sheet processing apparatus includes: a feature amount detecting unit; an identification unit; a recording number identifying unit; a memory unit; a determining unit; and a recognition algorithm updating unit; wherein the feature amount detecting unit is configured to detect The feature amount of the paper sheet is obtained by using the feature amount detected by the feature amount detecting unit, and identifying whether the paper sheet is a fake paper coupon by the attached recognition algorithm, and the recording number is recognized. The part is used to identify the record number marked on the paper sheet, and the memory part is a list of the fake paper coupons for storing the record number of the registered paper ticket, and the determination unit is used to identify the record number based on the record number Whether the record number recognized by the part is registered in the pseudo paper coupon list stored in the memory unit to determine whether it is a fake paper coupon, and the recognition algorithm update unit is configured to update the recognition algorithm, and the recognition algorithm The update unit is a record of the pseudo paper coupon that can be recognized as a fake paper coupon by deleting the recognition unit from the pseudo paper ticket list and updating the identification algorithm after updating the recognition algorithm. Number. The paper sheet processing apparatus according to claim 1, wherein the recognition algorithm update unit acquires the pseudo paper deleted from the pseudo paper coupon list together with the updated recognition algorithm. List of the deletion of the record number of the voucher. The paper sheet processing apparatus according to claim 2, wherein the deletion list is a recognition algorithm for making a recording number of a fake paper coupon and a pseudo paper coupon appropriately recognized as a fake paper coupon. Corresponding to the version, the recognition algorithm update unit extracts the record number of the fake paper coupon that the version of the recognition algorithm before the update is not recognized as a fake paper coupon from the deletion list, and the identification is updated. The version of the algorithm can be identified as the record number of the fake paper coupon of the fake paper coupon, and the record number of the fake paper coupon drawn here can be deleted from the above-mentioned fake paper coupon list. The paper sheet processing apparatus according to any one of the preceding claims, wherein the recording number identifying unit is a paper sheet that is recognized as a fake paper ticket by the identification unit, and does not A process of identifying the record number indicated by the sheet type is performed.