JPH0327951B2 - - Google Patents

Description

[Detailed description of the invention] The present invention relates to a barcode recognition device.

In POS systems, etc., product classification, price, etc. are recorded on labels using barcodes. The barcode above is the same as the JAN standard.
-7 standard is generally used. this
In the NW-7 standard, as shown in FIG. 1, one character is composed of a total of seven black and white bars, with thin bars being "0" and thick bars being "1". The barcode is read optically. As mentioned above, by using barcodes in a POS system, calculation processing such as totaling the price of sold products and aggregation can be done quickly and
It can be done reliably. Therefore, stationary scanners and hand scanners are generally used as barcode reading devices. Since the hand scanner manually scans the barcode, the scanning speed changes, and a length difference occurs between the barcode pattern and the electrical pattern after it is read, reducing the reading accuracy. In order to eliminate such problems, the bar width value has conventionally been determined as follows.

(a) Multiply the width value of the previous bar by a constant K to obtain the threshold value T _H , and multiply the width value of the previous bar by a constant 1/K to obtain the threshold value.
Find T _L.

(b) Compare the width value M of the bar just read with the threshold values T _H and T _L to determine the width as follows.

M<T _L ...Smaller than the previous bar.

M>T _H ...Bigger than the previous bar.

T _H > M > T _L ...Equal to the previous bar.

Therefore, depending on which threshold value is selected, there is a large difference in reading accuracy. That is, by setting the threshold value between the thin bar and the thick bar, reading accuracy can be improved. However, in the past, a constant constant K was always applied to the width value of the previous bar.
and 1/K to obtain the threshold values T _H , T _L
I'm looking for. Therefore, for example, if the previous bar width value is large, the smaller threshold value
Even if you set a constant so that T _L is an intermediate value between the thin bar and the thick bar, if the previous bar width value and the current bar width value are small, the threshold hold value T _L and the thin bar width The difference between the two values is small, that is, the margin is small and the reading accuracy is reduced.

The present invention has been made in view of the above points, and when recognizing a bar code in which the ratio of bar width values is x:1, the scanning speed is increased by increasing the tolerance value for speed fluctuation. It is an object of the present invention to provide a barcode reading device that can accurately read barcodes even if there are some changes, and can improve operability and reliability.

An embodiment of the present invention will be described below with reference to the drawings. In FIG. 2, reference numeral 11 denotes a hand scanner pen having an optical-to-electronic conversion function, and the output of this pen 11 is converted into a digital signal by an A/D conversion circuit 12 and sent to a pulse detection circuit 13. This pulse detection circuit 13 detects the leading edge and trailing edge of each pulse of the input signal, and outputs the detection signal to the CPU 14. This CPU 14 internally has registers D,
Z flag F _F , F _S , address register A _W , A _R , A _Z
It is equipped with The above address registers A _W , A _R
is for specifying the address of the memory 15 connected to the CPU 14, _AW specifies the write address, and _AR specifies the read add. Also,
The address register A _Z is for addressing register Z. A counter 16 is connected to the CPU 14 to constantly count the clock pulses φ from the CPU 14. Each time a pulse signal is input from the pulse detection circuit 13, the CPU 14 reads out the count contents of the counter 16 and writes them into the memory 15, and at the same time clears the counter 16. Further, a Y register 17, an X register 18, and a flip-flop 19 are connected to the CPU 14. Y register 17 above
, the bar width value is written this time, and the held data is sent to the comparator circuit 20. The width value of the previous bar is written in the X register 18, and the data held therein is input to the multiplication circuits 21 and 22. Further, the multiplier circuit 21 is given a constant K _H from the constant storage section 23 . The value of the above constant K _H is
When the width of the thin bar is "1" and the width of the thick bar is "x", it is set so that K _H =1+x/2.

The multiplication circuit 21 multiplies input data according to instructions from the CPU 14, and outputs the multiplication result to the comparison circuit 20 via the register 24. On the other hand, the flip-flop 19 is
It is set or reset under the control of the CPU 14, and its Q side output is sent to the AND circuit 2.
5, the side output is input to the AND circuit 26. Further, the constant K _L is inputted to the AND circuit 25 from the constant storage section 27 , and the constant K _L ' is inputted from the constant storage section 28 to the AND circuit 26 . The above constant K _L
is set so that K _L =1+x/2x, and the constant K _L ' is set so that K _L '=2/1+x. The outputs of the AND circuits 25 and 26 are input to the multiplication circuit 22 via the OR circuit 29. The multiplication circuit 22 performs calculations on input data according to instructions from the CPU 14, and outputs the multiplication results to the comparison circuit 20 via the register 30.
This comparison circuit 20 compares and judges the width value of the current bar held in the Y register 17 and the data held in the registers 24 and 30 as shown in _FIG . Output S ₃ to CPU 14. The CPU 14 receives the comparison result signals S ₁ to _{S 3}
The bar code is recognized in accordance with the bar code and finally written to register D. When the character recognition of a predetermined digit is completed, the barcode recognition result held in the register D is sent to the arithmetic processing unit (not shown) via the data line DL.

