JP2000293347A - Print presence/absence deciding method for medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To improve print presence/absence decision making by comparing a generated comparison print pattern with reference print patterns which are generated for a case wherein there is actually a print and stored, and deciding a line as a print line when there is a matching pattern. SOLUTION: Each time a scan is made, the numbers of pixel data of '1' by bytes are added and accumulated by multiple byte pixel counters 331 to 3328 and when one line is all scanned, the comparison print pattern of the line is generated on the basis of the contents of the counters 331 to 3328. The generated comparison print pattern is compared by a comparison part 39 with reference print patterns which are generated for a case wherein there is actually a print and stored in a 2nd table 38 and when there is a matching pattern, the line is decided as a print line.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a printer for printing characters, numbers, symbols, and the like (hereinafter, typically referred to as characters) on a line on which a printing medium such as a passbook (hereinafter, referred to as a passbook) is to be printed. The present invention relates to a printing presence / absence determining method for determining whether or not printing has already been performed on the line.

[0002]

2. Description of the Related Art Conventionally, a specific area (usually a date column) is scanned using a line sensor (CCD) while a passbook is being conveyed, and pixel information for determining the presence or absence of printing is acquired. The portion where the amount of light received in one scan is small, that is, the number of black pixels is counted, and the count exceeding a certain value is stored as a scan with printing, and this is repeated for one line. When the number of scans with print is equal to or greater than a predetermined value, the line was determined to be a print line.

[0003]

According to the above-mentioned conventional method, when the density of the print is reduced, the pixel information for determining the presence or absence of the print obtained by one scan is reduced and does not exceed a predetermined value. This makes it difficult to determine the presence or absence of printing. This will be specifically described below.

FIG. 22 is a plan view of a passbook in a double-page spread state. A passbook printed by a printer generally has a print format as shown in FIG.
That is, a predetermined range from the top end (form reference side) of the upper page P1 of the passbook A in the spread state is defined as an upper end print inhibition zone Z1, a predetermined number of print lines are provided following the upper end print inhibition zone Z1, and the binding line M is bounded. The area corresponding to one print line (total of two print lines) before and after that is defined as a central print prohibited zone Z2, the same number of print lines as the upper page P1 are provided on the lower page P2, and the lower page P2 following the last line is provided. A certain range up to the end is defined as a lower end printing prohibited zone Z3.

When printing the details of a transaction on a passbook, the presence or absence of printing is determined for each line, and the line without printing (unprinted line)
Must be printed on the first line of the. For this reason, the printer sets the position where the passbook has been conveyed by a predetermined feed amount corresponding to the upper end prohibition zone Z1 from the time when the leading end of the passbook inserted in the double-page spread state is detected, as the reading start position, Each time the bankbook A is conveyed one step in the direction F in FIG. 22 with a line sensor long in a direction parallel to the form reference side, the predetermined reading range up to the last line of the page is read in the horizontal direction (main scanning). The analog signal obtained by the scanning is converted into binary pixel data of “1” or “0” in 1-bit units, and the passbook is obtained during the transport (sub-scanning) of a predetermined step corresponding to one print line. The presence or absence of printing is determined by analyzing the value data by a predetermined analysis method.

An example of a conventional print presence / absence determination method will be described. FIG. 23 is a schematic diagram mainly showing a configuration of a mechanical portion of a printer of the passbook processor, and FIG. 24 is a block diagram showing a configuration of a control portion of the printer. The printer 1 performs a printing operation on the passbook A line by line. The printer 1 includes an insertion detector 2a for detecting that the passbook A has been inserted into the insertion / ejection port 2, a plurality of feed rollers 3 and a press roller 4 provided opposite to each other across the transport path L, and a feed roller. Motor 5 for rotating the bankbook 3 forward and backward, a leading edge detector 6 for detecting the leading edge of the inserted bankbook A, a line sensor 7 composed of a CCD group for scanning a predetermined area of the bankbook A and reading characters, and printing A print head 8 for performing printing while reciprocating along a row, a platen 9 provided opposite the print head 8, and a page provided at the end of the transport path L for turning pages of the passbook A It has a turning part 10.

As shown in FIG. 24, the printer 1 has a CPU 20 for controlling each section, a ROM 21 for storing a control program and the like, and an RA for providing a work area and the like.
M22, a transport unit 23 for controlling the motor 5 to transport the passbook A, a printing unit 24 for controlling the print head 8 and the platen 9 to perform printing on the passbook A, and reading the line sensor 7 It has a sensor control unit 25 for controlling operations and acquiring scan data (pixel information), and an interface unit (I / F unit) 26 for receiving print data from a host device (computer or the like). I have.
Further, the CPU 20 receives a detection signal from the leading edge detector 6 and detects that the passbook A has reached the position of the leading edge detector 6. The reading area Zr of the line sensor 7 is set in a position where printing is necessarily performed in each line, for example, a date column where a transaction date is printed.

As shown in FIG. 25, under the control of the sensor control section 25, the line sensor 7 controls
The main scanning is performed in the direction perpendicular to the transport direction every time the is transported by one step. When the amount of received light is small in each main scan, the logical value is “1”. When the amount of received light is large, the logical value is “0”.
That is, the data is converted into scan data composed of binary pixel data and output. Xd in FIG. X4 is a print dot pattern on the scanning line Lb of the line sensor 7, where a white circle indicates a portion where the amount of light received by the line sensor is large, that is, a portion where printing is not performed, and a black circle indicates a portion where the amount of light received is small, that is, The printed portion is shown.

More specifically, as shown in FIG. 26A, when the numeral "1" is printed at a sufficient density, the analog signal output from the CCD of the line sensor 7 by the main scanning is detected. The signal is detected at a predetermined level Vs by the detector (a in the figure), and a clock signal output from the clock generation circuit (c in the figure) while the gate is opened by a detection output when the level exceeds the predetermined level Vs (b in the figure) Pass through the gate. The output of the gate is input to the counter (d in the figure) and counted (e in the figure). The reading range of the passbook date column by the line sensor 7 is 2 per scan.
8 bytes, that is, 224 bits. Then, as shown in FIG. 26 (b), the gate passes four clock pulses for one dot of the print character. That is, the line sensor 7 outputs 4-bit "1" data at the timing of reading one dot of the character portion.

The transport section 3 transports the passbook A by one line (sub-scan).
If, for example, 24 steps are required to perform the scanning, the main scanning is performed 24 times. Then, in the binary data obtained in each main scan, if “1” is a predetermined value, for example, 12 or more, the scan is determined to be a scan with print, and if less than 12, the scan is a print scan. Judgment as scanning without
If the number of scans for the line (24 times) is determined to be scanning with printing is 8 or more, the line is determined to be a line with printing, and if less than 8, the line is determined to be a line without printing. I am trying to do it. The reason why the reference value of the print presence / absence determination is set to 12 or more based on the scan data in the main scanning is that, as shown in FIG. This is because the number of bits of the scan data at that time is 1 dot × 4 bits × 3 places = 12 bits if the vertical component of the print character is 1 dot.

