JP2000039881A - Character font storing device, character font output device and record medium for its program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To greatly change the basic contour and thickness of characters while increase in data quantity is suppressed and to enable expression of all sorts to be added by storing frequency font having a data structure in which character-forming contours are expressed by the aggregate of discrete spatial frequency components. SOLUTION: A CPU 1 transforms, to a specific frequency font, the character font which is read out from an ordinary font storing part 3-1 in a storage device 3 and the bit map which is read by an image reader in an inputting device 5. In other words, they are transformed to the frequency font having the data structure in which the contour of the existing character font (bit map font and outline font) is expressed by the aggregate of discrete spatial frequency components, in accordance with a development method by frequency analysis, i.e., the fast Fourier transformation speeding up the discrete Fourier transformation. After this frequency font is stored in the frequency font storing part 2-3 of a RAM 2, it is registered and stored in the storage device 3.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a character font storage device, a character font output device, and a program recording medium for defining a character outline.

[0002]

2. Description of the Related Art Conventionally, in a document data processing apparatus such as a word processor or a personal computer, when outputting input document data, a bitmap font expressing a character shape by a set of dots or a character outline is partially used. Each time, an outline (vector) font, which is expressed as a set of reference points and a straight line or Bezier curve connecting them, is read out and output / printed out, but the outline font is rotated and italicized compared to the bitmap font. There is an advantage that deformation such as enlargement / reduction is easy, and the contour is smooth after the deformation.

[0003]

However, even outline fonts are not suitable for changing the thickness of one stroke constituting a character or for changing the basic outline of a font greatly. If required, individual fonts had to be prepared in advance as special fonts. In such a case, there is a limit even if the data is compressed and stored structurally by making the font into parts,
The data capacity was large. Furthermore, an outline font generally has a small data capacity when the contour is smooth, but the data capacity explosively increases in a shape that changes (vibrates) finely. Was not prepared in advance. An object of the present invention is to express the outline of a character by a set of discrete spatial frequency components, thereby significantly changing the basic outline and thickness of the character while suppressing an increase in the amount of data, and performing various expressions. Is to be added.

[0004]

The means of the present invention are as follows. The invention according to claim 1 is a character font storage device used for a character font output device that draws and outputs a character font, and stores a character font to be output. The character font storage device is characterized by storing a frequency font having a data structure represented by a set of character fonts. Here, the characters are not limited to kana characters, numbers, kanji, alphabets, and the like, but have a broad meaning including symbols, figures, illustrations, and the like, and do not limit the objects. The present invention may be as follows. That is, (1) the frequency font may be one in which the outline of a character is represented by a set of discrete spatial frequency components in accordance with a discrete Fourier series expansion method. (2) The aggregate of the spatial frequency components may be numerical information indicating the magnitude of a real axis part and an imaginary axis part on a complex plane for each frequency component. (3) The aggregate of the spatial frequency components may be numerical information representing the value as an amplitude and phase component for each frequency component. (4) The aggregate of the spatial frequency components may be generated by sampling along the contour of the bitmap data and performing discrete frequency decomposition. (5) The collection of the spatial frequency components may be generated by sampling along a contour of a character developed based on an outline font and performing discrete frequency decomposition. (6) At the time of sampling while tracing the outline of a character, sampling may be performed for each stroke in units of one stroke of an element constituting the character. (7) The unit for performing sampling by following the outline of the character may be one element constituting the character, or may be for each line segment of a continuous stroke portion based on a straight line. (8) The sampling start position following the outline of the character is the position where the writing of the character starts or ends.
The sampling operation may be performed by following the outline of the character from that position. (9) The aggregate of the spatial frequency components may include only low-frequency components that govern the general shape of the character, and the amount of font data may be reduced by deleting all high-frequency components. (10) The aggregate of the spatial frequency components is composed of a combination of a low-frequency component that governs the approximate shape of the character and a specific high-frequency component, and is added to the contour by adding a specific high-frequency component. A fluctuation element may be provided. According to the first aspect of the present invention, since the outline constituting the character is represented by a set of discrete spatial frequency components, the basic outline and thickness of the character are significantly deformed, and various expression powers are added. For example, the character shape can be freely processed.

