JPH08263262A - Compression method for printing data - Google Patents

Abstract

PURPOSE: To provide a compression method by which compression can efficiently be executed even if data on a line which is to be converted and data on an immediately preceding line are hardly matched with each other. CONSTITUTION: An exclusive OR Ex (n, m) is obtd. between data on the line to be converted and data on the same bit position before n-lines (step 1-1). The head bit position variable s-bit which is not converted yet and the variable c-bit of a pointer showing a bit position, under retrieval are initialized (steps 1-2 and 1-3) and the number of continuous identical bits con(n) is initialized (step 1-4). Then, a sub-routine con-make is called, and the number of the continuous identical bits in respective lines preceding to the head bit position which is not converted yet is retrieved. Then, the result is stored (step 1-5) in con(n). The largest con (n) among generated pieces of con(n) is searched (step 1--6). The sub-routine data-make is called and converted data is generated(step 1-7) in accordance with the maximum number of continuous identical bits obtained as a result.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method of compressing print data to be printed by a printer, and more particularly to a method of compressing print data when it is stored.

[0002]

2. Description of the Related Art FIG. 2 is an explanatory diagram showing a print result, and FIG. 3 is an explanatory diagram showing print data for obtaining the print result of FIG. FIG. 2 shows a set of printable minimum physical size raw dots (dots), and FIG. 3 shows a dot as a set of binary values (information of 1 or 0, bits).

In FIG. 3, the 1st to 12th from the left,
The combination of 0 has not changed since the sixth row. Therefore, the data in each row is compared with the data in the immediately preceding row, and the portion of the same bit combination as in the immediately preceding row is “the combination of consecutive m bits from here is the same as the immediately preceding row”.
By converting to the information, and storing the portion where the bit combination is different from that of the immediately preceding row, the information is stored as "the information of consecutive n bits from here is the information of the following n bits", so that the number is smaller than that before the conversion. A bit string that can be reproduced with the number of bits can be generated. In such a compression method, a bit is required to indicate whether or not the following data shows the same information as the immediately preceding row, or how many bits are the same as the immediately preceding row.

[0004]

However, the above-described conventional compression method cannot compress data when printing dots as shown in FIG. FIG. 5 is a binary value table corresponding to the dots in FIG. As can be seen from FIG. 5, the data in each row hardly matches the data in the immediately preceding row. Therefore, when compressed by the above-mentioned conventional method, the number of bits is increased by the amount of bits required to indicate whether the following data shows the same information as the previous row or how many bits are the same as the previous row. Will be done.

[0005]

In order to solve the above problems, the compression method of the present invention determines whether data similar to the print data of the line to be converted is present in a plurality of lines preceding the line. The number of similar data that are consecutive in each of the preceding plurality of rows is searched, and the longest consecutive similar data holding row with the largest number of consecutive similar data in the preceding plurality of rows and the corresponding row When the number of consecutive similar data is equal to or more than a predetermined number, the data of the row is converted with the same data of the holding row.

[0006]

According to the present invention having the above configuration, whether or not the print data similar to the print data of the line to be converted is present in a plurality of lines preceding the line is compared, and the Since the data of the preceding row having the same continuous data as the print data is designated and compressed, the compression can be efficiently performed.

[0007]

Embodiments of the present invention will be described below with reference to the drawings. Note that elements common to the drawings are given the same reference numerals. FIG. 1 is a flowchart showing the operation of the embodiment according to the present invention.

First, in FIG. 5, the dot combinations in the third row are compared with those in the immediately preceding second row.
~ 29 odd columns, 30 and 32 columns are the same,
Compared to the first row, the bit strings from the 4th to 32nd columns are exactly the same. In this way, by expanding the range of comparison to a plurality of lines preceding the corresponding line, "the following n bits are m
If the same bit combination as the relevant row does not exist even after comparing up to the previous multiple rows, it is converted to the information "It is the same as the data at the same bit position in the row before the row." By storing it as information that "n bits are the information of the following n bits", a bit string that can be reproduced with a smaller number of bits than before conversion can be generated.