A method of comparing the number of "1" bits (hereinafter sometimes referred to as "black bits") in the scan data in one scan with a reference value to determine the presence or absence of printing is based on the fact that the density of the printed characters is appropriate. In that case, there is no problem. The problem occurs when the print density is low. That is, when the print density is low, the output level from the line sensor 7 is generally lower than the predetermined level Vs as shown in FIG. 27 (a in FIG. 27). In this case, as shown in FIG. 26 (a), scan data faithful to the print pattern of numeral 1 can be obtained, whereas when the print density is low, the scan data can be obtained as shown in FIG. 27 (a). Is deteriorated, and the fidelity to the print pattern is reduced. Therefore, the number of bits of the logical value “1” in the scan data obtained by one scan decreases as compared with the case where the fidelity is high, and It will be lower than the value (12).
As described above, as the PCS value decreases, the amount of pixel information for determining the presence or absence of printing from the line sensor decreases, and it becomes difficult to determine whether or not there is printing on the line. Further, the number of “1” bits obtained in one print line is also lower than the reference value, and there may be a case where there is an actual print but an erroneous determination that there is no print.

In view of the above, according to the present invention, when the print density is reduced, even if the amount of pixel information for print presence determination obtained by one scan decreases, the presence or absence of printing can be reliably determined without being influenced by the decrease. An object of the present invention is to provide a printing presence / absence determination method capable of determining.

[0013]

The principle of the present invention for solving the above-mentioned problems is as follows. That is, alphanumeric characters (mainly ANK numbers) printed in the date column are vertically long and include many components in the vertical direction. PCS
Even if the value decreases and the number of black bits in the scan data decreases, the proportion of the vertical component remaining is large, so the date column is vertically divided into certain ranges in the main scanning direction, and the vertical The direction component is extracted and used to determine the presence or absence of a print position.
That is, the number of black pixels of the pixel information obtained by scanning is added in the sub-scanning direction (vertical direction) in units of 1 byte, thereby expressing the vertical component of the character as an amount. Then, the obtained accumulated data for each byte sequence is patterned, the pattern is compared with a pattern with printing, and it is determined as a line with printing or a line without printing based on whether they match.

Focusing on the above principle, the print presence / absence determination method according to the present invention includes (a) a byte pixel counter for counting pixel data of "1" for each byte of scan data for one scan. , (B) each time a scan is performed
The number of pixel data of "1" for each byte is added and accumulated in each of the byte pixel counters. (C) When all the scans for one row are completed, based on the contents of each byte pixel counter, A comparative print pattern is created, and (d) the created comparative print pattern is compared with a reference print pattern created and stored when there is actually printing, and if there is a match, the line is replaced. It is characterized in that it is determined that the line has a print. Since each byte pixel counter counts the number of black pixels for each of the plurality of vertical sections of the area to be printed, the counted value reflects the vertical component of the character component. Therefore, even in the case of a character having a lowered PCS, as much pixel information as possible to determine the presence or absence of printing can be obtained, and the presence or absence of printing can be accurately determined. However, since the comparison print pattern is created in units of one byte, it can be created not for printing but also for a stain.

The types of the comparison print pattern and the reference print pattern can be the following two types depending on the accuracy of the judgment. That is, in the first type, the number of pixels of "1" in scan data obtained by scanning each row of the date column is added in the sub-scanning direction in byte units, and "1" in each byte is added. Classify which of the multi-level values the pixel count value of “
Assign a predetermined term type to the classification result,
A comparison print pattern is generated based on each of the four item types, and a reference print pattern is created corresponding to each single character. Then, the comparison print pattern is compared with a pre-created reference print pattern corresponding to a print pattern composed of a plurality of prints, and if at least one of the comparison print patterns matches the reference print pattern, the line is determined to be a printed line. How to In the case of the above method, since the comparative print pattern corresponds to a plurality of characters, the comparison is performed not in units of one character but in units of a fixed-length format, so that a stain can be eliminated.

The number of "1" pixels in scan data obtained by scanning each row of the date column is added in the sub-scanning direction in units of one byte, and the "1" pixels in each byte are added. Classify which of the multi-level values corresponds to
Assign a predetermined term type to the classification result,
A first comparative print pattern is generated according to the item type of each of the four items, the first comparative print pattern is collated with the first reference print pattern, and a second comparative print pattern is produced based on the collation results of all four items. A pattern is generated, the second comparative print pattern is compared with a second reference print pattern, and if there is a match, the line is determined to be a line with print. According to the above configuration, the format is compared on a line-by-line basis, so that the presence / absence of printing can be determined with higher accuracy.

In the printing pattern, the value X1 of each byte pixel counter is compared with two constant values a and b (a <b). When X1 ≦ a, “0” is set, and when b ≦ X1, Is set to "1", and when a <X1 <b, the value X is set.
1 is added to the value of the next byte pixel counter, and the value X2
Is greater than or equal to a constant value c, "1" is set, and the value X2
Is smaller than the fixed value c, the value of the next byte pixel counter is added to the value X2 in order, and the value X3 is set to a fixed value (c + n).
d: n is a value according to the number of byte pixel counters added). If the value is greater than or equal to “1”, “1” is set. If the value Xi of the byte pixel counter to be added becomes less than a, “0”
Is set to form a primary print pattern consisting of a combination of “1” or “0”, and the primary print pattern “1” and “0” are compressed to form a secondary print pattern,
Comparing the secondary print pattern with a basic print pattern,
When there is a match, the line is determined to be a line with printing. That is, the black bit count value X1 in 1-byte units is neither equal to or greater than a constant value a that can be determined to be “1” or “0”, and not less than a constant value b.
In the case of b, that is, when the count value is in the gray zone of the print presence / absence determination, the number of black bits of the byte next to the byte is added, and the addition result is compared with a constant value c, so that It is easy to determine whether or not there is a character component.

[0018]

Next, an embodiment of the present invention will be described with reference to FIGS. FIG. 1 is a block diagram showing a main configuration used to carry out the method of the present invention, FIG. 2 is a conceptual diagram schematically showing a relationship between a vertical section of a date column and scanning of each section, and FIG. FIG. 4 is a schematic diagram for explaining storage of scan data obtained by scanning for minutes.
Is a schematic diagram for explaining the operation of adding the number of black bits for each byte for 24 scans (vertically) and storing it, FIG. 5 is a schematic diagram for explaining the patterning of the number of black bits in each vertical section, and FIG. 5 is a flowchart illustrating a basic operation of the method of the present invention.
Since the basic configuration of the printer itself is the same as before,
FIG. 23 is referred to in the description related to the configuration of the printer.

In order to carry out the method of the present invention, as shown in FIG. 1, the OSU (line sensor 7 and sensor control unit 25) of the printer 1 shown in FIG. After the scan (24 times) for one row is completed, the scan data stored in the buffer memory is stored in the buffer memory 31 for storing 1-28 bytes in units of 1 byte for each scan. A read section 32 for sequentially reading and outputting from output terminals provided corresponding to 28 bytes individually, and black included in 1-byte scan data connected to each output terminal of the read section and output to the output terminal. 28 byte pixel counters 33 ₁ , 33 ₂ ,.
33 _28, and Kochi buffer 34 for storing capture the count value of each byte pixel counter, any class of a plurality of stages consisting of the number of bits ranges black total number of bits of each vertical segment stored is set in advance The conversion unit 35 converts the classification into a predetermined term type composed of a 4-bit binary number, and stores the multi-stage classification and the corresponding 4-bit term type according to A first table 36, a term type buffer 37 for storing the term type (pattern) output from the conversion unit in correspondence with the number (28) of the vertical sections, and each section is actually printed. A second table 38 storing the pattern of the case as a reference print pattern, and a term type (pattern) corresponding to each vertical section stored in the term type buffer
And compares it with the reference print pattern stored in the second table 38, and determines whether or not there is a match, and determines whether it is a line with printing or a line without printing and outputs a determination result. And a portion 39.

