JPH04140250A - Detecting device for folded corner of paper sheet - Google Patents

Abstract

PURPOSE:To detect a folded corner, distinguishing from a patched tape and double-ply sheets by observing a condition of falling of a detecting roller after bouncing at the leading end of a paper sheet so as to detect a folded corner. CONSTITUTION:With an example of tape is patched to the leading and trailing end parts of a paper sheet, sampling points #1 to #21 exhibit noise caused by vibration, but the zone #22 to #53 corresponding to a part of the paper sheet which is actually engaged with thickness detecting rollers are equally divided into 8 for every of four data. However, at point #22, the leading end of the paper sheet comes to the thickness detection rollers 20A, 20B and bumps against the latter so that the bouncing of the detection rollers is relatively large. Thereafter, they drop under reaction at point #27, then rebounce at point #30 by reflection, and thereafter they again drop at point #33. Accordingly, excepting the zone from #22 to #33, only five zones are observed. As a result, since the value of the zone #50 to #53 becomes minimum, it is estimated that this value is the thickness of the paper sheet. Further, if this value is out of a predetermined range, it is regarded the double ply sheets or a patched tape is present, so that it is rejected.

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial Application Field] The present invention relates to a device for detecting corner folds of paper sheets in equipment that processes paper sheets such as banknotes.

[Conventional technology]

2. Description of the Related Art In devices that classify, eject, etc. paper sheets such as banknotes, abnormality in the thickness of paper sheets is detected. This is because thickness abnormalities often occur when there are abnormalities such as two overlapping paper sheets or counterfeit tickets with tape attached.

Various proposals have been made to detect paper sheets with such abnormal thickness. For example, JP-A-1-1275
Publication No. 40 discloses that by setting the gap of the thickness detection roller to 0 from the beginning, the amount of displacement of the roller is different in the case of one sheet and the case of two sheets, and the amount of displacement of the roller is measured. A device has been disclosed that detects paper sheets that are stacked or pasted with tape.

In addition, in the copy of the specification originally filed for Utility Model Application No. 61-145748, the gap between the thickness detection rollers is slightly wider than the thickness of one sheet, and two sheets are overlapped, corners are bent, etc. A device has been disclosed in which a thickness detection roller is displaced when a paper sheet pasted with tape passes, thereby detecting paper sheets having an abnormal thickness.

[Problem to be solved by the invention]

However, with the former method, although it is possible to detect paper sheets that are double stacked or have tape pasted anywhere other than the edges, it is not possible to detect tape that is pasted to the edges or folded corners. This is because noise is generated at the edges of paper sheets due to mechanical vibrations, so the thickness detection signal is not seen at the edges.

On the other hand, in the latter case, the gap between the thickness detection rollers is narrower than the thickness of one sheet of normal paper, so
When paper sheets pass, the thickness detection roller is kicked up, and when folding corners and pasting tape, the amount of kick-up is greater than when stacking two sheets.Using this fact, folding corners and pasting tape can be done by stacking two sheets. It can be detected separately. However, even by observing the amount of kick-up, it is not possible to distinguish between bent corners and taped notes, and both are considered unfit notes.

By the way, there is a high possibility that tape-applied tickets are counterfeit tickets, so they should be avoided as much as possible, or folded corners do not indicate any defects in the paper sheet itself and are genuine bills. Therefore, eliminating non-defects such as corner folds reduces the efficiency of the equipment, so it is preferable to allow them to pass through.However, with conventional equipment, there is nothing that can distinguish between corner folds and taping, and it is difficult to distinguish between them. There is an increasing demand to do so.

The present invention has been made in view of these points, and an object of the present invention is to provide a folded corner detection device for paper sheets that can distinguish folded corners of paper sheets from two stacked sheets when taped. shall be.