In this embodiment, the formats shown in FIGS. 6 and 7 are used to mix the above two types of information. FIG. 6 shows the modalities and FIG. 7 shows the modalities. In the present embodiment, the maximum number of bits specified in one format is 16, and the maximum number of preceding lines to be compared is 8. The most efficient values of these two numerical values differ depending on the actual combination of dot rows to be converted.

In this embodiment, one bit is used to indicate whether or not the following data shows the same information as the preceding row. In the format 601, shown in FIG. 6, "1" indicates the same information as the preceding row, and "0" indicates a new bit combination. For example, in FIG. 6, since the format identification bit 602 is "1", the following data is the same information as the preceding row. Same as continuous 603
Indicates the number of bits to be stored + 1, which is the same combination as the preceding row. Further, the preceding row number 604 is a stored numerical value +1 and indicates how many rows before and the same. Therefore, as shown in the figure, the number of bits is 6
When 03 is 5 and the number of preceding rows 604 is 3, it indicates that the 6-bit data is the same information as the row 4 rows before.

In FIG. 7, in the format 701, the format identification bit 702 is "0". Insertion bit number 703
Indicates how many bits are consecutively inserted by the stored numerical value + 1. Inserted data 704 indicates the data itself to be inserted. That is, as shown in FIG. 7, when the number of insertion bits 703 is 4, it indicates that 5-bit data 704 is continuously inserted newly.

Since the data in the first row of the binary value table shown in FIG. 5 has no object to be compared, it is all in the format. Therefore, the format identification bit 702 is "0", the insertion bit number 703 is 15, and the insertion data is 704 is a binary value “01110101
01010101 ”, followed by insert data 704 in the same format
It becomes "0101010101010111". As a result, the data in the first row becomes as shown in FIG.

As for the second row, as compared with the first row, only 4 bits in which the bits are dispersed are similar to each other. Therefore, the format is the same as that of the first row, and the second row data has a result as shown in FIG.

Regarding the third row, the first 2 bits are the same as those in the second row, but when expressed in a format, 8 bits are required, and the number of bits is larger than 7 bits in the case of the format. Therefore, it is expressed in a format including the third bit. Since the 4th bit to the 32nd bit are all the same as those in the 1st row, they are expressed in a format. Since the 4th bit to the 19th bit have the content that "consecutive 16 bits are the same as two lines before",
The format identification bit 602 is “1”, and the number of bits is 1 as well.
5, the number of preceding lines is 1. Since the 20th bit to the 32nd bit have the content that "13 consecutive bits are the same as two lines before", the format identification bit 602 is "1", the number of bits is 12, and the number of preceding lines is 1. As a result, the third row data becomes as shown in FIG.

Now, let us consider how many bits are consecutive and the number of required bits is smaller when the format is adopted than in the same case. The format fixedly requires 8 bits. If consecutive x bits are expressed in a format, 5 bits are added to the x bit data itself, so the total (x
+5) bits are required. If the format is adopted when the number of bits is 8 or more, that is, x is 4 or more, that is, 4 bits or more are consecutive and similar to the preceding row, the required number of bits is the same as or smaller than the format.

FIG. 11 is a block diagram of a printer to which the embodiment of the present invention is applied. In the figure, the CPU 111 controls the entire operation of the printer and compresses print data in this embodiment. Bus 1 for CPU111
The ROM 113, the RAM 114, and the port 115 are connected via 12. A control program or the like is stored in the ROM 113, and print data is temporarily stored in the RAM 114. The upper interface circuit 116 and the printing mechanism unit drive circuit 117 are connected to the port 115, and the printing mechanism unit drive circuit 117 further includes the printing mechanism unit 1.
18 is connected. Upper interface circuit 116
An upper device (not shown) is connected to the printer, and print data is sent from this upper device to the printer.