The method of the present invention performs the basic operation as shown in FIG. 6 using the above-mentioned components shown in FIG. That is, a step (R1) of temporarily storing scan data obtained by scanning with a line sensor, counting the number of black bits for each byte of the obtained scan data, and counting the number of black bits for each byte of the scan data. Is added for 24 scans (for one row) to obtain a count value of black bits in each vertical section in units of 1 horizontal byte and 24 vertical scans (R2)
A step (R3) of classifying the obtained count values according to a plurality of preset classification criteria, a step (R4) of patterning according to the classification result, and a step of The step (R5) of comparing with the reference print pattern and determining the presence or absence of printing based on the result of the comparison is performed in this order. Hereinafter, each step will be sequentially described in detail.

[Capture of scan data (R in FIG. 6)
1)] Of the 1024 bits output by the OSU in one scan of the line sensor 7, scan data of 224 bits (28 bytes) is fetched only from the range corresponding to the date column and buffered for each byte. It is stored in the storage area at a predetermined address of the memory 31 as in the conventional case. Since the date column of one line is scanned 24 times, the first line is stored in the buffer memory 31 as shown in FIG.
Scan data from the scan to the 24th scan (d10 to
d127, d20 to d227,..., d240 to d2427) are stored. As illustrated in FIG. 4, the buffer memory 31 has 8 bits (1 byte) input from the line sensor 7.
2 represented by “1” or “0” depending on the amount of received light
It is stored as value pixel data.

[Addition of Black Bit for Each Byte (R2 in FIG. 6)] When the fetching of the scan data is completed, the reading unit 32 transfers the scan data by the first scan from the buffer memory 31 to the first to 28th bytes. It is counted by the read sequentially by one byte, the first byte pixel counter 33 respectively ₁ black bits included in the scan data for each 1 byte 28 byte pixel counter 33 _28. When the count of the black bits for scan data according to the first scan ends, followed by the black bits contained in each byte of the scan data according to the second scan, each of the first byte pixel counter 33 ₁ from the first 28 bytes pixel counter 3
3. Count and add according to ₂₈ . Hereinafter, similarly, the second
From the first byte to the second
For 8 bytes, the number of black bits per byte
Counts the number-add bytes pixel counter 33 ₁ by the 28 byte pixel counter 33 _28, the total number of the respective black bits from the first byte of one row to the 28th byte,
The item value buffer 34 stores the data in the storage areas at addresses (28) to (1) specified by the pointer. The number of black bits stored in each storage area of the term value buffer 34 may be referred to as a term value. That is, the number of black bits in each vertical section of the date column is stored in the term value buffer 34 as 0, 5, 20, 30 to 55, 15 as illustrated in FIG.

The counting of black bits for each vertical section can also be performed as follows. That is, after the end of the scan data capture, the read unit 32 reads the first data from the buffer memory 31.
The first byte data (d10) of the scan data (224 bits) in the scan is used as the first byte pixel counter 3
3 counts the black bits are input to _1, then the black bits counted cumulative enter the first byte of data in the scan data in the second scan (d20) to the first byte pixel counter 33 _1, below Similarly, the first byte of data (D240) in the 24 scan addition and cumulatively to the first counter 33 _1, by counting the total number of black bits of the first byte of one line,
The count value is stored in the term value buffer 34 in the storage area of the address (28) designated by the pointer. In the example of FIG. 4, the number (term value) of black bits in the first vertical section is zero. Subsequently, by counting the black bits by input from the reading unit 32 is a buffer memory 31 the data of the second byte in the scan data in the first scanning (d11) to the second byte pixel counter 33 _1, then the In the same way as for one byte of data, the total number of black bits of the second byte data (d11, d21,... D241) for one row is counted, and the counted value is stored in the term value buffer 34. The data is stored in the storage area at the address (27) specified by the pointer. In the example of FIG. 4, the number (term value) of black bits in the second vertical section is 5. Hereinafter, similarly, from the third byte to the 28th byte of the scan data in each scan stored in the buffer memory 31, (d12, d22, d242; to d12)
7, d227 to d2427) are counted for one row, and the counted values are respectively stored in the term value buffer 34 in the storage area of the address designated by the pointer. In this way, in the term value buffer 34, the number of black bits in each vertical section of the date column is set to 0, 0, as illustrated in FIG.
5, 20, 30 to 55, 15 are stored.

The capture of scan data (R1) and the vertical addition of black bits (R2) are realized by hardware as described above. Subsequently, another embodiment of the hardware will be described with reference to the drawings of FIGS.
FIG. 7 is a block diagram of the determination device, and FIG. 8 is a timing chart showing operation timing of the determination device.

In FIG. 7, an OSU is an optical reading device using a line sensor 7 constituted by a CCD group having a reading capability of, for example, 1024 bits. Based on a passbook insertion detection signal from an insertion detector 2a in FIG. When the conveyance unit 23 starts conveyance, and when the front end detector 6 detects the front end of the passbook and is conveyed by a predetermined number of steps, the CPU sends a signal (OSU ON) to the OSU to start scanning. And outputs the same signal to the first counter CTU.

The first counter CT1 is for extracting only 224 bits of effective scan data corresponding to a date column from 1024 bits of scan data (IMG) output from the OSU. The basic clock (SCLK) supplied from the CPU is counted, and the counted value is CP.
From the time when the signal U matches the designated value of the start of capture set via the bus B1, the time when the value coincides with the value of the designated capture stop, the Q1 output signal is output to open the AND gate & G. Then, the AND gate & G inputs the scan data output by the OSU to the serial / parallel conversion circuit SPC while the gate is open. The serial-to-parallel conversion circuit SPC converts 8-bit serial data into parallel data each time it is input and outputs the parallel data. The timing generation circuit TG supplies a latch signal (LAT1) to the first latch circuit L1 when a clock pulse necessary for 8-bit data to pass through the AND gate & G is input after inputting the Q1 output signal. The 8-bit parallel data output from the serial / parallel conversion circuit SP is converted into the next 8 bits.
Bit serial data is converted to a serial-parallel converter S
The data is held for a certain period of time (for example, 8 μS) until it is input to the PC.

The timing generation circuit TG sets the operation timing of each circuit based on the basic clock (SCLK) given from the CPU, and outputs the latch signal (LAT1) to the first latch circuit L1 at a predetermined timing. In addition, the latch signal (LAT2) to the second latch circuit L2, the first
A control signal (DIR) to the gate G1, a read / write switching signal (R / W) to the RAM 22, and a selection command signal (SEL) to the selector SE are generated and given, respectively. Control signal (UP-CLR) (CLR) for updating (-1) or clearing the data.