According to a twelfth aspect of the present invention, there is provided a font storing means for storing a frequency font having a data structure in which a contour constituting a character is represented by a set of discrete spatial frequency components, and the font storing means. And a conversion means for reading out the frequency font from the inside and converting this frequency font into coordinate data, and an output means for developing and outputting a character font based on the coordinate data converted by this conversion means. It is to be noted that a font generating means for reading out the existing ordinary character font, sampling the contour along the contour line, and generating the frequency font by discrete frequency decomposition is provided, and the frequency font generated by the font generating means is provided. May be written in the font storage means. According to the twelfth aspect of the present invention, a frequency font in which the outline of a character is represented by a set of discrete spatial frequency components is read out, and the frequency font is converted into a coordinate sequence by performing an inverse transform. Since the character font is output, the basic shape and thickness of the character can be significantly changed, and various expressions can be added. Therefore,
By expressing the outline of a character with a set of discrete spatial frequency components, it is possible to significantly change the basic outline and thickness of the character and add various expressions while suppressing an increase in the amount of data .

[0006]

DESCRIPTION OF THE PREFERRED EMBODIMENTS (First Embodiment) A first embodiment of the present invention will be described below with reference to FIGS. FIG. 1A is a block diagram showing the entire configuration of the document data processing device. The CPU 1 is a central processing unit that controls the overall operation of the document data processing device according to various programs loaded in the RAM 2. The storage device 3 includes a storage medium 4 in which an operating system, various application programs, data files, character font data, and the like are stored in advance, and a drive system thereof. The storage medium 4 is fixedly provided or removably mountable, and is constituted by a magnetic / optical storage medium such as a floppy disk, hard disk, optical disk, RAM card, or the like, and a semiconductor memory. The programs and data in the storage medium 4 are
The data is loaded into the RAM 2 under the control of the CPU 1 as needed. Further, the CPU 1 receives programs and data transmitted from other devices via a communication line or the like and stores them in the storage medium 4 or stored in a storage medium provided in other devices. Existing programs and data can be used via a communication line or the like. An input device 5, a display device 6, and a printing device 7, which are input / output peripheral devices, are connected to the CPU 1 via a bus line.
The CPU 1 controls these operations according to an input / output program.

The input device 5 includes a keyboard, a pointing device such as a mouse, and an image reader for inputting character string data and the like and various commands.
Here, when document data is input from the input device 5 at the time of document creation, the document data is displayed on the text screen of the display device 6 and the determined character string determined by the kana-kanji conversion is stored in the RAM 2. Note that the display device 6 is a liquid crystal display device, a CRT display device, a plasma display device, or the like that performs multicolor display, and the printing device 7 is a full-color printer device, which is a non-impact printer or an impact printer such as thermal transfer or inkjet. Output document data in color. As described above, the image reader in the input device 5 reads handwritten characters and printed characters as bitmap images.

In the document data processing device having such a configuration, a normal character font (bitmap font / outline font) resident in the system or externally supplied or a bitmap read by an image reader in the input device 5 is used. The image is converted to a frequency font specific to this embodiment based on the image. That is, in this embodiment, the outline of an existing character font (bitmap font or outline font) is developed in accordance with a development method based on frequency decomposition, that is, a development method of fast Fourier transform (FFT) in which discrete Fourier transform (DFT) is accelerated. Is converted into a frequency font having a data structure represented by a set of discrete spatial frequency components, and this frequency font is stored in the RAM 2 and then registered and stored in the storage device 3. In this embodiment, an existing normal font is subjected to FFT processing and converted into a frequency font. However, as a character font generator, a frequency font memory for storing a frequency font of the character together with a character code in advance in association with various characters is provided. It may be resident or externally supplied. A frequency font is a data structure expressed in accordance with a discrete Fourier transform expansion method. Each part constituting a character can be represented by a continuous function having a cycle of one cycle of the contour (closed curve). A frequency that vibrates once during one round is defined as a fundamental frequency of the FFT, and a frequency font is expressed by an aggregate of a combination of the fundamental frequency and a frequency that is an integral multiple of the fundamental frequency. Note that the maximum order of the conversion result obtained by the FFT processing corresponds to the sampling theorem.

