JP5336900B2 - Coin identification device and coin identification method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a coin discriminating device for correctly discriminating denominations of coins even between the denominations in which the coins are almost equal in diameter and thickness but the coins are made of different materials without being affected by deformation or dust attachment due to secular usage of the coins. <P>SOLUTION: Two types of material sensors being a mutual induction type diameter material sensor 5 and a resonance type material sensor 7 are arranged along a conveyance path 3 of a coin 1. A detection system of the mutual induction type is different from that of the resonance type, so that combinations of both signals indicating the material of respective coins disperse in a two-dimensional manner. Then, even when the respective coins are deformed, abraded, or the like, it is relatively easy to set a threshold for material determination, the materials of respective coins are correctly discriminated, and the denominations can be correctly discriminated. <P>COPYRIGHT: (C)2011,JPO&amp;INPIT

Description

  The present invention relates to a coin discriminating apparatus that detects the diameter, material, and thickness of a coin with a magnetic sensor and discriminates the denomination, authenticity, etc. of the coin. The present invention relates to a coin discriminating apparatus in which a material is accurately detected by a material sensor to improve a coin discriminating function.

  Conventionally, as a coin discriminating device that detects the diameter, material, and thickness of a coin with a magnetic sensor, there is one described in Patent Document 1. In this case, a diameter sensor, a resonance type first material / material thickness sensor, and a second material / material thickness sensor are arranged in order from the upstream in a conveyance path through which coins are conveyed by a conveyance belt. The material / material thickness sensor is arranged at a position to detect when any coin passes, and the resonance type second material / material thickness sensor is arranged at a position to detect only when a coin having a large diameter passes. As described above, the diameter of the coin is detected by a diameter sensor, and the material and thickness of the coin are well detected by two material / material thickness sensors regardless of the size of the coin to be detected.

  However, in the conventional coin discriminating apparatus, although the material of the coin is detected by two material / material thickness sensors, both of these sensors are of a resonance type that detects by a resonance method. In this resonance type sensor, resonance occurs between the coil L and the capacitor C, and when the coin passes through the magnetic field generated by the resonance current, the resonance frequency changes and the alternating current hardly flows. This is a method for detecting the change in voltage by utilizing the decrease in voltage at both ends, and has a characteristic that the output change amount of the sensor is larger as the conductivity of the material of the coin is lower and the thickness of the coin is thicker. Therefore, if the resonance type sensor is used, the material of the coin can be identified by the output change amount. However, in this resonance type sensor, when setting the threshold value for identifying the denomination of coins, particularly when the diameter and thickness of the coins are almost the same, and even when the denominations are different, only the material is different, Even so, if there are deformations due to aging of these coins or adhesion of dust, both denominations cannot always be accurately identified with high accuracy.

  Moreover, there exists what was described in patent document 2 as a structure of another coin identification apparatus. This includes a configuration relating to the identification of bicolor coins. The bicolor coin is a coin having a structure in which the metal of the core portion located in the center of the coin and the metal of the ring portion located in the periphery are different. Regarding the identification of this bicolor coin, in Patent Document 2, a reflective sensor is disposed on the sliding surface side of the coin conveyance path, and a high-frequency excitation signal of, for example, 250 kHz is supplied to the primary coil, from the secondary coil. When there is an inflection point in the signal waveform of the edge portion of the detection signal, that is, the electrical continuity at the junction between the core portion and the ring portion of the bicolor coin can be regarded as an increase / decrease signal of the surface current. In some cases, a configuration for determining that the coin is a bicolor coin is adopted.

However, the transfer speed of the coins being transferred through the transfer path is not always a uniform speed, and in practice, the transfer speed may be temporarily reduced. For this reason, in the technique described in Patent Document 2, an inflection point is detected in the signal waveform of the edge portion of the reflective sensor, but when the joint portion of the bicolor coin reaches the detection portion of the reflective sensor, the speed is just temporary. In the case of a decrease, it may be difficult to detect the inflection point in the signal waveform at the edge portion with high accuracy.
Japanese Patent No. 3718619 Japanese Patent No. 3908473

  As described above, the conventional coin discriminating apparatus can accurately discriminate even between similar denominations of coins having substantially the same diameter and thickness and different materials only in the coin material judgment using the resonance type material sensor. Even when the threshold value is set, there is a drawback that it is impossible to accurately identify denominations between similar denominations due to aging or deformation of such coins.

  In addition, the determination of bicolor coins has a drawback in that it may be difficult to identify with high accuracy, for example, due to variations in actual conveyance speed of coins in the conveyance path.

  An object of the present invention is to make it possible to accurately distinguish between denominations of different materials with certainty by always distinguishing the differences in the materials.

  In addition to the above object, the present invention eliminates fluctuations caused by variations in the coin conveyance speed included in the sensor, or accurately captures the value of the output signal at the ring portion of the bicolor coin, The goal is to identify color coins reliably.

  In order to achieve the above object, in the present invention, a material of a coin is detected with high accuracy by combining two types of a resonance type material sensor and a mutual induction type material sensor. Here, the mutual induction type material sensor excites the primary coil, receives the magnetic flux generated by the primary coil in the secondary coil, and when a coin enters between the two coils, an eddy current is generated and the secondary coil is generated. Since the magnetic flux received by the coil is reduced, this change in the magnetic flux is detected as a voltage change. The higher the conductivity of the coin material, the thicker the coin, the greater the output change of the sensor. . That is, the resonance sensor has a characteristic that the lower the conductivity is, the larger the output change amount of the sensor is. On the contrary, this mutual induction type material sensor has a characteristic that the higher the conductivity is, the larger the output change amount is. Pay attention and use the fact that when the output signals of these two types of material sensors are combined in two dimensions, the coin material of each denomination is distributed two-dimensionally, so the threshold for determining the material is relatively easy to set. To do.

  Further, in addition to the above-mentioned purpose, two feature amounts included in the sensor output signal are captured, and these feature amounts each include fluctuations caused by variations in the coin conveyance speed. If the ratio of the coin transport speed is eliminated or the output signal value at the ring part of the bicolor coin is extremely small, refer to the output signal at the ring part of another sensor. Know exactly and identify bicolor coins.

  Specifically, the coin identifying device according to the first aspect of the present invention is arranged to face the coin sliding surface of the transport passage and the transport passage for transporting the coins one by one with the coins being separated. The reflection type magnetic sensor, the first transmission type sensor disposed so as to sandwich one end of the conveyance path, and the conveyance path at a position facing the first transmission type sensor across the conveyance path. A mutual transmission type material sensor that detects the material of the coin by a mutual induction method, and a mutual induction type material sensor. A resonance-type material thickness sensor that detects the thickness of a coin by a resonance method at a position downstream of the conveyance path, and the mutual induction type at a position downstream of the conveyance path from the mutual induction type material sensor. Separately from the material sensor of the coin material Characterized in that a and a material sensor resonant detecting in a resonant manner.

  According to a second aspect of the present invention, in the coin identifying device according to the first aspect, the first transmission type sensor, the reflection type magnetic sensor, and the second transmission type sensor are linear in a direction perpendicular to the conveyance path. A common excitation coil for these three sensors is wound between the cores of the first transmission type sensor and the second transmission type sensor, and the mutual induction type material sensor is The mutual induction type diameter sensor that detects the diameter of the coin by a mutual induction method is also used, and excitation signals of a plurality of types of frequencies are applied to the excitation coil of the mutual induction type material sensor, and the mutual induction type Both the material sensor and the mutual induction type diameter sensor are operated.

