JPS6035716B2 - Print identification method - Google Patents

Description

[Detailed Description of the Invention] The present invention relates to a method for identifying printed matter such as banknotes and checks, and in particular, to sample and judge the detection force of a sensor obtained in response to a running printed matter to determine its suitability. Regarding the method.

When identifying printed matter, it is most common to determine its suitability by shining light on a specific area and detecting its transmittance or reflectance, or by detecting the magnetic component contained in the printing ink. It is.

However, according to this method, if the detection force of the inspection area happens to show the same value as in the case of proper printing even if the printed matter is not suitable, it may be considered suitable, and conversely, even if the printed matter is proper printed matter, the inspection location may be extremely dirty or damaged. There were drawbacks that made it inappropriate if On the other hand, if the suitability is determined based on the detection force sequentially generated by running the printed matter and passing measurement points, a part of the printed matter can be continuously inspected in the longitudinal direction, and the above-mentioned drawbacks can be solved.

However, since this method observes not as a point but as a line, each point has a different detection power from the proper case, but the sum of these forces, that is, the total detection power observed as a line, is There is an irrationality in determining that something is appropriate when the two forces are in agreement. In view of the above, the present invention relates to a method for determining the suitability of a running printed matter, and aims to improve identification accuracy by dividing the printed matter into sections and looking at the characteristics of each section based on the detection power of each section.

Furthermore, a problem that always arises in printed matter identification is a decline in identification performance due to fluctuations in detection power as the printed material becomes contaminated or worn out.

Regarding this point as well, the present invention aims to perform accurate identification by comparing the detection power of each section to cancel out the influence caused by this. An embodiment of the present invention will be described in detail below based on the drawings.

FIG. 1 shows a front view of a printing material separation device according to the present invention;
The printed material inserted through the insertion port 14 is carried between the belts 12 and 13 and conveyed in the direction of the arrow. Belt 12, 13
are stretched around pulleys 15 and 16 and pulleys 17 and 18, respectively, and are driven by a pulley 18 connected to a drive shaft of a motor 19. The detector 20 is disposed on the insertion row 4 and drives the motor 19 when it detects insertion of a printed matter. The identification sensor 1 is composed of a light emitting element 21 and a light receiving element 22, one of which is scanned in a strip shape in the longitudinal direction as the printed matter passes, and changes in the amount of light received by the light receiving element 22 are measured. Figure 2 shows the identification circuit.

The clock pulse generator 23 generates a clock pulse every 5 times when the printed material moves past the sensor in synchronization with the transport of the printed material.If the total length of the printed material to be identified is 15 pieces, 3 clock pulses are generated during the transport of the printed material. Output across. The light-receiving element 22 changes the amount of light it receives as the printed matter passes through it, depending on the density of the printing, and outputs an electrical signal corresponding to the printed pattern. The light receiving element 22 is connected to the A/D converter 2 via the intensifier 11, and the A/○ converter 2 samples the output of the light receiving element 22 every time a clock pulse is input to measure a digital quantity. Convert to data. The clock pulses are also introduced into a counter circuit 24, and the counter circuit 24 counts the clock pulses and determines whether the counted value is 10 or 2.
0 and 30, output signals a, b, and c are generated, respectively. The memory 7 has a storage capacity of 10 words and stores measurement data output from the A/D converter 2 every time a clock pulse occurs. The measurement data of the generated 1-bo weight will be stored. When the adder 3 receives any of the signals a, b, and c through the OR gate 25, it sums up the ten measurement data stored in the memory 7. memory 8
, 9, and 10 are written with summation data by the adder 3 in response to signals a, b, and c, respectively. The calculation discriminator 5 is operated by the signal c, calculates the difference between the data in the memory 8 and the memory 9, and determines the suitability based on this result. Further, the calculation discriminator 6 is operated by the signal c, calculates the difference between the data in the memory 8 and the memory 10, and determines the suitability based on this result. Furthermore, the discriminator 4 discriminates suitability based on each data in the memories 8, 9, and 10. Then, when any of the arithmetic discriminators 5, 6 and the discriminator 4 determines the appropriateness, the AND circuit 26 outputs a printed matter acceptance signal. With the above configuration, when a banknote inserted through the insertion slot 14 reaches the sensor, the clock pulse generator 23 outputs a clock pulse every time the printed matter is transferred every five strips, and the sensor 1 continuously detects while the printed matter passes. Generate strength.

