JPH0445877B2 - - Google Patents

Description

[Detailed Description of the Invention] (Industrial Application Field) The present invention relates to a method for use in a banknote discrimination device built into a banknote handling device such as an automatic transaction device installed in a financial institution, etc. The present invention relates to a technique for detecting data related to banknotes using a plurality of optical sensors arranged on a banknote conveyance path and setting the data extraction timing when determining the authenticity of banknotes.

(Prior art) Conventionally, in a banknote validating device, an optical sensor is used for a banknote traveling on a conveyance path to detect the amount of light such as reflected light from the banknote or transmitted light passing through the banknote, and converts this into an electrical signal. This electrical signal was then extracted at a predetermined time and used as banknote data to be analyzed for identification purposes.

In order to obtain this banknote data, a banknote data collection sensor (hereinafter referred to as a light receiving sensor) is actually used as an optical sensor that detects the amount of reflected light or transmitted light used for banknote identification and converts it into an electrical signal. ,
On the other hand, a plurality of banknote passage detection sensors are used, which are arranged on the banknote conveyance path at a certain distance.

This banknote passage sensor is a trigger sensor for detecting the passage of a banknote through the banknote validating device and timing the start of collecting electrical signals from the light receiving sensor. That is, the electrical signal from the light receiving sensor is amplified to an optimum value and converted into a digital signal at each reference timing synchronized with the running of the bill. Collection of this digital signal is started when the bill passage detection sensor detects a bill. Therefore, collection of digital signals is started before the time when the leading edge of the bill passes the light receiving sensor. The digital signals collected in this manner are sequentially stored in the memory for each light receiving sensor according to the reference timing. This digital signal is stored in the memory until the trailing edge of the bill passes the light receiving sensor.

It is clear that the digital signal string stored in the memory in this manner includes data that is not necessary for banknote identification at the beginning thereof. Therefore, it is necessary to extract this data to obtain banknote data from the leading edge of the banknote, which is necessary for banknote identification. For this reason, the conventional method is as follows. That is, the collected digital signal is compared with a predetermined threshold value by a comparator provided before the memory. This comparison operation is performed at a timing faster than the reference timing. Then, the time when the value of the digital signal first exceeds the threshold value is determined as the time when the leading edge of the bill is positioned on the light receiving sensor, and the digital signals are cut out at that time and stored in the memory sequentially. Then, the digital signals cut out in this manner are sequentially stored up to the rear end of the banknote, and the obtained digital signal string is used as banknote data for banknote discrimination.

(Problems to be Solved by the Invention) However, the above-mentioned digital signal extraction method causes the following problems.

As mentioned above, the digital signal is extracted by comparing the digital signal with a predetermined threshold value. Therefore, if the tip of the banknote is not damaged or soiled in any way,
Correct cutting is possible. However, if the tip of the banknote is partially damaged or dirty,
Correct cutting may not be performed. For example, when a damage such as a partially cut off leading edge is present on the scanning line of the light receiving sensor, the digital signal is below the threshold value at a time when the leading edge would be detected if it were a normal banknote. The tip of the damaged banknote is detected when the remaining portion of the damaged banknote passes through the light receiving sensor. Therefore, the detection time of the leading edge in this case is delayed when compared with a normal bill.
This delay has a negative impact on the formation of banknote data,
This makes the operation of determining authenticity incorrect. This is because the digital data is cut out as if the cut-off portion of the bill were the tip of the bill. In other words, the bill is processed as if it were traveling diagonally (assuming that the other light receiving sensors accurately detected the tip of the bill), even though the tip of the bill is actually perpendicular to the direction of travel. It will be done.

SUMMARY OF THE INVENTION Therefore, an object of the present invention is to solve the above-mentioned problems and provide a bill data cutout timing setting method that can correctly cut out digital signals even if the leading edge of the bill is partially damaged.

