JPH07210720A - Coin recognition device - Google Patents

Abstract

PURPOSE:To accurately perform coin recognition through simple constitution by binarizing image data and extracting discriminating and comparing the decision data with previously registered patterns and discriminating the denomination of a coin and whether or not the coin is genuine. CONSTITUTION:An image input part 3 has a function which generates an image by converting the intensity of reflected light of the coin 7 read by a read part 2 from analog to digital, writes it in a storage part consisting of a frame memory, etc., and extracts the external diameter of the coin 7 and its center point from the image. A binarization part 4 sets a threshold value for binarization based on the image data found by the image input part 3 and compares the image data with the threshold value of output the binarized image data. A decision part 5 compares the binarized data from the binarization part 4 with the registered patterns stored in a registered pattern storage part 6 and decides the denomination of the coin 7 and whether the coin is genuine. Here, the registered pattern storage part 6 is a storage part consisting of a ROM, RAM, or magnetic disk, etc.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a coin recognition device for discriminating the type and authenticity of coins.

[0002]

2. Description of the Related Art In financial institutions and the like, automated equipment such as an automatic transaction machine (ATM) is widely used for automating deposits and savings and transfer processing. In such an automated device, when a customer inserts a coin, the type and authenticity of the coin is recognized. This processing is performed by the coin discriminating unit in the automated device, and the recognition method is as follows.

For example, as disclosed in Japanese Examined Patent Publication No. 63-67714, a method of recognizing the outer diameter or material of a coin, or disclosed in Japanese Examined Patent Publication No. 3-63780, the thickness of a coin. Depending on the size, there is something that recognizes it. Further, as a device for detecting the outer diameter of a coin, for example, there is one described in Japanese Patent Publication No. 2-48951, and further, as a device for detecting the thickness of a coin, 4-1.
Some are disclosed in Japanese Patent No. 2518.

However, in the conventional coin recognition device as described above, when the outer diameter and the material of the coins are similar to each other, for example, foreign coins of 500 won coins and 500 yen coins are separated from each other. Diameter and material are almost similar, thickness is 1
Since the difference is only about 00 μm, a thickness detection sensor or a complicated determination circuit is required to detect this difference, which causes a cost increase. In addition, the material sensor changes its detection characteristics under the influence of temperature, so it requires a complicated auxiliary circuit to correct it.
This was also the cause of the cost increase. Therefore, there is a demand for a coin recognizing device that can accurately recognize coins at low cost.

[0004]

In order to solve the above-mentioned problems, a coin recognition device of the present invention comprises a light source for illuminating the surface of a coin and a reading section for reading the intensity of specular reflection light from the surface of the coin. Have. Further, based on the output from the reading unit, while having an image capturing unit that extracts the central point in the image data of the coin, set a threshold value for the specular reflection light intensity of the coin surface, and this threshold value and its reflection light intensity According to the comparison, each pixel in the image data of the coin has a binarizing unit that represents the binarized data. Furthermore, a determination unit is provided that compares the distribution state of the binarized data existing on the concentric circles in the image data of the coin with a preset registered pattern to determine the type and authenticity of the coin.

[0005]

In the coin recognition device of the present invention, the light source illuminates the surface of the coin. The specular reflection light intensity of the coin is read by the reading unit. The image capturing unit extracts the center point in the image data of the coin based on the intensity of the reflected light from the reading unit. The binarization unit sets a threshold value for the regular reflection light intensity read by the reading unit, compares the threshold value with the regular reflection light intensity, and converts each pixel in the image data of the coin into binary data. Represent The determination unit detects the distribution state of the binarized data existing on the concentric circles in the image data of the coin, compares the distribution state with a preset registered pattern, and determines the type and authenticity of the coin. .

[0006]

Embodiments of the present invention will now be described in detail with reference to the drawings. << First Embodiment >> FIG. 1 is a structural explanatory view showing a first embodiment of a coin recognition device of the present invention. The apparatus shown in the figure includes a light source 1, a reading unit 2, an image capturing unit 3, a binarization unit 4, a determination unit 5, and a registration pattern storage unit 6.

Further, 7 is a coin that has been inserted, and 8
Is a transport path for transporting the inserted coin 7 in the direction of arrow 9. The transport path 8 has a slit-shaped window 10 so that the pattern on the surface of the coin 7 can be seen from the lower side. Below the window 10 is a half mirror 11
Is installed diagonally, and the reading unit 2 is located below it. The reading unit 2 is composed of a line image sensor such as a CCD, and the coin 7 is passed through the half mirror 11.
It is configured to receive specularly reflected light from the surface. The light source 1 is installed beside the half mirror 11 via a slit 12. The light source 1 is composed of a line light source such as a fluorescent lamp or a line filament lamp.