Next, the print data compression operation will be described with reference to the respective flowcharts. The flowchart shown in FIG. 1 is a main routine showing the entire algorithm of this embodiment, and calls a subroutine in some steps. Here, the algorithm used in each subroutine will be described.

First, cur (m) represents the m-th bit of the row, and Bit (n, m) represents the m-th bit n rows before the row. That is, here, it is assumed that corresponding bit data is stored in advance in cur (m) and Bit (n, m). Ex
(n, m) is the exclusive OR of the data at the same bit position n rows before and the current row (symbol # is the exclusive OR operator), and Ex (n, m) is 0 , It can be said that the same data as the m-th bit of the row is n rows before. con (n) stores the number of bit combinations that are the same as those in the row and that continuously exist from the bit position being searched n rows before. s-bit is the conversion of the unconverted start bit position, and c-bit is the variable of the pointer that indicates the searched bit position. m-max represents the total bit width of the line. Since the format to be used is the same as the formats shown in FIGS. 6 and 7, the number of preceding rows 604 for searching similar bits is 8, and the maximum value of the number of consecutive bits 603 and the number of inserted bits 703 is 16. .. Also,
Exception handling when the number of rows preceding the row is less than 8 is not included in the operation described below.

First, in the flowchart of FIG. 1, a subroutine Ex-set is called to generate Ex (n, m). FIG.
2 is a flowchart showing the subroutine Ex-set.
In FIG. 12, first, the variable i representing the number of the preceding row is initialized to 1 (step 12-1), and then the variable j representing the bit position is initialized to 1 (step 12-2). In step 12-3, the exclusive OR Ex (i, j) of the data at the first bit position one row before and the row concerned is obtained, and step 1
At 2-4, the bit position is changed, and the exclusive OR Ex (i, j) of the data at the bit position one row before and the current row is obtained until the data at the last bit position (step 12-5). Exclusive OR Ex (i,
When j) is obtained, the exclusive OR Ex (i, j) of the data at each bit position two rows before is obtained (step 12-6). The above processing is performed for the data at each bit position 8 rows before (step 12-7).

In FIG. 1, when the subroutine Ex-set is completed, the unconverted leading bit position variable s-bit is initialized (step 1-2) and the pointer variable c-bit indicating the bit position being searched is initialized. (Step 1-
3). Next, call the subroutine con-ini (step 1-4).

FIG. 13 is a flowchart showing the subroutine con-ini. In the figure, first, the variable i of the number of the preceding row is initialized to 1 (step 13-1), and the bit number con (i) is initialized to 0 as in the continuous manner (step 1).
3-2). The above processing is performed for the preceding eight lines (steps 13-3 and 13-4). As a result, the number of bits co
All 0s are stored as n (i). When this process ends, the subroutine con-make is then called (step 1-5).

FIG. 14 is a flowchart showing the subroutine con-make. In the figure, first, the bit position variable j used in this subroutine is initialized with the unconverted head bit position variable s-bit (step 14-1). As a result, in this subroutine, the presence / absence of a bit is similarly searched from the s-bit position. Next, the number i of the preceding line is initialized to 1 (step 14-2). Next, it is judged whether the searched bit position data of the i-th preceding row is the same as the searched bit position data of the row (step 14-3). If not, the process branches to step 14-11.

In step 14-4, since one similar bit is found, the number of consecutive concatenated bits con (i) during the search is set to 1.
And After that, the number of consecutive bits in the i-th preceding row is searched for the same, and the result is stored in the con (i). Here, k is used as a variable indicating the bit position during the bit number search as in the case of continuation. That is, since the searched bit position is in the variable j, the variable k is initialized to the value of j in step 14-5,
Next, in order to check the bit position, it is incremented by 1 in step 14-6, and it is checked in step 14-9 whether the data is actually the same as that in the row. If it is determined in step 14-9 that they are the same, the number of bits con (i) is continuously determined in step 14-10.
Is incremented by 1, and an unconditional branch is made to step 14-6 to search for the next bit position. If it is determined in step 14-9 that the line is not the same as the line in question, the search loop is immediately exited, and the process branches to step 14-11 to shift to the process in the next preceding line.