[Means to solve the problem]

According to the corner fold detection device for paper sheets according to the present invention, there is provided a thickness detection means for detecting the thickness of the paper sheet being conveyed over the entire length from the front edge to the rear edge, and an output of the thickness detection means is sampled. A measured thickness storage means for storing, a thickness estimating means for estimating the thickness of the paper sheet based on the stored content of the measured thickness storage means, and a thickness estimating means for reading the stored content of the measured thickness storage means to determine the thickness of the transported paper sheet. corner fold detection means for detecting a corner fold of the paper sheet when the output of the thickness detection means becomes equal to or less than the thickness estimated by the thickness estimation means within a relatively short predetermined distance after reaching the thickness detection means; It is characterized by having the following.

[For production]

In the present invention, when the leading edge of a paper sheet reaches the thickness detection roller, the roller jumps up due to a sudden change in thickness, but it is noted that the subsequent fall height of the roller differs between taping and corner folding. I try to distinguish between these.

In other words, in the case of tape attachment, since the thickness of the tape is thinner than the thickness of the paper sheet, the jump height is not so large, and the position of the bottom dead center of the roller is lower than the estimated thickness of the paper sheet. It will never become. On the other hand, in the case of corner folding, the roller jumps up at the thickness of two sheets, so the falling distance is also large, and the sheet falls below the estimated thickness of the sheet. Therefore, by observing this situation, it is possible to distinguish between corner folding and tape attachment.

〔Example〕

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

FIG. 1 is a block diagram showing the general configuration of a corner fold detection device according to the present invention.

According to the figure, a thickness detecting means 100 that detects the thickness of the paper sheet being conveyed over the entire length from the front end to the rear end, a measured thickness storage means 200 measured by the thickness detecting means 100, and a storage means 200 for storing the measured thickness. A thickness estimating means 300 for estimating the thickness of a paper sheet from the contents stored in the means 200, and a corner fold detection means 400 for detecting a corner fold from the estimation result in the thickness estimation means 300 and the stored contents in the measured thickness storage means 200. ing.

FIG. 2 is a front view showing the thickness detection mechanism.

For example, paper sheet 1, which is a banknote, has pulleys IIA and 12A.
It is conveyed by a belt 13A stretched between them and a belt 13B stretched between pulleys IIB and 12B, and a magnetic head 2 for detecting magnetic ink and a light beam are installed in the middle of this conveyance path. An identification means 10 consisting of a light source 3 and an image sensor 4 for detecting a transmission pattern is provided to detect the denomination and fitness. A pair of thickness detection rollers 2OA, 20B and another set of belt conveyance mechanisms 14A, 15A, 16A are provided on the extensions of the belts 13A, 13B.

14B, 15B, and 16B are provided. This other one
The set of belt conveyance mechanisms are used to hold the paper sheet 1 that has passed the thickness detection sensor, and when there is no abnormality, it is further conveyed to the right in the figure, and when there is an abnormality, it is reversed and conveyed to the left. It is something. In addition, one sensor 5 made of a photo interrupter is provided on each side of the conveyance path near the pulleys 12A and 12B, which are intermediate positions between the thickness detection rollers 2OA and 20B and the image sensor 4. Generates a timing signal for detection.

FIGS. 3(a) and 3(b) are enlarged explanatory views showing details of the thickness detection mechanism related to the thickness detection roller.

As shown in FIG. 3(a), the thickness detection roller 2OA,
20B are arranged to face each other with a gap α slightly narrower than the thickness of one ordinary paper sheet. In other words, the thickness of normal banknotes is 75 to 100 μm, with a standard thickness of 85 μm.
The gap α is set to, for example, 60 μm. The thickness of the tape that is often applied to paper sheets is 50 μm in the case of sellotape (trade name) and 55 to 60 μm in the case of Scotch tape (trade name).

A cylindrical portion 21 having a smaller diameter than this is provided on the end face of the upper detection roller 2OB, and a small cylindrical driven body 22 is in contact with the surface of this cylindrical portion 21. This driven body 22 is attached to a rotation shaft 23, extends in a direction substantially perpendicular to the driven body 22 when viewed from the rotation axis, and is adapted to bias the driven body 22 against the cylindrical portion 21 with a spring 27. A pressing body 25 is attached to the tip of the arm 24,
This movement of the pressing body 25 is detected as an angle signal by a rotation sensor 26 consisting of a potentiometer that follows this movement.

