JPH0464001B2 - - Google Patents

Description

[Detailed description of the invention]

(Industrial Application Field) The present invention relates to a diameter determination method for determining the diameter of a circular object, such as a coin. (Prior Art) By the way, when classifying circular objects by their diameters, it is necessary to determine the diameters. One example is the classification of coins of different denominations with different diameters. Conventionally, in coin sorting machines that sort coins of various denominations using sorting holes provided for each denomination according to their diameter, each coin of each denomination is counted in the vicinity of each sorting hole. In the passage section before reaching the nearest sorting hole,
A method is known in which the diameter is determined optically. A light source and a photoelectric converter are installed between the passages, and at the moment when the coins that are fed one by one reach that position, the amount of light that is blocked by the coins is determined. This is to detect the diameter of the (Problem to be Solved) However, this device has the disadvantage that it is necessary to accurately detect the point at which the coin reaches the detection position using a position sensor, and it also There is a drawback that it cannot be detected if counterfeit coins are transported. To solve this problem, it is necessary to arrange a plurality of position sensors and detect the diameter each time each position sensor detects the diameter, which results in the inconvenience of complicating the apparatus. SUMMARY OF THE INVENTION Therefore, an object of the present invention is to provide a diameter determination method that can detect objects having a shape other than circular, such as counterfeit coins, even if they are transported. (Means for Solving the Problem) The present invention provides a repeating scanning type photoelectric conversion means arranged in a straight line so as to be orthogonal to the conveying direction of the circular object and facing the circular object,
The length of the light-shielding portion caused by the circular object being transported while being irradiated with light is detected by unit scanning in the photoelectric conversion means, and when a predetermined length or more is shielded from light by the circular object; The maximum value of the length of the detected light-blocking portion is obtained, and the number of scans by the photoelectric conversion means is measured, and if the measured number of scans is within a certain range, the length of the circular object is determined based on the maximum value. Determine the diameter of
If the number of scans is outside the certain range, the circular object is determined to be an abnormal object without determining the diameter. (Operation of the invention) If the number of times the photoelectric conversion means is periodically scanned is measured and it is within a certain range, it is considered normal, and if it is outside the certain range and the photoelectric conversion means is blocked by a predetermined amount or more, it is determined to be an abnormal object. It is something. (Embodiments of the Invention) Examples will be described below using a coin as an example of a circular object. An example using a one-dimensional image sensor as a photoelectric conversion means will be described with reference to the drawings.
In FIGS. 1 and 2, 1 is a passage bottom plate, 4,
Reference numeral 4' denotes a passage side wall, and the coin 5 as a circular object is continuously and forcibly conveyed in the direction of arrow A through a passage P formed by these walls by a conveyor belt (not shown) or the like. Rectangular holes 6 and 6' are provided in a straight line in the passageway bottom plate 1, and a light source 7 such as a light bulb and a lens system 8 are provided below the holes 6 and 6'. Further, above the holes 6 and 6', a light source 7 is provided.
a pair of image sensors 2, 2' to receive the light coming out from the lens system 8 and passing through the holes 6, 6'.
is provided. The image sensors 2, 2' are attached to the substrate 9, and are connected to the passage P from the passage side walls 4, 4'.
They are arranged so as to extend toward the center of the coin in a direction perpendicular to the coin conveyance direction A, and to be aligned with each other. Note that the substrate 9 is attached to the holding walls 10, 1
0', and the image sensors 2, 2' are provided so as to avoid positions facing the center of the coin. This means that when a coin with a hole in the center passes, such as a 5-yen coin or a 50-yen coin, the light passing through the hole in the coin passes through the image sensor 2,
This is to prevent light from being received by 2'. The image sensors 2 and 2' each have a photodiode array (hereinafter referred to as
It serially transmits an electrical signal corresponding to the amount of light received by each photodiode, that is, a signal of the "L" level when the light is shielded, and a "H" level signal when the light is not shielded, in the scanning order. Output. In this embodiment, scanning is performed in the order from the central end of each image sensor 2, 2' toward the outer end. Further, the ends of the image sensors 2, 2' are buried in the passage side walls 4, 4' so that at least eight photodiodes are illuminated by light when coins are conveyed. On the other hand, with respect to the flow direction of the coins 5, a well-known material sensor 3 is arranged slightly upstream of the image sensors 2 and 2'.
