JPS6066290A - Compression of character pattern data - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a method for compressing character pattern data. Specifically, the present invention relates to a method for efficiently compressing character pattern data displayed in dot matrix format.

In recent years, with the development of 8Fi information processing, electronic metering (
The results of various types of computer processing can be displayed on a CRT display in Japanese characters, including kanji, such as with word processors, etc., or printed out with a printer. i-processing equipment is becoming popular. The character patterns in the dot matrix format of the characters to be displayed or printed at this time are stored in a dedicated memory, from which desired character pattern data is recalled and displayed or printed.

However, as is well known, the number of characters in the E1 wooden language is extremely large, and there are approximately 3,000 types of JIS level 1 characters alone.

For this reason, the 24X2
A 4-dot dot matrix r stores 3000 characters, but requires 1728 bits (-216 bytes) of memory, and its capacity is 1? l becomes British<r. Therefore, the price of the device increases because memory is used sparingly, and when transferring character data, it takes time because there is a lot of data per character, which results in a slowdown of the overall processing speed of the device. It had drawbacks. For this reason, various methods have been devised to compress character patterns and store them in memory. However, both methods have drawbacks such as complex algorithms, requiring complex circuits to reproduce patterns, and slow reproduction speeds.

The present invention is based on this point (1), and data compression of character patterns is carried out in two stages: primary compression and secondary compression, and even secondary compression is performed according to the characteristics of the character pattern. For characters whose compression ratio is better, if you perform secondary compression, it will be 1
At the time of next compression, J: RIMO is distinguished from characters for which the compression rate is low, and compressed. The characteristics are as follows: First, in primary compression, a first matrix is created from a dot matrix representing a character pattern to show the points of change in white data and black data, and this first matrix is・Divide into n blocks with a predetermined number of dots in the horizontal direction, and if there is even one change point data 0r1 in this divided block, set it to bit 1, and if there is no change point data, set it to bit 0. Create the second 71~ risk. For the above-mentioned block made up of patterns 1 to 1, data consisting of patterns 1 to 1 that constitute the block are separately formed 1'. Next, J3 for secondary compression
If 7j' 1ri
l・M111 with specified number of dots in the horizontal direction, /1.1 of I
A third matrix is created in which bits are set to 1 if there is one [-6 front 11 self-curving point) - 0 in this divided block, and bits 1 and 0 are set to 0 otherwise. For a block whose bit is set to 1, separate data is created that indicates the bit pattern that constitutes the block. Performing this primary and secondary compression is the first step of this application.
It is a characteristic of And the second feature is that the data amount of the second 71-Risk created by the primary compression and the pitch of the third matrix and third i-Risk created by the secondary compression: ~1 Compare the data amount of the sum of the data consisting of the bit patterns that make up each block, and the character pattern where the former data amount is larger than the latter data amount is 2.
The first compression is performed until the next compression, and only the first compression is performed for other character patterns.

Hereinafter, the present invention will be explained in detail using the drawings.

In this case, a character pattern displayed in a 24×24 dot matrix will be explained as an example.

FIG. 1 is a flowchart briefly showing the processing flow of the present invention.

First, we will explain primary compression. The portion marked with △ in FIG. 1 is a portion showing the processing procedure of primary compression. Second
The figure shows an example of a character pattern for the kanji "F" in 24x2
It is represented by four dots.

From this figure 2, in the 2/1X24 drag 1~frix, the points corresponding to the points where white dots 1~ change to black dots, and the points where black dots change to white dots 1~ Create the first dot (~71 helix). This is shown in Figure 3. Next, the first dot 71~ helix, which is the change point just made, is divided by the solid line in Figure 3. J: Sea urchin ,Longitudinal direction,
Create blocks horizontally divided into n pieces by a predetermined number of dots (1 x 1 dots in this example, but it goes without saying that other divisions may be used. similar). Then examine each pattern of 1x4 totsu 1- from the direction of the arrow IJ in Fig. 3 sequentially from the direction of the arrow IJ in Fig.
If all is rOJ and all is [1], then it is 1, creating a second dratontrix as shown in Figure 4. Here, let 1~ be 1 and turn it into 1] Regarding the cock, it's the first part!
l) Separately, the bit pattern of the block consisting of 1×1 dots 1 and 1 in the figure is sequentially stored in the memory as first compressed data. To be more specific, all of the patterns from pitch 1 that make up the 4R blocks in Figure 3a are [O
[Therefore, b1 in Fig. 4 is also "O", and similarly, blocks a2 to a5 in Fig. 3 are also rOJ, so Fig. 4 is ~1.
) 5b "0", in the box a6 in Figure 3 [, 11
Since there are two "1"s, the bit pattern contents 0100 of block a6 in FIG. 3 are separately recorded as 3 in block b6 of FIG. −
Ru. The same process is continued until the end. 6.2 Finishing the primary compression for "lower character pattern" 6.2 Next, we will explain the secondary compression. This is the part that shows.

