JP3856715B2 - Paper sheet identification device and automatic transaction device - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
  The present invention relates to a paper sheet identification apparatus that performs banknote identification (denomination discrimination, authenticity determination), and an automatic transaction apparatus including the paper sheet identification apparatus.
[0002]
[Prior art]
  2. Description of the Related Art Conventionally, there is a bill discriminating device that discriminates the denomination or authenticity of a bill depending on whether signals from various sensors are within the permissible range of a discrimination standard.
[0003]
  In such a bill recognition device, when a bill is conveyed into the bill recognition device, is the image read by the image sensor (two-dimensional sensor) similar to a template image stored as a reference or more than a certain value? Identification is performed on the basis of whether a waveform of a sensor (one-dimensional sensor) that detects a feature of a bill such as a magnetic sensor or an infrared sensor is similar to a template waveform stored as a reference by a certain value or more.
[0004]
[Problems to be solved by the invention]
  However, the paper and ink used for the banknotes may differ depending on the fiscal year or lot in which they are issued. As a result, the transmittance of the emitted infrared light will be different even for the same type of banknotes, and the output of the magnetic sensor will change due to changes in the output of the infrared sensor and the characteristics of the magnetic ink depending on the year and lot. May change. For the purpose of preventing forgery or the like, this is often performed for features that cannot be visually confirmed, such as magnetic ink and infrared ink, and new features may be added. Therefore, the banknote after such a change in characteristics may have a lower recognition rate than a banknote before the change in characteristics.
[0005]
  In consideration of such circumstances, the conventional banknote recognition apparatus as a method for solving the above-described problem has means for widening the allowable range that is permitted as a genuine note, particularly in authenticity determination. There was a problem that performance deteriorated.
[0006]
  As described above, the present invention determines a true bill as a true bill regardless of the change even if it is a true bill whose characteristics have changed according to the issue year or lot, and reduces the identification performance (determination performance) of the counterfeit bill. The purpose of this paper is to propose a paper sheet identification device that prevents the occurrence of unforeseen disadvantages caused by counterfeit bills, etc., while maintaining the convenience of the user, and to contribute to the maintenance of healthy transactions in the economy and society.
[0007]
[Means for Solving the Problems]
  This inventionRead multiple optical features of paperA control unit that includes a reading unit, and that identifies a paper sheet from a reading result read by the reading unit and a predetermined identification standard, and that reads a management code recorded on the paper sheet A reading unit; and an identification standard changing unit that changes the identification standard based on the control symbol.Based on the control symbol, only the reading result readable by the plurality of reading means is identified by matching with the predetermined identification standard.It is a paper sheet identification device.
[0008]
  The reading means can be composed of a one-dimensional sensor such as a magnetic sensor or an infrared sensor, or a two-dimensional sensor such as an image reading device for reading an image. A combination of different types of sensors or a plurality of the same type of sensors can be used. Or using multiple sensors of the same type and combining different types of sensors. The reading means may be a reading means for determining authenticity, a reading means for determining type, or a reading means for determining authenticity and type.
[0009]
  The paper sheets can be composed of printed matter including value media such as banknotes, checks, and securities.
  The management symbol is used to individually manage paper sheets such as serial numbers, signs, year names, serial numbers, codes, codes, or signs recorded on paper sheets by a method such as printing, or for each denomination Includes composing with symbols used to manage.
[0010]
  The discriminating criterion is paper such as permissible width information indicating a permissible width that the identification result is true if it is within a predetermined range with respect to the reference value, a template representing the characteristics of the paper sheet, or a combination of these pieces of information. It can be configured with criteria for determining and identifying the denomination and authenticity of the leaf.
[0011]
  The identification reference changing means includes an allowable width changing means for changing the allowable width information based on the management symbol, a template selecting means for selecting the template to be used based on the management symbol, and the sensor based on the management symbol. It includes using sensor selection means for selecting designation information, or a combination of these means.
[0012]
  With the above-described configuration, it is possible to perform appropriate identification based on the management symbol recorded on the paper sheet, and it is possible to prevent deterioration of the discrimination accuracy particularly in authenticity determination. Thereby, when using a banknote as paper sheets, for example, the inconvenience for the user who judges that the banknote in which characteristics changed, such as an old true bill, is not a true bill can be eliminated.