The second counter CT2 includes a first latch circuit L
This is a counter that counts the number of black bits in the 8-bit scan data held at 1.
Output to the bus B2 in the form of a decimal number. The bus B2 has the second
First inputs of four 4-bit adders ADD1 to ADD4 so that a 16-bit count value obtained by adding an 8-bit count value output from the counter CT2 and 8-bit numerical data for numerical extension can be added. Side is connected. The second input side of each of the adders ADD1 to ADD4 is connected to the read side of the first gate G1 via the second latch circuit L2. The output side of each of the adders ADD1 to ADD4 is connected to the input side of the first gate G1 via a bus B3. The first gate G1 is a bidirectional gate, and is connected to the RAM 22 via the bus B4, the output side is connected to the bus B1 via the second latch circuit L2, and the input side is connected to the bus B3.

The RAM 22 stores the count value of black bits in the 1-byte scan data in the form of a 16-bit binary number.
It has a storage area capable of storing scan data (28 bytes). The new black bit count value output from the second counter CT2 and the black bit count value read from the storage area at the predetermined address of the RAM are added in the adders ADD1 to ADD4, respectively, to obtain a new black bit cumulative count value.
It is configured to be stored in the same storage area. A read / write switching signal from the timing generation circuit TG to the RAM for cumulative addition of the black bit count value and storage of the result.
(R / W) for controlling the first gate G1 so that data read from the RAM is output in the direction of the bus B1 and for switching data to be input in the direction of the RAM from the bus B3. The input / output direction control signal (DIR) is provided.

The RAM is connected to an address generation circuit AG and a CPU via a selector SE. The address generation circuit AG is for designating a storage area when writing or reading data to or from the RAM. The selector SE uses the selection command signal (SEL) received from the timing generation circuit TG to select the RAM and the selector S.
E, the address bus B5 is connected to the address bus B6 or the address bus B7, and the address generation circuit A is connected to the RAM.
An address from G or an address from CPU is provided.

The data held in the second latch circuit L2 is taken into the CPU via the data bus B1 into which the second gate G2 has been inserted. That is, the second gate G2 is opened by a read command signal from the CPU, and the CPU reads the cumulative black bit count value from the address specified by the address given to the RAM via the selector SE, and reads this from the first gate G1 and the first gate G1. Second latch circuit L2
Through the CPU.

The operation of the above configuration will be described with reference to the timing chart of FIG. The basic clock (SCLK) from the CPU is output at a constant cycle (for example, 1 μs), and the first counter CT
1 and to the timing generation circuit TG. While the AND gate & G is opened by the Q1 output from the first counter CT1, scan data for one scan output from the OSU is input to the serial-parallel conversion circuit SPC. Then, the first 1-byte (8-bit) data is converted into parallel data by the serial / parallel conversion circuit SPC and output. When the eighth bit has been input to the serial / parallel conversion circuit SPC, the timing generation circuit TG outputs the data. The output latch signal LAT1 is the first latch circuit L
1, the parallel data from the serial / parallel conversion circuit SPC is held for a predetermined time, during which the second counter CT2 counts the number of black bits contained in the 8-bit data, and the counted value is 8 bits 2 Output to the bus B2 in hexadecimal notation and four 4-bit adders ADD
It is input as 16 bits by adding 8 bits of "0" of dummy data for numerical value extension to 1 to ADD4.

In synchronization with the input of the first 8-bit scan data to the serial / parallel conversion circuit SPC, a clear signal (CLR) is given from the timing generation circuit TG to the address generation circuit AG, and the address for the RAM is initialized. Is done. At that time, the timing generation circuit TG outputs the read / write switching signal (R / W) to the RAM.
Since a read command signal is given and a control signal (DIR) for activating the output side gate is given to the first gate G1, RA
The accumulated count value (the accumulated count value is 0 at this time) stored in the storage area designated by the address output from the M address generation circuit AG is read out to the second latch circuit L2. Output. Now, since the second latch signal (LAT2) is given to the second latch circuit L2, the accumulated count value read from the RAM is input to the adders ADD1 to ADD4 via the bus B1. In this case, the address of the CPU is the address generation circuit A of the RAM.
This matches the address output by G.

Therefore, the adders ADD1 to ADD4 provide the expanded 16-bit black bit count value obtained from the first 8 bits (first 8 bits) of the scan data input from the OSU and the previous scan data from the RAM. And a black bit count value (16-bit data) obtained from the stored first 8-bit scan data to calculate a new cumulative count value, which is output to the bus B3. At this time, the timing generation circuit TG gives the write command signal of the read / write switching signal (R / W) to the RAM, and the address generation circuit AG still outputs the same address. TG is the first gate G1
Is supplied with a control signal (DIR) for activating the input side gate, the new accumulated count value of the adders ADD1 to ADD4 is stored in the storage area at a predetermined address of the RAM.

In the meantime, the second 8-bit scan data is continuously input from the OSU to the serial / parallel conversion circuit SPC via the AND gate & G, and when the input is completed, the first latch signal is output as in the previous case. (LAT
The parallel data is latched by 1), and the black bits contained therein are counted by the second counter CT2, output to the bus B2, and added to the adders ADD1 to ADD4.
Is input to Then, as in the case of the first 8 bits,
From the storage area specified by the address (updated) currently output by the address generation circuit AG of the RAM, the accumulated count value stored in this area (the accumulated count value is 0 at this time). Read out the second latch circuit L
Output to 2. Since the second latch signal (LAT2) is given to the second latch circuit L2, the cumulative count value read from the RAM is added to the adders ADD1 to ADD1 through the bus B3.
DD4. Therefore, the adders ADD1 to ADD
4 is a black bit count value obtained from the second 8-bit scan data input from the OSU, and a black bit count value obtained from the stored first 8-bit scan data obtained in the previous scan from the RAM. (16-bit data), and a new cumulative count value is calculated.
Output to At this point, the timing generation circuit T
G supplies a write command signal of the read / write switching signal (R / W) to the RAM, the address generation circuit AG still outputs the same address, and the timing generation circuit TG
Supplies a control signal for activating the input side gate to the second gate G2, so that the new accumulated count value of the adders ADD1 to ADD4 is stored in the storage area of the RAM at the predetermined address.

Thereafter, the third and fourth data input from the OSU 8
Similarly to the bit scan data, the black bit count value and the cumulative count value obtained by the previous scan and stored in the RAM are added to the adders ADD1 to ADD as in the previous case.
The new accumulated count value is stored in a predetermined address of the RAM by accumulating the data in step (4).

When the counting of the 28-bit black bits for the first scan and the storage of the count value are completed, the black bits of the scan data for the second scan are counted every 8 bits in the same manner as described above. This is added and recorded with the number of black bits of each byte obtained in the previous scan, and similarly, up to 24 scans (for one row), the counting and accumulation of black bits for each byte are similarly performed.

When the scanning for one row is completed, a selection command signal (SEL) is given from the timing generation circuit TG to the selector SE, and the RAM and the bus B7 are connected.
Gives a predetermined address to the RAM via the bus B6,
The number of black bits from the first byte to the twenty-eighth byte of one row, that is, the count value of the black bits corresponding to each vertical section of the date column is stored in the first gate G1 and the second latch circuit L2.
And via the bus B1 to the term value buffer 34 of FIG. The term value buffer 34 has 28 storage areas in which the count value of black bits in each 1-byte scan data is stored by term-betting, and is stored in each area from one end to the other end of the storage area. It has the number of terms to specify the area. After that, the CPU performs the value classification processing (R
3), patterning (R4), and determination processing (R5). Subsequently, these will be described.