Such thickness detection mechanisms are provided at two locations on both sides of the conveyance path.

Next, the operation in this embodiment will be explained.

FIG. 4 is a flowchart showing an outline of the operation.

First, a roller (not shown) takes in multiple sheets of paper for one transaction one by one, which are loaded into a hopper (not shown) at once.
is captured by rotation (step 511).

Data on the paper sheets taken in in this way is collected by the magnetic head 2 and image sensor 4 provided during transportation, and the denomination and authenticity are identified (step 512).

Next, when the leading edge of the paper sheet reaches the photosensor 5, sampling of the output value of the potentiometer 26 is started in synchronization with the conveyance of the paper sheet, for example, every time the paper sheet moves 2 mm. As mentioned above, the gap between the thickness detection rollers 2OA and 20B is smaller than the thickness of one sheet of paper, so in the initial state, the output value of the potentiometer 26 is close to 0, which indicates whether the paper sheet is Thickness detection roller 2OA2
When reaching 0B, the output value of the potentiometer 26 is sampled every movement 2III11, A/D converted, and stored in the memory corresponding to the measured thickness storage means 200 (step 813). At this time, potentiometer 26
are provided on both sides of the conveyance path, so the right channel and the left channel are alternately switched and stored until the sheet passes through the thickness detection rollers 2OA and 20B. For example, if the paper sheet is sampled 64 times from when it hits the photosensor 5 until it passes through the thickness detection roller, 128 pieces of sampling data for both channels will be stored.

In this case, in order to eliminate the influence of offset of the amplifier due to machine difference, the CPU calculates the value (0) from the stored sampling data until the paper sheet reaches the thickness detection roller.
Although it is a close value, it is often not exactly 0. ) is subtracted and stored as an accurate displacement signal.

Next, the obtained sampling data is divided into several regions in the paper sheet transport direction, the average value within the region excluding the edge of the paper sheet is determined, and the minimum value of the obtained average values is calculated. is estimated to be the thickness of the sheet (step 514). An example of this situation is shown in FIG. This example shows an example in which tape is attached to the leading and trailing ends of a paper sheet, and sampling points 1 to 21 are noise due to vibration. Here, the time points from the 22nd to the 53rd points when the sheet actually engages with the thickness detection roller are divided into eight equal parts for each four pieces of data. However, at the 22nd point in time, the leading edge of the paper sheet reaches the thickness detection rollers 2OA and 20B, but because it hits the thickness detection rollers 2OA and 20B with an impact, the detection rollers jump up by a large amount. At the 27th point it falls due to the reaction, at the 30th point it jumps up again due to the reflection, and at the 33rd point it falls again. Therefore, the three areas including the 22nd to 33rd points are excluded, and only five areas are observed. As a result, since the value of the area including the 50th to 53rd points is the minimum value, this is taken as the estimated thickness.

Next, this estimated thickness is compared with the reference thickness in step 515.
), check whether it is within a predetermined range (step 516).
. That is, if the estimated thickness is within a predetermined range, it is normal; if it is outside the range, it is assumed that two layers or the entire surface is taped, and the product is rejected (step 518). As mentioned above, the thickness of the tape is thinner than the thickness of paper sheets, so a reference value (TH2) for detecting double stacking and a lower reference value (THI) for detecting full-surface tape application are set. This is shown in Figure 6. Note that this tape pasting identification standard value is the standard thickness plus a predetermined value equivalent to the tape thickness, and the double-plying standard value is the standard thickness plus a predetermined value equivalent to the thickness of the other paper sheet. It was added.

Next, the estimated thickness is subtracted from the displacement signal from which the offset has been removed to obtain the actual displacement amount (step 517). The actual displacement amount obtained in this manner is shown in FIG. 7, and this value is also stored in the memory.

Then, the data of the entire range of paper sheets of the actual displacement amount obtained in this way is compared with the threshold value (step 520),
Check whether there is an abnormality in the thickness of paper sheets (step 521).
), and if there is an abnormality, reject it (step 518
).