[0013]
  Further, according to the above configuration, the reading means which is not necessary depending on the control symbol, that is, for example, certain When the paper of the control symbol is different in ink and a specific reading means (for example, magnetic sensor A) is not useful for identification, the reading means (that is, magnetic sensor A) is not used, so The time required for matching detection information (that is, a detected waveform or image) and reference data (for example, a waveform template or an image template) can be reduced, and the identification process can be speeded up.
[0014]
  As a preferred embodiment, as the identification criterion, tolerance width information indicating an identification tolerance range that is identified as being true with respect to a reference value is used, and the identification criterion changing unit is configured to perform the tolerance based on the management symbol. Allowable width changing means for changing the width information can be provided.
[0015]
  The allowable width information includes information indicating a width of a detection level by the reading unit that allows a difference from a reference level (reference value), and an error from a reference shape (reference value) of a waveform detected by the reading unit. It can be composed of information indicating the allowable width, or can be a composite of these pieces of information.
[0016]
  The allowable width changing means includes a control device that performs identification based on the changed allowable width information.
[0017]
  With the above configuration, for example, paper sheets produced in the old age are distributed for a long period of time based on the record management symbol on the paper sheets, and are characterized by causes such as wrinkles, kinks, dirt, ink wear, etc. Control code, such as setting a wide tolerance range for authenticity judgment because it is likely to have changed, or setting a narrow tolerance range for authenticity judgment of control symbols with many counterfeit paper sheets (for example, counterfeit bills) Considering the circumstances based on the above, it is possible to perform identification with higher accuracy.
[0018]
  Further, as a preferred embodiment, as the identification standard, a plurality of types of templates representing the characteristics of paper sheets are used, and the identification standard changing unit includes a template selection unit that selects the template to be used based on the management symbol. can do.
[0019]
  The template is a waveform template serving as a reference for identifying a waveform detected by the sensor, an image template serving as a reference for identifying an image detected by the image reading device, or a combination of the waveform template and the image template. Including waveform sample data, image sample data, or a set of these data, which serve as identification criteria for identifying a class.
[0020]
  The template selection means can be configured by a control device that selects the template to be used and executes identification.
[0021]
  With the above-described configuration, identification can be performed using an appropriate template that matches the characteristics of the paper sheet for each management symbol based on the management symbol, and in particular, it is possible to prevent a decrease in accuracy of true / false determination..
[0022]
  The present invention is also an automatic transaction apparatus comprising a display means for displaying information, an input means for inputting an operation, a paper sheet input / output means for inputting / extracting paper sheets, and a control means for controlling the means. Thus, an automatic transaction apparatus including the paper sheet identification apparatus can be provided.
[0023]
  The automatic transaction apparatus includes an apparatus that automatically executes transactions such as an ATM (automatic deposit payment apparatus), a ticket vending machine, or a currency exchange machine.
[0024]
  If the automatic transaction apparatus is an ATM,
The display means is constituted by a display using a cathode ray tube or a liquid crystal,
The paper sheet means is composed of an electrostatic or pressure sensitive touch panel for inputting coordinates touched with a finger or a touch pen,
The paper sheet loading / unloading means comprises a paper sheet loading / unloading apparatus that performs a paper sheet loading process for loading a paper sheet and a banknote ejection process for discharging the paper sheet,
The control means includes a control device comprising a CPU, a ROM, and a RAM.
[0025]
  If the automatic transaction apparatus is a ticket machine,
The display means comprises a display using a cathode ray tube or a liquid crystal, or a display lamp for displaying the amount of the ticket with a lamp,
The input means comprises an electrostatic or pressure sensitive touch panel for inputting coordinates touched with a finger or a touch pen, an amount input device for pressing an amount button corresponding to the amount of a ticket, or a combination thereof.
The paper sheet loading / unloading means comprises a paper sheet loading / unloading device that performs a paper sheet loading process for loading a paper sheet and a paper sheet discharging process for discharging the paper sheet,
The control unit includes a control device including a CPU, a ROM, and a RAM. The paper sheet loading / unloading means includes forming the paper sheet feeding part and the paper sheet discharging part integrally or separately.