[Classification of Item Value (R3 in FIG. 6)] After counting and storing the black bits in each vertical section of the date column, the value of each data item in the item value buffer 34, that is, , And the classification of the term values (the classification of the respective count values) is executed. As shown in FIG. 9, the term value classification refers to a plurality of black values included in scan data of a fixed unit amount, for example, 1 byte, which are set in advance stepwise according to whether the number of black bits is large or small. Is classified by attaching a mutually identifiable item type predetermined for the corresponding range according to which of the ranges falls. In the present embodiment, as shown in FIG. 9, a plurality of ranges set in advance in a stepwise manner have black bit numbers of 0 to 2, 3 to 9, 10 to 25, 26 to 31,
32 to 79,80 or more are set in 6 stages, and the corresponding item types are “0”, “3”, “6”, “7”, “E”
"F". Then, each term type is a 4-bit binary number, that is, 0000, 0011,
It is represented by 0110, 0111, 1110, 1111. 32 black pits
The term type is determined so that the most significant bit is "0" when the value is less than 32, and "1" when the value is 32 or more. This is because if the character to be printed is a normal alphanumeric typeface, it is considered that 32 can be used as a reference for determining the presence or absence of a character component. However, in the case of "-" (hyphen) and "." (Period), a value less than 32 is a criterion for determining the presence or absence of a character component. The first table 36 in FIG. 1 stores the above-described six-stage black bit numbers and the corresponding item types.

Next, the processing procedure of the classification of the term values will be described with reference to FIG.
This will be described with reference to the flowchart of FIG. FIG. 10 shows a part of the processing procedure for classifying the item values, and FIG. 11 shows the rest. To execute the term value classification routine (R3), the CPU 20 uses the term value buffer 34, the conversion unit 35, the first table 36, and the term type buffer 37 in FIG. When classifying the term values, first, the head pointer of the term value buffer 34 is set to specify the position of the storage area to be read from the term value buffer 34 (p
1) Subsequently, the head pointer of the term type buffer 37 is set to specify the position of the storage area where the data of the term type buffer 37 is to be stored (p2). And
A term counter (not shown) composed of a software down counter for counting the number of classification processes
Is set to the number of terms in both buffers (p3). In this embodiment, each buffer uses a storage area for 28 items corresponding to 28 vertical sections of one line in the date column of the passbook, and the item number counter is decremented every time the classification process for one item is completed. At first, the number of terms is the maximum value "28"
Is set.

Next, it is determined whether or not the classification is completed based on whether or not the value (number of items) of the item number counter is 0 (p
4). Since the classification has just started, the determination result is negative (No), and the term value is obtained from the storage area corresponding to the head pointer set earlier in the term value buffer 34 (p5). Then, it is determined whether or not the item value is 0 (p6), and in the case of affirmative (YES), the conversion unit 35 refers to the first table 36 to change the item type to “0” corresponding to the item value 0. It is converted and prepared for storage (p7).
Subsequently, the item type is stored in the storage area corresponding to the head pointer set earlier in the item type buffer 37 (p8). This completes the classification for the head pointer of the term value buffer.

When the classification for the head pointer is completed,
Next, the pointer of the term value buffer is updated (−1) (p
9) Similarly, the pointer of the term type buffer is also updated (p10), and the number of terms of the term number counter is also updated (−).
1) Performed (p11), and proceed to step p4.

If the term value is not 0 at step p6, the process proceeds to step p12, and it is determined whether the term value is 80 or more. If the determination is affirmative (Y), the conversion unit 35 refers to the first table 36 and converts the term type to “F” corresponding to a term value of 80 or more (p13). Term type is term type buffer 3
7, the pointers in the term value buffer 34 and the term type buffer 37 are updated (p9, p10) in steps 9, 10, and 11, as described above. After the number of items of the item number counter is updated (p11), the process returns to step p4.

On the other hand, if the decision result in the step p12 is negative (N), the process shifts to a step p14 to decide whether or not the term value acquired from the term value buffer 34 is 32 or more. In the case of affirmative (Y), it is converted to the term type "E" corresponding to the term value of 32 or more (p15), the process proceeds to step p8, and thereafter, the same processing as described above is performed. If the determination result in step p14 is negative (N), the process proceeds to step p16 (FIG. 11), and it is determined whether the term value is 26 or more, and if affirmative (Y), it corresponds to the term value 26 or more. After being converted to the term type “7” (p17), the process proceeds to step p8. If the determination result in step p16 is negative (N), the process proceeds to step p18, where the term value is 1
It is determined whether it is 0 or more, and if affirmative (Y), the item value 1
Is converted to the term type “6” corresponding to 0 or more (p1
9) The process proceeds to Step p8. If the determination result in step p18 is negative (N), the process proceeds to step p20, where it is determined whether the term value is 3 or more. If affirmative (Y), the term type corresponding to the term value 3 or more “ 3 "(p21), and the routine goes to Step p8. If not (N), it is converted to the term type "0" (p22), and the routine goes to Step p8.

When the number of terms in the term number counter is updated to 0 in step p11, the classification is determined to be completed in step 4, and the process returns to the main flow of FIG.

As described above, as shown in FIG. 5, the number of black bits in 28 vertical sections for one row stored in the term value buffer 34 is classified into six levels according to the magnitude of the number. Then, each term is converted to a preset term type, and each term type is represented as a 4-bit binary number and stored in the term type buffer 37.

1 and 2 stored in the term type buffer 37
The eight item types, individually or as a whole, constitute a kind of print pattern representing the print positions in the 28 vertical sections of the date column of the passbook. Therefore, a print pattern consisting of various 4-bit binary numbers or a combination thereof when there is actually printing is prepared in the second table 38 as a reference print pattern, and every time scanning of one line is completed, the print pattern is prepared. The comparison print pattern corresponding to the scan data is compared with the reference print pattern, and when they match, the line can be determined as a line with print. However, in the term type buffer 37, when the number of black bits (term value) is classified into six levels, the corresponding classification is represented by a 4-bit binary number for each term.
A total of 28 items are stored. Since the stages of the corresponding classification are variously represented, the collation determination between the patterns is complicated. In addition, the method of matching the reference print pattern for each item type, that is, for each character, is as follows.
Even when the medium is not printed and a blot is read, there is a possibility that the medium is erroneously determined to be printed. Therefore, in the preferred embodiment of the present invention, following the term value classification (R3), the term type stored in the term type buffer 37 is converted into a second comparative print pattern having a configuration capable of simplifying collation determination. (R4).

