JPS5816189B2 - Character pattern generation method - Google Patents

Description

[Detailed description of the invention] The present invention relates to a character pattern storage method.

In particular, the present invention relates to an improvement in a storage method for storing character patterns as binary information.

Advances in digital and memory technology have led to the adoption of dot-style output in printers and display devices.

The white and black parts of characters and figures are represented by dots in correspondence with the binary value 0.1.

For example, characters are represented by 32x32 dots.

At this time, when a certain character is expressed in a 32x32 format, it is difficult to create smaller characters, for example, 16x16 characters, simply from 32x32 characters.

For example, in Figure 1, figure a is the letter H represented by 12 x 12, but in order to represent it in 6 x 6, even numbered rows and even columns are mechanically thinned out, resulting in figure b, where the horizontal bar of H is has disappeared and can no longer be read as H.

Because of this, mechanical thinning methods do not work well.

When enlarging, there is no such thing as disappearing, but to put it more precisely, the balance of the characters may be disrupted.

Because of the above, generally when storing character patterns, even if the characters are of the same type but have different sizes, they are often interpreted as different characters and stored as character patterns of multiple types.

When a character pattern with a relatively large number of dots is doubled or tripled, the imbalance of the characters is small, and in that case, it is sufficient to mechanically enlarge the single pattern and smooth it if necessary.

However, it is generally necessary to store character patterns as separate characters for each character size, which is undesirable because it increases the required amount of memory (font memory) for storing character patterns and causes an increase in cost.

The object of the present invention is to obviate the above-mentioned problems of the prior art.

That is, the object of the present invention is to provide a character pattern generation method that enables reduction and enlargement by adding a relatively small amount of additional information to a type of character pattern, thereby reducing the required amount of font memory.

The feature of this invention is that additional information instructing addition/deletion is given to a character pattern of a standard size, and smaller or larger characters are created based on this additional information and the basic character pattern. That's true.

Examples are shown below.

Figure 2 shows the overall configuration.

The basic pattern memory is a memory that stores a 32x32 dot character pattern.

The additional information memory stores insertion thinning information for enlarging/reducing the basic sentence pattern.

A certain character pattern is obtained by adding an address to the memory and applying the output to a scaler, and the address information obtained includes information indicating the type and size of the character.

The principle of reduction is illustrated in FIG. Here, for simplicity, 12XI2 is used as a base, as in FIG. 1.

Figure 3a shows the pattern of 12x12
At the same time, additional information is shown.

The additional information has one bit in each row and column, and rows (columns) with 1'' indicate rows (columns) to be thinned out.

Due to the reduction by 1/2, the number of 1's in both rows and columns is 6.

A character pattern reduced to 1/2 according to this thinning information is shown in b.

Here, there are no defects that occurred in FIG.

This is because people thought about and added thinning information and thinned it based on that information.

In the case of enlargement, for example, 1.5 times enlargement is achieved with C in almost the same way.

A feature of the present invention is that it has additional information based on this principle at the same time as the basic character pattern.

FIG. 4 shows the row line subcircuit.

In the example, one line has 32 bits, of which 4
Shows bits.

The operation will be described regarding the basic pattern ab cd and thinning information 0110.

The leftmost output bit is either abcd, but since a is not thinned out, it is the leftmost gate G.

opens. The second bit output is ``+1'' when G: is opened from thinning.

The third and fourth bits are not thinned out, so they are not output and are zero.

In this way, data 1-G from the thinning information. It is clear that it is possible to construct a logic circuit and to make it 32 bits since it is known what to do by opening ~G9.

It is also understood that by doing the same thing, it is also possible to enlarge characters by inserting information.

Next, reduction in the column direction will be described.

But before that, it is necessary to explain how dots are output, for example in a printing device.

In a printing device, dot scanning is performed horizontally.

Thus, first the first line of a character is scanned, then the first line of another character, and in this way, after the first line of a character has been scanned, the second line of the character is scanned, and then the third line of the character is scanned. It will be done.

For this purpose, the memory for storing character patterns is configured to output one line of dot patterns with each access.

FIG. 5 shows a memory map of a font memory that stores character patterns.

In this way, the 32×32 dots constituting the character A are stored over 32 addresses from address a in a 32-bit depth memory, and then the font for the character B is stored over 32 addresses from b.

From the print control circuit, an address indicating the letter A-; address (
In other words, a), which line is being printed, and when is the tW report (
/ ) to access this font memory.

On the other hand, thinning information d regarding the column is obtained from the additional information memory based on the information indicating the character type and the information indicating the size of the character.

How to access the font memory using the information a, l, and d obtained in this way will be explained.

In Figure 6, when d (indicates 4 bits) is given, l
A table is shown for determining the value of the output S when the values are 0, 1, 2, and 3.

In Haratankomajiko, the value obtained by adding S and a becomes the address of the font memory.

However, in the embodiment, d is 32 bits and S is 6 bits.

Expanding FIG. 6 into 32 bits would result in a huge table, but since a restriction is placed on the reduction ratio, a considerable portion can be omitted.

For example, if d is 32 and /=7, assuming the reduction ratio is 1/2 and the number of lines to be thinned out between /=0 and 6 is 3 or less, the value of S will be 4 to 7. , 0 to 3 can be omitted.

When printing the second line of a certain character as described above, calculate the actual number of character lines S using the thinning line information d and l, add this to the starting address a on the font memory of the character, and add The memory is accessed and the read data is thinned out as shown in FIG. 4 using thinning column information to obtain a desired one-row pattern.

As mentioned above, characters can be reduced or enlarged by using a basic character pattern of a certain size as a standard and having additional information for reduction or enlargement. It is possible to double as

For example, the same additional information can serve as both 3/4 reduction and 3/2 enlargement.

The idea is to reduce it by 3/4 and then double the reduced character to get 3A.

However, at this time, the information thinned out by 3/4 is restored.

This will produce 3/2 characters of better quality than just 3/4 times.

In addition, as another application, instead of thinning out rows and columns, additional information can be provided in units of 1/2 rows, for example.

Of course, if enlargement or reduction is not required, a character pattern of the same size is generated without using the additional information.

As described above, the present invention provides the following effects.

(1) Since it is sufficient to have a small amount of additional information without having separate character patterns for each character size, the amount of memory required can be greatly reduced.

(2) Since the additional information is used to reduce or enlarge the characters, the quality of the resulting characters is good.

(3) Since additional information is given in advance, reduction and enlargement of characters can be processed at high speed.

(4) Memory can be further reduced if it also serves as additional information for reduction and enlargement.

(5) Since only characters of the basic size can be allocated on the font memory, there is no boundary variation between characters depending on the size, and the allocation is simple.

(6) Additional information allows for a large degree of freedom in transformation.

[Brief explanation of drawings]

Fig. 1 is an example of reduction by the conventional method, Fig. 2 is a block diagram of the present invention, Fig. 3 is an illustration of reduction/enlargement according to the invention, Fig. 4 is an illustration of reduction in the row direction, and Fig. 5 is a diagram of the font memory. Memory mapping FIG. 6 is a reduced explanatory diagram in the column direction.

Claims (1)

[Claims] 1 A storage means for storing character patterns represented by a dot matrix of basic size, and a reduction means for specifying rows and columns to be thinned out or repeated for each character pattern, which are set in advance according to the character pattern. Small
A character pattern generation method comprising a storage means for storing enlargement information, reads a basic character pattern and reduction/enlargement information, and generates a character pattern in which the basic character pattern is reduced/enlarged according to the reduction/enlargement information. .