  According to a third aspect of the present invention, in the coin identification device according to the first aspect, the resonance-type material thickness sensor is disposed at a predetermined position of the conveyance path on the coin pallet side of the conveyance path, and the conveyance The resonance-type material sensor is arranged on the opposite side of the conveyance passage in the direction orthogonal to the coin conveyance direction at a predetermined position of the passage, on the opposite side of the coin separation side.

  Invention of Claim 4 is a coin identification method using the coin identification device of the said Claim 2, Comprising: A conveyance path | route is used using the said mutual induction type | mold material sensor and diameter sensor, and the said reflection type magnetic sensor. The material of the coin that has been conveyed is detected, the denomination of the coin is tentatively determined, and then the tentatively determined coin is based on the output signals of the resonance-type material sensor and the resonance-type material thickness sensor. Is determined to be a true coin, and the denomination of the tentatively determined coin is determined.

  According to a fifth aspect of the present invention, in the coin identification device according to the first or second aspect, the reflective magnetic sensor has a core diameter that is greater than the diameter of the core portion of the bicolor coin that has been transported through the transport path. Is also formed small.

  Invention of Claim 6 is a coin identification method using the coin identification device of any one of the said Claim 1, 2, and 5, Comprising: A reflection type shape is provided in the lower surface of the conveyance path which conveys a coin. An output of the shape sensor is arranged based on an output waveform of the reflection type sensor and an output waveform of the reflection type material sensor with respect to the coin that has been transferred through the transfer path. The ratio between the distance between the first predetermined identical values at the rising and falling parts of the waveform and the distance between the second predetermined identical values at the rising and falling parts of the output waveform of the material sensor. The calculated ratio is used as a feature amount, and the feature amount is compared with a predetermined threshold value to identify whether or not the coin transported through the transport path is a bicolor coin.

  The invention according to claim 7 is a coin identifying method using the coin identifying apparatus according to claim 2 or 5, wherein the mutual induction type diameter sensor is arranged so as to sandwich one end portion of the conveyance path for conveying coins. A third transmissive sensor is disposed, and a fourth transmissive sensor of the mutual induction type diameter sensor is disposed so as to sandwich the other end of the transport passage, and a reflective type is disposed on the lower surface of the transport passage. Based on the output waveform of the mutual induction type diameter sensor and the output waveform of the reflection type material sensor with respect to the coin that has been arranged in the conveyance path, the peak value of the reflection type material sensor is arranged. The bimetal coin core material, the output signal value of the reflective material sensor at a predetermined percentage of the peak value of the output waveform of the mutual induction type diameter sensor as the ring metal material of the bimetal coin, And wherein the identifying Le coin.

  The invention according to claim 8 is a coin identifying method using the coin identifying device according to any one of claims 1, 2, and 5, wherein a resonance type material sensor is provided in a transport path for transporting coins. The distance between two peak values at the top of the output waveform of the resonance-type material sensor based on the output waveform of the resonance-type material sensor with respect to the coins that are disposed and transferred through the transfer path, and the output The ratio of the distance between the same predetermined values at the rising part and the falling part of the waveform is calculated, and the calculated ratio is used as a feature value, and the feature value is compared with a predetermined threshold value. It is characterized by discriminating whether the received coin is a bicolor coin or not.

  As described above, according to the first to eighth aspects of the present invention, since the output signals of the mutual induction type material sensor and the resonance type material sensor are combined in two dimensions, the diameters and thicknesses of the coins are substantially equal. However, since the combination of both signals indicating the material of such coins is two-dimensionally distributed, even if such coins deteriorate over time or are deformed, the threshold for determining the material is relatively set. It will be easier to identify these materials accurately.

  In particular, in the inventions according to claims 6 and 8, when bimetal coins are identified, two calculation values indicating characteristics are obtained from the output signal of the sensor, and one feature value is obtained from these two calculation values. Even if each value is affected by fluctuations in the coin transfer speed, the obtained feature value is absorbed and canceled by the fluctuations in the coin transfer speed. Can be accurately identified.

  Furthermore, in the invention of claim 7, the signal value at the core part of the bimetallic coin of the reflective material sensor is large, but the output signal value at the ring part is small, and what value is used as the signal value at the ring part. However, since the signal value of the mutual induction type diameter sensor at this ring part is large, the signal value at the ring part of the reflective type material sensor is based on the output signal of this mutual induction type diameter sensor. Can be accurately identified. Accordingly, it is possible to accurately identify the signal values of the core part and the ring part of the bimetal coin of the reflective material sensor, and to identify the bimetal coin well.

  As described above, according to the inventions described in claims 1 to 8, even if the coins are deteriorated over time or deformed between similar denominations having substantially the same diameter and thickness. It is possible to accurately distinguish and identify these materials.

  In particular, according to the inventions described in claims 6 to 8, even when there is a variation in the coin conveyance speed or the signal value at the ring portion of the bimetal coin of the sensor is small, it is not affected by them. It is possible to accurately identify bimetal coins.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

  FIG. 1 shows an overall configuration of a coin identifying module which is a coin identifying apparatus according to an embodiment of the present invention. In the figure, coins 1 are conveyed through the conveyance path 3 by the conveyance belt 2 at intervals from the left in the figure to the coin identification module. In the conveyance path 3, the coins are conveyed in a state of being shifted to the lower regulating surface in FIG. 1. A mutual induction type diameter material sensor 5 that magnetically detects the diameter and material of the coin 1 of the transport passage 3 is disposed at the front end of the coin identification module. Between the diameter material sensors 5, a pod-type reflection type magnetic sensor 6 disposed in the lower part facing the coin sliding surface of the transport passage 3 is incorporated.

  Further, on the rear side of the mutual induction type diameter material sensor 5 in the coin conveyance direction, a resonance type material sensor 7 that magnetically detects the material of the coin, and a resonance type material that magnetically detects the thickness of the coin. A thickness sensor 8 is arranged. The material thickness sensor 8 is disposed on the coin piece side (regulating surface side) at a predetermined position of the transport passage 3. On the other hand, the material sensor 7 is arranged in a direction opposite to the coin piece-shifting side on which the material thickness sensor 8 is arranged and in a direction orthogonal to the coin transport direction, but the coin of the material sensor 7 and the material thickness sensor 8 is arranged. The arrangement position in the transport direction is set to the same position. That is, since the material sensor 7 and the material thickness sensor 8 measure each part of the coin at the same timing, the feature quantity at the same timing of the two output signal waveforms is the feature quantity at the same part of the coin. Represents.

  Although not shown, an abnormal approach detection sensor (transmission type optical sensor) that also serves as a hole detection sensor is arranged on the front side of the entrance of the coin identification module, and the coins that have been transported through the transport path 3 are arranged. It is designed to detect optically.

  Next, the detailed overall structure of the diameter material sensor 5 built in the reflective sensor 6 is shown in FIG. The mutual induction type diameter material sensor 5 shown in the figure has a U-shaped cross section, and a coin conveyance passage 3 is formed at the bottom of the central space. The shield plate 4 is arranged. At one end and the other end (left end and right end in the figure) of the transport passage 3, two rectangular parallelepiped transmission sensors 51 and 52 (hereinafter referred to in the figure) so as to sandwich the end of the transport passage 3. The left transmissive sensor is referred to as a transmissive L sensor 51, and the right transmissive sensor in the drawing is referred to as a transmissive R sensor 52). An eddy current loss type reflective sensor 6 having a shape is disposed and incorporated. The transmission L sensor 51, the transmission R sensor 52, and the reflection type sensor 6 share the exciting coil 54 wound between the cores 53 and 59 respectively disposed below the left and right ends of FIG. The transmission L sensor 51 has a secondary coil 55 wound around a side core 58 disposed above the left end of the conveyance path 3 in FIG. 2, and the transmission R sensor 52 is disposed above the right end of FIG. The secondary coil 56 is wound around the side core 57. The reflective sensor 6 has a cylindrical pot core 61 disposed below the central portion of the transport passage 3, and a secondary coil 62 is wound and embedded on the inner peripheral surface of the pot core 61.

