JPH0564394B2 - - Google Patents

Description

[Detailed description of the invention]

(Industrial Application Field) The present invention relates to a character font enlarging device. (Prior Art) When printing a book or the like, a so-called block copy is created in an intermediate process. When this block copy is created using a laser printer or the like, it becomes necessary to generate characters of various sizes from a standard size character font. That is, horizontal (column) n dots,
Any size m x from the character font of the base size digitized with vertical (line) n' dots
It is necessary to convert it to m' dot font. Conventional methods for performing such conversion include the methods described below, but all of them have specific drawbacks that pose problems in practice. (1) Enlargement/reduction using contour vector font A contour vector font defines the shape of a character by straight lines connecting the contours of a character pattern, as shown in FIG. During conversion, expansion or reduction can be achieved by uniformly multiplying the coordinates of the straight line by m. However, although this conversion method allows smooth enlargement or reduction, there is a problem in that it is necessary to fill in the inside of the contour, and such contour vector fonts are difficult to obtain. (2) Enlargement/reduction using dot fonts The original character pattern is represented by n×n′ dots, and when enlarging or reducing it, the desired m×m′ dot points are When converted to the coordinates of a ' dot, it becomes the position of an x mark as shown in Figure 11. For example, for the density of a point marked □ x, the coordinates of the four dots surrounding that point (original n x n' dots) are ) is a method of determining the concentration of However, in this method, the character font is binary data (white and black; here, grid points with dots are abbreviated as black dots, and grid points without dots are abbreviated as white dots), and the conversion is smooth in terms of density. There are weaknesses that I can't do. (3) Enlargement/reduction in cathode ray tube (CRT) computer typesetting machines In most CRT type computerized typesetting machines, the font is held digitally, but it is possible to change the gain of the horizontal and vertical amplifiers of the CRT in an analog manner. Therefore, the text is enlarged or reduced. However, character patterns of the same size are always output digitally, and although the subsequent analog part is enlarged or reduced when outputting to a CRT, characters of the same size are always output when outputting to a laser printer. There is a problem in that only the pattern is output, and no enlargement or reduction is performed. SUMMARY OF THE INVENTION An object of the present invention is to solve these problems and to provide a character font enlarging device that enlarges a character font using run-length data. (Means for Solving the Problems) In order to achieve such objects, the present invention provides: (a) dots constituting a character are arranged at a large number of grid points specified by a plurality of lines and columns; The basic pattern of M x M dots (M is an integer) is
Each line is scanned in the column direction, and the number of grid points with no continuous dots and the number of grid points with continuous dots are listed in order, and the beginning of the grid point of the dot row with continuous dots is (b) a first memory that stores run-length data of the basic pattern; (c) ) a second memory for storing run-length data of the enlarged pattern; (d) reading out the run-length data stored in this first memory and converting the dot rows of each line of the basic pattern into N×N dots ( N is an integer, N
> M) to obtain run-length data by projecting it onto the enlarged pattern, and storing it in a second memory, as well as controlling each circuit to perform a predetermined process; When line i of the enlarged pattern is projected onto the basic pattern read from the memory of j and j+
When there is a grid point with a dot on line 1, find the intersection between line i' and the straight line passing through the first dot of the dot row on lines j and j+1, and project that intersection onto the enlarged pattern. At the same time, find the intersection of line i' with a straight line passing through the last dot of the dot array on lines j and j+1, and project the intersection onto the enlarged pattern. (f) an interpolation processing circuit that determines the end of the dot row in line i of the enlarged pattern to obtain run-length data from the enlarged pattern; (f) adding the enlarged pattern to the basic pattern read from the first memory; When line i is projected, line i' is j<i'<j+1 (where j=0, 1,...M-1), and at the same time
1, there is no dot row on one of the lines, j, j-1 or j+1, j where there is a dot row
Find the intersection point between the line i' and the line that passes directly through the point of the first dot of the dot string on the two lines +2, project that intersection onto the enlarged pattern, define it as the beginning of the bit string on line i of the enlarged pattern, j, j-1 or j+1, j with columns
Find the intersection point between the line i' and the straight line that passes through the last dot of the dot row on the two lines +2, apply that intersection to the enlarged pattern, define it as the end of the dot row on line i of the enlarged pattern, and create the enlarged pattern. (g) a lattice point with a dot row of the basic pattern read from the first memory that protrudes to the left or right when compared with an adjacent line; (h) a sag prevention processing circuit that prevents the interpolation from being performed when the number of dots is equal to or greater than a specified length; When it is detected that the ends of the columns are aligned on the same column, the number of dots that protrude to the left or right of the column is specified by comparing the dot column of the adjacent line adjacent to these multiple lines with the column. If the number of dots is less than or equal to the number of dots, the end of the protruding dot row is the starting point, and the midpoint between the end of the dot row of the outermost line of the plurality of lines adjacent to the line of the protruding dot row and the starting point. Find a straight line that ends at , and a straight line that starts from this midpoint and ends at the end of the dot row of the center line among the multiple lines, and use the points on these straight lines as the ends of the dot row. Project the run length onto the enlarged pattern.