Next, the operation of the above embodiment will be explained. When a barcode recorded on, for example, a product label is scanned with the pen 11, the barcode is converted into an electrical signal by the pen 11 and sent to the A/D conversion circuit 12. This A/D conversion circuit 12 is
The barcode signal read by the pen 11 is converted into a pulse signal having a time width corresponding to each pattern of the barcode, and is output to the pulse detection circuit 13. This pulse detection circuit 13 detects the rising edge and falling edge of each input pulse signal, and converts the detected signal into
Input to CPU14. This CPU 14 constantly supplies a clock pulse φ to the counter 16 to cause it to perform a counting operation, and each time a pulse detection signal is input from the pulse detection circuit 13, the CPU 14 reads out the count contents of the counter 16, sends it to the memory 15, and stores the count in the address register A _W The data is sequentially written to the addresses specified by , and the counter 16 is cleared at the same time. In this way, count values corresponding to the width of the barcode read by the pen 11 are sequentially written into the memory 15. Then, when the barcode is written into the memory 15 by scanning the pen 11, the CPU 14
Barcode recognition processing is started according to the flow shown in the figure.

First, as shown in step _A1 of Figure 4,
The CPU 14 writes "1" to the address register _AR , and then, as shown in step _A2 , clears the register Z, flags F _F and F _S , and resets the flip-flop 19 (F/F). Then address register A _R as shown in step _A3
Address the memory 15 according to the contents of
The held data is read out to the X register 18, and as shown in step _A4 , the multiplier circuits 21 and 2
Give a multiplication instruction to 2. In response to this multiplication instruction, the multiplication circuit 21 multiplies the bar width value held in the X register 18 by the constant K _H from the constant storage section 23 and writes the multiplication result into the register 24 . On the other hand, since the flip-flop 19 is reset at this time, the gate of the AND circuit 26 is opened, and the constant K _L ' from the constant storage section 28 is input, the multiplier circuit 22 inputs the constant K L ' held in the X register 18. The width value is multiplied by the constant K _L ', and the multiplication result is written into the register 30. However,
After outputting the multiplication instruction, the CPU 14 increments the contents of the address register _AR by "+2" as shown in step _A5 . Next, the process proceeds to step _A6 , where the contents of the memory 15 are read into the Y register 17 in accordance with the designated address of the address register _AR . As shown in step _A5 above, by incrementing the contents of the address register A _R by 2, the width value of the same type of bar as the white or black bar held in the X register 18 at that time is stored in the Y register 17. is memory 15
Read from. Then, the CPU 14 outputs a comparison command to the comparison circuit 20 as shown in step _A7 . By this comparison command, the comparison circuit 2
0 compares the contents of the Y register 17 and the registers 24 and 30 as shown in FIG. 3, and outputs signals S ₁ to S ₃ to the CPU 14 in accordance with the comparison result.
The CPU 14 performs the code determination process shown in step _A3 using the comparison result signals _S1 to _S3 , and writes the result to the register Z. Details of this code determination process will be described later. After completing the above code determination process, the process proceeds to step _A9 .
Checks whether the content of the address register A _R has become equal to or greater than "7n" (n is an integer that increases sequentially from 1), that is, whether recognition of a total of seven black and white barcodes according to the NW-7 standard has been completed. to decide. If it is determined in step _A9 above that the contents of address register A _R have not reached "7n", proceed to step _A10 , set the contents of address register A _R to "-1", and return to step _A3 . . That is, assuming that recognition processing is first performed on the barcodes held at addresses 1 and 3 of the memory 15, the address registers are recognized in step _A10 .
By subtracting "-1" from the contents of _AR , the barcodes held at addresses 2 and 4 of the memory 15 are next recognized. Thereafter, the barcodes held in the memory 15 are sequentially recognized in the same manner. Then, when the contents of address register A _R exceeds "7n",
The judgment result of step A ₉ is YES, and the step
A Proceed to ₁₁ . In this step A ₁₁ , register Z
The code recognition result held in is decoded, that is, converted into a decimal value, and written to register D as shown in step _A12 . And step A ₁₃
At this point, it is determined whether or not the processing for a predetermined character, that is, for one label, has been completed. If the processing has not been completed yet, the process returns to step _A2 and the recognition processing for the next code is started. And step A ₁₃
When it is determined that decoding for one label has been completed, the process proceeds to step _A14 , where the recognition result held in register D is transferred to the data processing section via data line DL. The process then proceeds to step _A15 , where the memory M and register D are cleared to complete the recognition process, and the process returns to step _A1 .