  If the automatic transaction apparatus is a money changer,
The display means comprises a display using a cathode ray tube or a liquid crystal display, or a guidance display unit displaying by printing that a 10,000 yen bill is exchanged for a thousand yen bill,
The input means comprises an electrostatic or pressure sensitive touch panel for inputting coordinates touched with a finger or a touch pen, or a button type input device for pressing whether to execute or cancel the exchange,
The paper sheet loading / unloading means comprises a paper sheet loading / unloading device that performs a paper sheet loading process for loading a paper sheet and a paper sheet discharging process for discharging the paper sheet,
The control unit includes a control device including a CPU, a ROM, and a RAM. The paper sheet loading / unloading means includes forming the paper sheet feeding part and the paper sheet discharging part integrally or separately.
  By the said structure, the identification rate of the paper sheets in ATM, a ticket machine, or a change machine improves by the said banknote insertion / extraction means, and can use the said banknote insertion / extraction means actually.
[0026]
  In the above solution, “identification” may be read as “authentication determination”. That is,
"Paper sheet identification device for identifying paper sheets", "Paper sheet authenticity determination device for determining authenticity of paper sheets",
The “identification sensor” may be a “authentication sensor”.
  Thereby, it can be set as the paper sheet identification device specialized in authenticity determination.
[0027]
【The invention's effect】
  According to the present invention, it is possible to prevent a decrease in paper sheet identification accuracy, particularly authenticity determination accuracy. In addition, it is possible to prevent a decrease in the accuracy of the authenticity determination even for paper sheets with many forgery cases such as US dollar bills, and to reduce the tolerance for management symbols with many forged bills. This improves the detection rate of counterfeit bills.
  Further, reading means that is not required depending on the control symbol, that is, for example, when the paper of a certain control symbol is different in ink and the specific reading means (for example, magnetic sensor A) is not useful for identification, the reading means. By not using the magnetic sensor A (that is, the magnetic sensor A), the time required for matching the detection information (that is, the detected waveform, image, etc.) by the reading means with the reference data (for example, the waveform template, the image template, etc.) is shortened. Processing speed can be increased.
[0028]
DETAILED DESCRIPTION OF THE INVENTION
  An embodiment of the present invention will be described below with reference to the drawings.
First, the external appearance of the money changer 1 will be described together with a perspective view of the money changer 1 equipped with the banknote recognition device 11 according to an embodiment of the present invention shown in FIG.
  The money changer 1 is installed in various stores such as banks, and has a coin outlet 2, a wrapping coin outlet 3, a tray 4, a speaker 5, a card insertion slot 6, a statement slip outlet 7, a banknote inlet / outlet 8 at the upper front. And a touch monitor 9 (display for both touch input) and the like, and a bill recognition device 11 is provided inside the bill entry / exit 8.
[0029]
  Moreover, the inside of the money-changing machine 1 is equipped with the banknote processing apparatus (illustration omitted) which sorts and stores a banknote for every money type, and performs storage and delivery of a banknote.
  With the above configuration, the money changer 1 executes banknote identification (denomination discrimination and authenticity determination) of banknotes by the banknote identification device 11 to enable banknote exchange transactions.
[0030]
  Next, the structure of the banknote handling apparatus will be described together with the control circuit block diagram of the banknote handling apparatus shown in FIG.
  The CPU 15 controls each circuit device in accordance with a program stored in the ROM 16 and stores the control data in the RAM 17 so as to be readable. The RAM 17 also stores, as operational data, a deposit transaction operational state that has been deposited at the time of exchange.
[0031]
  The CPU 15 controls the banknote transport unit 10, the banknote recognition device 11, the operation unit 12, and the display unit 13 via the I / O interface 14.
[0032]
  The banknote transport unit 10 drives the transport path in accordance with a control signal from the CPU 15 and transports the banknotes to the banknote recognition device 11.
[0033]
  The banknote identification device 11 is a device that executes authenticity determination after discriminating the denomination of the banknote, and details thereof will be described later.
[0034]
  The operation unit 12 allows the customer to input transaction data at the time of exchange from the touch monitor 9 (FIG. 1), accepts the customer's transaction use card from the card insertion slot 6 (FIG. 1), and further converts the banknote to be exchanged into the banknote. Accept from the entrance 8 (FIG. 1).
