JP2012083964A - Coin processor and coin processing method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To evaluate the unevenness on the surface of a coin with high accuracy.SOLUTION: Coins 100 are conveyed one by one and irradiated with slit pattern light penetrating a slit pattern 113 with a first region which a first color component penetrates and a second region which a second color component penetrates alternately arranged. An image of a coin irradiated with the slit pattern light is imaged by a camera 115, and a color separation unit 211 separates the imaging result into the first color component and the second color component. A height computation unit 213 generates height evaluation data from a change of a slit pattern shape in image data on the first color component and a change of a slit pattern shape in image data on the second color component, respectively, in order to obtain three-dimensional measuring data on the coin. A determination unit 206 determines authenticity, classification (denomination) and genuineness of the coin based on the three-dimensional measuring data.

Description

  The present invention relates to a coin processing apparatus and a coin processing method for evaluating a coin, and in particular, a coin processing for determining the authenticity, type (denomination), and correctness of a coin from three-dimensional measurement data for evaluating unevenness on the surface of the coin. The present invention relates to an apparatus and a coin processing method.

  2. Description of the Related Art Conventionally, a technique for obtaining a three-dimensional shape of an object from a sensing result by a magnetic sensor or an imaging result by a camera has been known, and is used for determining whether a coin is true or false, its type, and whether it is correct or not. For example, Patent Document 1 discloses a magnetic sensor in which a pair of detection magnetic pole portions detects a concave / convex shape of a coin by utilizing a change in magnetic flux.

  Further, in Patent Document 2, in order to enable discrimination of a camouflaged coin pasted with a copy of the coin itself, the occurrence frequency is counted for each output value for image data output from the image input unit, and the output value A technique for three-dimensionally recognizing the concave / convex pattern on the coin surface from the distribution of the occurrence frequency of the coin and determining the authenticity of the coin is disclosed.

JP 2001-188932 A JP-A-10-116368

  However, the conventional technique has a problem that it is not possible to highly accurately evaluate coins being conveyed at high speed. Specifically, according to Patent Document 1, for example, the presence or absence of a pearl pattern of a 500-yen coin can be detected, but since the detection portion is spot-like, it is confirmed whether the pattern covers the entire surface. It ’s difficult.

  Moreover, the thing of the said patent document 2 replaces the unevenness | corrugation of a coin with the light and shade of a reflected image, and when what copied the coin is stuck, there is no gentle inclination part on a reading surface. From the graph of the output value count result, the actual unevenness is not seen, but when the dirt of coins is generated, the shape of the graph may change and cannot be accurately determined.

  The present invention has been made to solve the above-described problems caused by the prior art, and provides a coin processing apparatus and a coin processing method for accurately evaluating the surface irregularities of a coin being conveyed. Objective.

In order to solve the above-described problems and achieve the object, the present invention provides a coin conveyance path that conveys coins one by one, and a first region that transmits a first color component to the coins on the coin conveyance path. And slit pattern light irradiating means for irradiating slit pattern light transmitted through the slit pattern in which the second regions transmitting the second color components are alternately arranged, and an image of the coin irradiated with the slit pattern light. Imaging means for imaging, three-dimensional data generation means for obtaining three-dimensional measurement data of the coin from a change in the slit pattern shape in the imaging result by the imaging means,
And determining means for determining at least one of true / false, type, and fitness of the coin based on the three-dimensional measurement data.

  Further, the present invention is the above invention, wherein the determination means extracts an edge of the three-dimensional measurement data by differential processing, and calculates an average value obtained from the extracted differential data and a standard average value of genuine coins. The authenticity of the coin is determined by comparing with the threshold obtained from the above.

  Also, in the present invention according to the present invention, the determination means determines that the coin is a genuine coin if the average value of the height data of the ring portion on the concentric circle of the coin is within a predetermined value range. It is characterized by.

  Further, in the present invention according to the above invention, the determination means obtains an average value of values indicating heights for a predetermined area of the coin, and determines that the coin is a genuine coin if the obtained average value is within a predetermined range. It is characterized by doing.

  Further, the present invention provides a storage for storing coins determined to be genuine coins by the determination means in the above invention, and reject means for transporting coins determined not to be genuine coins by the determination means. Is further provided.

  The present invention also includes a coin transport step for transporting coins one by one on the coin transport path, a first region that transmits the first color component to the coin on the coin transport path, and a second color component. A slit pattern light irradiation step of irradiating slit pattern light transmitted through slit patterns alternately arranged with second regions to be transmitted, an imaging step of capturing an image of the coin irradiated with the slit pattern light, and A three-dimensional data generation step for obtaining three-dimensional measurement data of the coin from a change in the slit pattern shape in the imaging result by the imaging means, and based on the three-dimensional measurement data, at least of true / false, type, and correctness of the coin And a determination step for determining any one of them.

  According to the present invention, the coins are conveyed one by one on the coin conveyance path, and the first area that transmits the first color component and the second area that transmits the second color component are alternately arranged. The image of the coin is imaged by irradiating the slit pattern light that has passed through the pattern, and based on the three-dimensional measurement data of the coin obtained from the change of the slit pattern shape in the imaging result, the authenticity, type, and harm of the coin Since at least one of them is determined, the unevenness on the surface of the coin can be evaluated with high accuracy.