Next, details of the code determination process in step _A8 will be explained in accordance with the flowchart of FIG. First, in step _B1 , the comparison circuit 20
If the comparison output is _S1 , that is, if the content of the Y register ₁₇ is smaller _than the data held in the register ₃₀ , proceed to step B. ₂ , if the comparison output is S ₂ ,
In other words, if the content of the Y register 17 is greater than the data held in the register 24, the process goes to step _B3 , and if the comparison output is _S3 , that is, if the content of the Y register 17 is the same as the previous bar, the process goes to step _S4 . It is determined whether the flag F _F is "0" or "1", that is, whether or not the code is determined for the first time. Therefore, in step _B4 above, if F _F is determined to be "0" (first code determination), proceed to step _B5 ,
A ``1'' is written in the address register A <sub>Z</sub> , followed by a ``0'' in the register Z addressed by the address register A _Z in step _B6 . Thereafter, the process proceeds to step _B7 , where the address register _AZ is incremented by 1, and the flag _FF is set to 1 in step _B8 . The program then proceeds to step _B9 and writes "0" into register Z addressed by address register A _Z.
That is, the first bar and the next bar are equal and “1”
If it is not possible to determine whether the
At B ₆ and B ₉ , "0" is once written into register Z. In addition, in step _B4 , if it is _determined that the _{flag F} It is determined whether or not the value has been determined to be “1” or “0”. If it is “0” (unconfirmed), proceed to step B ₉ above, and if it is “1” (confirmed), proceed to step B ₁₁ , and this time. Write the same value as the previous bar to register Z as a bar. Then, when the above step _B9 completes the process of _B11 , the step
Proceeding to step _B12 , it is determined whether the previous bar value held in register Z is "0" or "1", and if it is "0", the flip-flop 19 is reset in step _B13 ; If so, set it in step _B14 . Thereafter, as shown in step _B15 , the contents of address register _AZ are incremented by 1, and the process returns to step _A9 in FIG.

If the output of the comparison circuit 20 is S ₁ , that is, the current bar is smaller than the previous bar, the process goes to step B ₁ and then to step B ₂ , where flag F _F determines whether the code is determined for the first time or for the second time or later. It is done. If it is determined in step B ₂ that this is the first code determination, proceed to step B ₁₆ .
Write "1" to address register A _Z , and then in step B17, address register A _Z
Write "1" as the first bar (previous bar) into the register Z addressed by . Next, proceed to step _B13 and write "+" to address register A _Z.
1” and flag F _F at step _B19 .
Set "1" to "1". Thereafter, the process proceeds to step _B20 , where "0" is written as the current bar into the register Z addressed by the address register _AZ . Also, in step _B2 above, flag F _F is set to "1".
If it is determined that the code has been determined for the second time or later, the process proceeds to step _B21 , and the flag F _S is set to ``0'' or ``1''.
In other words, it is determined whether the previous bar has been determined to be "1" or "0". If the previous bar has not been determined, the previous data in the register Z is rewritten to "1" as shown in step _B22 , and the process then proceeds to step _B20 . i.e. step B ₂₂
If you proceed to step B ₆ , all the bars up to the previous time are equal and it is not possible to determine whether they are “1” or “0”.
At step _B9 , "0" has been written in register Z, but by outputting the comparison signal _S1 , it is determined that the current bar is smaller than the previous bar, so in step _B22 , the register Z is written. Rewrite all the bars up to the previous time to "1". If it is determined in step _B21 that the previous bar has been determined, the process immediately proceeds to step _B20 . In this step _B20 , as described above, "0" is written in the register Z as the current bar, and then in step _B23 , "1" is set in the flag F _S , and then in the above step _B12.
Proceed to.

Furthermore, if the output of the comparison circuit 20 is S ₂ , that is, the current bar is larger than the previous bar, the step
Proceeding to step _B3 via _B1 , a flag _FF is used to determine whether the code is determined for the first time or for the second time or later. If it is determined in step _B3 that the code has been determined for the first time, the process proceeds to step _B24 , where "1" is written in address register _AZ , and then in step _B25 , the register addressed by address register _AZ is Write “0” in Z as the first bar. Next, the program proceeds to step _B26 , where the address register _AZ is incremented by "+1", and in step _B27 , the flag _FF is set to "1". Thereafter, as shown in step _B28 , "1" is written in register Z as the current bar, and the process proceeds to step _B23 . Thereafter, code determination processing for each bar is performed in the same manner.