That is, with these half mirrors 11 and slits 12, the irradiation direction from the light source 1 to the surface of the coin 7 is on the vertical axis of the surface of the coin 7, and the reading unit 2 is located on the vertical axis, and the coin 7 It is configured to receive specularly reflected light from the surface. The specularly reflected light is the reflected light received by the reading sensor when the light source and the reading sensor are installed symmetrically with respect to the vertical axis with respect to the surface of the coin 7.

Then, the image capturing section 3 A / D-converts the reflected light intensity of the coin 7 read by the reading section 2 to create an image, and writes the image in a storage section such as a frame memory (not shown), It has a function of extracting the outer diameter of the coin 7 and its center point from this image. The binarization unit 4 calculates 2 from the image data obtained by the image capturing unit 3.
The threshold value for performing the binarization is set, and the binarized image data is output by comparing the threshold value with the image data. The determination unit 5 has a function of comparing the binarized data from the binarization unit 4 with the registered pattern stored in the registered pattern storage unit 6 to determine the denomination and authenticity of the coin 7. ing. The registration pattern storage unit 6 is a ROM
And a RAM or a magnetic disk device.

Next, the operation of the coin recognizing device having the above structure will be described. Now, when the coin 7 is inserted into the coin recognizing device, the light source 1 is turned on and the light passing through the slit 12 is reflected by the half mirror 11. Then, the lower surface of the coin 7 is illuminated.
The light specularly reflected from the surface of the coin 7 passes through the half mirror 11 and enters the reading unit 2.

Here, the pattern of the surface of the coin 7 is such that the uneven pattern of the surface is clearly extracted by the principle described below. Figure 2
It is this principle explanatory drawing. First, in FIG. 2A, the coin 7
The reflected light 102 travels in the direction opposite to the direction of the irradiation light 101 in order to irradiate the flat part of the light. Therefore, the reflected light 10
2 passes through the half mirror 11 and enters the reading unit 2. On the other hand, in the case of FIG. 2B, the reflected light 102 is directed in a direction different from the direction of the irradiation light 101, because the unevenness is irradiated to the upper left portion in the drawing. Therefore, the reflected light 102 is in a direction different from the direction of the half mirror 11 and therefore does not enter the reading unit 2.

Further, in the case shown in FIG. 2C, since the flat portion of the apex of the unevenness is irradiated, the reflected light 102 goes in the direction opposite to the irradiation light 101. Therefore, the reflected light 102 passes through the half mirror 11 and enters the reading unit 2 as in the case of FIG. Further, in the case shown in FIG. 2D, the reflected light 102 is directed in a direction different from that of the irradiation light 101, because the concave and convex portions on the left side in the drawing are irradiated. Therefore, the reflected light 102 is the same as in FIG.
Does not enter.

In this way, the pattern of the surface of the coin 7 becomes bright by the specular reflection light when the surface of the coin 7 is flat, while the obliquely inclined portion becomes a shadow and appears dark. Therefore, the uneven pattern on the surface of the coin 7 can be read.

FIG. 3 shows a state in which an image is written in a storage unit (not shown). In the figure, the portion 103 is image data to the storage unit. Then, the image 104 of the white portion in the image data 103 is the image of the coin 7, and the uneven pattern exists therein. Here, the image 10 in the xy direction
The left and right end points X _s and X _{e of 4} are detected, and the upper and lower end points Y _s and Y _e of the image 104 in the y direction are detected.
From these data, the outer diameter of the coin 7 can be extracted. Further, the center point (X _c , Y _c ) is extracted from these data. The outer diameter data and the center point coordinates (X _c , Y
_c ) is output to the determination unit 5. Then, the determination unit 5 uses the outer diameter data to determine the denomination of the coin 7, and such an operation will be described later.

The number of images written in the storage unit (not shown) is 2
Output to the digitization unit 4. The binarization unit 4 reads the image written in the storage unit and extracts the density occurrence frequency representing the gray value. Then, in order to binarize the image of the coin 7, it is necessary to determine an appropriate threshold value so as not to be caught by fluctuations in output due to dirt, rust, etc. for each coin 7 and changes in the line sensor over time. desirable. Therefore, in this embodiment, in order to determine the threshold value, the average value of the concentration distribution is calculated and this is determined as the optimum threshold value.

FIG. 4 is a diagram showing the operation of the binarization unit 4. That is, the vertical axis 105 in FIG. 4 represents the density occurrence frequency, and the horizontal axis 106 represents the density of the target image. In this case, the average value of the densities is determined as the threshold value 107, and when the distribution shape of the density occurrence frequency has two peaks, the valley portion of the peaks may be determined as the threshold value 107. Whether the threshold value 107 is obtained by any one of these methods is empirically obtained depending on the nature of the coin 7 actually targeted.

The image binarized in this manner is output to the determination unit 5. The determining unit 5 first reads out the binary image output from the binarizing unit 4 based on the center point (X _c , Y _c ) output from the image capturing unit 3 and outputs the center point (X
_The black (white) bits existing at the distance from _c , Y _c ) are counted, and the frequency of occurrence of the black bit or the white bit of the binary image with respect to the distance from the center point (X _c , Y _c ) is calculated. Create a black (white) bit distribution sequence to represent.