In step 14-7, it is indicated that the loop is exited when the number of bits exceeds the maximum value 16 as in the case of continuous search. Since the variable k represents the bit position, its value at the time of retrieval does not need to exceed the maximum bit width m-max. In step 14-8, the loop is executed when the bit position variable k exceeds the maximum bit width m-max. It indicates that it will come out. As described above, the search result of the number of bits is stored in con (n) in the same manner as in the preceding rows from the unconverted leading bit position. When the processing of the subroutine con-make is completed, the subroutine con-max is called next (step 1-
6).

FIG. 15 is a flowchart showing the subroutine con-make. This subroutine performs a process of searching for the largest number of bits con (i) as in the case of continuation of each preceding row. In the figure, i indicates the number of the preceding line being searched, and p is a variable holding the preceding line number of a line having the same number of bits as the maximum continuous number up to the i-th preceding line search time point.
First, the variable i of the preceding line number is set to 1 representing the line immediately preceding the line.
(Step 15-1), and the bit holding row number p is set to 0 (step 15-2) in the same manner as longest continuous. As a result, the bits are set so that there is no bit in any row at the start of the search. Next, at step 15-3, it is determined whether the number of bits is 0, which is the same as the continuation of the line being searched. If it is 0, the process branches to step 15-7 to move the search target to the next preceding line.

If the number of consecutive bits of the row being searched is not 0, the process branches to step 15-4 to compare with the known number of consecutive longest bits. In step 15-4, it is determined whether or not there is a known number of bits as in the longest consecutive case (if the bit holding row number p is 0 as in the longest consecutive case,
If there is no known longest consecutive bit number as well), and if there is no known longest consecutive bit number, then step 15-
In order to branch to 6 and make the row being searched for the bit holding row number in the same manner as the longest consecutive row, the row number i being searched is assigned to p.

If there is a known longest consecutive bit number con (i), the known longest consecutive bit number con (p) and the row contiguous bit number con (i) of the row being searched are found in step 15-5. If it is compared, and if the number of consecutive bits of the row being searched con (i) is larger than the known number of longest consecutive bits of con (p), in step 15-6, the row of the row being searched is given the same number of consecutive bits. In order to use the holding row number, the number i of the row being searched is assigned to p. Consecutive number of bits for the line being searched con (i)
Is not larger than the known longest consecutive bit number con (p), the longest consecutive bit holding row number p is not updated, and the process branches to step 15-7 to move the search target to the next preceding row. . In step 15-8, it is judged whether or not the search of the preceding line to be searched is completed, depending on whether or not the number i of the line being searched exceeds 8, and if the number is equal to or less than the number of lines to be searched, the next preceding line is searched. The process branches to step 15-3 to search for lines, and when the number of lines to be searched exceeds 8, this subroutine is finished. At the end of this subroutine, the number of the bit holding row is stored in the variable p as in the longest continuous. When the same bit is not included in any preceding row to be searched, the variable p is 0. When the processing of the subroutine con-max is completed, the subroutine data-make is then called (step 1-7).

FIG. 16 is a flow chart showing the subroutine data-make. This subroutine data-make
The conversion data is generated according to the same number of bits as the longest continuous obtained by the subroutine con-max. Before explaining this subroutine data-make, another subroutine called in this subroutine data-make
Explain make-same and subroutine make-new. FIG. 17 is a flowchart showing the subroutine make-same.