In the present invention, the constant storage unit 2 that stores the constant K _L
7 and a constant K _L ', and sets or resets the flip-flop 19 depending on whether the previous bar was "1" or "0", and stores the constants K _L and K _L '. are making choices. Figure 6 shows the correspondence between the threshold values T _L and T _H and the recognition width value when the previous bar is "1" and Figure 7 when the previous bar is "0". , FIGS. 6a and 7a are for the constants K _H and K _L , and FIGS. 6 b and 7 b are for the constants K _H and K _L '.

Therefore, when the width of the thin bar (“0”) is “1” and the width of the thick bar (“1”) is x, the intermediate value K _H between the thick bar and the thin bar is K _H = 1 + x/2. becomes. If the width value of the previous bar is small, the larger threshold value T _H can be obtained by multiplying by the width value of the previous bar. That is, T _H =1×K _H =1+x/2. Also, if the previous bar width value is large,
If K _L is a constant such that the smaller threshold value T _{L is equal to the above intermediate value 1+x/2, then T L =} x×K _L =1+x/2 ∴K _L =1+x/2x. .

Furthermore, if the bar width value is small this time,
And whether the previous bar is small or large, the difference β
Let K _L ′ be a constant for making them equal, β=1−1×K _L ′=x·K _L ′−1 ∴K _L ′=2/1+x=1/K _H . Note that in the NW-7 standard, the width x of the thick bar is set to x=2.2.

Therefore, in the present invention, two types _{of constants K L ,}
K _L ' is stored in the constant storage units 27 and 28, and when the previous bar width value was large (“1”), the constant K _L is selected as shown in FIG. 6a and the threshold value T _{L is} set.
, and when the previous bar width value was small (“0”), the constant K _L ′ is selected as shown in FIG. 7b to find the threshold value T _L . By selecting the constants K _L and K _L ' according to the width value of the previous bar in this way, an ideal threshold value can be obtained no matter how the bar codes are combined.

In the above embodiment, the constants K _H , K _L , K _L ′ are fixed, but in other cases, for example, the ratio of the size of the barcode is detected and the constants K _H , K _L ,
The value of K _L ′ may also be made variable. With this configuration, even if the barcode printed on a label has a code width that deviates from the specified value due to errors during printing, the barcode can be recognized accurately. .

Furthermore, the present invention can be implemented not only in hand scanners but also in stationary scanners, in which case barcodes can be read accurately even if the scanning drive section is not designed with high precision.

Further, in the above embodiment, the case where an optical barcode is read is shown, but the present invention can be implemented even when a magnetic barcode is used.

As described above, according to the present invention, when recognizing a barcode in which the ratio of bar width values is x:1, the threshold value setting for setting the threshold value is determined according to the previous bar width value. Since a constant is selected, an ideal threshold value can always be found, and barcodes can be recognized accurately.

[Brief explanation of the drawing]

Fig. 1 is a diagram showing a bar code according to the NW-7 standard, Figs. 2 to 7 show an embodiment of the present invention, Fig. 2 is a circuit configuration diagram, and Fig. 3 is a comparison of comparison circuits. FIGS. 4 and 5 are flowcharts showing the operation contents, and FIGS. 6a and 7b and 7a and 7b are diagrams showing the setting state of the threshold value. 11... Hand scanner pen, 14...
CPU, 15...Memory, 16...Counter, 1
7, 18, 24, 30...Register, 20...Comparison circuit, 23, 27, 28...Constant storage section, D,
Z...Register, F _F , FS...Flag, A _W ,
A _R , A _Z ...Address register.

Claims (1)

[Claims] 1. A barcode recognition device that recognizes a barcode in which the ratio of bar width values is x:1, comprising a reading means for reading a bar width value, and a read bar to be recognized. recognition bar storage means for storing the width value of the recognition bar, a recognition bar storage means for storing the width value of the recognition bar whose width value has already been recognized, and a constant 1+x/2x depending on the size of the recognition bar. or 2/1+x; multiplying the width value in the recognition bar storage means by the constant selected by the selection means to obtain a first threshold value; multiplying means for multiplying the width value in the bar by a constant 1+x/2 to obtain a second threshold value; and multiplying the width value in the recognized bar storage means and the first and second threshold values. and a comparison means for determining the size of the recognized bar with respect to the recognition bar, and after determining the recognized bar, the width value of the recognized bar is stored in the recognized bar storage means, and a signal indicating the size thereof is sent to the recognized bar storage means. Guide you to the means of selection,
A barcode recognition device that recognizes the next bar to be recognized read by the reading means.