Since the inserted coin 7 is a circular object, the directionality of the uneven pattern is not constant at the actual reading stage. Therefore, it is desirable to extract the discrimination data that does not change the directionality of the uneven pattern. Therefore, the determination unit 5 obtains a black (white) bit distribution sequence existing at a distance from the center point of the coin 7 and makes a determination based on this.

FIG. 5 is a diagram showing the operation of such a determination unit 5. The horizontal axis 108 in FIG. 5 indicates the center point (X _c , Y
_c )), and the vertical axis 107 indicates the frequency of black (white) bit occurrence. That is, the horizontal axis 108 indicates the radius of the concentric circles. Therefore, the black (white) bit distribution 109, that is, the distribution of black (white) bits representing the frequency of occurrence of black bits or white bits of the binary image with respect to the distance from the center point (X _c , Y _c ) is uneven. It will have a constant shape regardless of the directionality of.

The determination unit 5 determines the denomination of the coin 7 based on the outer diameter data output from the image capturing unit 3.
That is, using the outer diameter data of the coin 7, it is determined whether the coin 7 is domestic 500 yen, 100 yen, 50 yen, 10 yen, 5 yen, or 1 yen. Also, with foreign coins,
If data other than the outer diameter of these domestic coins is extracted, this is excluded.

When the denomination of the coin 7 is discriminated in this way, the outer diameter data is the most approximate of the reference distributions of some black (white) bits stored in the registered pattern storage unit 6 in advance. The reference distribution pattern of the black (white) bits is extracted, and the registered pattern of the reference is compared with the black (white) bit distribution 109 shown in FIG. 5 to determine the authenticity of the target coin. Therefore, even if a foreign coin is close to the outer diameter of a domestic coin, the unevenness pattern on the surface of the coin 7 is used to determine the authenticity.
Judgment can be performed with high accuracy. Moreover, since the authenticity determination is performed based on the result of the denomination determination based on the outer diameter data, unnecessary determination processing can be omitted, the processing can be simplified, and the recognition processing can be speeded up. Moreover, since the outer diameter data can be obtained from the image data of the coin 7, no other sensor or the like is needed, and the configuration can be simplified and the cost can be reduced.

In the first embodiment, the light emitted from the light source 1 passes through the slit 12 and the half mirror 11 and the coin 7
However, the specularly reflected light is transmitted through the half mirror 11 and is incident on the reading unit 2. However, other configurations may be used, and this embodiment will be described below. Figure 6
[Fig. 6] is an explanatory view of the configuration. In the figure, 13 is an optical lens, such as an aspherical lens or a self-fog lens. The focal position of this optical lens 13 is coin 7
And the light receiving surface of the reading unit 2. Therefore, the light source 1
The light from is reflected by the half mirror 11 to illuminate the surface of the coin 7, and the reflected light is focused by the optical lens 13 and is incident on the reading unit 2. Since the subsequent operation is the same as that of the above-mentioned embodiment, the description thereof will be omitted.

Thus, in the first embodiment, the coin 7
The image data of the coin 7 obtained by detecting the regular reflection light on the surface is binarized, and the type and authenticity of the coin 7 are determined from the distribution state of the binarized data existing on the concentric circles in the image data of the coin 7. Since the discrimination is made, the coin can be accurately recognized without being influenced by the rotation angle of the coin 7. In addition, the readable length read by the reading unit 2 is
The coin 7 with the largest diameter among the coins 7 to be recognized
Since it is set to be equal to or larger than the diameter of, all the necessary images can be extracted without being affected by the displacement of the target coin 7.

<Second Embodiment> FIG. 7 is a block diagram showing a second embodiment of the present invention. In the figure, 7 is a coin that has been inserted, and this coin 7 has a transport path 8 having a window 10.
The upper part is conveyed in the direction of arrow 9 as in the first embodiment. In addition, below the transport path 8, through the window 10,
Two light sources 1 for illuminating the surface of the coin 7 from two directions
4, 15 are installed. These light sources 14 and 15 are
The irradiation angles on the surface of the coin 7 are set to be equal to each other, and the light sources 14 and 15 are configured to be alternately turned on.

Further, in the second embodiment, the reading unit 1
6, a binarization unit 17, a determination unit 18, a registration pattern storage unit 19
Is equipped with. The reading unit 16 includes a line image sensor and the like, and reads the reflected light emitted from the light sources 14 and 15 to the coin 7. The binarizing unit 17 relatively compares the reflected light intensities of the light source 14 and the light source 15 read by the reading unit 16 and outputs a binary image. Details of this comparison will be described later.
The determination unit 18 compares the binarized data output from the binarization unit 17 with the registered pattern stored in the registered pattern storage unit 19 to determine the denomination and authenticity of the coin 7. Is.