It is characterized by comprising a vertical line smoothing processing circuit that calculates data. (Example) The present invention will be described in detail below using specific examples.
FIG. 1 shows an example of an 8×8 dot font character pattern (hereinafter referred to as the basic pattern). In the present invention, this basic character font is converted into data called run length as shown in Table 1.

[Table] This is a lattice point with no bits (simply called white) on the 8x8 lattice point shown in Figure 1.
The number of grid points where dots are located (abbreviated simply as black) is expressed for each line.
It is a row of values that count how many whites and how many blacks there are in the horizontal direction. For example, line 7 represents eight consecutive whites, line 6 represents first four whites, then two blacks, and then two whites. Furthermore, line 5 indicates that two whites, five blacks, and one white are lined up. Now, when such a character font is enlarged (in this case, N>8) or reduced (in this case, N<8) to N×N dots, in the present invention, line i
Calculate the run length data for (i=0, 1, . . . , N-1) using the data in lines 0-7 of Table 1. The calculation method will be explained below. However, in order to make the explanation easier to understand, the explanation will be made in correspondence with character fonts and figures. (1) Let us take as an example the case where a basic character pattern of 8×8 dots is expanded to an N×N dot pattern.
Now, we are trying to find N×N as shown in Figure 2.
If line i of the dot pattern (herein referred to as an enlarged pattern) is projected onto the original 8.times.8 dot pattern, it will be located on line i'. If this relationship is expressed in a formula, it becomes i'=i.8-1/N-1...(1). Using this formula, all lines of the N.times.N dot pattern are projected onto an 8.times.8 dot pattern. In addition, the basic pattern is generally M×
Assuming that the pattern is made of M dots, the above equation (1) can be expressed as i'=i(M-1)/(N-1). (2) For projected line i′, j≦i′<j+1
(Here, j = 0, 1, ..., 7. However, Fig. 2 shows the case of j = 1.) When line j of the 8 x 8 dot pattern is the run length of line j + 1, Refer to the data and
If the i' line does not match the line of the 8x8 dot pattern, interpolation or extrapolation is performed to determine the starting coordinates of the black point on the i' line (the beginning of the dot row) and the ending coordinates of the black point (the end of the dot row). ,... and so on in order. Note that when line i' matches a line of an 8x8 dot pattern, the dot pattern of the basic pattern is converted to (N-) without interpolation or extrapolation.
The image is multiplied by 1)/(8-1) and projected onto the enlarged pattern to obtain run length data. (3) To improve the appearance, perform corrections to prevent sagging and smooth vertical lines. (4) For each line in the 8×8 dot pattern space obtained by the above procedure, the black point starting coordinates and the black point ending coordinate group are (N-1)/
(8-1) Obtain an enlarged character font by multiplying and moving it onto the N×N dot pattern space. Note that run-length data is stored in memory in the vector data format shown in Figure 3, so the starting coordinates (column) of the first sunspot are: X ₁ The ending coordinates (column) of the first sunspot are : X ₁ +X ₂ -1 The starting coordinates (column) of the second sunspot are obtained as: X ₁ +X ₂ +X ₃ , etc. Conversely, if the starting and ending coordinates of the sunspot are known, run-length data can be created from the above relationship. The basic character font can be enlarged in this way. Next, each method of interpolation, extrapolation, sagging prevention, and vertical line smoothing will be explained in detail. 1 Interpolation As shown in Figure 4, for example, a basic character font with an 8 x 8 dot pattern is converted into a 22 x 22 dot pattern.
shall be expanded into a pattern. Now pay attention to line i=16 of the enlarged pattern. line 16 to 8
Line when projected on ×8 dot pattern
i′ is calculated as i′=8−1/22−1×16=5.33. (1) How to find the starting coordinates of the black dot on line i (=16) Find the starting coordinates of the black dot in the 8×8 dot pattern from the run length data in Table 1.