[Patterning Process (R4 in FIG. 6)] Next, the patterning process will be described with reference to FIGS.
This will be described with reference to FIG. FIG. 12 is a conceptual diagram of patterning, and FIGS. 13 to 17 are partial views of one flowchart. When the patterning process is performed, in addition to the buffer memory 31, the term value buffer 34, and the term type buffer 37 of FIG.
A patterning element buffer 40 is provided. The patterning element buffer 40 has a storage area of 7 bytes (* 1 to *
7). Then, for each vertical section of the date column, four items (16 bits, that is, 2 bits) of the item type buffer 37 storing 4-bit data (item type) are stored.
Each byte is converted into one patterning element and stored in each storage area of the buffer 40. The patterning element is "1" or "0" when represented by binary "1" or "0" depending on whether or not the pattern corresponds to a 4-bit reference print pattern when printing is performed. Further, when performing the patterning process, various second reference print patterns when printing is performed are stored in the second table 38 in advance. The second reference print pattern is based on the black bit count value obtained by scanning the date field when ANK alphanumeric characters are actually printed according to the embodiment shown in FIGS. 18 and 19 are patterns obtained with pattern printing and include patterns as shown in FIGS. The data of the pattern in the storage area corresponding to each item of the second table 38 is constituted by a 4-bit binary number. However, in FIGS. 18 and 19, the data is represented by a decimal number for easy understanding. ing. Further, the second reference print pattern is slightly different depending on the typeface of characters used for printing the date.

In the patterning process, first, the storage area of the patterning element buffer 40 is cleared (p23). Then, in order to specify the storage area to be read from the term type buffer 37, the pointer of the term type buffer 37 is set. It is set to the head pointer (p24). Then, the number of items in both buffers is set to an item number counter (not shown) configured by a software down counter for counting the number of checks of the item type for the patterning process (p25). In this embodiment, since the term type buffer 37 uses a storage area for 28 terms, the number of terms is set to the maximum value “28”.

Next, whether or not the item type check has been completed is determined based on whether or not the number of items in the item number counter is 0 (p26). If the determination result is negative (N), the item type is obtained from the storage area of the pointer (head pointer) set earlier in the item type buffer 37 (p27).
Then, it is checked whether the item type is 0 (p28).
If affirmative (Y), the pointer of the term type buffer 37 is updated (-1) (p29), and the number of terms in the term counter is updated (-1) (p30), and step p26 is executed.
Return to

If the decision result in the step 28 is negative (N), the process moves to a step 31 (FIG. 14), and the type of the four items from the item type buffer, that is, for example, from the 28th item to the 25th item in FIG. 16-bit (2 bytes) data is obtained. Then, the most significant bit of each of the obtained item types is obtained, and a second comparative print pattern in which each item is represented by 4 bits is generated (p32). That is, in the example of FIG. 5, the second comparative print pattern of 0000, 0000, 0000, 0000 obtained by ANDing 1000 to the highest order 0 (in decimal notation according to FIGS. 18 and 19, “0 ,
0,0,0 ").

Next, the head address of the second table 38 is designated (p33), and subsequently, the primary reference print patterns as illustrated in FIG. A pattern matching for matching the primary reference print pattern with the second comparison print pattern corresponding to the four items of the item type buffer generated earlier, that is, a pattern matching routine (R6) is executed. Then, in the acquired primary reference print pattern, the second comparative print pattern corresponding to the first four items of the item type buffer, that is, the pattern corresponding to the first to fourth vertical sections of the date column of the passbook is matched. It is determined whether or not there is something to do (p34).

If they match (if affirmative in P34), a patterning element setting routine is executed (R
7). Here, the patterning element setting routine (R7) will be described with reference to FIG. If there is a match in the first comparison print pattern and the first reference print pattern matching step in the main flow of FIGS. 13 to 16, the patterning element setting routine (R
7) is activated. First, the number of items currently checked in the item type buffer 37 is obtained (p71). Next, it is checked whether the number of items is 28 to 25, that is, whether it is the first four items (p72). If affirmative (Y), the pattern is stored in the first storage area (* 1) of the patterning element buffer 40. The storage element "1" is stored (p73). If negative (N) in step p72, it is checked whether the number of terms is 24 to 21, that is, whether it is the second four terms (p74), and affirmative (Y)
Then, the patterning element "1" is stored in the second storage area (* 2) of the patterning element buffer 40 (p7).
5). If negative (N) in step p74, it is checked whether the number of terms is 20 to 17, ie, whether it is the third fourth term (p76). If positive (Y), the patterning element buffer 40 The patterning element "1" is stored in the third storage area (* 3) (p77). Hereinafter, similarly,
It is checked whether the number of terms is 16 to 13 (p78), 12 to 9 (p80), or 8 to 5 (p82). If the result is affirmative, the fourth storage of the patterning element buffer 40 is performed. Stored in the area (* 4), the fifth storage area (* 5) or the sixth storage area (* 6) (p79, p81, p83),
If negative (N) in step p82, the patterning element “1” is stored in the seventh storage area (* 7) (p
84). When the storage of the patterning element “1” in any of the storage areas is completed, the processing returns to the main flow from the patterning element setting routine.

When the patterning element setting routine (R7) after step p35 in FIG. 14 is completed, step p3
In 6, the pointer of the term type buffer 37 is updated (+4), the storage area of the data to be acquired from the term type buffer 37 is specified, and then the number of terms is updated (−
4) Then, the number of remaining terms is set (p37). continue,
The next four item types are obtained from the item type buffer 37 (p38), and whether the item type is E (1110) or more is determined.
That is, it is determined whether the number of black bits is larger than a certain value (32 in the embodiment) or more (p39). What is printed in the passbook date column includes alphanumeric characters 0-9, "-" (hyphen) and "." (Period), and the alphanumeric characters and symbols depend on the size of the number of black bits. It is identifiable. Based on this viewpoint, in this embodiment, for the purpose of distinguishing both,
The above E is used as a term type identification criterion.

If the decision result in the step p39 is affirmative, it is determined whether or not the number of terms in the term type buffer indicating the position of the term is 5 or more (p40). If the determination result of step p40 is negative, that is, if the number of print items is 4 or less,
In the next step p41, the above-described patterning element setting routine (R7) is executed, and the patterning element is set to "1" in the sixth storage area * 6 of the patterning element buffer 40.

When the patterning element setting routine (R7) in step p41 ends, the pointer of the term type buffer 37 is updated (-1) (p42) and the term number counter is updated (+1) (p43). . Subsequently, after the number of terms is updated (−1) (p44), it is determined whether or not the term type check has been completed based on whether the value of the term number counter is 0 (p45). If the judgment result is negative,
Returning to step p38, the type of the next item is obtained from the item type buffer 37, and step 3 is performed in the same manner as described above.
Move to 9. On the other hand, if the determination result in step p45 is affirmative, the process returns to the basic flow in FIG.

If the decision result in the step p40 in FIG. 14 is affirmative, the process shifts to a step 46 in FIG. 15 to determine whether or not the number of terms is 9 or more. If not, in the next step p47, a patterning element setting routine (R7) is executed, and the patterning element is set to "1" in the fifth storage area * 5 of the patterning element buffer 40. Thereafter, the process proceeds to step p42 in FIG. If the determination result in step p46 is affirmative, it is determined whether the number of terms is 13 or more (p
48). If not, in the next step p49,
The patterning element setting routine (R7) is executed, and the patterning element is set to "1" in the fourth storage area * 5 of the patterning element buffer 40. Thereafter, the process proceeds to step p42 in FIG. If a positive determination result is obtained in step p48, it is next determined whether the number of terms is 17 or more (p50). If not, in the next step p51, a patterning element setting routine (R7)
Is executed, and the patterning element is set to "1" in the third storage area * 5 of the patterning element buffer 40. Thereafter, the process similarly proceeds to step p42 in FIG. If a positive determination result is obtained in step p50, it is next determined whether or not the number of terms is 21 or more (p52). If not, in the next step p53, the patterning element setting routine (R7) is similarly executed, and the patterning element is set to "1" in the second storage area * 2 of the patterning element buffer 40. Thereafter, the process similarly proceeds to step p42 in FIG. Further, if a positive determination result is obtained in step p52, it is next determined whether the number of terms is 25 or more (p54). If not, in the next step p55, the patterning element setting routine (R7) is similarly executed, and the patterning element buffer 4
The patterning element is set to "1" in the first storage area * 1 of 0. Thereafter, the process similarly proceeds to step p42 in FIG. If the decision result in the step p54 is affirmative,
The process directly proceeds to step p42 in FIG.