  The transmission L sensor 51 and transmission R sensor 52 of the mutual induction type diameter material sensor 5 and the shared excitation coil 54 of the reflection type sensor 6 are supplied with a combined excitation signal of high frequency (for example, 250 KHz) and low frequency (for example, 8 KHz). Is done. Magnetic flux is generated by the supply of the composite excitation signal, which alternates with the secondary coils 55, 56, and 62 of the transmission L sensor 51, transmission R sensor 52, and reflection type sensor 6. When a coin enters the magnetic flux, a magnetic field and an electric field based on an excitation signal component of high frequency (250 KHz) are greatly attenuated at a shallow position of the coin, that is, a surface portion of the coin. The high frequency (250 KHz) component of the output signals of the 52 secondary coils 55 and 56 is a signal including information on the shape of the coin, particularly the diameter of the coin. On the other hand, since the magnetic field and electric field based on the excitation signal component of low frequency (8 KHz) are attenuated greatly only after entering the internal position of the coin, the output signals of the secondary coils 55 and 56 of the transmission L sensor 51 and the transmission R sensor 52 The low frequency (8 KHz) component of the signal is a signal including information on the material of the coin. Similarly, the low frequency (8 KHz) output of the output signal of the secondary coil 62 of the reflective sensor 6 becomes a signal including information on the material of the coin, cross-sectional material of the coin, and magnetic / nonmagnetic information.

  In this way, the composite excitation signal of the high frequency (250 KHz) and the low frequency (8 KHz) is supplied to the common excitation coil 54, and the transmission L sensor 51 and the transmission R sensor 52 of the mutual induction type diameter material sensor 5 are mutually connected. The first and second transmissive sensors of the inductive material sensor and the two transmissive sensors of the mutual inductive diameter sensor are shared.

  In the present embodiment, the high frequency supplied to the transmission L sensor 51 and the transmission R sensor 52 is 250 KHz, and the low frequency supplied to both the transmission sensors 51 and 52 and the reflective sensor 6 is 8 KHz. Various frequencies can be used. For example, 4 kHz may be used as the low frequency. The diameter material sensor 5 detects the diameter of the coin, but may be a sensor for detecting the radius.

  FIG. 3 shows windings of the resonance type material sensor 7 and the material thickness sensor 8. A material sensor 7 is arranged on the right side of the figure and a material thickness sensor 8 is arranged on the left side of the figure. In the material sensor 7, two resonance coils 71 and 72 are embedded in the holding member 73 on the upper surface of the carrying passage 3 and the lower surface main body of the carrying passage 3. The resonance coil 71 and the resonance coil 72 are connected in series so that the direction of the generated magnetic flux is the same direction, and a current having a frequency of, for example, 330 KHz is supplied to the resonance coils 71 and 72 to form a resonance circuit. It has become.

  In such a configuration, the material sensor 7 generates an alternating magnetic flux by applying a sine wave to the resonance coils 71 and 72, and when a coin enters between the upper and lower cores, the magnetic flux is attenuated by an eddy current loss caused by the coin. . Therefore, the apparent mutual inductance is reduced, and the resonance frequency is increased. In addition, the higher the conductivity of the material of the coin 1, the easier the eddy current corresponding to the change in the magnetic flux that passes through it, and the magnetic lines of force are absorbed.

  The configuration of the resonance type material thickness sensor 8 is substantially the same as that of the material sensor 7 and includes two resonance coils 81 and 82 disposed above and below the conveyance path 3. The resonance coils 81 and 82 are connected in series so that the direction of the generated magnetic flux is opposite to that of the material sensor 7. In the resonance-type material thickness sensor 8, a resonance circuit is configured by flowing a current having a frequency of, for example, 180 KHz, which is different from the frequency (for example, 320 KHz) of a current flowing through the resonance coils 71 and 72 of the material sensor 7.

  The detection principles of the resonance-type material sensor 7 and the material thickness sensor 40 can be summarized as follows: (a) As the coin becomes thicker, the amount of attenuation of the magnetic flux increases, and as the resistivity decreases, the output change increases. (B) Frequency When the coin thickness is increased, the resistivity, that is, the influence of the coin material is reduced. (C) Since the output of the detection coil is proportional to the diameter of the coin, the coin is inserted between the upper and lower cores. The change amount ΔΦ of the magnetic flux can be expressed by ΔΦ = f (frequency ω, resistivity ρ, permeability μi, coin thickness T). Therefore, the material or thickness of the coin can be detected from the amount of change in the sensor output when the coin passes.

  For each output signal of the mutual induction type diameter material sensor 5 and the reflection sensor 6, and the resonance type material sensor 7 and the material thickness sensor 8, the characteristic amount of the coin to be identified is previously provided for each coin. The true / false of the coin is identified by comparing with the threshold area. Although the signal difference due to the material is small for high-frequency output signals, the attenuation rate is determined by the material of the surface layer, and for low-frequency output signals, the material of the intermediate layer is also affected, so the attenuation at each frequency can be determined in advance. The coins can be identified by comparing with the threshold region.

  In the present embodiment, excitation and detection signal processing are performed on the coin identification module with the circuit configuration shown in FIG. That is, a clock having a low frequency (for example, 8 KHz) and a high frequency (for example, 250 KHz) obtained by frequency-dividing a square wave oscillation signal from the vibrator 100 is output from the CPU 101, and a predetermined voltage (for example, 5V) is output from the clock voltage converter 102. ) Is converted into a sine wave through low-pass filters (LPF) 103 a and 103 b of the respective frequencies and input to the voltage booster circuit 104. In the voltage booster circuit 104, the two sine waves having different frequencies are combined and converted into a constant current, and the combined wave is supplied to the common excitation coil 54 of the diameter material sensor 5 and the reflection sensor 6. By supplying this synthetic excitation signal, a magnetic flux indicated by a broken line in the figure is generated from the excitation coil 54, and the two secondary coils 55 and 56 of the diameter material sensor 5 and the reflection sensor 6 are separated from the coin conveyance path 3. When a coin is inserted into the magnetic flux that intersects the secondary coil 62, the magnetic flux changes.

  The magnetic flux change in one of the secondary coils 55 of the diameter material sensor 5 is amplified in voltage by the first stage amplifier 110, and then the frequency is changed to a low frequency (8 KHz) and a high frequency (250 KHz) by the two frequency separators 111a and 111b. Separated, half-wave rectified by the subsequent half-wave rectifiers 112a and 112b, converted to direct current by the DC conversion circuits 113a and 113b, and taken into the CPU 101 as, for example, a 10-bit digital value by the A / D converter 114 of the CPU 101. It is.

  Similarly, the magnetic flux change in the other secondary coil 56 of the diameter material sensor 5 is also amplified in voltage by the first stage amplifier 120 and then low frequency (8 KHz) and high frequency (250 KHz) by the two frequency separators 121a and 121b. Are separated by the half-wave rectifiers 122a and 122b in the subsequent stage, converted into direct current by the DC conversion circuits 123a and 123b, converted into a digital value by the A / D converter 114 of the CPU 101, and stored in the CPU 101. Is taken in.

  On the other hand, the magnetic flux change in the secondary coil 62 of the reflection sensor 6 is voltage amplified by the first-stage amplifier 130, and then only one low frequency component (8 KHz) is frequency-separated by one frequency separator 131. It is half-wave rectified by the rectifier 132, converted to direct current by the DC circuit 133, converted to a digital value by the A / D converter 114 of the CPU 101, and taken into the CPU 101.