For example, it has a primary coil and a secondary coil placed close to the passing coin, and the material (magnetic properties) of the passing coin and the area of the coin facing the primary coil and secondary coil. Outputs a voltage signal that varies depending on the In other words, even if the coin is made of the same material, the output voltage value will differ between when the area is completely occupied by the coin, as in the area surrounded by the broken line 3 in Figure 1, and when it is only partially occupied. In addition, the output voltage value also differs when the region indicated by the broken line 3 is wholly or partially occupied by the hole portion of a coin such as a 5-yen coin or a 50-yen coin.
For this reason, the material of the coin is determined at the timing when the diameter of the coin is determined by the image sensors 2, 2'. FIG. 3 is a block diagram showing the control system of the present invention, in which a basic clock generation circuit 11 outputs a basic clock signal CP, and a timing pulse generation circuit 1
2 is a four-phase clock signal φ1,
Outputs φ2, φA, φB and start pulse SP. In addition, CPU (Central Processing Unit or
MicroProcessor Unit) 13 is the basic clock signal
Based on the CP, data is written to or read from the RAM 15 based on input signals from each input/output port (described later) according to the program stored in the ROM 14, and various arithmetic processes are performed, and various signals are output from the output port. It's starting to output. Here, the image sensor sections 25, 25'
are the clock signals φ1, φ2, φA, φB and the start pulse from the timing pulse generation circuit 12.
SP is input to detect the light received by the photodiode, and the process is shown in the configuration diagram in Figure 4 and Figure 5
This will be explained with reference to the time charts of ~G. The image sensor 2 used here is a self-scanning image sensor, and includes a shift register 201 as one scanning circuit driven by a start pulse SP and clock pulses φ1 and φ2, and scanning generated at each stage of this shift register 201. A shift switch 202 for temporally shifting the pulses using clock pulses φA and φB, and a shift switch 202 for noise compensation.
A capacitor 203 made of a MOS transistor,
It is composed of an address switch 204 and a PDA 205. The operation is to automatically scan each photodiode by applying a start pulse SP to the shift register 201, convert the optical input into an electrical signal for each photodiode, and then output it as a continuous pulse train to the video line VS. ,
Input to differential amplifier 206. This differential amplifier 2
06 is for removing noise components from the pulse signal flowing through the video line VS to obtain an appropriate signal level, and the output of this differential amplifier 206 becomes the video signal VDS (or VDS'). Note that the image sensors 2, 2' and the differential amplifiers 206, 20
6' constitutes image sensor sections 25, 25', and the video signals VDS, VDS' output from the image sensor sections 25, 25' are sent to the next stage light-shielded portion length measuring circuits 24, 24', respectively. is input. Therefore, these light-shielded portion length measuring circuits 24, 2
4' receives the video signals VDS, VDS' from the image sensor parts 25, 25', and sends the image sensor part 2
It outputs a signal corresponding to the length of the portion of 5, 25' that is shielded from light by the coin.To explain the circuit configuration specifically, the counting circuit 17 receives the basic clock signal CP from the basic clock generating circuit 11.
Count. In addition, the counting circuit 16 receives the "H" level video signal VDS from the image sensor section 25.
is counted, and when the counted value reaches a predetermined value (for example, "8"), a coincidence signal CUS indicating this is generated, and the subsequent video signal from the image sensor section 25 is
Avoid inputting VDS and match signal
CUS is applied to the LD terminal of the latch circuit 18 as a control input. The latch circuit 18 is a match signal CUS
When received, the count value of the counting circuit 17 at that time
CUD is stored, and the output MCD of the latch circuit 18 becomes the output of the shaded portion length measuring circuit 24, and the output signal MCD is a value obtained by adding, for example, “8” to the number of photodiodes in the shaded portion. It's summery. In principle, it is possible to recognize that the light-shielding part has ended when the first "H" level signal appears in each scan, but considering noise etc., it is possible to Only when the output is input a predetermined number of times (eight times in this example) is it recognized that the light-shielding portion has ended. When one scan is completed, the counting circuits 16 and 17 output the reset signal RST from the output port 20.
will be reset. In addition, the light-shielded portion length measurement circuit 2
4' is also constructed similarly to the circuit 24. however,
Since the coin does not necessarily pass directly between the image sensors 2 and 2', the output MCD' of the circuit 24' is not necessarily exactly the same as the output MCD of the circuit 24.
However, if the diameter of the coin is the same, the measurement circuit 24 when the coin passes under the image sensors 2 and 2'
and 24' vary according to the same pattern and their maximum values are also the same. In addition, the input ports 19 and 19' are connected to the shielded portion length measuring circuits 24 and 2.
4', that is, the latch outputs MCD and MCD' from the latch circuits 18 and 18', respectively.
It is designed to be sent to RAM14 etc. On the other hand, the material detection circuit 21 inputs the output signal MTS from the material sensor 3, amplifies it appropriately, performs half-wave rectification, integrates the half-wave signal, and generates a predetermined reference voltage for each denomination. Compares with level and outputs material signal MSD. In addition, this material signal MSD
is output as a 4-bit signal, for example, as shown in Table 1 below.