If the comparison print pattern does not match the primary reference print pattern illustrated in FIG. 18 at step p34 in FIG. 14, the process proceeds to step 57 in FIG. 16, and the item type is less than E in step p39 in FIG. Also, the process proceeds to step p56 in FIG. In step p56, the following four types of ties are obtained from the item type buffer 37, and the routine goes to step 57.
In step 57, a secondary reference print pattern ("-" pattern) of a hyphen, which is one type of a separator character as illustrated in FIG. 19, is selected from the second reference print patterns.
Of the acquired secondary reference print pattern and the second comparison print pattern (R6), and it is determined whether they match (p).
59).

If they match, the patterning element setting routine (R7) is executed (p60), the patterning element is set to "1" in the storage area corresponding to the number of items at that time, and the pattern is stored. Is done. Subsequently, after the pointer of the term type buffer is updated (+4) (p61) and the number of terms is updated (-4) (p62), the process proceeds to step p26 in FIG.

If a negative determination result is obtained in step p59, a secondary reference printing pattern ("." Pattern) of a period, which is another type of separator character, is used.
Of the obtained secondary reference print pattern and the second comparison print pattern (R64) are executed (p64), and it is determined whether or not they match (p64).
65). If they match, the patterning element setting routine (R7) is executed (p66), the patterning element is set to "1" in the storage area corresponding to the number of terms at that time, and the pattern is stored. Subsequently, the pointer of the term type buffer is updated (+4) (p67) and the number of terms is updated (-4) (p68).
The process proceeds to Step p26 of Step 3. On the other hand, if they do not match in step p65,
Move to 29.

In summary, the patterning element buffer 40
Stores the patterning element “1” or “0” based on the number of terms and the term value in the term value buffer as follows. That is, when it is determined that the number of items 1 to 4 has a character (when the item value is 32 or more) or no character (when the item value is 31 or less), "1" or "1" is stored in the first storage area * 1. 0 "is stored. If it is determined that there are characters or no characters in the item numbers 5 to 8, “1” or “0” is stored in the second storage area * 1. When it is determined that there is a character or no character in the number of items 9 to 12, "1" or "0" is stored in the third storage area * 1. If it is determined that the number of items 13 to 16 has a character or no character, the fourth storage area * 1 stores
“1” or “0” is stored. If it is determined that the number of items 17 to 20 has a character or no character, the fifth storage area *
1 stores “1” or “0”. Number of terms 21 to 24
Is determined to have a character or no character, "1" or "0" is stored in the sixth storage area * 1. 25 terms
When it is determined that there is a character or no character in
"1" or "0" is stored in the storage area * 1.

In the above embodiment, whether or not the number of black bits is 32 or more is used as a criterion for determining whether or not there is a character component in 1-byte scan data. However, depending on the form of the character, that is, the typeface, the normal character or the separator character, even if there is a print character, the number of black bits in one byte varies with a certain width. Then, even when the medium has not only print characters but also some stains, black pixels are detected. Under such circumstances, it is difficult to make a highly accurate print presence / absence determination using only one determination criterion.

In view of the above circumstances, the present invention has attempted to use two boundary values as the printing presence / absence criterion as follows. That is, two constant values a and b (where a> b) are set as the determination boundary values, and when the count value of the byte pixel counter (that is, the number of black bits in one byte) x is x ≦ b, "0", which means no printing, is set in the register if x≥a, "1", which means that there is printing. When using such a two boundary values, distinguishes characters and blemishes, clarity although that can be determined indium, when circle simply etc. stop thereto, x is from between the two boundary values If it belongs to the gray zone, it becomes impossible to make a determination, and the determination accuracy is impaired. For example, as illustrated in FIG. 20, when the print range of the number “0” is divided into three equal parts in the vertical direction, the black bit numbers x1 and x3 of the first vertical section a1 and the third vertical section a3 are:
Since it is larger than the determination boundary value a, it is set to “1”. On the other hand, the number of black bits x2 of the second vertical section a2 falls between the determination boundary values a and b. And cannot be determined.

In order to solve such a problem, in the method of the present invention, the count value x of one byte pixel counter
When 1 falls between two boundary values, the count value x2 of the next byte pixel counter is sequentially added to the count value x1 of the byte pixel counter, and the addition result (x1 + x2) becomes another constant value. If it is equal to or more than c (where c = a + d, d is a small difference), “1” is set and the added value (x1 + x
If 2) is less than the fixed value c, the value (x3) of the next byte pixel counter is added to the value in order, and the added value is equal to or more than a fixed value (c + nd: n is a value according to the number of added byte pixel counters). Is set to “1”, and when the value xm of the byte pixel counter to be added becomes smaller than “a”, “0” is set and the primary comparison consisting of a combination of “1” or “0” is performed. A print pattern is configured.

As illustrated in FIG. 21, the number of black bits included in the scan data of each byte stored in the buffer memory or the term value buffer BF of FIG. .., and assuming that the relationship with respect to the two boundary values a and b is as shown in FIG. 3C, “0” is set for the first and second bytes. The third byte is the boundary status a, b
, The number x3 of the byte pixel counter of the third byte is added to the number x4 of the byte pixel counter of the next fourth byte.
Is added, and the addition result is equal to or more than c.
The third byte is set to "1". Therefore, as illustrated in FIG. 20, even when the number of black bits in the intermediate portion belongs to between two boundary values due to insufficient print density or the like when “0” is originally printed, the above-described addition is performed. Since the intermediate portion is also set to "1" by the processing, the determination is corrected to the determination according to the actual print pattern, and the determination with high reliability is performed.

In this way, for each byte shown in FIG. 21A, "0", "0", "1", "1" and "1" are stored in the register.
"0", "0", "0", "1",... Are set, and a primary comparison print pattern is generated. 28 sets of data "0" or "1" individually corresponding to all bytes are used as a primary comparison print pattern, compared with the primary reference print pattern, and the final print presence / absence is determined based on whether they match. It is also possible to make a determination. However, the amount of data to be processed is C
In order to reduce the burden on the PU, it is desirable that the number be as small as possible. Therefore, in a preferred embodiment of the present invention, FIG.
(C) As shown in (d), when a plurality of the same logical values of “1” or “0” of the primary comparison print pattern in the register are consecutive, the number is compressed to one and the secondary comparison is performed. Composes a print pattern, compares the secondary comparison print pattern with a secondary reference print pattern consisting of compressed logical values when there is a print, and if there is a match, determines that line as a print line I did it.