That is, for the line i′=5, (5,
2), for the line i′=6 (6, 4)
Know. and point (5,2) and point (6,4)
The intersection point between the straight line l passing through the two points and the i'=5.33 line can be found from the following equation. x _n -2=4-2/6-5 (y _n -5) where x _n is the horizontal position in the 8x8 dot pattern, and y _n is the vertical position. From the above equation, on the line y _n = i′ = 5.33, x _n
=2.66. Next, apply this x _n to the enlarged pattern. Using equation (1), the starting point of the black dot on line 16 of the enlarged pattern, X _o (calculated as an integer by rounding off the decimal places), is x _o = 22-1/8-1 x _n ≒ 8. Desired. (2) How to find the coordinates of the end of the sunspot on line i (=16) Use the same method as above. That is, first, the last point (5, 2 + 5 - 1) and the point (6, 4 + 2 - 1) of the black points of the 8 x 8 dot pattern.
Knowing this, find the intersection of the line segment k connecting these two points and the i'=5.33 line. From the relationship x _n -6=5-6/6-5 (y _n -5), x _n =5.67 is obtained for y _n =5.33. Therefore, the coordinate x _o of the end of the black point on line 16 of the enlarged pattern is determined as x _o =22-1/8-1·X _n ≈17. (3) In this way, on line 16 of the enlarged pattern, only the area between x _o =8 and x _o =17 becomes a black point. 2. Extrapolation As shown in FIG. 5, black points are extrapolated for the projection lines above line 6 on the basic pattern. For example, when the i=19 line of the enlarged pattern is projected onto the basic pattern, it becomes i'=8-1/22-1×19=6.33. Line 6 for line 7 in the basic pattern
Since there is no data that can be connected to the starting point of the black dot and the ending point of the black dot, an error will occur if the above interpolation method is used. Therefore, in this case, extrapolation is performed from the data of lines 6 and 5 of the basic pattern. (1) How to find the starting point of the sunspot Using the same procedure as in the interpolation method above,
Straight line l passing through point (5,2) and point (6,4)
Find the intersection between and the i′=6.33 line. From x _n -2=4-2/6-5 (y _n -5), when y _n =6.33, x _n =4.66 is obtained. Starting point of black dot on line 19 of enlarged pattern x _o
obtains x _o ≒ 14 from x _o = 22-1/8-1 x _n . (2) How to find the coordinates of the end of the sunspot The following relational expression can be obtained from the run length data. x _n -6=5-6/6-5 (y _n -5) From the above equation, when y _n =6.33, x _n =4.67. The end point x _o of the black point on line 19 of the enlarged pattern is x _o =14 from x _o =22-1/8-1 x _n . (3) Line 19 of the enlarged pattern according to (1) and (2) above
Obtain that there is only one black point at x _o = 14 above. 3 Preventing sagging Preventing sag means preventing excessive smoothing and does not perform interpolation. For example, the portion indicated by the double connecting line in FIG. In other words, the outline of the point where the horizontal and vertical dot rows intersect in the dot pattern should maintain the same rectangular concave outline as before enlargement, even if it is enlarged.