The A/D converter 2 samples this detection force every time a clock pulse is input, and converts it into a digital quantity. Store data. 5 printed items
When the weight is transferred, that is, the first clock pulse is generated and the counter circuit 24 counts 10 and generates the signal a, the sum of the measurement data of the first weight stored in the memory 7 is added to the adder 3.
The calculation result is written into the memory 8. Therefore, data representing the characteristics of the front section area from the leading edge of the printed matter to the fifth scale is stored in the memory 8. When the printed matter is further transferred, the memory 7 is sequentially rewritten with the measurement data output from the A/D converter 2 at the 11th to 2nd clock pulse, and the counter circuit 25 counts 20 and generates the signal b. and adder 3 from the 11th to 2G fleas. The total sum of the measured data from the clock pulses is calculated and stored in the memory 9. Therefore, the memory 9 stores data indicating the characteristics of the middle section area from the 5th to the 10th scale of the printed material. Subsequently, the printed material is transferred, and the memory 7 is rewritten with the measurement data output from the A/D converter 2 at the 21st to 3rd clock pulse, and when the counter circuit 24 counts 30 and generates the signal c, it is added. The total sum of the measurement data of the first weight that has been rewritten in the memory 7 by the device 3 is calculated and stored in the memory 10'. Therefore memory 01
This stores data indicating the characteristics of the rear section area from the 10th proboscis to the rear end of the printed matter. When the trailing edge of the printed material finishes passing the sensor 1, this transporting operation stops. In the discriminator 4, measurement data regarding the front, middle, and rear sections obtained when a proper printed material is inspected under the same conditions is set as reference data, and when the measurement data is stored in the memories 8, 9, and 10, It compares with the corresponding reference data and outputs a match signal if all data are within the allowable error range.

Further, the difference between the measurement data regarding the front section and the middle section obtained when proper printed matter is inspected under the same conditions is set in the calculation discriminator 5 as reference data, and the memory 8, 9
This reference data is calculated by calculating the difference between the measured data stored in the
If it is within the allowable error range, a matching signal is output. In the calculation discriminator 6, the difference between the measurement data regarding the front section and the rear section obtained when a suitable printed material is inspected under the same conditions is set as reference data, and is recorded in the memories 8 and 10. The difference between the measured data is calculated and compared with this reference data, and if the difference is within the allowable error range, a match signal is output. When the signal c arrives, the AND circuit 26 outputs an acceptance signal indicating the suitability of the printed material if the arithmetic discriminators 5, 6 and the discriminator 4 all generate matching signals. Obtaining measurement data for each section in this manner allows the characteristics of each section of the printed matter to be grasped, thereby ensuring reliable identification.

That is, as long as the measurement data obtained for the entire printed matter satisfies the standard regardless of whether it is inappropriate, the above-mentioned irrationality of determining that the printed matter is appropriate even if different measurement data are shown for each part can be avoided. Furthermore, even if a sampling point on a printed matter is damaged and sufficient data cannot be obtained, the measurement data for each section is the sum of several sampling data, so the effects of this can be smoothed out and even if some parts are damaged. Printed matter will not be deemed inappropriate. Therefore, it is desirable that the allowable error range be set somewhat roughly when comparing and determining the measurement data for each section. In addition, by making judgments based on the differences in the measured data for each section, even if the surface of the printed matter is uniformly dirty and the sensor's detection power fluctuates, this effect can be canceled out by calculating the difference in the measured data. This avoids a decrease in identification accuracy. Further, in this example, the printed matter is divided into three parts and the measurement data is viewed, but the number is not limited to this. However, if the division is too finely divided, even if the printed matter is properly printed, a certain sampling point is damaged and sufficient data cannot be obtained, which would be affected by the situation, making it impossible to take advantage of the above-mentioned characteristics. According to the present invention described in detail above, the measurement data obtained by the relative movement of the printed matter and the sensor is sequentially stored in the memory, and the data is calculated for each section to see the characteristics of each section of the printed matter, thereby improving identification accuracy. .

Furthermore, even if the surface is dirty or partially damaged, it can be reliably identified if it is proper, and is particularly effective in identifying banknotes.

[Brief explanation of the drawing]

FIG. 1 is a front view of a printed matter identification device according to the present invention, and FIG. 2 shows a control circuit. 1...Sensor, 2...A/○ converter, 3
... Adder, 4 ... Discriminator, 5, 6 ...
・Arithmetic discriminator, 7”HH memory. Figure 1 Figure 2

Claims (1)

[Claims] 1. Means for generating a detected force in response to relative movement with the printed matter, means for sampling the detected force at predetermined intervals, and means for dividing the printed matter and summing the sampling values for each section. , a means for comparing the summed value with a reference value set in advance for each section, and a means for comparing the difference between the two summed values with another reference value set in advance, and which comparison is made. A method for identifying printed matter, characterized in that the printed matter is determined to be appropriate if the means match within an allowable error range.