  Further, according to the present invention, the edge of the three-dimensional measurement data is extracted by differential processing, and the average value obtained from the extracted differential data is compared with the threshold value obtained from the standard average value of genuine coins. Since the authenticity of the coin is determined, the authenticity of the coin can be determined with high accuracy.

  Further, according to the present invention, if the average value of the height data of the ring portions on the concentric circles of coins is within a predetermined value range, it is determined that the coin is a genuine coin. The accuracy can be evaluated.

  Further, according to the present invention, the average value of the values indicating the height of a predetermined area of the coin is obtained, and if the obtained average value is within the predetermined range, it is determined as a genuine coin. False can be evaluated with high accuracy.

  In addition, according to the present invention, since it includes a storage for storing coins determined to be genuine coins, and reject means for transporting coins determined to be not genuine coins by the determination means, Based on the determination result, coins can be stored and rejected.

FIG. 1 is an explanatory diagram illustrating the configuration of the main part of the coin processing device according to the present embodiment. FIG. 2 is an explanatory diagram of the three-dimensional data generation unit. FIG. 3 is a configuration diagram illustrating an outline of the internal configuration of the coin processing device according to the present embodiment. FIG. 4 is an explanatory diagram for explaining a combination of colors of the first region and the second region used in the lattice plate. FIG. 5 is an explanatory diagram for explaining the deformation of the slit pattern depending on the height of the measurement surface. FIG. 6 is an explanatory diagram for explaining the height calculation from the stripe transition. FIG. 7 is an explanatory diagram for explaining a combination of colors used in the slit pattern (part 1). FIG. 8 is an explanatory diagram for explaining a combination of colors used in the slit pattern (part 2). FIG. 9 is an explanatory diagram for explaining the control of the material of the coin and the light source. FIG. 10 is a flowchart showing the processing operation of the coin processing apparatus. FIG. 11 is an explanatory diagram for explaining creation of three-dimensional measurement data. FIG. 12 is a flowchart illustrating the three-dimensional data generation process (step S109) shown in FIG. FIG. 13 is a flowchart showing a determination process when edge extraction is used for three-dimensional measurement data. FIG. 14 is an explanatory diagram when edge extraction is used for three-dimensional measurement data. FIG. 15 is a flowchart of a determination process in the case of using height comparison by matching for three-dimensional measurement data. FIG. 16 is an explanatory diagram in a case where height comparison by matching is used for three-dimensional measurement data. FIG. 17 is a flowchart of a determination process in the case where height comparison using a profile is performed on three-dimensional measurement data. FIG. 18 is an explanatory diagram when height comparison using profiles is performed on three-dimensional measurement data.

  Exemplary embodiments of a coin processing device and a coin processing method according to the present invention will be described below in detail with reference to the accompanying drawings.

  FIG. 1 is an explanatory diagram for explaining a configuration of a main part of the coin processing device according to the present embodiment. The coin processing device according to the present embodiment includes an identification unit 34 and a control unit 16. The identification unit 34 presses the coins 100 one by one with the pressing belt 101 and transports them on the coin transport path, and images the surface of the coins with the imaging unit 110.

  The identification unit 34 includes the magnetic sensor 102, and the material determination unit 201 in the control unit 16 determines the material of the coin based on the sensing result of the magnetic sensor. The illumination adjustment unit 202 inside the control unit 16 controls the light amount of the light source 111 inside the imaging unit 110 based on the material of the coin.

  In the identification unit 34, the light source 111, the condensing lens 112, the lattice plate 113, and the projection lens 114 constitute a slit pattern light irradiation unit. The light emitted from the light source 111 is collected by the condenser lens 112, passes through the lattice plate 113, becomes slit pattern light, and is irradiated onto the coin 100 through the projection lens 114. Here, the lattice plate 113 is a slit pattern in which first regions that transmit the first color component and second regions that transmit the second color component are alternately arranged.

  The camera 115 images the reflected light of the slit pattern light from the coin 100 and stores it in the image storage unit 204 inside the control unit 16. Note that the passage timing of the coin 100 is detected by the sensor 116, and the position detection unit 203 inside the control unit 16 determines the imaging timing of the camera 115 based on the detection result of the sensor 116.

  The three-dimensional data generation unit 205 of the control unit 16 is a processing unit that calculates the three-dimensional measurement data of the coin 100 from the image data stored in the image storage unit 204. The three-dimensional data generation unit 205 includes a color separation unit 211, a density normalization processing unit 212, and a height calculation unit 213.

  The color separation unit 211 separates the image data stored in the image storage unit 204 into a first color component, a second color component, and a common color component that passes through both the first region and the second region. The density normalization processing unit 212 performs a process of normalizing by dividing the density of the first color component by the density of the common color component. Further, the density normalization processing unit 212 performs a process of normalizing by dividing the density of the second color component by the density of the common color component.