[0035]
  The display unit 13 displays and guides input items and operation guides on the touch monitor 9 (FIG. 1), an attached display (not shown), and an attendant panel (not shown).
[0036]
  The clerk panel is provided with input permission only for the clerk, and allows the clerk panel to input arbitrary setting data for setting the operation mode.
  With the above structure, the money changer 1 executes a currency exchange process for exchanging the banknotes received from the banknote entry / exit 8 based on the transaction data input by the touch monitor 9.
[0037]
  Next, the structure of the banknote identification device 11 will be described together with a circuit block diagram of the banknote identification device 11 shown in FIG.
  The banknote recognition device 11 includes a CPU 21, a ROM 22, a RAM 23, a communication device 24, and an I / O driver 25, and various sensors (26 to 29) via the I / O driver 25.
[0038]
  CPU21 selects the optimal image template and waveform template based on the serial number printed on the banknote, and performs the process matched with the detection data (after-mentioned) stored in RAM23. As a result of the matching, if the degree of similarity is within the appropriate range, it is determined as a genuine note, and if it is out of the range, it is determined as a false note, and the determination result is transmitted to the CPU 15 (FIG. 2) of the banknote processing device via the communication device 24. .
[0039]
  The ROM 22 stores a program for operating the banknote recognition apparatus 11, denomination discrimination image data, and template data (image template data and waveform template data). The template data is template data including an appropriate range, and is determined to be a genuine note if it is within the appropriate range (within an allowable range) of the template data, and a false note if it is outside the appropriate range.
[0040]
  In addition, use sensor data is also stored as data for selecting detection data to be used among detection data detected by the infrared sensors 29f to 29j in the authenticity determination process described later.
[0041]
  The RAM 23 temporarily stores data detected by the various sensors (27 to 29). Here, the detected data refers to detection of serial numbers, detected image data, detected waveform data, and the like detected by various sensors described later, that is, serial number reading sensor 27, image sensor 28, and unique feature sensor 29 (29e to 29j). Refers to data.
[0042]
  The RAM 23 stores template data as necessary (ie, when template data needs to be added).
[0043]
  The communication device 24 executes communication with the banknote processing device (described above) which is a host device, and transmits the above-described determination result to the banknote processing device.
[0044]
  The I / O driver 25 is a driver for connecting various sensors (26 to 29), and relays transmission and reception of signals between the various sensors and the CPU 21.
[0045]
  The trigger sensor 26 detects that a banknote has been conveyed to the banknote recognition device 11 and transmits a detection signal to the CPU 21. In response to the detection signal, the CPU 21 starts identification processing.
[0046]
  The serial number reading sensor 27 is a sensor that reads a serial number printed on a banknote, and includes a scanner and an OCR (Optical Character Reader) program. The read serial number is transmitted to the CPU 21 as serial number data.
[0047]
  The image sensor 28 reads an image of a banknote and transmits the image data as detection data to the CPU 21. This makes it possible to discriminate denominations of banknotes.
[0048]
  The unique feature sensor 29 (29e to 29j) is a sensor group that detects (or reads) a unique feature of the bill, and includes an infrared light image sensor 29e and five infrared sensors 29f to 29j. The detected image data detected by the infrared light image sensor 29e and the detected waveform data detected by the infrared sensors 29f to 29j are transmitted to the CPU 21 as detection data. Note that the configuration of the unique feature sensor 29 is not limited to this, and the type and quantity of sensors can be configured in various patterns, such as adding a magnetic sensor.
[0049]
  With the above structure, the image data detected by the image sensor 28 and the image data for denomination discrimination are matched to perform denomination discrimination, and the detection data from the unique feature sensor 29 and the template corresponding to the serial number are compared for accuracy. Can be executed.
[0050]
  Next, the relationship and configuration of banknotes, infrared sensors 29f to 29j, image template data, and waveform template data will be described with reference to the template explanatory diagrams shown in FIGS.
[0051]
  The banknote 40 is configured by printing a money amount 41, a serial number 42, a pattern frame 43, a pattern 44, and a portrait 45 as shown in FIG.