【table】

[Table] In addition, the material sensor 3 and the material detection circuit 21 can be constructed using known technology, and the material sensor 3 and the material detection circuit 21
The 4-bit configuration of the MSD is also not limited to this. And the material signal MSD is input port 2
2 to the RAM 15, etc. The output port 23 outputs denomination signals m1 (1 yen coin), m2 (50 yen coin), m3 (5 yen coin), m
4 (100 yen coin), m5 (10 yen coin), m6 (500 yen coin), m7, m8 (counterfeit coin), and outputs an abnormal signal AL.
These input ports 19, 19', 22, output ports 20, 23, CPU 13, ROM 14, RAM 1
5 are connected to each other by an address bus AB, a data bus DB, and a control bus CB. Next, the principle of denomination discrimination using an image sensor will be explained. By the way, in a MOS image sensor, photodiodes are arranged in a straight line, for example, in units of 28 μm, and as shown in Figure 6, the image sensor 2
By irradiating parallel light PL onto the image sensor 2 and blocking the light with the slit 30, there are areas on the image sensor 2 that are illuminated by the light and areas that are not illuminated by the light. The image sensor 2 converts this one-dimensional position information based on light into a temporal electrical signal VDS, and one bit of this electrical signal corresponds to a length of 28 μm.
The principle of measuring the diameter of a coin using this image sensor 2 will be explained with reference to FIGS. 1 and 2. In order to reduce the influence of dust and dirt, the image sensors 2 and 2' are installed in the upper part of the coin passage section P with the light receiving windows 207 and 207' facing down, and the parallel light PL from the lens system 8 is irradiated from below. . Then, the coin 5 is inserted into the light receiving hole 6 of the image sensor 2, 2'.
By passing through 6', the parallel light PL is blocked. By determining the number of bits n1 and n2 of the video signal VDS in the portion where this light is blocked, the length l where the light is blocked is determined as l=a+28[μm]×(n1+n2)...(1) be able to. Then, the length l when n1+n2=maximum value MAX (2) can be regarded as the diameter of the coin. This measurement method makes it possible to determine the denomination based on the diameter when the coin 5 passes through the line of the light receiving holes 6, 6' of the image sensors 2, 2'. Here, it is easier to calculate if both the diameter l and the length a are expressed in bit numbers, so in the following, the diameter l
and the length a is expressed as the number of bits L [bit] and A [bit], which are converted into 1 bit 28 [μm], and L [bit] = A [bit] + (n1 + n2) ... (3). By the way, regarding the A bit, the distance a between the image sensors 2 and 2' is attached to the substrate 9, and there may be variations in assembly or the distance a between the image sensors 2 and 2'.
The value of the A bit differs from device to device due to variations in the arrangement of photodiodes within 2', and the value of the A bit must be determined before using the device. Therefore, how to determine the A bit will be explained below as a setting mode for the initial reference value BIAS. First, we poured eight 1 yen coins as samples,
The average value of (n1+n2), which can be regarded as the diameter, is calculated. Here, since the diameter l of a 1 yen coin is 20 [mm], the number of bits corresponding to this diameter is 20 [mm] ÷ 28
[μm] = 714 bits, so A bit = L bit - (n1 + n2) = 714 - (n1 + n2) ...(4) Find the A bit, and set the A bit value to RAM15.
is stored in the initial reference value memory BIAS. Then, if the A bit value stored in this initial reference value memory BIAS is determined and the maximum value MAX of formula (2) for each coin can be measured, L [bit] = A value of BIAS + (n1 + n2) max
...(5) The diameter of the coin can be measured. In addition,
The coin shown here (1 yen) and the number of samples (8
times) can be changed arbitrarily. Here, the relationship between the basic number of bits of each coin and the denomination identification bit range is shown in Table 2 below.