  The height calculation unit 213 generates first height evaluation data from the change in the slit pattern shape in the normalized image data of the first color component. Further, the height calculation unit 213 generates second height evaluation data from the change in the slit pattern shape in the normalized image data of the second color component. Then, the height calculation unit 213 obtains coin three-dimensional measurement data from the first height evaluation data and the second height evaluation data. As an example of such three-dimensional measurement data, an average value of the first height evaluation data and the second height evaluation data can be used.

  The determination unit 206 inside the control unit 16 compares the height calculated by the height calculation unit 213 with the determination reference value stored in the determination reference value storage unit 207, and determines whether the coin 100 is true or false. The determination result output unit 208 is a processing unit that outputs the determination result of the determination unit 206. The determination reference value is, for example, a height threshold obtained from a standard average value of genuine coins. Moreover, the average value of the data obtained by differentiating the height data calculated by the height calculation unit 213 can be compared with a threshold value obtained from the standard average value of genuine coins.

  FIG. 2 is an explanatory diagram for explaining the three-dimensional data generation unit 205. In the example shown in FIG. 2, the lattice plate 113 has a first color component of red (R), a second color component of blue (B), a first region of yellow, and a second region of cyan. It shall be. The first region and the second region transmit green (G) as a common color component.

  A yellow slit pattern and a cyan slit pattern will be described as a line width L and a space width S. The line width L and the space width S are equal, and the period T is L + S. The period T is a coin projection plane, and is 0.3 to 0.4 mm as an example. Since the yellow slits and the cyan slits are alternately arranged, in the synthesized lattice pattern, the yellow slit pattern and the cyan slit pattern have a half-cycle phase shift. Since both cyan and yellow transmit green, the green component can be used as density correction data.

  By using such a lattice plate, it is possible to simultaneously capture two slit image data and density correction data in which the irradiated slit patterns are phase-shifted from each other by a half cycle without increasing the number of times of imaging. it can.

  In this way, the identification unit 34 irradiates the coin with a slit pattern synthesized so that the slit patterns of the same frequency component (line width = space width) with two different color components have a half-cycle phase shift. By separating the color image captured by the color camera for each RGB color component, two types of slit images with different phases (inverted) and density correction data for canceling illumination unevenness, extraneous light, etc. are obtained once. It can be acquired simultaneously by imaging.

  In the example shown in FIG. 2, when the color separation unit 211 separates an image of a coin imaged by irradiating a slit pattern of cyan and yellow into RGB, the phase is inverted between the red (R) component and the blue (B) component. A slit image is obtained. As a green (G) component, an image in which density unevenness due to illumination variation is detected with little influence of the slit pattern is obtained.

  The density normalization processing unit 212 corrects the B component data and the R component data with the green (G) component data, and the height calculation unit 213 calculates from the normalized B component data and the R component data. Height evaluation data is calculated for each, and the results are averaged to obtain three-dimensional measurement data.

  FIG. 3 is a configuration diagram illustrating an outline of the internal configuration of the coin processing device according to the present embodiment. As shown in FIG. 3, the coin processing device 10 includes a substantially rectangular parallelepiped housing 12, a slot 14 for throwing coins into the inside from the outside of the casing 12, and a coin thrown into the slot 14. Are supplied to a coin storage and feeding unit 70, which will be described later, and a coin storage and feeding unit 70 that stores the coins supplied from the supply unit 20 and feeds the stored coins. The insertion port 14, the supply unit 20, and the coin storage and feeding unit 70 constitute a receiving unit for receiving coins inside the coin processing apparatus 10.

  The coin storage and feeding unit 70 is connected to a transport unit 30 that transports the coins fed from the coin storage and feeding unit 70 inside the housing 12. An identification unit 34 for identifying loss, authenticity, etc. is provided. Further, a sorting unit 32 (a part of the coin guide unit) is connected to the downstream side of the transport unit 30, and the coins transported from the transport unit 30 by the sorting unit 32 are included in the identification result of the identification unit 34. Based on the denomination or mixed state, the storage location of coins is selected.

  Specifically, a plurality of (for example, three) openings 36a, 36b, and 36c (a part of the coin guide part) are provided in the coin conveyance path 31a (a part of the coin guide part) in the sorting part 32. ing. Each opening 36a, 36b, 36c is sent to the reject coin chute 62 and the chute 32a, 32b, respectively.

  Furthermore, an opening 36d is provided further downstream of each opening 36a, 36b, 36c at the downstream end of the transport path 31a. The opening 36d communicates with the chute 32c. When the coins conveyed by the conveyance path 31a do not enter the openings 36a, 36b, 36c, the coins are conveyed to the downstream end of the conveyance path 31a and enter the opening 36d. . Coins that have entered the opening 36d are sent to the chute 32c.

  As described above, the reject coin chute 62 is connected to the sorting unit 32, and a coin that cannot be identified by the identifying unit 34 or a coin that is identified as not a normal coin by the identifying unit 34 is selected as a reject coin. 32 is sent to the reject coin chute 62 from the opening 36a. In addition, a reject coin take-out port 60 that is accessible from the outside of the housing 12 is provided at the downstream end of the reject coin chute 62, and a reject coin from the reject coin chute 62 is sent to the reject coin take-out port 60. . Thereby, the operator can take out the reject coin from the reject coin outlet 60. Further, the object sent from the coin storage and feeding unit 70 to the discharge chute 64 is also sent to the reject coin takeout port 60.