[0052]
  When the banknote 40 is read by the infrared light image sensor 29e, for example, if the serial number 42 is a DDDDDDD banknote 40, only the amount 41 and the portrait 45 can be read as shown in FIG. The other serial number 42, the pattern frame 43, and the pattern 44 cannot be read because they are printed with ink that cannot be read with infrared light. Therefore, as image template data for identifying the banknote 40 whose serial number 42 is DDDDDDD, image data displaying only the amount 41 and the portrait 45 is selected as shown in FIG.
[0053]
  Moreover, as shown to the top view of (C) in a figure, the infrared sensor 29f-29j is provided in the position corresponding to each component of the banknote 40. FIG. That is, the infrared sensor 29f corresponds to the amount 41, the infrared sensors 29g and 29i correspond to the pattern frame 43, the infrared sensor 29h corresponds to the pattern 44, and the infrared sensor 29j corresponds to the person image 45. Yes.
[0054]
  In addition, since the banknote 40 passes above each infrared sensor 29f-29j toward the back side from the front side, the reading range of each infrared sensor 29f-29j becomes a line of the front-back direction (vertical direction of a figure). .
[0055]
  For this reason, when the image template data shown in (B) is used in the authenticity determination process described later, only the detected waveform data by the infrared sensor 29f and the infrared sensor 29j is valid, and the use sensor data in this case is The information uses only the detected waveform data by the infrared sensor 29f and the infrared sensor 29j.
[0056]
  When the banknote 40 having the serial number 42 in which all of the amount 41, the pattern frame 43, the pattern 44, and the figure 45 described above react to infrared rays is read by the infrared sensors 29f to 29j, the waveform of the detection data is as follows. become that way.
[0057]
  That is, the waveform detected by the infrared sensor 29f becomes a waveform 51f indicated by a phantom line in FIG. 5D, and the waveform template data for determining as a genuine note has a predetermined appropriate range ( It is formed by waveform template data 52f in which an allowable width is set.
[0058]
  Similarly, from the infrared sensor 29g to the waveform template data 52g shown in (E), from the infrared sensor 29h to the waveform template data 52h shown in (F), from the infrared sensor 29i to the waveform template data 52i shown in (G), the infrared sensor The waveform template data 52j shown in FIG.
[0059]
  In addition, the waveform 51 (51f-51j) shown to (H) from (D) has shown the average waveform at the time of reading the banknote 40, and when forming the waveform template data 52 based on this, If the variation of the genuine note waveform tends to protrude upward on the graph, an appropriate appropriate range is provided for the infrared sensor, such as widening the upper allowable width as shown in (D).
[0060]
  As described above, the image template data and the waveform template data 52 (52f to 52j) are created and stored in the ROM 22 described above.
[0061]
  Next, the operation of the banknote recognition apparatus 11 will be described together with the processing flowchart of the identification processing shown in FIG. 6 and the processing flowchart of the authenticity determination processing shown in FIG.
  The CPU 21 waits until the trigger sensor 26 is turned on. When the bill 40 is conveyed and the trigger sensor 26 is turned on, the CPU 21 receives the detection signal and starts the identification process (step n1).
[0062]
  The CPU 21 causes the serial number reading sensor 27 to read the serial number 42 of the banknote 40, temporarily stores the serial number 42 in the RAM 23 (step n2), reads the image of the banknote 40 with the image sensor 28, and reads the read image data. The detected image data is stored in the RAM 23 (step n3).
[0063]
  The CPU 21 compares the detected image data with the image data for denomination discrimination stored in the ROM 22 to determine whether it is a one dollar bill (step n4), a five dollar bill (step n5), a ten dollar bill (step n6), Whether the bill is a 20 dollar bill (step n7) or the other is determined, and true / false determination processing (steps n8 to n11) with the determined denomination is executed.
[0064]
  If it is not any denomination, error processing is performed, and an error signal is transmitted to the banknote processing apparatus which is the host apparatus (step n12).
[0065]
  The CPU 21 determines whether or not the matching results obtained by the authenticity determination process are all determined to be within the appropriate range (step n13), and all of the matching results are within the appropriate range with respect to the banknote processing device that is the host device. If it is out of the proper range, a determination result that it is a “fake ticket” is transmitted (step n15).