[Table] The lower value (mm) of the basic bit value in Table 2 indicates the diameter of the coin, and the CN column is 5
Since the diameters of yen coins and 100 yen coins are close to each other, and it is difficult to distinguish between the two due to wear and other factors, the coins are distinguished by their materials. In addition, the CPU 13 calculates the amount of light blocking of the image sensors 2 and 2' by the coin 5 bit CAD (=n1+n2).
However, when the scanning start value reaches “250”, it is assumed that the coin is passing, and the light shielding amount bit is set.
Measure the maximum value of CAD and calculate the number of scans PFG until CAD decreases to the scan end value "200"
Also measured. Therefore, this number of scans PCG is 100
In the above cases, the coin is not judged to be normal and an abnormality signal is signaled.
AL is output, and even if the number of scans PFG is 3 or less, the coin is not determined to be normal and moves on to measuring the diameter of the next coin. On the other hand, the scanning time of the image sensors 2 and 2' is, for example, 512 [μs], and if the passing speed of the coin is 800 [mm/s], the distance that the coin moves in one scanning time is approximately 0.4 [mm]. Become. The material signal MSD is always read into the CPU 13 at intervals of 512 [μs] in synchronization with the scanning time of the image sensors 2 and 2'. That is, the material signal MSD is read into the CPU 13 every time the coin moves 0.4 [mm] along the passage. Then, the read material signal MSD is sequentially written into the material detection memories n1 to n15 of the RAM 15, and when a new material signal is written to the material detection memories n1 to n15, the nth previous material signal is erased. Here, the coin first passes over the material sensor 3,
It moves a certain distance and reaches the light receiving line of the image sensor 2, 2'. When the number of bits of the video signal VDS reaches "250" on the light receiving line of the image sensor 2, 2', the most material signal among the 15 material detection memories n1 to n15 is stored. The coin is determined to be made of material having a large number of material signals n·max. Incidentally, the distance that the coin 5 travels from passing over the material sensor 3 to reaching 250 bits on the light receiving line of the image sensors 2 and 2' can be considered to be approximately 6 mm. Therefore, the material detection memories n1 to n15 are set to 6/0.4.
= 15, which is 15 pieces. Next, an example of the memory map of the ROM 14 will be explained with reference to FIG. It is designed to memorize bit values. In addition,
The reason why the minimum and maximum values are provided with an allowable range is to cope with coin wear, variation, etc. There is a scan start value memory for setting and storing the bit value at which the image sensors 2 and 2' start scanning ("250" in this example), and a scanning memory for setting and storing the bit value at the end of scanning ("200" in this example). It is provided with an end value memory and a scanning number setting memory for storing the maximum value (“100” in this example) and minimum value (“3” in this example) of the scanning number PFG. In addition, the RAM 15 includes a material detection memory, an initial reference value memory (BIAS), a light shielding amount memory (CAD),
It has a scanning number memory (PFG), an addition memory (MAD), and a maximum coin diameter memory (COL), and the material detection memory has 15 memory areas n.
It has an area for storing 1 to n15 and the maximum value n·max therein. Here, the overall operation will be explained with reference to the flowcharts shown in FIGS. 9A and 9B. First, when the coin discrimination operation starts, the basic clock signal CP from the basic clock generation circuit 11 is
The image sensors 2 and 2' are driven, and 1-bit scanning is performed first (step S1,
S2), the video signals VDS, VDS' output from the image sensor sections 25, 25' are outputted from the counting circuits 16, 1
6' (step S3). Further, the counting circuits 17, 17' count the basic clock signal CP (step S4), and determine whether the count value of the counting circuits 16, 16' is "8" (step S5). and,
If the count value of the counting circuits 16, 16' is not "8", the above operation is repeated, and if the count value is "8", the coincidence signals CUS, CUS,
CUS' is outputted (step S6), whereby the latch circuits 18, 18' latch the count values CUD, CUD' of the counting circuits 17, 17' (step S7). The CPU 13 then determines whether one scan has been completed (step S8), and adds "+1" to the scan number memory PFG in the RAM 15 (steps S8, S9). The material of the coin is detected by the material sensor 3, and the material signal MSD from the material detection circuit 21 is sent via the input port 22 and the CPU 13.
The information is stored in the material detection memories n1 to n15 of the RAM 15 (step S10). Furthermore, latch circuit 1
The count values MCD and MCD' latched at 8 and 18' are added and stored in the addition memory MAD of the RAM 15 via the input ports 19 and 19' and the CPU 13 (step S11), and the value of the addition memory MAD is is “250” set in the scan start value memory of the ROM 14 (step
In step S12), if the scanning start value is not "250", each part is reset (step S13) and the process returns to the original state.
Therefore, when the value of the addition memory MAD is the set scanning start value "250", the material signal with the largest number among the material detection memories n1 to n15
Store MSD in memory n・max (step 14),
It is determined whether the amount of light shielding CAD is greater than or equal to the additional value MAD (step S15), and if it is smaller, the amount of light shielding is
Set CAD to the additional value MAD (step 16). Next, the value of the scanning number memory PFG is set in the ROM14.
The maximum value stored in the scanning number setting memory is “100”
It is determined whether or not the maximum value is greater than "100" (step S17), and if the maximum value is greater than "100", an abnormal signal AL is output from the output port 23 (step S19),
Reset each part (step S31). However,
If the number of scans PFG is less than "100", it is further determined whether it is greater than or equal to the set minimum value "3" (step S18), and if it is greater than "3", the additional value is
It is determined whether MAD is smaller than "200" stored in the scan end value memory of the ROM 14 (step S20). If the added value MAD is ``200'' or more, each part is reset (step S13), and if it is smaller than ``200'', it is reset to an initial reference value indicating the distance a between the image sensors 2 and 2'.
Add BIAS and the amount of shading CAD that indicates the length of the shading part (n1+n2), and use this added value as the coin diameter COL.
The data is stored in a predetermined area of the RAM 15 (step S21). The maximum and minimum values for each coin are stored in the ROM 14, and are compared with the maximum coin diameter COL stored in the RAM 15 (step S22). In this case, it is first determined whether the value is in the CN column (step S23), and if it is not the value, 1 yen coin ~
It is determined whether or not the 500 yen coin has the corresponding value (step S24), and if the corresponding denomination exists, it is further compared with the denomination detected by the material sensor 3 (step S25).
When the identification based on the diameter and the identification based on the material match, the corresponding denomination signals m1 to m6 are output from the output port 23 (step S26), and when they do not match, a counterfeit currency signal m8 is output (step S26).
S30). Further, in step S24, if there is no corresponding value, a counterfeit currency signal m7 is output from the output port 23. On the other hand, if the diameter in the CN column is determined in step S23, it is determined whether the material n・max in the material detection memory is a 5 yen coin (CuZn) or a 100 yen coin (CuNi) (step S23). S28,
S29), if it is determined to be 5 yen or 100 yen, the corresponding denomination signal m3 or m4 is output from the output port 23 (step S26). In addition, 5 yen,
If the material does not correspond to the 100 yen material, a counterfeit currency signal m8 is output from the output port 23 (step
S30). Note that the contents of the ROM described above can be changed arbitrarily, and the type and size of coins can also be changed arbitrarily. Furthermore, although the effect has been exemplified as a circular object in the above description, the same applies to other circular objects (for example, tokens). (Effect of the invention) As described above, according to the diameter determination method of this invention,
By appropriately setting the number of scans depending on the denomination of coins in circulation, the coin diameter can be accurately determined.