In the above embodiment, it is possible to determine the presence or absence of each type of a predetermined character or symbol at a predetermined position. In addition, since the presence or absence of printing can be determined not in units of one character but in units of format composed of one line of characters, the content to be printed such as a date column is determined in advance within a certain range. In such a case, it is possible to determine the presence / absence of printing with high accuracy by creating a secondary reference printing pattern corresponding to the printing content.

[0068]

As described above, according to the method of the present invention, a scan is performed by providing a byte pixel counter for counting "1" pixel data for each byte for scan data for one scan. Each time one byte of pixel data of "1" is added and accumulated in each of the byte pixel counters, and when all the scans for one row are completed, the row is read out based on the contents of each byte pixel counter. A comparison print pattern is created, and the created comparison print pattern is compared with a reference print pattern created and stored when there is actually printing, and if there is a match, the line is printed. Because it was decided to be a line, vertical characters,
Even if the PCS value decreases and the number of black bits in the scan data decreases, it is possible to determine the presence or absence of printing with high reliability by utilizing the characteristic that the ratio of remaining vertical components is large.

According to the third aspect of the present invention, it is possible to determine the presence or absence of printing not in units of one character but in units of format composed of one line of characters. If the contents to be printed are predetermined in a certain range, it is possible to determine the presence / absence of printing with high accuracy by creating a secondary printing pattern according to the printing contents.

Further, according to the fourth aspect of the present invention, even when the number of bytes in a print line is large, the determination data is compressed and determined, so that the processing load on the CPU is reduced, and the set of characters in a certain range is reduced. Alternatively, it is possible to determine the presence or absence of printing for each format.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a main configuration used to carry out a method of the present invention.

FIG. 2 is a conceptual diagram conceptually showing a relationship between a vertical section of a date column and scanning of each section.

FIG. 3 is a schematic diagram illustrating storage of scan data obtained by scanning one row.

FIG. 4 is a schematic diagram illustrating an operation of adding the number of black bits per byte for 24 scans (vertically) and storing the sum.

FIG. 5 is a schematic diagram illustrating patterning of the number of black bits in each vertical section.

FIG. 6 is a flowchart illustrating a basic operation of the method of the present invention.

FIG. 7 is a block diagram of a determination device.

FIG. 8 is a timing chart showing the operation timing of the determination device.

FIG. 9 is a conceptual diagram showing the relationship between the number of black bits and the term type in term value classification.

FIG. 10 is a flowchart showing a part of the processing procedure for term value classification.

FIG. 11 is a flowchart showing the rest.

FIG. 12 is a conceptual diagram of patterning.

FIG. 13 is a flowchart showing a part of a patterning process.

FIG. 14 is a flowchart showing another part of the patterning process.

FIG. 15 is a flowchart showing still another part of the patterning process.

FIG. 16 is a flowchart showing the rest of the patterning process.

FIG. 17 is a flowchart showing a subroutine for setting a patterning element.

FIG. 18 is a conceptual diagram illustrating an example of a reference print pattern when it is determined that printing is performed.

FIG. 19 is a conceptual diagram showing an example of a secondary reference print pattern when it cannot be determined that printing is performed.

FIG. 20 is a schematic diagram illustrating a relationship between a character form, a change in the number of black bits, and a determination criterion.

FIG. 21 is an explanatory view showing another embodiment of the present invention.

FIG. 22 is a front view showing a print format of a generally used passbook.

FIG. 23 is a schematic configuration diagram of a printer.

FIG. 24 is a block diagram of a printer.

FIG. 25 is a conceptual diagram illustrating a scanning principle of a date column.

FIG. 26 is an explanatory diagram showing scan data when scanning “1” printed at an appropriate density.

FIG. 27 is an explanatory view showing scan data when scanning “1” printed at an inappropriate density.

[Explanation of symbols]

OSU optical reader 31 buffer memory 32 readout unit 33 _{1 to} 33 _28- byte pixel counter 34 term value buffer 35 conversion unit 36 first table 37 term type buffer 38 second table 39 comparison unit

Continuation of front page (72) Inventor Yuji Urushida 4-3, Jonai, Hanamaki-shi, Iwate F-term in Shinko Seisakusho (reference) 2C059 FF03 FF07 5B021 AA01 BB02 CC06 DD01 EE01 GG03 9A001 HH21 JJ35 KK58

Claims (4)

[Claims] 1. A scan data comprising binary pixel data of "1" and "0" outputted corresponding to a black pixel and a white pixel when a line sensor scans a line of a print medium on which printing is to be performed. In the print presence / absence determination method of determining whether or not printing has been performed on the line by performing a predetermined analysis on the scan data, (a) scanning data for one scan every one byte (B) Each time scanning is performed, the number of “1” pixel data for each byte is added to and accumulated in each of the byte pixel counters. c)
At the time when the scanning of one row is completed, a comparative print pattern of the row is created based on the count value of each byte pixel counter, and (d) the created comparative print pattern is actually printed. Comparing with a reference print pattern created and stored for each case, and if there is a match, determine the line as a line with printing (C1) The number of pixel data of "1" in each byte of scan data obtained by scanning each line of a date column of a print medium is added in the sub-scanning direction in byte units. Then, it classifies whether the count value of the pixel data of “1” in each byte corresponds to a value of a plurality of levels, assigns a predetermined item type to the classification result, and assigns each of the four items. A comparative print pattern is generated by the term type of
1) The comparison printing pattern is compared with a reference printing pattern, and when at least one comparison printing pattern matches the reference printing pattern, the line is determined to be a printed line. Printing presence / absence determination method. (C21) The number of "1" pixel data in each byte of scan data obtained by scanning each line of the date column of the print medium is added in the sub-scanning direction in byte units. Then, it classifies whether the count value of the pixel data of “1” in each byte corresponds to a value of a plurality of levels, assigns a predetermined item type to the classification result, and assigns each of the four items. (D21) collating the first print comparison pattern with the first reference print pattern, and (c22) comparing the second print pattern with the first reference print pattern by using the collation result for every four terms. Generate a comparative print pattern and (d
22) The second comparative print pattern is compared with a second reference print pattern created corresponding to one line of various print columns, and if there is a match, the line is regarded as a printed line. The method according to claim 1, wherein the determination is made. (C31) The print pattern compares the value x1 of each byte pixel counter with two constant values a and b (a> b), and when x1 ≦ b, “0” is set, and a ≦ x1 When a>x1> b, a value of the next byte pixel counter is added to the value x1 when a>x1> b, and when the value x2 is equal to or more than a fixed value c (c = b + d), , "1", and when the value x2 is less than the constant value c, the value x
2, the value of the next byte pixel counter is added in order, and when the value xk is equal to or greater than a fixed value (c + nd: n is a value according to the number of added byte pixel counters), "1" is set and the addition is attempted. When the value xm of the byte pixel counter becomes smaller than a, “0” is set and “1” or “0” is set.
To form a primary comparison printing pattern composed of
32) When a plurality of “1” and “0” of the primary comparative print pattern are respectively continuous, they are compressed into one to form a secondary comparative print pattern, and (d31) the secondary comparative print pattern is used as a reference. 2. The print presence / absence determination method according to claim 1, wherein the print pattern is compared with a print pattern, and if there is a match, the line is determined as a print line.