[0066]
  The authenticity determination process (steps n8 to n11) will be described in detail. As shown in FIG. 7, the CPU 21 first detects the detected image data and the detection by the unique feature sensor 29 (infrared light image sensor 29e, infrared sensors 29f to 29j). Detection data such as waveform data is acquired and stored in the RAM 23 (step p1).
[0067]
  Next, the serial number 42 stored in the RAM 23 is read out, whether AAAAAA ≦ serial number 42 <BBBBBBBB (step p2), BBBBBBB ≦ serial number 42 <CCCCCC (step p3), CCCCCC ≦ serial number 42 <DDDDDD Or not (step p4).
[0068]
  If AAAAAAA ≦ number 42 <BBBBBBB, as shown in (I) of the image comparison explanatory diagram of FIG. 8, the CPU 21 detects the image data detected by the infrared image sensor 29e and the image template data A (stored in the ROM 22). AAAAAAA ≦ the serial number 42 <BBBBBBB image template data) is compared, whether or not the difference is within a predetermined range is matched, and the result is stored in the RAM 23 (step p5).
[0069]
  Further, as shown in (M) of the sensor usage explanatory diagram of FIG. 9, the waveform data read by the infrared sensor 29 f is used by using the waveform data read by all the infrared sensors 29 f to 29 j based on the usage sensor data. It is matched whether or not it is within the range of the template data 52f ((D) in FIG. 5). Similarly, the waveform data read by the infrared sensor 29g is waveform waveform data 52g and the waveform data read by the infrared sensor 29h is a waveform. The template data 52h, the waveform data read by the infrared sensor 29i are matched with the waveform template data 52i, and the waveform data read by the infrared sensor 29j are matched with the waveform template data 52j, respectively, and whether or not all the matching results are within the range. Is stored in the RAM 23 (step p6).
[0070]
  Similarly, if BBBBBBB ≦ number 42 <CCCCCC in step p3, as shown in FIG. 8J, the CPU 21 detects image data detected by the infrared image sensor 29e and image template data stored in the ROM 22. B is compared, whether or not the difference is within a predetermined range is matched, and the result is stored in the RAM 23 (step p7).
[0071]
  In this case, as shown in FIG. 9 (N), only the infrared sensors 29g, 29i, and 29j are used, and the waveform data read by the infrared sensors 29g, 29i, and 29j are waveform template data 52g, 52i, and 52j ( Whether or not they are within the ranges of (E), (G) and (H) in FIG. 5 is matched, and whether or not all the matching results are within the range is stored in the RAM 23 (step p8).
[0072]
  Similarly, if CCCCCC ≦ the serial number 42 <DDDDDDD in step p4, image template data C ((K) in FIG. 8) is used (step p9) and matching is performed using infrared sensors 29f to 29i. Processing is performed (step p10).
[0073]
  Similarly, if none of the conditions in steps p2 to p4 match, image template data D ((L) in FIG. 8) is used (step p11), and matching processing is performed using infrared sensors 29f and 29j. (Step p12).
  With the above operation, the CPU 21 can perform true / false determination with high accuracy using an image template matched to the serial number 42 for each denomination, and only uses necessary waveform template data for matching based on the waveform template data. Thus, matching can be performed, and the time required for matching can be shortened to make high-speed determination.
[0074]
  In steps p6, p8, p10, and p12 described in conjunction with FIG. 7, matching is performed using waveform data detected by all infrared sensors 29f to 29j and corresponding waveform template data at any step. May be.
[0075]
  In this case, the waveform template data when there is a response and the waveform template data when there is no response may be stored in the ROM 22 and the waveform template data to be used may be selected based on the sensor data used.
[0076]
  Thereby, although the time required for the matching process is extended, it is possible to detect a case where there is a response to an infrared sensor which should not have responded originally, and it is possible to improve the authenticity determination capability.
[0077]
  Next, the second embodiment will be described with the waveform template data explanatory diagram shown in FIG. 10 and the processing flow diagram shown in FIG.
  In this embodiment, as shown in FIG. 10, waveform template data having different allowable widths is used based on a standard combined waveform 51x which is a standard form of a combined waveform obtained by combining all the waveforms detected by the infrared sensors 29f to 29j.