  A temporary holding unit 40 is provided below the sorting unit 32. The temporary holding unit 40 includes a plurality (for example, three) of temporary holding parts 40a, 40b, and 40c that temporarily hold coins in a denomination state or a mixed state. The coins sorted by the sorting unit 32 are temporarily reserved 40a, 40b, 40c via the chutes 32a, 32b, 32c (a part of the coin guide) corresponding to the temporary holding portions 40a, 40b, 40c, respectively. Will be sent to 40c.

  A storage unit 50 is further provided below the temporary storage unit 40. The storage unit 50 includes a plurality of (for example, three) cassettes 50a, 50b, and 50c that store coins in a denomination state or a mixed state. Coins temporarily held in the temporary holding portions 40a, 40b, and 40c are transferred to the cassettes 50a, 50b, and 50c through the chutes 42a, 42b, and 42c (part of the coin guide portion). It is sent to cassettes 50a, 50b and 50c. The temporary holding unit 40 is provided so that the entire amount can be returned when a trouble occurs during the deposit transaction.

  FIG. 4 is an explanatory diagram for explaining a combination of colors of the first region and the second region used for the lattice plate 113. When RG (yellow) and RB (magenta) are used, red (R) becomes a common color component, and a slit pattern of green (G) and blue (B) is obtained. When RG (yellow) and GB (cyan) are used, green (G) is a common color component, and a red (R) and blue (B) slit pattern is obtained. When RB (magenta) and GB (cyan) are used, blue (B) is a common color component, and slit patterns of red (R) and green (G) are obtained.

  Next, height calculation will be described with reference to FIGS. FIG. 5 is an explanatory diagram for explaining the deformation of the slit pattern depending on the height of the measurement surface. When the measurement surface is irradiated with light from the light source 111 through the lattice plate 113 and imaged by the camera 115, the stripe pattern of the lattice plate is deformed depending on the height of the measurement surface, and is captured as a modified fringe image.

FIG. 6 is an explanatory diagram for explaining the height calculation from the stripe transition. As shown in FIG. 6, the height of the surface to be measured is h, the angle between the projection direction of the fringe (slit) pattern and the observation direction (imaging direction) is θ, and the deformation (transition) amount of the fringe is ΔX. Then
ΔX = h × Tanθ
Thus, the height h of the measurement surface can be obtained from the stripe deformation amount ΔX.

  Next, a combination of colors used for the slit pattern will be described with reference to FIGS. As shown in FIG. 7, in the lattice (B1 + W) in which black (B1) and white (W) slits are combined, the black (B1) slits block all light, and the white (W) slits all. Transmits light. For this reason, when separated into RGB, information cannot be obtained from the black (Bl) slit portion, and information about all of RGB can be obtained with respect to the white (W) slit. Therefore, the three-dimensional shape can be detected from one image (one shot) corresponding to the white (W) slit, but two images (two shots) whose phases are shifted cannot be obtained. Further, the black (Bl) slit serves as a light-shielding portion and is not suitable for fine detection. Further, since there is no image information in the black (Bl) slit, the contamination cannot be detected. Therefore, it is difficult to apply the lattice (B1 + W) in which the black (B1) and white (W) slits are combined to the disclosed apparatus.

  Consider the combination of RGB slits. Fine detection is possible if there is a primary color common to the two types of slits irradiated on the entire surface of the coin, and in order to see the fouling, a primary color common to the two types of slits and another color are required. Find the ratio to detect dirt.

  When a grid (GB) that combines green (G) and blue (B) slits is used, if the imaging result is separated into RGB, fine detection is impossible because there is no common primary color, and each slit is a single color. As a result, it is impossible to detect fouling. Therefore, it is difficult to apply to the disclosed apparatus. The lattice (RG) that combines the slits of red (R) and green (G) has no common primary color, so fine detection is impossible, and since each slit is a single color, stain cannot be detected. Therefore, it is difficult to apply to the disclosed apparatus.

  Similarly, the lattice (RB) combining the slits of red (R) and blue (B) cannot be detected finely because there is no common primary color, and stains cannot be detected because each slit is a single color. . Therefore, it is difficult to apply to the disclosed apparatus.

  A combination in which any of RGB and white (W) slits are alternately arranged will be described. When using a grid (BW) that combines blue (B) and white (W) slits, if the imaging results are separated into RGB, two images with different phases for R and G cannot be obtained. Since the B component image is obtained on the entire surface, fine detection can be performed. However, since the blue (B) slit portion has only one color, the stain cannot be detected. Therefore, it is difficult to apply to the disclosed apparatus.