(Means for Solving the Problems) The present invention provides a bill data cutout timing setting method for obtaining timing for detecting bill data under the control of a central processing unit using a plurality of optical sensors arranged on a bill conveyance path. It depends.

Two transmissive sensors are arranged perpendicularly to the running direction of banknotes at an interval l ₁ , each of the transmissive sensors detects the presence of a banknote, and a counter is provided. Counting is started when a banknote is detected, and counting is stopped when a banknote is subsequently detected by the other transmission sensor, and the time difference between banknote detection is calculated. The reflection sensor is arranged downstream of the passage detection sensor at an interval _l2 different from the sensor interval _l1 , and each of the reflective sensors detects the presence of a banknote.

The memory includes a memory that stores the bill detection time difference of the transmissive sensor and the bill detection data from the reflective sensor, and the memory stores the bill data from the reflective sensor at a time when one of the transmissive sensors detects the bill first. Let's do it from

The central processing unit controls as follows.

A first skew amount θ of a banknote is calculated based on the banknote detection time by the transmission sensor stored in the memory and the interval _l1 between the transmission sensors, and the banknote detection data from the reflection sensor stored in the memory is calculated. As a result, the time difference between banknote detection by each of the reflective sensors is calculated, and the second skew amount θ ₁ of the banknote is calculated from the calculated time difference and the interval l ₂ between the reflective sensors.

On the other hand, the central processing unit determines the leading direction of the banknote based on which output of the transmission sensor comes first.

Furthermore, the central processing unit determines that the banknote is not defective when the calculated first skew amount θ and second skew amount θ ₁ are equal, and determines that the banknote is not defective, and determines that the banknote is not defective at the time when either one of the transmission sensors detects the banknote. The time when each of the reflection sensors detects a banknote is the starting point, and when the calculated first skew amount θ and the second skew amount θ ₁ are different, Assuming that there is a defect, determine in which position of the reflective sensor the defect has occurred based on the magnitude relationship with the first and second skew amounts θ and θ ₁ and in combination with the determined leading direction, and The time difference and the interval of bill detection of each permeable sensor
The product with the ratio of l ₁ and l ₂ is used as a correction value at the time detected by the reflection sensor located at the position where the defect occurs, and the added or subtracted time is set as the timing to start cutting out banknote data.

(Function) Transmissive sensors and reflective sensors are arranged on the banknote transport path at different intervals l ₁ and l ₂ in the direction perpendicular to the banknote transport direction. ). Each transmissive sensor and reflective sensor detects the presence of a banknote by its passage. Bill data from the reflective sensors is stored in the memory from the time when one of the transmissive sensors first detects a bill.

The central processing unit calculates the first skew amount θ of the banknote based on the banknote detection time difference detected by the normal detection sensor and the interval between the transmission sensors. Further, in the same manner, the central processing unit calculates the time difference between banknote detection by the reflective sensor, and calculates the second skew amount θ ₁ based on the time difference and the interval l ₂ between the reflective sensors. Further, the central processing unit detects which of the left and right sides of the banknote is running first based on which of the transmission sensors outputs first.

Here, when θ=θ ₁ =0, it can be seen that the banknote runs without being skewed, and there is no damage or the like at the tip. Furthermore, when θ=θ ₁ ≠0, the banknote is running obliquely, but it can be seen that the banknote has no damage or the like at its tip. Therefore, when θ=θ ₁ =0 and θ=θ ₁ ≠0, the output generation time of each reflective sensor is set as the cutting timing, starting from the time when one of the transmissive sensors detects a banknote.

On the other hand, when θ≠θ ₁ , that is, when the skew amounts are different, the central processing unit determines that there is a defect or the like at the leading edge of the banknote. Then, the central processing unit determines in which position of the transmission sensor the defect has been corrected based on the magnitude relationship between θ and θ ₁ and the combination of the determined leading direction. Furthermore, the central processing unit calculates the product of the time difference between the respective banknote detections of the transmission sensors and the ratio of the intervals l ₁ and l ₂ ,
As a correction value at the time detected by the reflection sensor located at the position where the defect occurs, the time added or subtracted is set as the timing to start cutting out the banknote data.