Next, since the authenticity identification result in step S12 has already been obtained at this point, the quality is checked (step 522), and if there is an abnormality, the result is rejected (step 518).

The above steps are continuously repeated until all loaded paper sheets have been taken in (step 823).
).

Next, the process moves to the standard thickness update procedure. For this reason, it is checked whether the current transaction involves a predetermined number of coins, for example, 16 or more (step 524). Note that when the number of traded coins is small, a sufficient average value has not been obtained, so the process ends immediately.

If the number is equal to or greater than the predetermined number, a reference thickness update calculation is performed (step 525). This calculation is an average calculation of the estimated thickness of normal paper sheets. For this reason, the estimated thicknesses of the obtained normal paper sheets are totaled and divided by the number of sheets. Then, the updated value is obtained by summing the previous reference thickness and the current reference thickness and dividing the sum by 2 (averaging).

Next, it is checked whether the updated value of the reference thickness obtained in this way is within a predetermined range (step 526), and if it is outside the predetermined range, the system is stopped as an error (step 528). However, the transaction itself will not be invalidated.

This is because if the reference value fluctuates too much, it is possible that some kind of problem, such as gap fluctuation, has occurred on the apparatus side. On the other hand, when the updated value is within the predetermined range, it is updated and stored as a new reference thickness (step 527).

FIG. 5 is a flowchart showing details of the thickness abnormality detection step of step S20.

First, read the actual displacement data from the memory. Step 5
201), each value is compared with a high threshold THon (step 5202), and a counter is started when the value of this high threshold THon is exceeded (step 3).
203). Then, it is compared with a low threshold value THoff (step 5204), and if it is less than this value, the counter is stopped (step S205). The reason why two threshold values are used in this way is to absorb noise caused by mechanical vibration.

Next, after the counter starts, the continuous count value until reaching the low threshold THoff is determined (step 5206).
), and a portion where this continuous count is a predetermined number, for example, 5 or more is detected (step S207). Then, it is checked whether the number of rising portions is greater than or equal to a predetermined number (step 3208), and if not, the process jumps to step 5212 since there is no possibility of corner bending. In addition, the count is 1
This is done for all data on a sheet of paper.

On the other hand, if it is equal to or greater than a predetermined value, there is a possibility of corner bending, so this is detected.

First, it is checked whether there is a part where the actual displacement becomes 0 up to the third region (step 5209), and if there is a part, the corner bend flag is turned on (step 5210). This is because the thickness detection roller is thought to have jumped up high due to the corner fold, and as a result fell to a deep position, resulting in a portion that is less than the thickness of the paper sheet. This situation is shown in Figures 8 and 9.
This is explained in detail in the figure. If there is a corner bend only at the right end in the conveyance direction, the thickness data after subtracting the offset value will be as shown in FIG. 8, and there are places where the thickness data is 0 in the second region.

This shows the data of the actual displacement after subtracting the estimated thickness.
It is even more clearly visible in the figure. Therefore, T.H.
Corner bending may be detected based on whether there is a location where the actual displacement becomes 0 in the second region R2 after the state of on is exceeded. However, if there is a corner bend and skew, the point where the actual displacement is 0 will be missed, so the range to observe when the actual displacement becomes 0 is expanded as shown in Figure 9, and it is observed in the range of R2'. You should do it.

Next, in order to check for defects other than corner bends, the data of the rising portion is ignored (step 5211). On the other hand, if there is no part where the actual displacement becomes O up to the third area, it is determined that tape is to be pasted, and the tape pasting flag is turned on (step 8230).

Next, when the count value is greater than a predetermined number in a part other than the rising edge (step 821.2), the tape pasting flag is also turned on (step 5230), and if it is less than the predetermined number, the average value for each segmented area is Among the values, read out the remaining 5 areas excluding the stable ones, that is, the unstable 3 areas of the first kick part (Step 3213)
, these are compared with the tape pasting identification reference value (step 5214). Then, it is checked whether there is a predetermined number of items, for example, 3 or more, that are larger than this tape application identification reference value (THl in FIG. The flag is turned on (step 3230).