  Similarly, when a grid (GW) in which green (G) and white (W) slits are combined is used, it is impossible to obtain two images with different phases if the imaging results are separated into RGB. Therefore, it is difficult to apply to the disclosed apparatus. When a grid (RW) that combines slits of red (R) and white (W) is used, it is impossible to obtain two images with different phases if the imaging results are separated into RGB. Therefore, it is difficult to apply to the disclosed apparatus.

  In FIG. 8, the combination of cyan (C), magenta (M), and yellow (Y) slits is considered. When a grid (MY) in which magenta (M) and yellow (Y) slits are combined is used and the imaging result is separated into RGB, an R component is obtained for the entire target surface, and G and B have different phases. A striped image is obtained. Since information is obtained with the R component as a common primary color for the entire surface of the object, fine detection is possible, and the common primary color red (R) and another color (green (G) or blue (B)) are obtained. Therefore, it is possible to detect contamination. As described above, the grating (MY) combining the slits of magenta (M) and yellow (Y) can acquire two images with different phases, can acquire the entire image, and can perform fine detection and contamination detection. Therefore, it is suitable for the disclosed apparatus.

  Similarly, when a grid (CY) in which slits of cyan (C) and yellow (Y) are combined is used, when the imaging result is separated into RGB, a G component is obtained for the entire surface of the object, and the phase for R and B is obtained. Two striped images with different values are obtained. Since information is obtained with the G component as the common primary color for the entire surface of the object, fine detection is possible, and the common primary color green (G) and another color (red (R) or blue (B)) are obtained. Therefore, it is possible to detect contamination. In this way, the lattice (CY) that combines the slits of cyan (C) and yellow (Y) can acquire two images with different phases, acquire the entire image, and can perform fine detection and contamination detection. It is suitable for the disclosed apparatus.

  Further, when a grid (CM) in which cyan (C) and magenta (M) slits are combined is used, when the imaging result is separated into RGB, a B component is obtained for the entire surface of the object, and R and G have different phases. Two striped images are obtained. Since information is obtained with the B component as a common primary color for the entire surface of the object, fine detection is possible, and blue (B) and another color (red (R) or green (G)) that are common primary colors are obtained. Therefore, it is possible to detect contamination. As described above, the grating (CM) combining the slits of cyan (C) and magenta (M) can acquire two images with different phases and the entire image, and can perform fine detection and contamination detection. It is suitable for the disclosed apparatus.

  A combination in which one of CMY and white (W) are alternately arranged will be described. When a grid (CW) in which cyan (C) and white (W) slits are combined is used, an image of a slit pattern can be obtained for only the R component when the imaging result is separated into RGB. In addition, images of the G component and the B component are obtained for the entire surface of the object. Since the common primary colors blue (B) and green (G) can be obtained, fine detection can be performed, and since the ratio of blue (B) and green (G) can be obtained, contamination can also be detected. However, it is a problem that two images having different phases cannot be obtained, and it is difficult to apply to the disclosed apparatus.

  Similarly, when a grid (MW) in which magenta (M) and white (W) slits are combined is used, it is impossible to obtain two images having different phases when the imaging results are separated into RGB. Therefore, it is difficult to apply to the disclosed apparatus. When a grid (RW) that combines yellow (Y) and white (W) slits is used, if the imaging results are separated into RGB, two images with different phases cannot be obtained. It is difficult to apply to the disclosed apparatus.

  Next, control of the type of coin and the light emission amount of the light source 111 will be described. Since the reflectivity varies depending on the type of coin, especially the material, the brightness of the light source 111 can be lowered when imaging a coin with a high reflectivity, and the brightness of the light source 111 can be increased when capturing a coin with a low reflectivity. desirable. Specifically, as shown in FIG. 9, as a result of the discrimination by the material discriminating unit 201, the amount of light emission is set to level 1 for a copper-based or aluminum coin (100 yen, 50 yen, 1 yen), and copper or brass For the coins (5 yen, 10 yen) of the system, the light emission amount is set to level 2 which is larger than the level 1, and the nickel brass coin (500 yen) is set to level 3, which is another level.

  Next, the processing operation of the coin processing apparatus 10 will be described. FIG. 10 is a flowchart for explaining the processing operation of the coin processing apparatus 10. For the sake of explanation, the processing of one coin will be taken as an example. The coin processing device 10 first transports one coin (step 101), and identifies the type of coin by the material discrimination unit 201 (step 102).

  Thereafter, the position detection unit 203 detects the arrival of a coin (step 103), and performs pre-imaging before the coin passes, that is, before the coin enters the imaging range of the camera 115 (step 104). The illumination adjustment unit 202 adjusts the amount of light emitted from the light source 111 based on the type of coin, the brightness state in the image obtained by the pre-imaging, and controls the amount of light emitted from the slit pattern (step 105). The slit pattern light is irradiated (step 106).

  Thereafter, the camera 115 captures the entire surface of the coin at the timing when the coin enters the imaging range of the camera 115 (step 107), and stores the captured coin in the image storage unit 204. The color separation unit 211 reads an image from the image storage unit 204 and separates it into RGB (step 108).