(Example) Hereinafter, an example of the present invention will be described in detail with reference to the drawings.

FIG. 1 is a block diagram of one embodiment of the present invention. In the figure, 1-A and 1-B are transmission sensors that constitute a transmission-type passage detection sensor for detecting the amount of skew of banknotes on the conveyance path, and 2-A and 2-B are the printed patterns (patterns) of banknotes. This is a reflective sensor that constitutes a reflective light receiving sensor that photoelectrically converts reflected light from ). These sensors are arranged as shown in the plan view of Figure 2b. In addition, as shown in the side view of FIG. 2a, the transmission sensors 1-A and 1-B are connected to light sources 1-A' and 1-B' which are installed at opposing positions across the conveyance path of the banknote P, respectively. , are arranged at a predetermined interval l ₁ perpendicular to the banknote transport direction. On the other hand, the reflective sensors 2-A and 2-B are transmissive sensors 1-A and 1- that constitute the passage detection sensor.
They are arranged on the downstream side of B in the conveyance direction and perpendicular to the conveyance direction with a predetermined interval l ₂ different from the interval l ₁ (in this example, l ₁ <l ₂ ).

Returning to FIG. 1, 3-A and 3-B are comparators that compare the signals from the transmission sensors 1-A and 1-B, respectively, with a predetermined threshold value SL.
Convert into binary data. The outputs of the comparators 3-A and 3-B are supplied to one input of a logic circuit (AND) 14 via a logic circuit (EX-OR) 13. The other input of the logic circuit 14 is supplied with a reference timing synchronized with the running of the banknote. 4 is a counter which starts counting the reference timing upon receiving a signal supplied to the input when either one of the transmission sensors 1-A or 1-B turns on, and when the other transmission sensor also turns on. As a result, the counting operation is performed until the logic circuit 14 is closed. A register 5 temporarily stores the count value of the counter 4. 12 is an I/O port, and comparators 3-A, 3-
A respective output signal of B is applied to bus 15.

Amplifiers 8-A and 8-B amplify the output voltages photoelectrically converted by the reflection sensors 2-A and 2-B to an optimum value, respectively. A multiplexer (MPX) 9 selects signals obtained from other reflection sensors not shown in FIGS. 1 and 2 via amplifiers. 10 is an A/D converter which converts the analog output signal of the multiplexer 9 into a digital signal according to reference timing and outputs the digital signal to the bus 15. A register 11 temporarily stores the digital signal from the A/D converter 10. Reference numeral 6 denotes a central processing unit (hereinafter referred to as CPU), which organically controls each part shown in the figure as described later. 7 is memory,
Via the CPU 6, the count value stored in the register 5 is stored, and the digital signal stored in the register 11 is stored for each reflective sensor.

Next, the operation of this embodiment will be explained with reference to FIGS. 3 to 5 in addition to FIGS. 1 and 2.

As shown in FIG. 3, the banknotes processed in this embodiment have a black printed part Q shown by diagonal lines in the figure, and a missing part R shown by broken lines at the tip of the banknote.
P' (on the other hand, it also has a black printed part Q, but
Let P be a banknote with no missing parts. ).

First, when the banknote P' travels on the banknote transport path and its tip comes directly under the transmission sensors 1-A and 1-B, the transmission sensors 1-A and 1-B are turned on, and their respective output signals are compared. 3-A and 3-B. Comparators 3-A and 3-B each compare the input electrical signal with a threshold value, and if this electrical signal exceeds the threshold value, it binarizes this electrical signal and sends it to the logic circuit 13 and I/O port. Output to 12. The logic circuit 13 is the comparator 3-
It turns on when a signal is output from either A or 3-B, and then turns off when the other signal is also output. Therefore, the counter 4 counts the reference timing input via the logic circuit 14 while the logic circuit 13 is on. here,
The value counted at count 4 is n, and the banknote
If the amount of skew during conveyance of P' is θ, and the amount of bill travel per reference timing is α, then the counter value n represents the difference in bill detection time between the transmission sensors 1-A and 1-B, and the following It is expressed as follows.

n=l ₁ tan θ/α ...(1) However, as mentioned above, l ₁ is the transmission sensor 1-
This is the distance between A' and 1-B. The count value n thus obtained is temporarily stored in the register 5 and then stored in the memory 7 via the CPU 6.