  The resonance-type material sensor 7 is connected to the LC resonance circuit 140 having a capacitor inside in a state where the primary coil 71 and the secondary coil 72 are connected in series so that the generated magnetic flux is in the same direction. By applying a sine wave to the coils 71 and 72 by the LC resonance circuit 140, an alternating magnetic flux is generated. When a coin enters the magnetic flux, the magnetic flux changes, and the apparent mutual inductance is small. Thus, the resonance frequency is increased. The output of the LC resonance circuit 140 is half-wave rectified by a half-wave rectifier 141, then converted into a direct current by a DC conversion circuit 142, and analog / digital converted by an A / D converter 145, thereby being multi-bit (for example, 10 bits). It becomes a digital value and is taken into CPU101. The DC circuit 142 receives the adjustment voltage from the digital-analog converter (DAC) 146, and the DAC 146 receives the control signal from the CPU 101 and changes the adjustment voltage. In other words, the DC circuit 142 also has a function of giving an offset voltage to the output of the half-wave rectifier 141 so that the DC output value of the LC oscillation circuit 140 does not become a negative value during standby when no coin enters the magnetic flux.

  The resonance type material thickness sensor 8 is connected to an LC resonance circuit 150 having a capacitor inside in a state where the primary coil 81 and the secondary coil 82 are connected in series so that the generated magnetic flux is in the opposite direction. The As described above, the material thickness sensor 8 is different from the material sensor 7 in that it is connected in series so that the directions of two generated magnetic fluxes are opposite to each other. The LC resonance circuit 150 applies a sine wave to both the coils 81 and 82 to generate an alternating magnetic flux, and the output of the LC resonance circuit 150 changes according to the change in magnetic flux when a coin enters the magnetic flux. The output of the LC resonance circuit 150 is half-wave rectified by the half-wave rectifier 151, then converted to direct current by the DC conversion circuit 152, A / D converted by the A / D converter 145, and converted into a multi-bit digital value. Is taken in. Similar to the circuit configuration of the material sensor 7, the DC circuit 152 receives an adjustment voltage from the DAC 146 and applies an offset voltage to the output of the half-wave rectifier 151 so that the DC output value does not become a negative value during standby. Also have.

  Moreover, in FIG. 4, the abnormal approach sensor 170 arrange | positioned just before the entrance of a coin identification module has the light emitter 170a and the light receiver 170b. A predetermined constant current is supplied to the light emitter 170 a by a constant current circuit 176 that receives a control signal from a D / A converter 175 provided in the CPU 101. Further, the amount of light received by the light receiver 170 b is converted into a current and then taken into the CPU 101 through the Schmitt circuit 177.

  The outputs of the diameter material sensor 5, the reflection sensor 6, the material sensor 7, the material thickness sensor 8, and the abnormal approach sensor 170 captured by the CPU 101 are subjected to a coin identifying process by the CPU 101. Connected to the CPU 101 is an EEPROM 160 for storing data used for identifying coin denomination, authenticity, etc., threshold values, various adjustment values, and the like. That is, the CPU 101 uses the threshold value which is the stored data of the EEPROM 160 in identifying the denomination, authenticity, etc. based on the outputs of the five types of sensors.

  Next, denomination determination using the diameter material sensor 5, the reflection sensor 6, the material sensor 7, the material thickness sensor 8, and the abnormal approach sensor 170 will be described. FIG. 5 shows a flowchart of the denomination discrimination. In the figure, after checking the operation of each sensor in step S1 in advance, starting in step S2, the coins that are sequentially conveyed in the conveyance path 3 in step S3 are the current 500 yen, 100 yen, A temporary denomination of 50 yen, 10 yen, etc. is made. This temporary denomination is determined by detecting the approach of the coin 1 based on the output of the abnormal approach sensor 170 and then the output of the mutual induction type diameter material sensor 5 (specifically, the output of the secondary coil 55 of the transmission L sensor 51) This is based on the sum of the output of the secondary coil 56 of the transmission R sensor 52 and the output of the reflective magnetic sensor 6. For example, as shown in FIG. 6, the output of the diameter material sensor 5 is a low frequency with the horizontal axis indicating the attenuation factor of the high frequency component output (output of the 250 KHz frequency separator 111 b) (diameter) of the transmission L sensor 51 and the transmission R sensor 52. Taking the attenuation rate of the component output (output of the 8 KHz frequency separator 111a) (material) on the vertical axis, combining these outputs and setting seven relatively wide threshold regions surrounded by solid lines in the figure, 1 yen, 5 yen ~ The seven types of denominations of the current 500 yen and the old 500 yen can be roughly identified. The “attenuation rate” of the sensor output is indicated by a value obtained by dividing the change voltage of the sensor output by the standby voltage of the sensor output. Hereinafter, the term “attenuation rate” has the same meaning.

  For coins for which the denomination has been tentatively determined, the authenticity of the diameter and material of the coin is determined for each coin in steps S4a to S4g. This true / false determination is performed by comparing the both outputs (diameter and material) of the diameter material sensor 5 with a predetermined narrow threshold region within each of the wide threshold regions. Next, in steps S5a to S5g, the material and thickness of the coin are determined for each coin based on both outputs of the material sensor 7 and the material thickness sensor 8, and the authenticity is determined. Further, in steps S6a to S6g, additional logic is set for each coin so as to exclude other coins that are very similar to the coin. This additional logic is performed by setting a threshold region that combines two or more outputs among all the diameter material sensors 5, the reflection sensor 6, the material sensor 7, and the material thickness sensor 8 provided in the coin identification module.

  If all the tentatively determined coins enter all the threshold areas in steps S3 to S6, it is determined in step S7 that the tentatively determined coin is a true coin.

  On the other hand, in the above-described steps S3 to S6, when deviating from each threshold region, the coin is rejected, and the cause of the rejection is stored and output in step S8.

  Next, the effect of this embodiment will be described. FIG. 7 shows the identification result of the 10 yen coin. In the figure, the horizontal axis represents the attenuation rate of the output of the resonance type material sensor 7, and the vertical axis represents the attenuation rate of the low frequency component output (output of the 8 KHz frequency separator 111a) (material) of the mutual induction type diameter material sensor 5. is there. As can be seen from the figure, the output of the resonance type material sensor 7 alone cannot distinguish the 10 yen true coin from the 10 FILS coin of Bahrain that has a conductivity slightly higher than the 10 yen true coin. The 10FILS coin is also mistakenly identified as a 10-yen coin, but the 8 KHz component output (material) of the mutual induction type diameter material sensor 5 and the output of the resonance type material sensor 7 are combined and surrounded by a solid line. If it is set, only the 10 yen coin can be identified well, and the 10 FILS coin of the similar Bahrain country can be surely excluded. Therefore, the authenticity determination of the material of the 10 yen coin using the 8 kHz component output (material) of the mutual induction type diameter material sensor 5 in step S4e in FIG. 5 and the resonance type material sensor 7 in step S5e are performed. When the authenticity determination of the material of the used 10 yen coin is used, it becomes possible to identify the 10 yen coin satisfactorily while effectively eliminating similar coins of the 10 yen coin.

  In addition, in this embodiment, the resonance type material sensor 7 and the resonance type material thickness sensor 8 are linearly arranged in a direction perpendicular to the coin transport direction. That is, the arrangement positions of the resonance type material sensor 7 and the material thickness sensor 8 in the coin conveyance direction are set to the same position. Accordingly, since the material sensor 7 and the material thickness sensor 8 measure each part of the coin at the same timing, the feature quantity at the same timing of the two output signal waveforms is the feature quantity at the same part of the coin. It is easy to handle both outputs of the resonance type material sensor 7 and the material thickness sensor 8, and the coin can be accurately identified.