[0078]
  That is, for the banknote 40 with serial number 42 that is frequently counterfeited, the waveform template data 52x having a narrow allowable width is set as shown in (Q), and the banknote with serial number 42 with a large variation in detection results such as the old banknote 40 is set. 40, waveform template data 52y having a wide tolerance range is set as shown in (R), and the bill 40 having the serial number 42, which is a variation in the average state, is normally set as shown in (S). Is set and stored in the ROM 22.
[0079]
  Since other configurations and structures are the same as those in the first embodiment described above, detailed description thereof is omitted.
  Also in this embodiment, the identification process is executed as described with reference to FIG. 6, but the processing contents of the authenticity determination process (steps n8 to n11) are set as follows.
[0080]
  FIG. 11 shows a process flow diagram of the authenticity determination process in this embodiment. The CPU 21 first detects detected image data, detected waveform data, etc. by the unique feature sensor 29 (infrared light image sensor 29e, infrared sensors 29f to 29j). Is detected and stored in the RAM 23 (step q1).
[0081]
  CPU21 reads serial number 42 memorize | stored in RAM23, is it a serial number 42 with many forged banknotes (step q2), or is it a serial number 42 with many variations in detection results, such as an old banknote 40 (step q3)? Determine if it is not.
[0082]
  If the serial number has a large number of counterfeit bills, the image data detected by the infrared light image sensor 29e and the image template data A (FIG. 8) are matched and the determination result is stored in the RAM 23 (step q4). The detected waveform data 29f to 29j and the waveform template data 52x are matched and the determination result is stored in the RAM 23 (step q5).
[0083]
  Similarly, in the case of serial number 42 with many variations, matching is performed using image template data A (step q6), and matching is performed using waveform template data 52y (step q7).
[0084]
  Similarly, in the case of the serial number 42 indicating the other normal banknote 40, matching is performed using the image template data A (step q8), and matching is performed using the waveform template data 52z (step q9).
[0085]
  With the above operation, the strictness of the authenticity determination can be changed corresponding to the state of the banknote 40 based on the serial number 42 (whether there are many counterfeits or variations, etc.), and a highly accurate authenticity determination can be performed. It can be carried out. That is, when the counterfeit banknote is matched, as shown in FIG. 10 (T), the detected waveform 53 of the counterfeit banknote protrudes from the waveform template data 52x as a protruding portion 53a as a protruding portion 53a. If it is an old banknote (a banknote with a large variation), as shown in (U), the protrusion 53b (protrusion due to a dirty part or the like) of the detected waveform 53 is within the appropriate range of the waveform template data 52y. And can be determined to be a “genuine ticket”.
[0086]
  In this embodiment, the image templates used in steps q4, q6, and q8 are all template A. However, as in the first embodiment, the image template may be selected based on the serial number. good. As a result, it is possible to perform true / false determination with high accuracy at high speed.
[0087]
  In each of the above embodiments, when authenticating a bill printed with a signature without a serial number 42, a sign reading sensor is provided in place of the serial number reading sensor 27, and depending on the type of the read signature, Steps p2, p3, and p4 described with reference to FIG. 7 or steps q2 and q3 described with reference to FIG. 11 may be performed, or a bill having both a serial number and a signature may be used. For example, a sign reading sensor may be provided in addition to the serial number reading sensor 27, and the case may be classified according to the read serial number and the type of the signature.
[0088]
  The serial number reading sensor 27 may be configured to read not only the number but also the color of the serial number 42. Thereby, the serial number based on the serial number is attached, and when the upper limit of the number is reached, the printing color of the serial number 42 is changed, and the banknote 40 that has been assigned the serial number from the beginning again is added to the banknote 40. On the other hand, appropriate image template data and waveform template data can be created and used without being affected by overlapping numbers.
[0089]
  In the first embodiment, the appropriate range may be changed for each of the infrared sensors 29f to 29j when matching is performed using waveform template data as in the second embodiment.
[0090]
  In this case, error tolerance range data (for example, an error of +3 to -3 points with respect to the level of the waveform template data is allowed for each of the classification numbers 42 (AAAAAAA to BBBBBBB, BBBBBBB to CCCCCC, CCCCCC to DDDDD). Data) is set and stored in the ROM 22, and the error range allowed by the error tolerance range data corresponding to the section with respect to the standard waveforms 51f to 51j (FIG. 5) serving as the reference of the waveform template data 52f to 52j. What is necessary is just to comprise so that it may match.