On the other hand, when the bill P' comes directly under the reflective sensors 2-A and 2-B, the reflective sensors 2-A and 2-B output electrical signals corresponding to the amount of reflected light. These electrical signals are amplified to optimum values by amplifiers 8-A and 8-B, respectively, and then input to multiplexer 9. Now, assuming that the multiplexer 9 is sequentially selecting the output signals of the amplifiers 8-A and 8-B, these output signals are A/D converted by the A/D converter 10 and temporarily stored in the register 11. After that, CPU6
is stored in a predetermined area of the memory 7 for each reflection sensor.

Here, the digitized output signals (corresponding to the signals stored in the memory 7) of the reflection sensors 2-A and 2-B are shown in FIG. Here, a in the same figure shows the waveforms of the output signals of the reflection sensors 2-A and 2-B when the banknote P (without the missing part R) travels without skewing, and b shows the waveform of the output signal of the reflection sensor 2-A and 2-B when the banknote P runs skewly. A waveform is shown when the right side of the banknote runs in advance, and c shows a waveform when a banknote P' having a missing portion R at the tip runs without being skewed. In the case shown in figure a, the time
At t _p , output signals are obtained from the reflection sensors 2-A and 2-B, but in b, if the right side of the bill is in front, the reflection is detected at time t _q , which is later than the time t _p at which this is detected. Sensor 2-A detects the tip. Furthermore, due to the influence of the defective part R on the scanning line of the reflective sensor 2-A, as shown in _{FIG .} For convenience, the reflection sensor 2-A produces an output at the same time as in the case of FIG. Note that t=0 in the figure is the same time as the time when either one of the transmission sensors 1-A and 1-B detects a banknote.

The data stored in the memory 7 as described above is processed by the CPU 6 as shown in FIG.
The digitized signals of the reflection sensors 2-A and 2-B are extracted.

First, in step 101, the CPU 6 reads the count value n stored in the memory 7, and calculates the skew amount θ of the banknote at the transmission sensors 1-A and 1-B as described above.
Calculate using formula (1). That is, θ=
Find the skew amount θ using tan ^-1 nα/l ₁ .

Next, in steps 102 and 103, the CPU 6 executes the reflection sensors 2-A and 2-A stored in the memory 7.
The digital signal of B is read out and compared with the threshold value V=d, and the digital signal is detected at time t=
After 0, detect the time when the threshold value is exceeded for the first time. This time is assumed to be t _a for the reflective sensor 2-A and t _b for the reflective sensor 2-B. Time difference between t _a and t _b |
The following relational expression holds true between t _b −t _a | and the distance l ₂ between the reflection sensors 2-A and 2-B.

｜t _b −t _a ｜=l ₂ tanθ ₁ /α ...(a) Then, in step 104, the CPU 6 calculates the skew amount θ ₁
is calculated as θ ₁ = tan ^-1 α|t _b −t _a |/l ₂ .

Next, in step 105, the CPU 6 transmits the transmission sensor 1-A, which is sent via the I/O port 12,
The leading direction of the banknote is detected from the output signal of 1-B.

Then, the CPU 6 determines the leading direction in step 106, and operates as follows depending on whether it is right leading or left leading.

For example, considering the case of FIG. 4b, times t _a and t _b become t _a =t _q and t _b =t _p , respectively. Therefore, θ ₁ is calculated as θ ₁ =tan ⁻¹ αt _q −t _p /l ₂ .