Next, the operation of the second embodiment will be described. now,
When the coin 7 is inserted, first, the light source 14 is turned on to illuminate the surface of the coin 7. The reflected light from the surface of the coin 7 enters the reading unit 16. After this, the light source 15 lights up,
The surface of the coin 7 is illuminated. The reflected light from the surface of the coin 7 enters the reading unit 16. Here, on the flat surface of the coin 7, the irradiation light amounts of the light sources 14 and 15 are adjusted so that the reflected light outputs of the light sources 14 and 15 become equal to each other. For the pattern on the surface of the coin 7, the characteristic data based on the uneven pattern on the surface is clearly extracted based on the principle described below.
That is, the reflection intensity of the irradiation light of the light source 14 and the irradiation light of the light source 15 greatly differs depending on the shape of the surface of the coin 7.

FIG. 8 is an explanatory view of this principle. First, FIG.
As shown in (a), on the flat surface of the coin 7, from the relationship between the irradiation angle of the irradiation light of the light source 14 and the irradiation light of the light source 15 and the inclination angle of the surface of the coin 7, the magnitude relationship of the reflected light intensity by each light source is determined. , The irradiation angle of the light source 14 and the irradiation angle of the light source 15 with respect to the surface of the coin 7 are equal, and thus the reflected light amounts are substantially equal.

Next, as shown in FIG. 8 (b), when the projection surface of the coin 7 is tilted at an upward slope, the relationship between the irradiation angle of the light source 15 and the tilt angle of the surface of the coin 7 is close to regular reflection. Therefore, the intensity of the reflected light from the light source 15 increases, whereas the relationship between the irradiation angle of the light source 14 and the inclination angle of the protrusion surface is in the opposite direction to the regular reflection, so the intensity of the reflected light from the light source 14 decreases. .

Further, as shown in FIG. 8C, on the flat surface on the protruding surface of the coin 7, as in the case of FIG. 8A, the light source 1
From the relationship between the irradiation angle of the irradiation light of No. 4 and the irradiation angle of the irradiation light of the light source 15 and the inclination angle of the surface of the coin 7, the magnitude relationship of the reflected light intensity by each light source is determined by the irradiation angle of the light source 14 with respect to the surface of the coin 7 and the light source. Since the irradiation angles of 15 are equal, the amounts of reflected light are almost equal.

Further, as shown in FIG. 8 (d), when the projection surface of the coin 7 is inclined with a downward slope, the relationship between the irradiation angle of the light source 14 and the inclination angle of the surface of the coin 7 is close to regular reflection. While the intensity of the reflected light from the light source 14 increases, on the contrary, the relationship between the irradiation angle of the light source 15 and the inclination angle of the projection surface is opposite to that in the regular reflection, so the intensity of the reflected light from the light source 15 decreases To do.

In this way, by relatively comparing the reflected light intensities of the light source 14 and the light source 15, the shape of the surface of the coin 7 can be easily detected. That is, when the two reflected light intensities are substantially equal, it is a flat surface, and when one of them is larger, it is an inclined surface, and the inclination direction of the protruding surface of the coin 7 can be extracted depending on the magnitude relationship between the two.

Next, the operation of the binarization unit 17 will be described. FIG. 9 is a diagram showing the operation. Vertical axis 201 of FIG. 9
Is the reflected light intensity when the light source 14 is turned on, and the horizontal axis 2
Reference numeral 02 represents the reflected light intensity when the light source 15 is turned on. In this space, the relationship between the respective reflected light intensities input as elements can be represented by a point C.

For example, as described above, when the surface of the coin 7 is flat, the reflected light intensities of the light source 14 and the light source 15 become substantially equal, and the diagonal line 2 rising to the right in FIG.
A point C appears around 03. When the surface of the coin 7 is inclined with the upward slope of the protrusion, the reflected light intensity when the light source 15 is turned on becomes relatively higher than the reflected light intensity when the light source 14 is turned on. Point C appears at. When the surface of the coin 7 is inclined with the descending slope of the protrusion, the reflected light intensity when the light source 14 is turned on becomes relatively higher than the reflected light intensity when the light source 15 is turned on.
A point C appears in the area indicated by 205.

In this way, the binarization unit 17 outputs the output "1" when the point C appears in the area 205 and the area 204 in the space of FIG. 9 and outputs "0" otherwise. Image data is binarized.

The binarization unit 17 detects the outer diameter of the coin 7 from the output of either the reflected light intensity when the light source 14 is turned on or the reflected light intensity when the light source 15 is turned on. That is, the maximum detection length detected by the reading unit 16 in the sub-scanning direction (the direction orthogonal to the arrow 9 in the conveying direction) on the surface of the coin 7 and the reading unit 16 in the conveying direction of the coin 7 (the direction parallel to the arrow 9) are The outer diameter of the coin 7 is detected from the detected maximum detection length. Also, from these data, coins 7
Extract the coordinates of the center point of. Then, the center point coordinates and the outer diameter data are output to the determination unit 18.