  Furthermore, another example will be described as an effect of the present embodiment. FIG. 8 shows the distribution of Japanese one-yen coins and British old one-penny coins. A large number of 1-yen coins indicated by x and a large number of 1-penny coins indicated by ◯ are distributed and distributed with a certain degree of spread due to aging and dust adhesion. The horizontal axis of FIG. 8 is the attenuation factor of the 250 KHz component output (diameter) of the mutual induction type diameter material sensor 5, and the vertical axis is the attenuation rate of the output of the resonance type material sensor 7. The scales on the vertical axis and the horizontal axis indicate the secondary during standby when no coin has entered the magnetic flux between the primary coils 54 and 72 and the secondary coils 55, 56 and 71 of the diameter material sensor 5 and the material sensor 7. This is a value obtained by standardizing the cross magnetic flux amount in a state where a coin has entered, assuming that the cross magnetic flux amount of the coil is 1. In the distribution of FIG. 8, when the output of the 250 kHz component (diameter) of the mutual induction type diameter material sensor 5 and the output of the resonance type material sensor 7 are combined, the distribution of 1 yen coin and the distribution of 1 penny coin are clear. It can be seen that they can be distinguished.

  For comparison with FIG. 8, FIG. 9 similarly shows the distribution of 1-yen coins and 1-penny coins, and the horizontal axis indicates the attenuation rate of the 250 kHz component output (diameter) of the diameter-inductive diameter sensor 5. The vertical axis shows a distribution diagram in which the attenuation factor of the 4 kHz component output (material) of the mutual induction type diameter material sensor 5 is shown. In the figure, the threshold area for discriminating the denomination of 1-yen coins with a certain degree of tolerance is shown in the figure with a certain degree of tolerance so as not to mistakenly identify true 1-yen coins as fake coins. When set, there is a high possibility that a large number of 1 penny coins are included in this threshold region, and the 1 penny coin is erroneously identified as a 1 yen coin.

  On the other hand, in the distribution diagram shown in FIG. 8, if the area surrounded by the broken line in the figure is set as a threshold area that does not misidentify true coins as fake coins, almost one penny coin will not enter this threshold area. It can be seen that, compared with the distribution of FIG. 9, a threshold region for identifying denominations and counterfeits can be easily set. Accordingly, the true / false determination of the diameter of the one-yen coin using the 250 KHz component output of the mutual induction type diameter material sensor 5 in step S4g of FIG. 5 and 1 using the resonance type material sensor 7 in step S5g. When the authenticity determination of the material of the yen coin is used, it is possible to successfully identify the true 1-yen coin while effectively excluding similar coins of the 1-yen coin.

  However, also in the distribution diagram of FIG. 8, one penny coin is entered in the upper right corner of the threshold region indicated by the broken line in FIG. Therefore, it is necessary to take measures to prevent erroneous identification even at this corner. This countermeasure is additional logic for eliminating similar coins in step S6g of FIG. A flowchart showing this additional logic is shown in FIG.

  FIG. 10 shows a determination flow for the CPU 101 to determine that the threshold area located at the upper right corner of the threshold area surrounded by the broken line in FIG. 8 is a fake coin (abnormal 1-yen coin). In the figure, in step S1, it is determined whether or not the high frequency (250 kHz) component output X of the diameter material sensor 5 of the mutual induction method is 0.647 or less (see FIG. 8), and if X ≦ 0.647. Are determined to be 1 yen true coins in step S2. In step S3, it is determined whether or not the 330 KHz output Y of the resonance type material sensor 7 is 0.636 or less, for example. If Y ≦ 0.636, it is determined in step S2 that it is a 1 yen true coin. In step S4, a determination line indicated by a solid diagonal line in FIG. 8 is set, and the point determined by the combination of the high frequency (250 kHz) component output X of the diameter material sensor 5 and the 330 kHz output Y of the resonance type material sensor 7 is determined. It is determined whether or not it is less than this determination line, and if it is less than this determination line, it is determined to be a 1 yen true coin in step S2. Is judged to belong to the corner of the upper right corner of the threshold region shown in FIG.

  As described above, denominations having substantially the same diameter and material thickness can be obtained by combining the high frequency component output (diameter sensor output) X of the mutual induction type diameter material sensor 5 and the output Y of the resonance type material sensor 7. In this case, misidentification may occur with a low probability, but by reviewing the threshold area according to the determination flow, it is possible to reliably determine that the false threshold (material abnormality) is present in the threshold area where misidentification occurs. , Improving the ability to eliminate fake coins.

  FIG. 11 shows a distribution map of the seven types of coins in Japan. In the figure, the vertical axis represents the attenuation rate of the low frequency (8 kHz) component output of the mutual induction type diameter material sensor 5, and the horizontal axis represents the output attenuation rate of the resonance type material sensor 7. As can be seen from the figure, with only the attenuation rate of the low frequency (8 kHz) component output (output of the material sensor) of the diameter material sensor 5 of the mutual induction method, the output attenuation rate is 1 circle located near 0.500, It is difficult to identify the three types of coins of 5 yen and the current 500 yen. On the other hand, with only the output attenuation rate of the resonance-type material sensor 7, the output attenuation rate is 50 yen, 100 located near 0.700. It is difficult to identify three types of coins, yen and old 500 yen. However, when the low frequency (8 kHz) component output of the mutual induction type diameter material sensor 5 and the output of the resonance type material sensor 7 are combined, the low frequency (8 kHz) component signal of the mutual induction type diameter material sensor 5 is combined. The coins of 1 yen, 5 yen, and current 500 yen that are difficult to identify by themselves can be reliably identified by the resonance type material sensor 7, and 50 yen, 100 yen, which are difficult to identify only by the resonance type material sensor 7, It can be seen that the old coins of 500 yen can be reliably identified by the low frequency (8 kHz) component signal of the diameter material sensor 5 of the mutual induction method, and all these seven types of coins can be identified as good.

  In the present embodiment, a jagged sensor may be arranged along the transport path 3 so as to recognize a jagged carved around the coin to be identified. If a Giza sensor is arranged, it is possible to distinguish well between 100 yen coins and their fake coins, 50 yen coins and their fake coins, old 500 yen and 500 WON.

(Second Embodiment)
Next, a coin identifying device according to a second embodiment of the present invention will be described.

  The coin identification device of the present embodiment has the same overall configuration as the coin identification module shown in FIG. The difference is that the reflective sensor 6 is used as a shape sensor and a material sensor, and bicolor coins are detected with high accuracy based on the outputs of both sensors. That is, even when the conveyance speed of the coin 1 on the conveyance belt 2 varies, the variation in speed is canceled out by a combination of both signals of the shape sensor and the material sensor, and the bicolor coin is recognized with high accuracy. As shown in FIG. 13, the bicolor coin 180 includes a core portion 180a located at the center thereof and a ring portion 180b located around the core portion 180a, and the material of the core portion 180a and the ring portion 180b. Is different.

  FIG. 12 shows a circuit configuration and excitation and detection signal processing of the reflective sensor 6 in this embodiment. In the same figure, the common excitation coil 54 of the reflective sensor 6 is supplied with a composite excitation signal of high frequency (for example, 250 KHz) and low frequency (for example, 8 KHz) as in FIG. After the voltage is amplified by the amplifier 130, both the high frequency (250 KHz) component and the low frequency (8 KHz) component are frequency separated by the two frequency separators 131a and 131b, and then the two half wave rectifiers 132a and 132b are half separated. Wave rectified, converted into direct current by the two DC conversion circuits 133a and 133b, converted into a digital value by the A / D converter 114 of the CPU 101, and taken into the CPU 101.