  The density normalization processing unit 212 normalizes two-color image data that is a slit pattern having different phases among the separated RGB, and the height calculation unit 213 determines from the normalized data. Height calculation is performed to obtain three-dimensional measurement data (step 109). The determination unit 206 compares the three-dimensional measurement data with the determination reference value stored in the determination reference value storage unit 207 to determine the authenticity of the coin, and the determination result output unit 208 outputs the determination result (step 110). ), The process is terminated.

  FIG. 11 is an explanatory diagram for explaining creation of three-dimensional measurement data. FIG. 11 illustrates a case where cyan and yellow slit patterns are used. When the color stripe image captured by the camera 115 is separated into RGB, blue (B) stripe image data, green (G) image data, and red (R) stripe image data are obtained. Blue (B) stripe image data, green (G) image data, and red (R) stripe image data are density distribution data, respectively.

  The blue (B) fringe image data is normalized by dividing by the green (G) image data to obtain normalized blue (B) fringe image data. Similarly, the red (R) stripe image data is normalized with the green (G) image data to obtain normalized red (R) stripe image data. And height is calculated | required from the transition of the fringe of normalization blue (B) fringe image data, and blue (B) height evaluation data are obtained. Similarly, the height is obtained from the transition of the stripes in the normalized red (R) stripe image data to obtain red (R) height evaluation data. The blue (B) height evaluation data and the red (R) height evaluation data are averaged to obtain three-dimensional measurement data.

  FIG. 12 is a flowchart for explaining in detail the three-dimensional data generation process (step 109) shown in FIG. The three-dimensional data generation unit 205 performs the process of FIG. 12 on the two-color fringe image data, which are slit patterns having different phases, among the RGB separated by the color separation unit 211.

  The density normalization processing unit 212 of the three-dimensional data generation unit 205 acquires a fringe image (step 201), creates fringe image density correction data from the common color component data (step 202), and performs density correction (step 202). Step 203).

  The height calculation unit 213 performs two-dimensional Fourier transform on the density-corrected fringe image data (step 204), performs frequency filtering that extracts only the fringe component (step 205), and performs two-dimensional inverse processing. Perform Fourier transform. Thereby, the phase value is obtained (step 206), and then the phase difference from the reference height is obtained (step 207). Thereafter, the sensitivity based on the difference in optical path length at the time of projection of the slit is corrected (step 208), the corrected phase difference is converted into a height value (step 209), and the processing for one coin is completed.

  The determination process (step 110) using the three-dimensional measurement data shown in FIG. 10 uses a profile which is a graph shape of the average value of the heights of the ring portions on the concentric circles by edge extraction, height comparison by matching. This can be done by comparing heights.

  FIG. 13 is a flowchart showing a determination process when edge extraction is used for three-dimensional measurement data. The determination unit 206 takes in the three-dimensional measurement data (step 301). The numerical value of the captured three-dimensional measurement data indicates the height of the coin surface at that position. The determination unit 206 cuts out the entire surface of the coin as a target area (step 302). Thus, the outline of a coin can be detected by discriminating with the edge information of the entire area of the coin. The determination unit 206 performs edge extraction processing (step 303) on the cut out area. Edge extraction uses primary differentiation, secondary differentiation, or the like. As an example, Kirsch, Laplacian, or the like can be used.

  The determination unit 206 calculates an evaluation value used for determination from the extraction processing result (step 304). The average value of the extracted edges can be used as the evaluation value. The determination unit 206 compares the evaluation value with the determination reference value stored in the determination reference value storage unit 207 (step 305). As a result, when the coin is a forged coin (step 306, Yes), the determination unit 206 performs a reject process of discharging the coin to the reject coin outlet 60 (step 307), and ends the process. On the other hand, when the coin is not a forged coin (step 306, No), the determination unit 206 counts the coin, performs a storing process (step 308), and ends the process.

  FIG. 14 is an explanatory diagram for explaining a case where edge extraction is used for three-dimensional measurement data. FIG. 14 shows a specific example of the three-dimensional measurement data and a distribution example of the average value of the edges obtained from the three-dimensional measurement data.

  As shown in the specific example of the three-dimensional measurement data, the numerical value of the data indicates the height at that position on the coin surface. If the entire area of the coin is targeted, the outline of the coin becomes the object of determination. When discriminating by limiting the part of the coin, it is only necessary to obtain the outline and center of the coin, detect the rotational position, and apply the area mask.

  In the specific example of the average value of the edge, the distribution of the average value of the edge of the counterfeit coin is smaller than the distribution of the edge of the true coin. By using a value that separates this distribution as a determination threshold, that is, a determination reference value, forged coins can be identified.

  FIG. 15 is a flowchart illustrating a determination process in a case where height comparison by matching is used for three-dimensional measurement data. The determination unit 206 takes in the three-dimensional measurement data (step 401). In the captured three-dimensional measurement data, the numerical value of the data indicates the height at that position on the coin surface. The determination unit 206 performs noise processing on the three-dimensional measurement data using, for example, a smoothing filter (step 402). Thereby, a noise component generated at the boundary portion of the outer diameter of the coin can be removed.