The determination unit 18 determines the binary image output from the binarization unit 17 and the registration pattern stored in advance in the registration pattern storage unit 19 as the center point coordinates output from the binarization unit 17. Based on this, comparison is made by a generally known pattern matching method (for example, rotation matching or the like) to determine the authenticity of the coin 7. That is, the determination unit 18, similar to the determination unit 5 of the first embodiment, stores some of the data stored in the registration pattern storage unit 19 based on the outer diameter data output from the binarization unit 17. Of the reference registration pattern data, the authenticity of the target coin is determined by using the reference data having the closest outer diameter data.

As described above, in the second embodiment, the coin 7 is separately irradiated with light from two different directions, the optical reading is performed twice at the same position on the surface of the coin 7, and based on the two read outputs. Since the binarization is performed to extract the pattern of the coin 7 due to the unevenness, as in the first embodiment, the coin 7 is not affected by dirt, rust, color, etc. on the surface to be read, A stable recognition can always be performed.

Further, as in the first embodiment, since the authenticity determination is performed based on the result of the denomination determination based on the outer diameter data, unnecessary determination processing can be omitted and the processing can be simplified.
The recognition processing can be speeded up. Moreover, since the outer diameter data can be obtained from the image data of the coin 7 in the binarization unit 17, there is no need for other sensors,
The configuration can be simplified and the cost can be reduced.

<< Third Embodiment >> FIG. 10 is a block diagram showing a third embodiment. In this third embodiment, coin 7
Since the transporting configuration of 1 and the configuration of irradiating the coin 7 from the light source 1 are the same as those in the first embodiment, the description thereof is omitted here. Further, the specularly reflected light from the coin 7 is read by the reading unit 2 as in the first embodiment, and the unevenness pattern, the outer diameter and the center point of the coin 7 are obtained by the image capturing unit 3. Then, the image data obtained by the image capturing unit 3 is stored in a storage unit (not shown), and the outer diameter data and the center point data thereof are sent to the determination unit 23 described later.

In the third embodiment and the first embodiment, the smaller the angle between the irradiation light and the reflection light on the coin 7 with respect to the vertical axis from the surface of the coin 7, the smaller the coin 7 is. Since highly reliable image data can be obtained without being affected by the rotation angle, it is desirable to determine the positional relationship between the light source 1, the reading unit 2, the half mirror 11, and the slit 12 so as to have such a relationship.

On the other hand, the third embodiment has a filtering unit 20, a binarization unit 21, a look-up table 22, a judgment unit 23, and a registration pattern storage unit 24 as a configuration different from that of the first embodiment. ing. Filtering unit 20
Is for emphasizing the change in the regular reflection light intensity due to the uneven pattern of the coin 7 with respect to the regular reflection light intensity data of the coin 7 stored in the storage unit, using a 3 × 3 template described later. The binarization unit 21 sets a threshold for binarization on the data output from the filtering unit 20, and compares the pixel data with the threshold to binarize each pixel data into white / black. It has a function to do.

The look-up table 22 stores two pixels for each pixel.
9 is a table showing a plurality of different pixel patterns represented by values. The determination unit 23 sets an image within a circle smaller than the outer diameter of the coin 7 as the determination region based on the center point data and the outer diameter data obtained by the image capturing unit 3,
The binarized data is compared with each pixel pattern of the look-up table 22, a histogram showing the frequency of occurrence of the pixel pattern of the look-up table 22 in the determination area is created, and a reference point is set from this histogram. Then, based on the reference point, the registered pattern stored in the registered pattern storage unit 24 is compared to determine the denomination and authenticity of the coin 7. The registration pattern storage unit 24 is a storage unit composed of a ROM, a RAM, a magnetic disk, or the like, like the registration pattern storage units 6 and 19 of the above-described embodiments.

Next, the operation of the third embodiment will be described. Now, when the coin 7 is inserted, the light source 1 is turned on, and the light passing through the slit 12 is reflected by the half mirror 11 to illuminate the surface of the coin 7. The light specularly reflected from the surface of the coin 7 passes through the half mirror 11 and enters the reading unit 2. This operation is similar to that of the first embodiment. Also,
As for the pattern of the surface of the coin 7, the uneven pattern on the surface is clearly extracted by the principle shown in FIG. 2 of the first embodiment. And
The image data extracted by the image capturing unit 3 is stored in a storage unit (not shown), and the image written in the storage unit is output to the filtering unit 20.