  As shown in FIG. 14, the reflective sensor 6 is formed such that the diameter of the core 6 a is smaller than the diameter of the core portion 180 a of the bicolor coin 180 to be recognized. When passing through the position, an output signal for only the core portion 180a is obtained.

  FIG. 15 shows a bicolor coin identification flowchart by the reflective sensor 6 (reflective shape sensor and reflective material sensor). In the figure, sampling of both data of the high frequency (250 KHz) component and the low frequency (8 KHz) component from the secondary coil 62 of the reflective sensor 6 is started at predetermined time intervals in step S1. This data sampling is repeatedly performed using a clock synchronized with the coin transport, triggered by a signal that the abnormal approach sensor 170 detects the entry of the coin. An example of the result of this data sampling is shown in FIG. This figure shows the signal waveform of a monometal (uniform material) coin sample (for example, the current 500 yen (nickel brass coin)), and the signal waveform of the high frequency component is used as information on the shape of the coin, as shown by the solid line It has a flat part followed by a high value, and the length of this flat part represents the diameter of the coin. The signal waveform of the low frequency component is used as information on the material from the surface of the coin to the inner center, and has a peak value in the vicinity of the coin center position as indicated by a broken line. FIG. 17 shows the signal waveform of a bicolor coin sample, and the signal waveform of the high frequency component has an inflection point D due to the influence of different materials in the coin ring and core, as shown by the solid line, and the low frequency component. As shown by the broken line, the signal waveform of FIG. 4 is affected by the difference in material between the ring part and the core part, and the slope of the waveform changes suddenly near the boundary between the edge and the core part.

  Returning to the coin discrimination flow of FIG. 15, in step S <b> 2, peak values in each sampled high frequency component and low frequency component data are detected. Specifically, the current sampling value is compared with the previous value, and if the current value> the previous value, the value is stored in a RAM (not shown) built in the CPU 101 and the peak value is repeatedly updated. By doing. Thereafter, when the data sampling is completed in step S3, the time when the high frequency component is changed by more than ½ value of the peak value using a large number of sampling data and the detected peak value in step S4 is the same as the first predetermined value. In addition to calculating the distance between the values, the low frequency component is similarly calculated as the distance between the second predetermined identical values by the time when the peak value has changed by a half value or more. This time is indicated by reference numerals tms, tmq, tbs, and tbq in FIGS. After that, in step S5, evaluation is performed for each monometal coin and bicolor coin with a zone value obtained by dividing the low frequency component (material) peak 1/2 hour by the high frequency component (shape) peak value 1/2 hour as an evaluation value Z. The values Zm = a · tmq / tms and Zb = b · tbq / tbs (a and b are constants) are calculated. Then, the evaluation values Zm and Zb are compared with a predetermined threshold value to determine whether the metal is a monometal coin or a bicolor coin.

  When the shape and material of the coin are detected by the reflective sensor 6, if the transport speed of the coin 1 on the transport path 3 varies, the signal waveform of the high-frequency (shape) component in FIG. In some cases, it is difficult to distinguish from the signal waveform of the high-frequency (shape) component of the bicolor coin shown in FIG. However, in the present embodiment, the two peak ½ hours of the high frequency component (shape) and the low frequency component (material) similarly vary according to the variation in the coin conveyance speed, and the two peaks ½. Since the evaluation values Zm and Zb are expressed by the ratio between the times, the evaluation values Zm and Zb are values that cancel and eliminate the variation in the coin conveyance speed. Therefore, even if there is a variation in the coin conveyance speed, it is possible to accurately identify the monometal coin and the bicolor coin.

  FIG. 18 is calculated for a large number of two kinds of monometal coin samples (current 500 yen coin and old 500 yen coin) and four kinds of bicolor coin samples (samples A, B, C, and D). A distribution diagram of the evaluation values Zm (= a · tmq / tms) and Zb (= b · tbq / tbs) is shown. In the figure, the distribution of each sample is a normal distribution, but the evaluation value at the apex (maximum number) is “755” in the zone ratio in the current 500 yen coin, whereas “735” in the old 500 yen coin. The evaluation value is “575” for the bicolor coin sample A, “620” for the sample B, and “445” for the samples C and D. Therefore, between these samples, when the threshold value is set around “675”, the current 500 yen coin and the old 500 yen coin (monometal coin sample) and the bicolor coin samples A, B, C, and D are accurately obtained. It can be seen that they can be distinguished. In creating the distribution diagram of FIG. 18, since the change in the signal is opposite to that in the non-magnetic coin when the coin is ferromagnetic, the evaluation value is calculated as an absolute value.

  In the present embodiment, when calculating the evaluation value Z, two peak 1/2 times of the high frequency component (shape) and the low frequency component (material) of the reflection sensor 6 are used. It is not necessary to use time, and it is of course possible to use, for example, a peak 1/3 hour as long as it is possible to eliminate variations in the coin conveyance speed.

(Third embodiment)
Then, the coin identification device of the 3rd Embodiment of this invention is demonstrated.

  The coin identification device of the present embodiment has the same overall configuration as the coin identification module shown in FIG. The difference is that the core portion and the ring portion of the bicolor coin are basically discriminated using the reflection type sensor (material sensor) 6, but the feature amount of the ring portion is extracted by the reflection type sensor (material sensor) 6. The bicolor coins are discriminated with high accuracy by reinforcing with the high frequency component output (diameter) of the mutual induction type diameter material sensor 5 shown in FIG.

  Also in the present embodiment, the reflective sensor (material sensor) 6 is formed such that the diameter of the core 6a is smaller than the diameter of the core portion 180a of the bicolor coin 180 to be recognized, as shown in FIG. When the bicolor coin 180 passes the position of the reflective sensor 6, an output signal for only the core portion 180a is obtained.

  Next, FIG. 19 shows a flowchart for identifying a bicolor coin in the present embodiment. In step S1, the high frequency component output (diameter) and reflection of the mutual induction type diameter material sensor 5 are triggered by using a clock synchronized with the coin transport, using the signal detected by the abnormal approach sensor 170 as the trigger. Sampling of the low frequency (8 KHz) output of the mold sensor (material sensor) 6 is started. An example of the result of this data sampling is shown in FIG. In the figure, the signal waveform of the attenuation amount of the high frequency component (250 KHz) of the mutual induction type diameter material sensor 5 is used as information of the coin material × thickness × diameter, and as shown by a solid line, a symmetrical bell shape It becomes the apex in the central part of the coin. On the other hand, the low-frequency (8 KHz) attenuation signal waveform of the reflective sensor (material sensor) 6 is used as material information from the surface of the coin to the vicinity of the inner core, and as shown by a broken line, a symmetrical bell. However, the gradient of the waveform change is gentler, and the signal waveform of the attenuation amount of the high frequency (250 KHz) component of the mutual induction type diameter material sensor 5 is steeper.

  In step S2 of FIG. 19, the high frequency component (250 KHz) attenuation amount of the diameter material sensor 5 and the low frequency (8 KHz) attenuation amount of the reflective sensor (material sensor) 6 are the same as in the second embodiment. The peak value is detected. Thereafter, when the data sampling is completed in step S3, the peak value Rp (see FIG. 20) of the attenuation amount of the low frequency signal of the reflection type sensor (material sensor) 6 is set as the feature amount of the coin core in step S4. In step S5, the feature amount of the coin ring portion is calculated. In the calculation of the characteristic amount of the ring portion, the sampling value reduced to 30% with respect to the detected peak value among the sampling values of the high-frequency component (250 KHz) component of the diameter material sensor 5 (symbol T30 shown in FIG. 20). And the signal value (attenuation amount) Rr of the reflective sensor (material sensor) 6 sampled at the same timing as the 30% value of the peak value is used as the feature amount of the coin ring. That is, in the reflection type sensor (material sensor) 6, since the output value at the coin ring portion is small and the inclination of the signal change is small, it is difficult to accurately grasp the output value at the coin ring portion. The attenuation of the high frequency (250 KHz) component output of the diameter material sensor 5 having a large diameter and a sharp signal change is used, and the attenuation of the output of the reflective sensor (material sensor) 6 at the same timing as the 30% value is determined as a coin. This is the feature value of the ring part.