  The determination unit 206 performs position correction processing on the data from which noise has been removed (step 403). In the position correction process, the position of the coin is detected and corrected to the same reference position as the template. Specifically, coin contour detection, center detection, center alignment, rotational alignment, tilt correction, and offset correction are performed in order. The contour detection can be detected at the rising edge of the outer diameter or the inner peripheral portion of the rim.

  The determination unit 206 cuts out the target area for the data subjected to the position correction (step 404). A mask may be applied when the area is limited to an area. The extracted target area is compared with the template (step 405), and an evaluation value is calculated (step 406). In comparison with the template, a difference from the template is obtained for each pixel. The template is preferably average data of a number of current currencies with position correction. In calculating the evaluation value, the average value may be obtained by dividing the total value of the differences from the template by the number of pixels.

  The determination unit 206 compares the evaluation value with the determination reference value stored in the determination reference value storage unit 207 (step 407). As a result, when the coin is a forged coin (step 408, Yes), the determination unit 206 performs a reject process of discharging the coin to the reject coin outlet 60 (step 409), and ends the process. On the other hand, when the coin is not a forged coin (step 408, No), the determination unit 206 counts the coin, performs storage processing (step 410), and ends the processing.

  FIG. 16 is an explanatory diagram in a case where height comparison by matching is used for three-dimensional measurement data. FIG. 16 shows a specific example of the three-dimensional measurement data, an example of the center obtained from the rise of the outer diameter, an example of the area cut-out mask, and an example of the distribution of the average value of differences from the template.

  As shown in the specific example of the three-dimensional measurement data, the height is indicated by the concentration distribution with respect to the surface of the coin. From the rise of the outer diameter of the coin, the center position of the coin can be obtained. By obtaining the center position, correcting the rotational position and applying the area cutout mask, it is possible to select a characteristic area of the coin as a determination target.

  In the distribution example of the average value of the difference from the template, the distribution of the average value of the difference from the template of the counterfeit coin is larger than that of the genuine coin. By using a value that separates the distribution as a determination threshold value, that is, a determination reference value, forged coins can be identified. As a matching method, a method using a density image can be used similarly.

  FIG. 17 is a flowchart of a determination process in the case where height comparison using a profile is performed on three-dimensional measurement data. The determination unit 206 takes in the three-dimensional measurement data (step 501). The captured three-dimensional measurement data is data in which each density value on the surface of the coin corresponds to the height at that position. The determination unit 206 performs noise processing on the three-dimensional measurement data (step 502). Thereby, a noise component generated at the boundary portion of the outer diameter of the coin can be removed.

  The determination unit 206 performs center position detection processing on the data from which the noise has been removed (step 503). In this center position detection process, the position of the coin is detected and corrected to the same reference position as the template. Specifically, the coin outline detection and the center detection are performed in this order. The contour detection can be detected at the rising edge of the outer diameter or the inner peripheral portion of the rim.

  The determination unit 206 calculates an evaluation value for the data for which the center position has been detected (step 504). Specifically, the total value (average value, profile) of the height data of the ring portions taken at equal intervals on the concentric circle is obtained, and the height value of the ring portion of the characteristic part is obtained. For example, a pearl-like pattern provided on the coin surface is used as the characteristic part.

  The determination unit 206 compares the evaluation value with the determination reference value stored in the determination reference value storage unit 207 (step 505). As a result, when the coin is a forged coin (step 506, Yes), the determination unit 206 performs a reject process for discharging the coin to the reject coin outlet 60 (step S507), and ends the process. On the other hand, when the coin is not a forged coin (step S506, No), the determination unit 206 counts the coins, performs a storing process (step S508), and ends the process.

  FIG. 18 is an explanatory diagram when height comparison using profiles is performed on three-dimensional measurement data. FIG. 18 shows a specific example of three-dimensional measurement data, a specific example of an average value of height data, and a distribution example of pearl height.

  As shown in the specific example of the three-dimensional measurement data, the numerical value of the data indicates the height at that position on the coin surface. Finding the center of this coin and taking the average value of the height data of each ring part on the concentric circle, the average value of the ring part height at the radius position where the outer edge of the coin and the pearl pattern are located is a profile The peak is getting above. Among these, the average value of the height in the ring part which becomes a pearl pattern part is used as the evaluation value as the pearl height.

  In the distribution example of the pearl height, the distribution of counterfeit coins is smaller than the distribution of genuine coins. By using a value that separates the distribution as a determination threshold value, that is, a determination reference value, forged coins can be identified.

  As described above, in this embodiment, the coins are conveyed one by one, and the first region that transmits the first color component and the second region that transmits the second color component are alternately arranged. The slit pattern light that has passed through the slit pattern is irradiated. Then, an image of the coin irradiated with the slit pattern light is captured, and the imaging result is separated into the first color component and the second color component, and the change of the slit pattern shape in the image data of the first color component First height evaluation data is created, second height evaluation data is generated from a change in the slit pattern shape in the image data of the second color component, and the first height evaluation data and the second height evaluation data are generated. 3D measurement data of coins is obtained from height evaluation data. Since two types of slit pattern images having different phases are obtained by one imaging, it is possible to obtain unevenness information of a moving coin without being affected by fluctuations in the conveyance speed.