In FIG. 17, the subroutine make-same
Is a subroutine for creating conversion data in the case where similar data is continuously present in the preceding line. In this subroutine, the parallelogram processing box indicates the processing for storing the conversion data in the area in the memory (RAM 114 shown in FIG. 11). First, in step 17-1, "1" is stored in the memory as the style identification bit 602.
This clearly indicates that it is a form. Next, in step 17-2, a pointer variable c-bi indicating the searched bit position
The difference between t and the unconverted leading bit position variable s-bit is stored in the memory as the number of bits 603, which is the same as continuous. Continued Step 1
In 7-3, the number of preceding lines 604 of the style is stored, but since the variable p has the number of the immediately preceding row set to 1, p-1 is stored to correspond to the style.

FIG. 18 is a flowchart showing the subroutine make-new. This sub-routine make-new is a sub-routine that creates converted data in a format when there is no similar data in any preceding row to be searched. Also in this subroutine, the parallelogram processing box stores the converted data in the memory (RAM 1 shown in FIG. 11).
14) shows the process of storing in the area inside. First step 1
8-1, "0" is stored in the memory as the format identification bit 702.
Is stored and the format is clearly specified. Step 18
-2, a pointer variable c-bit indicating the bit position being searched
And the unconverted leading bit position variable s-bit is stored in the memory as the insertion bit number 703. (C-bit)-(s-bit) can be used as the number of inserting bits 703 of the format by using the pointer variable c-bit indicating the bit position during the search.

In steps 18-3 to 18-6, the format insertion data 704 is stored. The variable i is used as a pointer to the bit position to be stored,
A pointer that is initialized by substituting the unconverted bit position variable s-bit in step 18-3, stores the bit data cur (i) of the row in step 18-4, and is incremented by 1 in step 18-5. And the pointer is
Store continuously in memory until it exceeds c-bit.

Next, the subroutine data-make will be described with reference to the flowchart of FIG. In this subroutine, if the number of bits con (p) exceeds 4, the converted data is stored in a format and the unconverted start bit position variable s-bit is updated, otherwise, the number of unconverted bits is changed. When ((c-bit)-(s-bit + 1)) tries to exceed the maximum bit length 16 specified in the format, the converted data is stored in the format and the unconverted start bit position variable s-bit is set. It is updated, and if it does not exceed, it simply increments c-bit by 1.

First, at step 16-1, it is judged whether or not there is a similar bit in the preceding row to be searched, depending on whether the variable p is 0 (if it does not exist) or 1 (if it exists). If it does not exist, the process branches to step 16-2. If not, the process branches to step 16-3. In step 16-2,
As described above, if the number of bits is 4 or less as in continuous, the number of bits after conversion is smaller in the format. If the number of bits is 4 or less as in continuous, the process branches to step 16-3 and the same bit exists. The same processing is performed as when not performing.

In step 16-2, the number of bits is 4 as in the continuous manner
If it exceeds, the process branches to step 16-4, and it is determined whether or not the unconverted leading bit position is equal to the searching bit position. If it is determined that they are not equal, the data must be converted according to the format up to the unconverted head bit position and one bit before the bit being searched, so according to the format in steps 16-5 to 16-8. To be converted. In step 16-5, the searching bit position pointer variable is decremented by 1 in order to convert the data up to 1 bit before the searching bit position according to the format. This prepares to call the subroutine make-new in the next step 16-6. In step 16-6, the subroutine make-new is called, and the subroutine described in FIG. 18 is executed.
Perform make-new processing and convert data according to the format. In step 16-7, the contents of the c-bit corrected in step 16-5 are restored, and in step 16-8, the restored c-bit value is substituted into s-bit to correct the unconverted leading bit position. To do.

If it is determined in step 16-4 that the unconverted leading bit position is equal to the searched bit position, the process branches to step 16-9. In step 16-9, the subroutine make called in the next step 16-10
In order to match the meaning of the variable c-bit used in the -same process, con (p) is added to c-bit and further decremented by 1, so that the content of c-bit is the bit trailing edge position.
In step 16-10, the subroutine make-same is called to store the format conversion data as described with reference to FIG. 17, and the steps 16-11 and 16-
At 12, the s-bit and c-bit are updated.