[0091]
  Alternatively, waveform template data having a different appropriate range may be set for each section and stored in the ROM 22, and matching may be performed by this. You may match by.
[0092]
  Further, instead of determining whether or not each matching is within an appropriate range, each matching determines the similarity with the template data, and step n13 (depending on whether or not the similarity in all matchings is within a predetermined range. The “true ticket” and “fake ticket” determination of FIG. 6) may be performed.
[0093]
  In correspondence between the configuration of the present invention and the above-described embodiment,
The automatic transaction apparatus of the present invention corresponds to the money change machine 1 of the embodiment,
Similarly,
The paper sheet loading / unloading means corresponds to the banknote loading / unloading port 8,
The display means and input means correspond to the touch monitor 9,
The paper sheet identification device corresponds to the banknote identification device 11,
The control means corresponds to the CPU 15,
Identification standard changing means, allowable width changing means,as well asTemplate selection meansIsCorresponding to CPU21,
The management symbol reading means corresponds to the serial number reading sensor 27,
The reading means corresponds to the unique feature sensor 29 (infrared light image sensor 29e, infrared sensors 29f to 29j),
Paper sheets correspond to the banknote 40,
The control symbol corresponds to the serial number 42,
The identification criteria correspond to the waveform template data 52f to 52j, 52x to 52z, the image template data A to D, the sensor data used, and the error tolerance data,
The template corresponds to the waveform template data 52f to 52j, 52x to 52z, and the image template data A to D,
Permissible width information corresponds to waveform template data 52x-52z and error tolerance dataYes,
The present invention is not limited only to the configuration of the above-described embodiment, and many embodiments can be obtained.
[Brief description of the drawings]
FIG. 1 is a perspective view showing an appearance of a money changer.
FIG. 2 is a circuit block diagram of the banknote handling apparatus.
FIG. 3 is a circuit block diagram of a true / false determining apparatus.
FIG. 4 is an explanatory diagram of a template.
FIG. 5 is an explanatory diagram of a template.
FIG. 6 is a processing flowchart of identification processing.
FIG. 7 is a processing flow diagram of authenticity determination processing.
FIG. 8 is an image comparison explanatory diagram.
FIG. 9 is an explanatory diagram of a sensor used.
FIG. 10 is an explanatory diagram of waveform template data according to the second embodiment.
FIG. 11 is a processing flowchart of authenticity determination processing in the second embodiment.
[Explanation of symbols]
    1 ... Currency exchange machine
    8. Banknote entry / exit
    9 ... Touch monitor
    11 ... Bill discriminating device
    15 ... CPU
    21 ... CPU
    27: Serial number reading sensor
    29 ... Unique feature sensor
    29e ... Infrared light image sensor
    29f-29j ... Infrared sensor
    40 ... banknotes
    42 ... serial number
    52f to 52j, 52x to 52z, waveform template data
    A to D: Image template data

Claims (4)

A paper sheet identification device comprising a plurality of reading means for reading optical characteristics of a paper sheet, and for identifying a paper sheet from a reading result read by the reading means and a predetermined identification standard,
Control symbol reading means for reading the control symbol of the record on the paper sheet,
An identification standard changing means for changing the identification standard based on the management symbol ;
The paper sheet identification apparatus according to claim 1, wherein only a read result readable by the plurality of reading means is identified by matching with the predetermined identification standard based on the management symbol . As the identification criterion, using tolerance width information indicating an identification tolerance range for identifying the true value with respect to the reference value,
2. The paper sheet identifying apparatus according to claim 1, wherein the identification reference changing unit is an allowable width changing unit that changes the allowable width information based on the management symbol. As the identification standard, a plurality of types of templates representing the characteristics of paper sheets are used,
3. The paper sheet identification apparatus according to claim 1, wherein the identification reference changing unit is a template selection unit that selects the template to be used based on the management symbol. Display means for displaying information and input means for inputting operations;
Paper sheet entry / exit means for entering and exiting paper sheets,
An automatic transaction apparatus comprising control means for executing control of the means,
The automatic transaction apparatus provided with the paper sheet identification apparatus as described in any one of Claim 1 to 3 .

Images