As shown in the second dot matrix created by first-order compression and divided by solid lines in Figure 4, a predetermined number of 1 to m dots (2 x 1 dots in this example) are formed in the vertical and horizontal directions. Create a block divided into . and then the first dot mark 1~ in Figure 3.
Same as the process I did on Rikus in the evening and 1, 2 x 1 dots)
If the two dots I in the pattern of ~ are all "0" then O
If there are even ``1'', make the third dora 1-71 helix as shown in Figure 6 with ``1''. 1. For the block whose bit is set to 1 here, the patterns of the block consisting of 2×1 dots 1 and 1 are sequentially stored as shown in FIG. This completes the secondary compression.

This completes the compression work, and if only the data shown in FIGS. 5, 6, and 7 are finally stored, the original character pattern can be reproduced.

However, for some characters, the number of bits required by secondary compression becomes larger than the number of bits required when compressed to 1 by only primary compression. Therefore, for primary compression 83
The amount of data for storing the second matrix as shown in FIG. 4 (in this embodiment, 144
The third matrix shown in the figure is the block 1 to 1 of the J3 matrix.
The data in Figure 7 consisting of the pattern (11 nodes I')
Compare 71 and
From the matrix in Figure 7, turn 11- again). - As a result, when the former is popular, the 71 heritas in Figures 6 and 7 is 144 bits J: is less than 1 in the row i.
The matrix obtained by 111 memory (-1 for character patterns up to secondary compression) (Fig. 5.6.7) is stored as character pattern data.

When the rear right side is large (sJ, there is no effect of secondary compression (
11 is reserved for "primary compression", and the 71-x in FIGS. 4 and 5 is stored as character pattern data.

This J is a sea urchin, and η all the letters are -51 and 2.
Next compression (i)
i < L/, J, it is possible to memorize characters with less data amount f1 < 7. Ki'fu jr'M I:I
・・□′!・In the case of the character "', 72 bits in Figure 3, 36 bits in Figure 7, 108 bits in #t,
It can be seen that it can be stored in memory with less than 144 bits, so the character pattern "lower" is quadratic compression j-
After going there, all you have to do is remember J゛.

Furthermore, according to the present invention, the total number of bits (212 bits 1~) required for the "lower" character pattern after performing secondary compression, and 576 bits 1~ when no compression is performed.
It is possible to memorize about 37% of the numbers compared to the previous one. For reference, the results of compressing other characters using the present invention are shown in the 8th page.
As shown in the figure.

When reproducing character patterns, if they have been subjected to secondary compression, figures 6 and 7 to figure 4 can be reproduced, figures 4 and 5 to figure 3 can be reproduced, and so on. From Figure 3, you can play back Figure 2 with the desired character pattern.1
In the case of only the following compression, it is possible to reproduce from FIG. 4 and FIG. 5 to FIG. 3, and from FIG. 13 to FIG.

Although the present invention has been described above, the present invention is not limited to this. For example, after extracting change points in both x and y directions, similar processing may be performed, or it may be further developed. Also, extract the change points for each character from the direction with the highest compression ratio (Is it higher to extract the change points only in the x direction, or is it higher to extract the change points only in the yh direction? ), and when extracting changing points while looking in the X direction,
It is possible to implement various methods such as comparing each character to see which dora has a higher compression ratio in both x and y directions (fj 11, r) and overlapping 31.3 billion in a method that can be memorized with a small number of pins. be.

As explained above, according to the present invention, the number of steps required to memorize character patterns can be greatly reduced, so the number of memories used can be reduced, and the device size can be reduced. It is possible to lower the price. J. It is a simple method that compresses character patterns using an algorithm, and has great effects such as a high reproduction speed of character patterns.

[Brief explanation of drawings]

Figure 1 4. L A flowchart showing an example of the processing procedure of the present invention,
Figure 2 shows an example of the character pattern. Figure 3 shows the changing points of black and white data. Figure 4 shows the second 71 helices. Figure 5 shows the divided blocks. Figure 6 is a diagram showing the data of the third 71~Risk, Figure 7 is a diagram showing the data of the divided lζ block, and Figure 8 is a diagram showing the data of the divided lζ block. It is a figure showing the result at this time. Patent Applicant Co., Ltd. Shaseki H Figure 2 Figure 3 Figure 4 Figure 1 Figure 5

Claims (2)

[Claims] (1) From the dot 71 helices that express character patterns using white data and black data, create a first matrix showing the points of change in white data and black data, and divide this first 71 helix in the vertical direction. In the horizontal direction, n blocks [
If there is the change point data in this divided block, create a second matrix with bits 1 to 1, otherwise bits are set to 0, and each block is also set to the bit minus 1. Primary compression is performed to create data consisting of the pattern forming the C1 block, and the second matrix is compressed in the vertical direction.
Divide horizontally into m blocks by a predetermined number of 1 to
This division 1. If there is the change point data in the nib block, create a third matrix with bits 1-1 and otherwise with bit O, and (lit!) A method for compressing character patterns, characterized in that character pattern data is compressed by performing secondary compression to create data consisting of bit patterns constituting each block. (2) The data of the second matrix created by the primary compression, the third matrix created by the secondary compression, and the third matrix constitute each block designated as pill-1. Compare the sum of the data consisting of P1-patterns, and perform secondary compression on character patterns where the former data is larger than the latter data, and perform only primary compression on other character patterns. [Claims of Claims @(j)] A method for compressing character pattern data according to J'i.