The filtering unit 20 reads the image written in the storage unit (not shown) and rewrites the value of the pixel of interest based on the relationship between the values of the pixel of interest and its peripheral pixels. FIG. 11 is a diagram showing an operation example of the filtering unit 20. For example, as shown in this figure, 3 × 3
With the center pixel of the template as the target pixel, the image written in the storage unit (not shown) is read out, and the target pixel (x,
y) is multiplied by 7, and the pixels (x, y-i), (x, y + 1), (x-1, y) (x
+1 and y) are each multiplied by 1.5, and these are counted and output. Therefore, the output value becomes the following formula. Output value = 7 (x, y) -1.5 (x, y-1) -1.5
(X, y + 1) -1.5 (x-1, y) -1.5 (x +
1, y) By thus converting the image by the filtering unit 20, the edge portion of the image formed by the unevenness can be emphasized.

The filtering unit 20 writes this output value in a storage unit (not shown). The binarization unit 21 reads an image written in a storage unit (not shown) and creates a density distribution series representing the frequency of occurrence of the density value. Similar to the binarizing unit 4 of the first embodiment, in the binarizing unit 21, in order to binarize the image data, the output fluctuation due to dirt, rust, etc. for each coin, the change over time of the line sensor, etc. It is desirable to determine an appropriate threshold so as not to be bound by the fluctuation of
Therefore, also in this third embodiment, in order to determine the threshold value, the average value of the concentration distribution is calculated and this is determined as the optimum threshold value.

FIG. 12 is a diagram showing the operation of the binarization unit 21. That is, in this figure, as in the case of FIG. 4 of the first embodiment, the vertical axis 301 represents the density occurrence frequency and the horizontal axis 302 represents the density of the target image. In this case, in addition to determining the average value of the density as the threshold value 303, when the distribution shape of the density occurrence frequency has two peaks, the valley portion of the peak may be determined as the threshold value 303. Whether the threshold value 303 is obtained by any one of these methods is empirically obtained depending on the nature of the coin 7 actually targeted. In addition, by combining with the operation of the filtering unit 20, other than the pattern formed by the unevenness of the surface of the coin 7,
For example, the effects of rust and dirt can be completely removed. The image binarized in this manner is output to the determination unit 23.

The determining unit 23 first determines the reading area of the binary image output from the binarizing unit 21 based on the center point (X _c , Y _c ) of the coin output from the image capturing unit 3. To do. FIG. 13 is a diagram showing the concept of the reading area.
In FIG. 13, 304 is a binary image written in a storage unit (not shown). 305 is an image of the target coin. 306 is an area designated by a radius smaller than the radius of the coin 7 based on the center point (X _c , Y _c ) of the coin 7 output from the image capturing unit 3.

In this way, the area for judgment is defined by the coin 7
The reason why the area is smaller than the outer diameter is that, for example, in the case of foreign coins having the same outer diameter, the shapes of the outer circles may be the same. This is because it is desirable to extract the discrimination data with respect to.

The determination unit 23 uses, for example, the following 3 ×
Feature data is extracted using 16 types of lookup tables 22 configured in 3. FIG. 14 is a diagram showing the lookup table 22. In this lookup table 22, "1" is a black pixel and "0" is
It is a white pixel. Further, "?" Means that either black pixels or white pixels may be used.

These look-up tables 22
Has a number in parentheses () and is configured such that the positional relationship between white and black is gradually rotated based on the central pixel in a 3 × 3 table from a small value to a large value. Further, this lookup table 22 has 16
The number is not limited to 32, and may be 32 or 64. Furthermore,
The configuration is not limited to the 3 × 3 configuration, and may be 5 × 5 or 7 × 7. Here, the binary image output from the binarization unit 21 is read out, and if any of the pixel patterns in the lookup table 22 matches, the value (number) of the matched pixel pattern is output.

Below, the lookup table 22 is 3 × 3.
It will be explained using the one configured in. The determination unit 23, based on the value attached to the matching lookup table 22,
Create the following histogram. FIG. 15 is an explanatory diagram of the histogram. In the figure, the horizontal axis 308 shows the value of the lookup table 22 that is output, and the vertical axis 307 shows the frequency. Further, 309 is a reference point.

FIG. 16 is an output example of the histogram when the data of the same coin 7 is used, but when the coin 7 passes through the conveying path 8 at a rotation angle different from that shown in FIG. As is clear from FIGS. 15 and 16, the coin 7
When is rotated, the directions are different, and the histogram is shifted to the left and right.

The determination unit 23 extracts a peak value, that is, the reference point 309 from this distribution. Based on the reference point 309 extracted from this target coin 7, FIG.
5 and FIG. 16 are converted as described below. FIG. 17
FIG. 6 is an explanatory diagram of a histogram after the conversion. FIG. 17
In the histogram in, the distribution shape having the reference point 309 as a starting point has a constant shape regardless of the direction of the uneven pattern of the target coin 7.