In the case of Figure 4b, the right lead is leading, so step 107
Proceed to. In step 107, the CPU 6 compares the skew amount θ calculated in step 101 with the skew amount _θ1 calculated in step 104. When θ=θ ₁ , that is, when n/l ₁ = t _q − _{t p} /l ₂ ...(3) holds, the CPU 6 determines that there is no damage to the banknote, and at time t _q , t _p is used as the cutting timing t _a and t _b of the digital signals related to the reflection sensors 2-A and 2-B, respectively (step 108).

On the other hand, if it is determined in step 107 that θ≠θ ₁ , the CPU 6 determines the magnitude relationship between θ and θ ₁ , that is, the magnitude relationship between n/l ₁ and (t _q − t _p )/l ₂ . (Step 109). In step 109, if n/l ₁ > t _q - t _p /l ₂ , the CPU 6 determines that the defective part R exists at the position of the reflective sensor 2-B, and sets t _p so that equation (3) holds true. Correct to t _p ′ as follows.

t _p ′=t _q −l ₂ /l ₁ n (4) Then, the times t _q and t _p ′ are the cutting timings t _a and t _p of the digital signals related to the reflection sensors 2-A and 2-B, respectively. (Step 110).

On the other hand, in step 109, when n/l ₁ <t _q −t _p /l ₂ , the CPU 6 determines that the defective portion R exists at the position of the reflective sensor 2-A, and the equation (3) holds true. Correct t _q to t _q ′ as follows.

t _p ′=l ₂ /l ₁ n+t _p ...(5) Then, the times t _q ′ and t _p are measured by the reflection sensor 2.
-A and 2-B are used as the cutout timings _ta and _tb of the digital signals (step 111).

On the other hand, if it is determined in step 106 that the driver is leading on the left, the CPU 6 operates θ in the same way as in step 107.
and θ ₁ are compared and calculated (step 112). In this case, let time t _a and t _b be t _a =t _x and t _b =t _y , respectively.
Then, in the case of θ=θ ₁ , that is, when n/l ₁ = t _y −t _x /l ₂ ...(6) holds, the CPU 6 determines that the banknote is not damaged, and sets the time t _x , t _Reflection sensor 2-
It is used as the cutout timings _ta and _tb of the digital signals related to A and 2-B (step 113).

On the other hand, if it is determined in step 112 that θ≠θ ₁ , the CPU 6 determines the magnitude relationship between θ and θ ₁ , that is, the magnitude relationship between n/l ₁ and ( _ty − t _x )/l ₂ ( Step 114). In step 114, if n/l ₁ > t _y - t _x /l ₂ , the CPU 6 determines that the defective part R exists at the position of the reflection sensor 2-A, and sets t _x so that equation (6) holds true. Correct to t _x ′ as follows.

t _x ′=t _y −l ₂ /l ₁ n ...(7) Then, the times t _x ′ and t _y are measured by the reflection sensor 2.
-A and 2-B are used as the cutting timings _ta and _tb of the digital signals (step 115).

On the other hand, in step 114, when n/l ₁ <t _y −t _x /l ₂ , the CPU 6 determines that the defective portion R exists at the position of the reflective sensor 2-B, and the equation (6) is In order to hold, t _y is corrected to t _y ′ as follows.

t _y ′=l _{2 /} l ₁ n+t _x ...(8) Then, the times _t
It is used as the cutout timings _ta and _tb of the digital signals related to A and 2-B (step 116).

Reflection sensor 2-A obtained as above,
Extraction timing of digital signal related to 2-B
According to t _a and t _b, the CPU 6 extracts the digital signal stored in the memory 7 to obtain banknote data, and the banknote data is subjected to authenticity determination together with the banknote data related to other reflective sensors obtained in the same manner.

One embodiment of the present invention has been described above. In this embodiment, by detecting the amount of skew of the banknote at the rear end of the banknote as well, it is possible to set an accurate cutting timing of the reflective sensor.