If the number is less than the predetermined number (3), read the displacement signal data after subtracting the offset value (step 5216);
Continuous data larger than the two-sheet stacking reference value (TH2 in FIG. 6) (step S217).

Then, it is checked whether the number is a predetermined number, for example, 20 or more (step 8218), and if it is 20 or more, it is determined that two sheets are stacked and the double stack flag is turned on (step S240). If the number is less than 20, read out the displacement signal data again after subtracting the offset value (step 5219), and find the number of continuous data smaller than a separately preset thickness reference value (step 5220).
).

Then, it is checked whether the number is a predetermined number, for example, 20 or more (step 5221), and if it is 20 or more, it is determined that the thickness is too small and the too-thickness flag is turned on (
Ste Mbu 5250). This occurs when the paper sheet itself is thin (for example, a paper sheet that has been peeled into two sheets), or when the gap between the detection rollers fluctuates and widens due to some reason.

Such abnormality determination is performed for both the left and right channels,
Based on the results, a final judgment is made based on the following judgment criteria to determine the number of pieces.

If the thickness is too small, DS is used, taping is TP, and double stacking is Do, the left and right judgment results are as shown in the table below.

In the examples above, each channel is independently judged to be taped when five or more tapes are pasted in a row, but for example, a smaller number, for example, four consecutive tapes are pasted at approximately the same location (nearby position within 10+u+). When this is detected, tape attachment may be detected as if a thin tape was attached in a direction perpendicular to the conveyance direction.

In addition, in this embodiment, it is also possible to detect skew feeding of paper sheets.

FIG. 10 shows an example of a normal paper sheet, and the right channel and left channel show almost similar waveforms.

FIG. 11 shows an example of such normal paper sheets being skewed, and the waveforms of the right and left channels are extremely different. A characteristic of skew feeding is that the waveform rises earlier than in the case of normal paper sheets. Therefore, as mentioned above, for normal paper sheets, 22
The output does not rise at the second sampling point, but as shown in FIG. 11, the output rises at that point in the case of skewed paper sheets. Therefore, in order to prevent erroneous detection of tape attachment, data is deleted from the 22nd to 24th sampling points.

By doing so, the number of data items from exceeding THon to falling below THoff will be less than the predetermined number, so there will be no confusion with tape attachment.

[Effects of the Invention] As described above, according to the present invention, corner folds are detected by observing how the detection roller at the leading edge of the paper sheet falls after it jumps up. It becomes possible to clearly distinguish corner folds from overlaps and detect corner folds.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing the general configuration of the device according to the present invention, FIG. 2 is a front view illustrating the thickness detection means, FIG. 3 is an explanatory diagram showing the detailed configuration and operation of the thickness detection means, and FIG. 4 is a flowchart explaining the operation of the device according to the present invention,
FIG. 5 is a flowchart explaining the details of the thickness abnormality detection operation in FIG. 4, and FIGS. 6 to 11 are waveform diagrams showing examples of thickness detection waveforms of banknotes in various states. 1... Paper sheet, 2... Magnetic head, 4... Image sensor, 5... Photo sensor, 2OA, 20B...
・Thickness detection roller, 100...Thickness detection means, 200
... Measured thickness storage means, 300 ... Thickness estimation means, 400 ... Corner fold detection means.

Claims (1)

[Scope of Claims] Thickness detecting means for detecting the entire thickness of paper sheets being conveyed from the front edge to the rear edge; Measured thickness storage means for sampling and storing the output of this thickness detecting means; This measurement Thickness estimating means for estimating the thickness of the paper sheet based on the stored content of the thickness storage means, and comparison after reading the stored content of the measured thickness storage means and after the conveyed paper sheet reaches the thickness detecting means Corner fold detection means for detecting corner folds of paper sheets when the output of the thickness detection means becomes equal to or less than the thickness estimated by the thickness estimating means within a predetermined distance close to the target. Detection device.