  In step S6, whether the coin on the transport path 3 is a monometal coin or a buy-by coin is determined by the combination of the two feature amounts (peak value Rp and high frequency component peak 30% timing value Rr) of the reflective sensor (material sensor) 6. In addition to determining whether it is a color coin, the type of bicolor coin is also determined.

  FIG. 21 shows the result of actually measuring the combination of the two feature amounts Rp and Rr of the reflective sensor (material sensor) 6 with respect to six types of coins. In the same figure, the measurement result about the monometal coin sample (the current 500 yen coin and the old 500 yen coin) and four types of bicolor coin samples A, B, C, and D is shown. In the figure, the horizontal axis is the peak value Rp of the attenuation rate of the reflective sensor (material sensor) 6, and the vertical axis is the position of the 30% value of the peak value of the attenuation amount of the high-frequency component (250 KHz) of the diameter material sensor 5. This is the attenuation rate Rr of the output of the reflective sensor (material sensor) 6. As can be seen from the figure, in the old 500 yen coin and the bicolor coin sample B, the peak value (core material) Rp of the reflective sensor (material sensor) 6 is in the same range, but the high frequency component peak 30% timing value. (Ring material) Since the range of Rr is different, the monometal coin (old 500 yen coin) and the bicolor coin sample B can be clearly identified. In the old 500 yen coin and the bicolor coin samples A, C, and D, the range of the high frequency component peak 30% timing value (ring material) Rr is the same range, but the range of the peak value (core material) Rp is Since they are different from each other, they can be identified. Thus, by combining the peak value Rp of the reflective sensor (material sensor) 6 and the high frequency component peak 30% timing value Rr, it is possible to distinguish between monometal coins and bicolor coins and between bicolor coins. It is possible to carry out with high accuracy.

  In the present embodiment, the high frequency component peak 30% timing value Rr of the reflective sensor (material sensor) 6 is used as the feature amount of the coin ring portion, but the present invention is not limited to this 30% timing value. In addition, as long as the sampling value of the high frequency component (250 KHz) of the diameter material sensor 5 reliably represents the output value at the coin ring, the high frequency component peak 20% timing value or the high frequency component peak 40% timing value is adopted. Of course, you may do.

(Fourth embodiment)
Then, the coin identification device of the 4th Embodiment of this invention is demonstrated.

  The coin identification device of the present embodiment has the same overall configuration as the coin identification module shown in FIG. The difference is that the output signal of the resonance type material sensor 7 is specially combined to favorably identify the bicolor coin. That is, by combining the signal of the resonance type material sensor 7 at the core portion 180a of the bicolor coin 180 and the signal at the ring portion 180b, the bicolor coin regardless of the variation in the conveyance speed of the coin 1 on the conveyance path 3. Is identified with high accuracy.

  The identification flow of the bicolor coin of this embodiment is shown in FIG. In the figure, in step S1, sampling of the output of the resonance-type material sensor 7 is started using a signal synchronized with the coin transport, using the signal detected by the abnormal approach sensor 170 as the trigger of coins. An example of the result of this data sampling is shown in FIG. In the figure, a signal waveform (attenuation amount of output) of the resonance type material sensor 7 is used as information of coin material × thickness, and in a monometal coin (for example, 100 yen coin), a waveform corresponding to a peripheral portion of the coin. Each of the corners x has the highest value before and after the high value region. On the other hand, as shown in FIG. 24, bicolor coins (for example, Kenyan 10 shilling coins) have corners x before and after the high-value region of the waveform. The distance between the two corners is shown in FIG. It has a feature that is shorter than the distance between the corners of the waveform of the monometal coin shown.

In step S2 of FIG. 22, data sampling is performed on the output of the resonance type material sensor 7 in the same manner as described above, and the peak values of the corners x located before and after the high value region of the waveform are detected. Thereafter, when the data sampling is completed, a distance A between the two corners is calculated and output as shown in FIGS. 23 and 24 based on a large number of sampling data and the peak value x of the corner in step S3. A distance (number of zones) B between two sampling values whose attenuation amount is attenuated by 100 or more, for example, is calculated as an AD value, and using this distance A, B, a feature amount C is expressed by the following equation:
Feature C = Distance A / Distance B × 1000
Calculate by It should be noted that the distance B may be a distance that varies in accordance with variations in the conveyance speed, and it is particularly desirable that the distance B varies according to the variation in the distance A.

  Thereafter, in step S4, the peak value x of the corner and the feature amount C are compared with respective threshold values to identify whether the coin on the transport path 3 is a monometal coin or a bicolor coin.

  When variation occurs in the coin conveyance speed, the output waveform of the resonance-type material sensor 7 also expands or contracts in the time axis direction, and the distance A between both corners of the output waveform also varies. Accordingly, as can be seen from FIG. 23 and FIG. 24, the distance A is generally shorter for monometal coins (100 yen coins) and bicolor coins (Kenyan 10 shilling coins) than for Kenyan 10 shilling coins. When only A is used as a feature amount for coin identification, when the waveform extends in the time axis direction, a Kenyan 10 shilling coin may be erroneously identified as a 100 yen coin.

  However, in this embodiment, since the ratio of the distance A and the distance B that expands and contracts according to the variation in the coin transport speed is the feature amount C, the feature amount C does not depend on the coin transport speed, and the speed Variations can be absorbed. Therefore, it is possible to identify the monometal coin and the bicolor coin with high accuracy.

  FIG. 25 shows the results of actually measuring the influence on the feature amount C due to the variation in the coin conveyance speed for the 100 yen coin and the Kenyan 10 shilling coin. In the figure, the horizontal axis represents a feature amount (attenuation rate) C, and the vertical axis represents a peak value (attenuation rate). Moreover, the conveyance speed was actually measured with two types, 525 mm / s and 630 mm / s. As can be seen from the figure, the variation range of the feature value C of the 100 yen coin is within the same range even if the transport speed is different, and the variation range of the feature value C of the Kenyan 10 shilling coin is also within the same range. It can be seen that the feature amount C does not depend on the coin conveyance speed. In the actual measurement result of FIG. 25, if 550 is adopted as the threshold value of the feature amount C and 608 is adopted as the threshold value of the peak value of the corner, the feature amount C <550 and the peak value> 608 are obtained. With the exclusion logic, Kenyan 10 shilling coins can be reliably excluded and identified from 100 yen coins.

  As described above, the present invention accurately determines the material of these coins even if they are deteriorated or deformed between similar denominations whose coin diameter and thickness are approximately equal. Since it is possible to distinguish and identify, and it is possible to accurately identify bimetal coins regardless of variations in the coin conveyance speed, it is useful as a coin identifying device.