This is a location where interpolation should not be applied to create a smooth curve. For example, in characters such as "year" and "black", there are many places where horizontal and vertical dot rows intersect, but the contours of these places should not be interpolated into smooth curves. Now, whether or not to perform interpolation is determined based on the following conditions. l ₁ , l ₂ , and l ₃ are the numbers of dots that protrude to the left or right from the edge of the adjacent dot row compared to the adjacent dot row. Interpolation is not performed when l ₁ , l ₂ , l ₃ ≧L. When l ₁ , l ₂ , l ₃ <L, interpolation is performed. Note that the value of L needs to be determined by the number of dots of the original character (basic pattern), the typeface, and the character itself. Note that FIG. 6 shows the case where L=3. 4 Vertical Line Smoothing Figure 7 shows the 8x8 dot pattern in Figure B expanded to a 22x22 dot pattern (Figure A) using the various correction methods described above. The area marked with a circle on the right side of the dot does not form a smooth curve. This is because interpolation is only performed using lines 1 and 2, lines 4 and 5 when M=8. Therefore, it can be seen that if interpolation is performed so that the end of the black point is at the position of the solid line L shown in FIG. 8B, a smoother curve can be obtained. Such interpolation, that is, vertical line smoothing, is performed by the following procedure. The conditions for performing vertical line smoothing are that there is a predetermined number or more of lines in which the ends of the dot rows are consecutively aligned on the same column, and that the line adjacent to the outermost line of the plurality of lines is This is a case where the dot row protrudes to the right or left of the line, and the number of protruding dots is less than or equal to a predetermined number. (1) Detect whether the ends of black points in adjacent lines are on the same column. In FIG. 7, for example, in lines 3 and 4, it is detected that the black points end in column 5 and are on the same column. In performing the following processing, line 3 is temporarily set as the starting point. (2) Going upwards from starting line 3, count the line where the black dot ends in column 5. In this case, there is only line 4, so the number of lines (N _L ) is 1. (3) Going downwards from starting line 3, count the line where the black dot ends in column 5. In this case, since there are line 3 and line 2, the number of lines (N _L ) is 2. (4) Therefore, the center of the line where the black dot ends in column 5 is calculated to be the line of (2+4)/2=3. The line L for correction is determined as follows. As shown in Figure 8, the midpoint of the end of the black dots between line 5 and line 4 is a (in this case, it is the point (4.5, 5.5)), and the midpoint of the end of the black dot between line 2 and line 1 is a. Let the point be b (in this case, it is the point (1.5, 5.5)). and,
The point at the end of the black dot on line 5...point a...line 3
The end point of the black dots... point b... The line obtained by connecting the end points of the black dots of line 1 with a straight line is the line L to be found. (5) Correct the end point of the black point for each line based on the above line L. For example, line
The ends of the black dots at 3.33 (corresponding to line 10 of the enlarged pattern) are point (3,5) and point (4.8,
Point c on the straight line connecting 5.5). Therefore, this point c is projected onto the enlarged pattern (this transformation is performed in the same manner as described in the section on interpolation or extrapolation). However, in this case,
Since the data is converted into integers, it becomes column 15 on the enlarged pattern as before correction. Vertical line smoothing is performed by such a method, and the dot pattern shown in FIG. 7A becomes a smooth curve as shown in FIG. 8A. Expansion is performed in this way. however,
In the examples, dots are used to simplify the explanation.
Although the explanation has been made using patterns, actual correction processing is performed by numerical calculations based on run-length data without using dot patterns. Furthermore, although the embodiment has been described in the case of enlargement, the same method can be used in the case of reduction. Furthermore, the magnification of enlargement or reduction is not limited to the embodiments, and can be arbitrarily selected. FIG. 9 shows an embodiment of an apparatus for carrying out the enlarging method based on the algorithm as described above.
In the figure, 1 is a conversion circuit that converts basic patterns such as characters and symbols into run-length data; 2
3 is an interpolation processing circuit that performs the interpolation processing; 4 is an extrapolation processing circuit that performs the extrapolation processing; Reference numeral 5 denotes a sag prevention processing circuit that performs the sag prevention; 6 a vertical line smoothing processing circuit that performs the vertical line smoothing; 7 a second memory in which run length data of the enlarged dot pattern is stored; 8 is a control that projects the dot rows of each line of the basic pattern onto the enlarged pattern using a magnification based on the ratio of the number of dots between the basic pattern and the enlarged pattern to obtain run length data, and also provides necessary control signals etc. to each part. It is a circuit. The operation in such a configuration will be explained next.
When the conversion circuit 1 is given a basic pattern to be enlarged, it converts it into run-length data as shown in Table 1. The data is stored in the first memory 2. The write signal and address at this time are given by the control circuit 8. Now, based on the information on the number of dots in the case of enlarging (this information is given in connection with the provision of the basic pattern), the control circuit 8 sets a magnification based on the ratio of the number of dots in the basic pattern and the enlarged pattern. The dot rows of each line of the basic pattern are projected onto the enlarged pattern to obtain run length data, and the data is transferred from the interpolation processing circuit 3 to the vertical line smoothing processing circuit 6.