For example, when the character "100" of a 100-yen coin is vertical, the frequencies (5) and (13) of the lookup table 22 are frequently applied, and the histogram as shown in FIG. The value of the vertical axis 307 increases with "5" and "13" on the scale of the axis 308. On the other hand, when the character "100" of the 100-yen coin is horizontal, (1) of the lookup table 22 is selected.
(9) is applied more frequently, and in the histogram as shown in FIG. 15, the values on the vertical axis 307 are large at the scales “1” and “9” on the horizontal axis 308.

Thus, the relationship between the case where the character "100" of the 100-yen coin is vertical and the case where it is horizontal is 5-1 = 4, 13-9 = 4, and the difference between them is 4 Is. Therefore, the rotation angle of the coin 7 can be corrected by correcting this difference. That is, using the reference value 309 in FIG. 15 as a starting point, conversion is performed as shown in FIG.
In the case of FIG. 16 in which the rotation angle is different from the case of FIG. 15, the reference value 309 is similarly used as the starting point for conversion as shown in FIG. By doing this, coins 7
Even when reading, the rotation angle can always be corrected as described above, and determination data having a constant shape as shown in FIG. 17 can be obtained.

After that, the judging section 23 compares the distribution shape with the reference histogram registered in advance in the registration pattern storing section 24 to judge the authenticity of the target coin 7. That is, the determination unit 23 determines the denomination based on the outer diameter data output from the image capturing unit 3, and based on the denomination determination result, some criteria registered in advance in the registration pattern storage unit 24. From the histogram distribution shapes of the above, the authenticity of the target coin is determined by comparing with the reference histogram distribution shape to which the outer diameter data is most approximate.

In the third embodiment described above, the irradiation light from the light source 1 irradiates the coin 7 through the slit 12 and the half mirror 11, and the specular reflection light is transmitted through the half mirror 11 and the reading unit 2 Although it is configured to be incident on the beam, other configurations may be used, and this example is shown below. FIG.
[Fig. 6] is an explanatory view of the configuration. In FIG. 18, 25
Is an optical lens, and is composed of an aspherical lens, a self-fog lens, and the like. The focal position of the optical lens 25 is set on the surface of the coin 7. Therefore, even in such a configuration, the specularly reflected light of the coin 7 is incident on the reading unit 2, so that the uneven pattern on the surface of the coin 7 can be read as in the third embodiment. Since the operation thereafter is the same as that of the third embodiment, the description thereof is omitted.

As described above, in the third embodiment, the image of the surface of the coin 7 is read, and the change in the density level caused by the unevenness of the surface of the coin 7 is emphasized from the density image, and the image data is converted into 2 by the result. The binarized data is used, and a histogram showing the frequency of occurrence of pixel patterns corresponding to the binarized data is created, and the type of coin 7 and the authenticity determination are performed based on this histogram. As a result, the standardized data for discrimination can be set only in the histogram, and the reliability of recognition can be improved.

In each of the above embodiments, the coin recognition is performed by the image data of one side of the coin 7, but the coin recognition may be performed by the image data of both sides. Further, in each of the above-described embodiments, light is emitted from the lower surface side of the coin 7 through the window 10 and the specularly reflected light is received by the reading unit 2, but the present invention is not limited to this configuration. For example, the light source 1 and the reading unit 2 may be provided on the upper surface side of the coin 7.
However, in this case, it is necessary that the specularly reflected light from the coin 7 is read. The reason for this is the irregular reflection light from the surface of the coin 7, that is, the vertical axis from the surface of the coin 7,
When the angle of the light source 1 and the angle of the reading unit 2 are different, the unevenness pattern of the surface of the coin 7 received by the reading unit 2 changes depending on the rotation angle of the coin 7, which makes it difficult to determine the authenticity. Is.

Further, also in the first and second embodiments,
Similar to the third embodiment, the determination area of the coin 7 may be an image within a circle smaller than the outer diameter of the coin 7. With this configuration, even in the case of foreign coins having the same outer diameter, it is easy. It is possible to determine the type of coin 7. Further, in each of the above-described embodiments, the type of the coin 7 is determined by using the outer diameter data of the coin 7, but in addition to this, the material data, the thickness data, the weight data, the presence or absence of jaggedness of the outer peripheral edge, and the like. It may be.

[0061]

As described above, according to the coin recognition device of the present invention, the pattern on the surface of the coin is imaged, the image data of the coin is taken out, the image data is binarized, and the discrimination data is extracted. However, since the discrimination data is compared with the registered pattern registered in advance to discriminate the type and authenticity of the coin, it is possible to perform the coin recognition only with the image data of the coin, thus simplifying the configuration. The cost can be reduced and the coins can be recognized accurately.

[Brief description of drawings]

FIG. 1 is a configuration explanatory view showing a first embodiment of a coin recognition device of the present invention.

FIG. 2 is an explanatory view of the principle of detecting the uneven pattern in the first embodiment of the coin recognition device of the present invention.

FIG. 3 is an explanatory diagram of read image data in the first embodiment of the coin recognition device of the present invention.

FIG. 4 is a coin recognition device according to the first embodiment of the present invention;
It is explanatory drawing of operation | movement of a digitization part.