(Effects of the Invention) As explained above, according to the present invention, it is determined whether or not a banknote is damaged based on the skew amount of the banknote and the leading direction of the banknote, and when it is determined that there is no damage, each reflective sensor The time detected by each is the timing for cutting out the banknote data, and if there is a defect, the magnitude relationship between the amount of skew detected by the transmission sensor and the amount of skew detected by the reflection sensor, and the leading direction of the banknote are determined. The time detected by the reflective sensor is corrected based on the ratio between the detection time difference by the transmission sensor and the interval between the transmission sensors, and the ratio between the detection time by the reflection sensor and the interval between the reflection sensors. Therefore, it is possible to set the bill data cutout timing for correctly reading bill data according to the physical conveyance state of the bill, even if the bill is partially damaged or the like.

[Brief explanation of drawings]

FIG. 1 is a block diagram of one embodiment of the present invention, and FIG.
The figures are diagrams showing the arrangement of the transmission sensor and the reflection sensor in this embodiment. Figure a is a side view, figure b is a plan view, and Figure 3 is a plan view showing banknotes processed in this embodiment. FIG. 4 is a diagram showing a signal obtained by digitizing the output signal of the reflection sensor in this embodiment, and FIG. 5 is an operation flowchart for setting the cutout timing of the operations of this embodiment. 1-A, 1-B...Transmission sensor, 2-A, 2-
B... Reflection sensor, 3-A, 3-B... Comparator,
4...Counter, 5...Register, 6...Central processing unit (CPU), 7...Memory, 8-A, 8-B
...Amplifier, 9...Multiplexer (MPX), 1
0...A/D converter, 11...Register, 12...
...I/O port, 13...Logic circuit (EX-
OR), 14... logic circuit (AND), 15... bus.

Claims (1)

[Claims] 1. A bill data cutout timing setting method for obtaining bill data detection timing under control of a central processing unit using a plurality of optical sensors disposed on a bill conveyance path, comprising: sensors are arranged at intervals l ₁ perpendicular to the running direction of the banknotes, each of the first sensors detects the presence of the banknote, and a counter is provided, and the counter is configured to Counting is started when a banknote is detected by the other first sensor, and counting is stopped when a banknote is detected later by the other first sensor, and the time difference between banknote detections is calculated. and arranged to be located downstream of the passage detection sensor at an interval _{l 2} different from the interval l 1 of the first sensor, and detect the presence of a banknote with each of the second sensors, a memory for storing a bill detection time difference of one sensor and bill detection data from a second sensor; the memory stores bill data from the second sensor when either one of the first sensors Starting from the time when the banknote was first detected, the central processing unit determines the banknote detection time difference by the first sensor and the interval l ₁ of the first sensor stored in the memory.
calculate a first skew amount θ of the banknote; calculate a time difference between the banknote detections of the second sensors from the banknote detection data from the second sensor stored in the memory; A second skew amount θ ₁ of the banknote is calculated from the calculated time difference and the interval l ₂ of the reflective sensor, and the leading direction of the banknote is determined based on which output of the first sensor comes first. and the calculated first skew amount θ and second skew amount θ ₁
If they are equal, the banknote is considered to be free of defects, and the first
The starting point is the time when one of the sensors detects the banknote, and the time when each of the second sensors detects the banknote is the timing to start cutting out the banknote data, and the calculated first skew amount θ and the first skew amount θ 2 skew amount θ ₁
If they are different, it is assumed that there is a deficiency and the first
and the second skew amount θ and θ ₁ and the combination with the determined leading direction, it is determined in which position of the second sensor the defect has occurred, and each of the first sensors is determined. The product of the time difference in banknote detection and the ratio of the intervals l ₁ and l ₂ is added or subtracted as a correction value at the time when the second sensor located at the position where the defect occurred is added or subtracted from the banknote data. A method for setting a timing for cutting out banknote data, the method comprising setting the timing to start cutting out.