It is a figure which shows the whole schematic structure of the coin identification module which is a coin identification device of the 1st Embodiment of this invention. It is sectional drawing which shows the internal structure of the mutual induction type diameter material sensor with which the coin identification module is equipped. It is a figure which shows schematic structure of the resonance type material sensor and material thickness sensor with which the coin identification module is equipped. It is a block diagram which shows the excitation of various sensors with which the coin identification module is provided, and a detection signal processing system. It is a flowchart figure of denomination identification in the same coin identification module. It is a figure explaining the setting of the threshold area | region for temporary money type determination. It is a figure which shows the effect of this embodiment and demonstrates that a 10-yen coin and a Bahrain 10FILS coin can be clearly distinguished using a mutual induction type material sensor and a resonance type material sensor. It is a figure which shows another effect of this embodiment, and shows distribution of 1 yen coin and 1 penny coin in both outputs of a mutual induction type diameter sensor and a resonance type material sensor. It is a figure which shows distribution of 1 yen coin and 1 penny coin in both outputs of a mutual induction type diameter sensor and a mutual induction type material sensor. It is a flowchart figure which shows the additional logic for clearly distinguishing 1 yen coin and 1 penny coin in FIG. It is a figure which shows distribution of the Japanese coin 7 denomination in both outputs of a resonance type material sensor and a mutual induction type material sensor. It is a block diagram which shows the excitation and detection signal processing system of a reflective sensor with which the coin identification module of the 2nd Embodiment of this invention is equipped. It is a figure which shows the structure of a bicolor coin. It is a figure which shows the magnitude relationship between the core of the reflective sensor of the embodiment, and a bicolor coin. It is an identification flowchart figure of the bicolor coin by the reflective sensor (a reflective shape sensor and a reflective material sensor) of the embodiment. It is a signal waveform diagram of the data sampling result of the monometal coin sample using the reflection type sensor. It is a signal waveform diagram of the data sampling result of the bimetal coin sample using the reflection type sensor. It is a figure which shows the effect which can distinguish clearly both 2 types of monometallic coin samples and 4 types of bicolor coin samples in the same embodiment. It is a flowchart figure which identifies a bicolor coin using the reflection type sensor with which the coin identification module of the 3rd Embodiment of this invention is equipped, and a mutual induction type diameter sensor. It is a signal waveform diagram in a bimetal coin sample using the same reflection type sensor and a mutual induction type diameter sensor. It is a figure which shows the effect which can distinguish clearly between 2 types of monometallic coin samples and 4 types of bicolor coin samples in the same embodiment. It is a flowchart figure which identifies a bicolor coin only using the resonance type material sensor with which the coin identification module of the 4th Embodiment of this invention is equipped. It is a signal waveform diagram of the data sampling result of the monometal coin sample (100 yen coin) using the resonance type material sensor. It is a signal waveform diagram of the data sampling result of the bimetal coin sample (Kenyan 10 shilling coin) using the resonance type material sensor. It is a figure explaining the effect which 100 yen coin and Kenyan 10 shilling coin can distinguish favorably irrespective of the dispersion | variation in coin conveyance speed in the same embodiment.

DESCRIPTION OF SYMBOLS 1 Coin 2 Conveyance belt 3 Conveyance path 5 Mutual induction type diameter material sensor 51 Transmission L sensor 52 Transmission R sensor 54 Excitation coil 55, 56 Secondary coil 6 Reflection sensor 6a Core 62 Secondary coil 7 Resonance type material sensor 71, 72 Resonance Coil 8 Resonance type material thickness sensor 81, 82 Resonance coil 101 CPU
140, 150 LC resonance circuit 180 Bicolor coin 180a Core part 180b Ring part

Claims (8)

A transport path for transporting the coins one by one with the coins being separated,
A reflective magnetic sensor disposed facing the coin sliding surface of the transport passage;
A first transmissive sensor disposed so as to sandwich one end portion of the transport passage, and the other end portion of the transport passage are sandwiched at a position facing the first transmissive sensor across the transport passage. A mutual inductive type material sensor that detects the material of the coin by a mutual induction method.
A resonance-type material thickness sensor that detects the thickness of a coin by a resonance method at a position on the downstream side of the conveyance path from the mutual induction-type material sensor;
A resonance type material sensor for detecting the material of coins by a resonance method is provided separately from the mutual induction type material sensor at a position downstream of the mutual induction type material sensor. A coin identification device characterized by the above. The coin identification device according to claim 1,
The first transmission type sensor, the reflection type magnetic sensor, and the second transmission type sensor are linearly arranged in a direction perpendicular to the transport path, and a common excitation coil for these three sensors is the first excitation sensor. And is wound between the cores of the transmission type sensor and the second transmission type sensor,
The mutual induction type material sensor also serves as a mutual induction type diameter sensor that detects the diameter of the coin by a mutual induction method, and excitation signals of a plurality of types of frequencies are present in the excitation coil of the mutual induction type material sensor. Applied to the mutual induction type material sensor and the mutual induction type diameter sensor, both operations are performed. The coin identification device according to claim 1,
The resonance-type material thickness sensor is arranged at a predetermined position of the transport passage on the coin shunting side of the transport passage,
The coin identification device, wherein the resonance-type material sensor is arranged on the opposite side of the transport passage in the direction orthogonal to the coin transport direction at a predetermined position of the transport passage. A coin identifying method using the coin identifying device according to claim 2,
Using the mutual induction type material sensor and the diameter sensor and the reflection type magnetic sensor, the material of the coin that has been transported through the transport path is detected, and the denomination of the coin is provisionally determined,
Then, based on the output signals of the resonance type material sensor and the resonance type material thickness sensor, it is confirmed that the tentatively determined coin is a true coin, and the denomination of the tentatively determined coin is determined. Coin identification method characterized by the above. In the coin identification device according to claim 1 or 2,
The reflective magnetic sensor is
The coin identification device, wherein the diameter of the core is smaller than the diameter of the core portion of the bicolor coin that has been conveyed through the conveyance path. A coin identifying method using the coin identifying device according to any one of claims 1, 2, and 5.
A reflective shape sensor and a reflective material sensor are arranged on the lower surface of the transport passage for transporting coins,
Based on the output waveform of the reflection-type shape sensor and the output waveform of the reflection-type material sensor for the coin that has been transferred through the transfer path, the rising and falling portions of the output waveform of the shape sensor Calculating the ratio between the distance between the first predetermined identical values and the distance between the second predetermined identical values at the rising and falling parts of the output waveform of the material sensor;
Using the calculated ratio as a feature amount, comparing the feature amount with a predetermined threshold value, it is identified whether or not the coin that has been transported through the transport path is a bicolor coin. A coin identifying method using the coin identifying device according to claim 2 or 5,
A third transmission type sensor of a mutual induction type diameter sensor is disposed so as to sandwich one end portion of the transport passage for transporting coins, and the mutual induction type diameter is disposed so as to sandwich the other end portion of the transport passage. Arranging a fourth transmissive sensor of the sensor;
A reflective material sensor is placed on the lower surface of the transport path,
Based on the output waveform of the mutual induction type diameter sensor and the output waveform of the reflective material sensor for the coin that has been transported through the transport path, the peak value of the reflective material sensor is determined based on the core portion of the bimetallic coin. Identifying the bimetal coin using the output signal value of the reflective material sensor at a predetermined percentage of the peak value of the output waveform of the mutual induction type diameter sensor as the material of the ring portion of the bimetal coin. A characteristic coin identification method. A coin identifying method using the coin identifying device according to any one of claims 1, 2, and 5.
A resonance type material sensor is arranged in the conveyance path for conveying coins,
Based on the output waveform of the resonance-type material sensor with respect to the coins conveyed through the transfer path, the distance between two peak values at the top of the output waveform of the resonance-type material sensor and the rise of the output waveform Calculating the ratio of the distance between predetermined identical values at the part and the falling part,
Using the calculated ratio as a feature amount, comparing the feature amount with a predetermined threshold value, it is identified whether or not the coin that has been transported through the transport path is a bicolor coin.

Images