FIG. 5 is an explanatory diagram of the operation of the determination unit in the first embodiment of the coin recognition device of the present invention.

FIG. 6 is a configuration explanatory diagram of another example of the coin recognition device of the first embodiment of the present invention.

FIG. 7 is a structural explanatory view showing a second embodiment of the coin recognition device of the present invention.

FIG. 8 is an explanatory diagram of the principle of detecting the uneven pattern in the second embodiment of the coin recognition device of the present invention.

FIG. 9 is a coin recognition device according to a second embodiment of the present invention;
It is explanatory drawing of operation | movement of a digitization part.

FIG. 10 is a structural explanatory view showing a third embodiment of the coin recognition device of the present invention.

FIG. 11 is an explanatory diagram of the operation of the filtering unit in the third embodiment of the coin recognition device of the present invention.

FIG. 12 is an explanatory diagram of the operation of the binarizing unit in the third embodiment of the coin recognition device of the present invention.

FIG. 13 is an explanatory diagram of a reading area in the coin recognition device according to the third embodiment of the present invention.

FIG. 14 is an explanatory diagram of a lookup table in the third embodiment of the coin recognition device of the present invention.

FIG. 15 is an explanatory diagram of a histogram in the third embodiment of the coin recognition device of the present invention.

FIG. 16 is an explanatory diagram of histograms at different rotation angles in the third embodiment of the coin recognition device of the present invention.

FIG. 17 is an explanatory diagram of a converted histogram in the third embodiment of the coin recognition device of the present invention.

FIG. 18 is a structural explanatory view showing another example of the third embodiment of the coin recognition device of the present invention.

[Explanation of symbols]

 1, 14, 15 Light source 2, 16 Reading unit 3 Image capturing unit 4, 17, 21 Binarization unit 5, 18, 23 Judgment unit 6, 19, 24 Registration pattern storage unit 7 Coin 20 Filtering unit 22 Look-up table

Claims (7)

[Claims] 1. A light source that illuminates the surface of a coin, a reading unit that reads the intensity of specularly reflected light on the surface of the coin, and an image capture that extracts the center point of the image data of the coin based on the output from the reading unit. Section and a threshold value is set for the specular reflection light intensity on the surface of the coin, and by comparing the threshold value with the specular reflection light intensity, each pixel in the image data of the coin is represented by binarized data. And a determination unit for comparing the distribution state of the binarized data existing on the concentric circles in the image data of the coin with a preset registered pattern to determine the type and authenticity of the coin. A coin recognition device characterized by being provided. 2. A plurality of light sources that irradiate the surface of a coin from different directions, a reading unit that reads the intensity of reflected light from the plurality of light sources to the coin, and a reflected light from the plurality of light sources by the reading unit. Based on the difference in intensity, a binarizing unit that binarizes the reflected light intensity of the coin, data binarized by the binarizing unit, and a preset registration pattern are compared, A coin recognition device, comprising: a discriminating unit for discriminating the type and authenticity of coins. 3. A light source that illuminates the surface of the coin, a reading unit that reads the intensity of specularly reflected light on the surface of the coin, and an image capturing unit that extracts image data of the coin based on the output from the reading unit. With respect to the regular reflection light intensity of the coin, a filtering unit that emphasizes a change in the regular reflection light intensity due to the uneven pattern of the coin, and the data output from the filtering unit,
By setting a threshold value and comparing the threshold value with the data, a binarization unit that represents each pixel as binarized data, and a lookup table that indicates a plurality of different pixel patterns that represent each pixel in binary, Image data of the coin extracted by the image capturing unit is used as a determination area, and the binarized data in the determination area is compared with each pixel pattern of the lookup table,
Creating a histogram showing the frequency of occurrence of pixel patterns of the lookup table in the determination area, and
Coin recognition, characterized in that a reference point is set from this histogram, comparison is made with a preset registered pattern based on the reference point, and a discrimination unit for discriminating the type and authenticity of the coin is provided. apparatus. 4. The image capturing unit detects the outer diameter of the coin based on the output from the reading unit, and the determining unit determines the type of the coin based on the outer diameter of the coin detected by the image capturing unit. The coin recognition device according to claim 1 or 3, wherein a registered pattern corresponding to this type is selected for comparison. 5. The binarizing unit detects the outer diameter of the coin based on the output from the reading unit, and the determining unit determines the type of the coin based on the outer diameter of the coin detected by the binarizing unit. The coin recognition device according to claim 2, wherein a registered pattern corresponding to this type is selected, and the registered pattern is selected as a comparison target. 6. The readable length of the reading unit is set to be equal to or larger than the diameter of the coin having the largest diameter among the coins to be recognized. Coin recognition device. 7. The coin recognition device according to claim 1, wherein the determination unit sets the determination region of the coin as an image in a circle smaller than the outer diameter of the coin.