The above correction processes are performed sequentially as appropriate. Each processing circuit performs the processing described in sections 1 to 4 above. Note that each circuit can appropriately refer to the run-length data stored in the first memory. The run-length data that has undergone these processes is stored in the second memory 7 and read out as needed. In addition, in the example, the basic pattern and
Although an embodiment has been described in which data is received one by one, converted into run-length data, and stored in the first memory, the present invention is not limited to this. It is also possible to convert this into a file and store it in the first memory. In this case, when enlarging, the run length data of the desired basic pattern is called and various processes are performed in the same way. As explained above, the present invention has the following effects. Enlargement conversion can be easily performed using run length data easily obtained from dot fonts, and data that can be directly output and printed on a laser printer or the like can be easily obtained. Processing such as interpolation, extrapolation, sag prevention processing circuit, and vertical line smoothing performed during conversion is also run-length
This can be easily done by calculation based on data. Since the target font is handled as run-length data, the amount of data can be compressed compared to the conventional case where it is handled as a dot pattern or the like. Note that this invention uses MH (Modified Huffman), which is an international standard in facsimile.
A run-length method can be used. Because run-length data can be processed, there is no need to determine whether data expanded into dots is black or white, making high-speed processing possible. There is no need for a memory area developed into dots, and the device can be made smaller and less expensive.

[Brief explanation of the drawing]

Figure 1 shows an example of a character pattern in 8x8 dot font, Figure 2 shows the correspondence between basic character patterns and enlarged patterns, and Figure 3 shows the format of run-length data. , FIG. 4 to FIG. 8 are diagrams for explaining each process, FIG. 9 is a block diagram of main parts showing an embodiment of an apparatus for carrying out the method of the present invention, and FIG. 10 is a diagram for explaining the contour vector FIG. 11 is a diagram for explaining a dot font. 1... Conversion circuit, 2... First memory, 3...
Interpolation processing circuit, 4... Extrapolation processing circuit, 5... Sag prevention processing circuit, 6... Vertical line smoothing processing circuit, 7...
...Second memory, 8...Control circuit.

Claims (1)

[Scope of Claims] 1 (a) A basic pattern of M×M dots (M is an integer) in which dots constituting a character are arranged at a large number of grid points specified by a plurality of lines and columns, The number of grid points with no consecutive dots and the number of grid points with consecutive dots are listed in order, and the beginning and end of the grid points in the row of dots with consecutive dots are scanned in the column direction for each column. (b) a first memory for storing the run-length data of the basic pattern; (c) an expansion circuit; a second memory for storing run-length data of the pattern; (d) reading out the run-length data stored in this first memory and dividing the dot rows of each line of the basic pattern into N×N dots (N is An integer, N
> M) to obtain run-length data by projecting it onto the enlarged pattern, and storing it in a second memory, as well as controlling each circuit to perform a predetermined process; When line i of the enlarged pattern is projected onto the basic pattern read from the memory of j and j+
When there is a grid point with a dot on line 1, find the intersection between line i' and the straight line passing through the first dot of the dot row on lines j and j+1, and project that intersection onto the enlarged pattern. At the same time, find the intersection of line i' with a straight line passing through the last dot of the dot array on lines j and j+1, and project the intersection onto the enlarged pattern. (f) an interpolation processing circuit configured to determine run length data of the enlarged pattern by determining the end of the dot row in line i of the enlarged pattern; (f) adding the enlarged pattern to the basic pattern read from the first memory; When line i is projected, line i' is j<i'<j+1 (where j=0, 1,...M-1), and at the same time
1, there is no dot row on one of the lines, j, j-1 or j+1, j where there is a dot row
Find the intersection point of line i' with a straight line passing through the point of the first dot of the dot row on the two lines of j, j-1 or j+1, j with columns
Find the intersection point between the line i' and the straight line that passes through the last dot of the dot row on the two lines +2, apply that intersection to the enlarged pattern, define it as the end of the dot row on line i of the enlarged pattern, and create the enlarged pattern. (g) a lattice point with a dot row of the basic pattern read from the first memory that protrudes to the left or right when compared with an adjacent line; (h) a sag prevention processing circuit that prevents the interpolation from being performed when the number of dots is equal to or greater than a specified length; When it is detected that the ends of the columns are aligned on the same column, the number of dots that protrude to the left or right of the column is specified by comparing the dot column of the adjacent line adjacent to these multiple lines with the column. If the number of dots is less than or equal to the number of dots, the end of the protruding dot row is the starting point, and the midpoint between the end of the dot row of the outermost line of the plurality of lines adjacent to the line of the protruding dot row and the starting point. Find a straight line that ends at , and a straight line that starts from this midpoint and ends at the end of the dot row of the center line among the multiple lines, and use the points on these straight lines as the ends of the dot row. A character font enlarging device equipped with a vertical line smoothing circuit that projects onto an enlarged pattern to obtain run length data.