  A slit pattern in which the first color component is green and the second color component is blue, and the first region and the second region are yellow and magenta regions that transmit red in common is preferable. A slit pattern in which the first color component is red and the second color component is blue, and the first region and the previous two regions are yellow and cyan regions that transmit green in common is also suitable. The first color component may be red, the second color component may be green, and the first area and the second area may be a slit pattern that is a magenta and cyan area that transmits blue in common.

  In the present embodiment, the separating unit separates the imaging result obtained by the imaging unit into a first color component, a second color component, and a common color component that the first region and the second region transmit in common, The three-dimensional data generation means uses the image data of the common color component for density correction of the image data of the first color component and the image data of the second color component. For this reason, density unevenness and illumination unevenness can be corrected and highly accurate determination can be performed. Pattern matching from image information is also possible.

  Further, in this embodiment, the timing at which the coin enters the imaging range is detected, and the irradiation light amount of the slit pattern light is controlled based on the received light amount of the common color component that is transmitted through the first region and the second region in common. Thus, illumination can be performed according to the type and material of the coin. Furthermore, by performing pre-imaging before the image of the coin and controlling the light quantity of the slit pattern light, the coin can be appropriately illuminated to obtain the three-dimensional measurement data.

  Based on the three-dimensional measurement data obtained in this way, it is possible to improve the determination accuracy by determining the authenticity, type (denomination), and correctness of the coin. That is, since the feature of the edge shape of the concavo-convex pattern in the coin can be detected from the three-dimensional measurement data, the authenticity determination can be performed with high accuracy. In addition, the accuracy of discriminating damage (distorted coins, worn coins, or damaged ones) is improved by the three-dimensional measurement data on the concave / convex pattern of the coins.

  As described above, the coin processing device and the coin processing method according to the present invention are useful for acquiring three-dimensional measurement data for evaluating the unevenness of the surface of the coin and for determining the coin.

DESCRIPTION OF SYMBOLS 10 Coin processing apparatus 12 Case 14 Slot 16 Control part 20 Supply part 30 Conveyance part 31a Conveyance path 32a, 32b, 32c Chute 32 Sorting part 34 Identification part 36a, 36b, 36c, 36d Opening part 40 Temporary reservation part 40a, 40b , 40c Temporary holding part 42a, 42b, 42c Chute 50a, 50b, 50c Cassette 50 Storage part 60 Reject coin outlet 62 Reject coin chute 64 Discharge chute 70 Coin storage and delivery part 100 Coin 101 Belt 102 Magnetic sensor 110 Imaging unit 111 Light source 112 Condensing lens 113 Slit pattern 113 Grid plate 114 Projection lens 115 Camera 116 Sensor 201 Material discrimination unit 202 Illumination adjustment unit 203 Position detection unit 204 Image storage unit 205 Three-dimensional data generation unit 206 Determination Unit 207 determination reference value storage unit 208 determination result output unit 211 color separation unit 212 density normalization processing unit 213 calculation unit

Claims (6)

A coin transport path for transporting coins one by one;
Irradiating the coins on the coin transport path with slit pattern light that has passed through a slit pattern in which first regions that transmit the first color component and second regions that transmit the second color component are alternately arranged. Slit pattern light irradiation means to perform,
An imaging means for capturing an image of the coin irradiated with the slit pattern light;
Three-dimensional data generation means for obtaining three-dimensional measurement data of the coin from the change in the slit pattern shape in the imaging result by the imaging means;
A coin processing apparatus comprising: a determination unit that determines at least one of true / false, type, and fitness of the coin based on the three-dimensional measurement data.   The determination means extracts the edge of the three-dimensional measurement data by differential processing, compares an average value obtained from the extracted differential data with a threshold value obtained from a standard average value of genuine coins, and The coin processing apparatus according to claim 1, wherein authenticity of the coin is determined.   The said determination means determines with a genuine coin if the average value of the height data of the ring part on the concentric circle of the said coin exists in the range of a predetermined value, It is characterized by the above-mentioned. Coin processing equipment.   The said determination means calculates | requires the average value of the value which shows height about the predetermined area | region of the said coin, and will determine with a genuine coin if the calculated | required average value is in a predetermined range. The coin processing apparatus according to any one of the above.   The storage device for storing coins determined as genuine coins by the determination means, and reject means for transporting coins determined as non-authentic coins by the determination means. The coin processing apparatus as described in any one of 1-4. A coin transport step for transporting coins one by one on the coin transport path;
Irradiating the coins on the coin transport path with slit pattern light that has passed through a slit pattern in which first regions that transmit the first color component and second regions that transmit the second color component are alternately arranged. Slit pattern light irradiation step to perform,
An imaging step of capturing an image of the coin irradiated with the slit pattern light;
A three-dimensional data generation step of obtaining three-dimensional measurement data of the coin from a change in the slit pattern shape in the imaging result by the imaging means;
And a determination step of determining at least one of true / false, type, and fitness of the coin based on the three-dimensional measurement data.

Images