In step 16-3, the pointer variable c-bit indicating the bit position being searched is the unconverted head bit position variable.
If there are 15 more s-bits (there are no 16 consecutive bits in the same manner), step 16-1 is performed in order to convert the data up to the currently searched row bit.
Branch to 3 and call the subroutine make-new to convert the data according to the format. And step 16-1
In step 4 and step 16-16, the s-bit and c-bit are updated.

If the difference between the variable c-bit and s-bit is less than 15 in step 16-3, the process branches to step 16-15,
It is determined whether or not the pointer variable c-bit indicating the bit position being searched reaches the maximum bit width. If it has reached, the processing is terminated, and if it has not reached, the process branches to step 16-16 and c- Update the bit.

Step 1-8 in FIG. 1 is a step for exiting the loop when the unconverted bit reaches the maximum bit width, and branches to step 1-9 when the unconverted bit reaches the maximum bit width. If not, the process branches to step 1-4. In step 1-9, if there are unconverted bits, a subroutine make-new for converting them into a format is called and processing is performed.

Based on the conversion algorithm described above, FIG.
FIG. 19 shows the result obtained by converting the binary value table shown in FIG. When converted according to this embodiment, 960-bit data is 69
It is converted to 6 bits and compressed to about 72%.

[0040]

As described above in detail, according to the present invention, the data of the row is compared with the data of the preceding rows, and the row data having the longest bit similar to the row is selected from the data. Since the designated data is compressed, the print data that hardly matches the data of the immediately preceding line can be efficiently compressed so that the number of bits is reduced.

[Brief description of drawings]

FIG. 1 is a flowchart showing the operation of an embodiment of the present invention.

FIG. 2 is an explanatory diagram showing a print result.

FIG. 3 is an explanatory diagram showing print data for obtaining the print result of FIG.

FIG. 4 is an explanatory diagram showing another print result.

FIG. 5 is an explanatory diagram showing print data of FIG.

FIG. 6 is an explanatory diagram showing a format of compressed data.

FIG. 7 is an explanatory diagram showing a format of compressed data.

FIG. 8 is an explanatory diagram showing first row data.

FIG. 9 is an explanatory diagram showing second row data.

FIG. 10 is an explanatory diagram showing third row data.

FIG. 11 is a block diagram of a printer to which the embodiment is applied.

FIG. 12 is a flowchart showing a subroutine of the embodiment.

FIG. 13 is a flowchart showing a subroutine of the embodiment.

FIG. 14 is a flowchart showing a subroutine of the embodiment.

FIG. 15 is a flowchart showing a subroutine of an embodiment.

FIG. 16 is a flowchart showing a subroutine of the embodiment.

FIG. 17 is a flowchart showing a subroutine of the embodiment.

FIG. 18 is a flowchart showing a subroutine of the embodiment.

FIG. 19 is an explanatory diagram showing a result of conversion according to the example.

[Explanation of symbols]

 111 CPU 114 RAM 601 Style 602 Style identification bit 603 Same number of consecutive bits 604 Number of preceding lines

Claims (2)

[Claims] 1. A method for compressing print data, wherein whether or not data similar to the print data of the line to be converted is present in a plurality of lines preceding the line, and each of the preceding lines is searched. In the above, the number of consecutive similar data is calculated, and the longest consecutive similar data holding row with the largest number of consecutive similar data among the preceding plurality of rows and the number of consecutive similar data of the row is a predetermined number or more. If the same, the print data compression method is characterized in that the data of the row is converted with the same data of the holding row. 2. The conversion of the data of the row with the same data of the data holding row similar to the longest continuous, a bit indicating that conversion is performed with the similar data of the holding row, and a bit indicating a number of bits similar to the continuous row, The method for compressing print data according to claim 1, wherein the method is performed by using a bit indicating the number of preceding rows of the held row.