[Brief explanation of the drawing]

Figure 1 is a mechanical diagram of the coin diameter measurement of this invention, Figure 2
The figure is a cross-sectional view of FIG. 1, FIG. 3 is a block diagram showing the circuit system of this invention, FIG. 4 is a configuration diagram of an image sensor used in this invention, and FIGS. Figure 6 is a diagram to explain the principle of measurement using the image sensor, Figure 7 is the ROM memory map, and Figure 8 is the time chart shown.
The RAM memory map, FIGS. 9A and 9B, is a flowchart showing an example of operation. 1... passage bottom plate, 2, 2'... image sensor,
3... Material sensor, 4, 4'... Passage side wall, 5...
...Coin, 6,6'...hole, 7...light source, 8...lens system, 13...CPU, 14...ROM, 15...
...RAM, 16, 16', 17, 17'... Counting circuit, 24, 24'... Shading portion length measuring circuit, 25,
25'...Image sensor section.

Claims (1)

[Claims] 1. A repeating scanning type photoelectric conversion means is disposed in a straight line so as to be orthogonal to the conveyance direction of the circular object and opposite to the circular object, and the photoelectric conversion means is arranged in a straight line so as to be perpendicular to the conveying direction of the circular object, and the photoelectric conversion means is arranged in a straight line so as to be perpendicular to the direction of conveyance of the circular object, and the photoelectric conversion means is arranged in a straight line so as to be perpendicular to the conveyance direction of the circular object and to face the circular object. The length of the light-blocking portion on the photoelectric conversion means that occurs when the light is blocked by the circular object is detected by unit scanning of the photoelectric conversion means, and the length of the detected light-blocking portion when a predetermined length or more is blocked by the circular object. While obtaining the maximum value of the length, the number of scans by the photoelectric conversion means is measured, and if the measured number of scans is within a certain range, the diameter of the circular object is determined based on the maximum value, and the number of scans by the photoelectric conversion means is determined. A method for determining the diameter of a circular object, characterized in that the circular object is determined to be an abnormal object without determining the diameter if the number of times is outside the certain range.