FIG. 1B shows a main configuration of the RAM 2, and various memory areas are allocated to the RAM 2. The character string storage unit 2-1 temporarily stores a character string to be subjected to the FFT processing. 3-1 is retrieved, and the corresponding character font is fetched. Normal font storage 3
-1 stores a bitmap font or an outline font in association with various characters, and the CPU 1 performs FFT processing on the character font read from the normal font storage unit 3-1 for each part. The work part 2-2 is used. Frequency conversion work part 2-2
Is a work area for temporarily storing the conversion result for each part,
The contents are stored in the frequency font storage section 2-3.
The frequency font storage unit 2-3 stores the frequency font generated by the FFT processing for each character.
U1 generates a deformed font in which the outline of the character is deformed by processing data constituting the frequency font,
Write to the modified font storage section 2-4. The deformed font storage unit 2-4 stores a variety of deformed fonts, such as changing the thickness of a character outline or basic character shape, giving a fine vibration to the outline, or giving a fluctuation as if handwritten by a person. The CPU 1 generates the original coordinate data also by performing the inverse Fourier transform on the deformed font, and then develops the bitmap data in the print buffer 2-5 based on the coordinate data.

FIG. 2 is a diagram for explaining the data structure of the frequency font. The contents of the frequency font storage section 2-3 are as shown in FIG. In the case of a normal outline font, the number of parts is defined and attributes such as a reference point coordinate value, a straight line, and a Bezier curve are defined for each part (each image). In the case of the frequency font in the embodiment, the coordinates of the outline font and the description part of the attribute are a table of the number of spatial frequency components. That is, the frequency font storage unit 2-3
Is configured to store a numerical table describing a DC component, a fundamental frequency component, and a frequency component obtained by multiplying the DC component, a fundamental frequency component, as an FFT processing result for each part in association with the character code, the number of parts, and the part number. Here, the processing result by the FFT is expressed as a complex number, and its size is discretely obtained on a complex plane and a real axis part and an imaginary axis part. In the figure, “Xrl” is a complex plane Above, the value of the real axis portion in the X-axis direction is shown, and “Xim” indicates the value of the imaginary axis portion. Similarly, “Yrl”, “Yi
“m” is the value of the real axis part and the value of the imaginary axis part in the Y-axis direction. In the figure, “F0” is a DC component, and the center part coordinates of the contour are described in the real part thereof. Note that a value of “zero” is always described in the imaginary axis part of “F0”. "F1" is 1 when making one round of the contour of the part.
This is a fundamental frequency component representing the frequency of vibrating, and the size and direction of the part are governed by this value. "F
"2" is a frequency component (secondary high-frequency component) that vibrates twice when making one round of the contour of the part, and the shape of the part is governed by this value, and this value increases in a part that is greatly bent. Subsequent "F3", "F4" ... are tertiary,
It is a fourth-order high-frequency component, and its maximum order is half the number of coordinates when sampling the contour of the part to be converted. This is because the value of the second half can be easily derived only by performing the inversion of the positive / negative and the reverse order from the value of the first half.

FIG. 3 is a diagram conceptually showing a process of creating a frequency font. One part of the font to be processed (in this case, an outline font) is drawn, and the outline of this bitmap is plotted. Here, the number of points for plotting one part is determined depending on how finely the conversion is desired. In the case of the discrete Fourier transform, “2” is used.
, For example, "512", and the contour of the part is plotted with this number, and the points P0, P1, P2
XY coordinate values are fetched every time, and the coordinate values are subjected to FFT processing. As a result, a value obtained by converting the plot coordinate value into a frequency distribution is obtained, and this value is stored in the frequency font storage unit 2-3 as font data. It should be noted that the transform result by FFT is reversible unless the calculation accuracy is strictly determined, and can be returned to the original coordinate value by Fourier transform. On the other hand, in the present embodiment, as shown in FIG.
3, but the data structure of the frequency font may be rearranged into the amplitude and the position as shown in FIG. Since this is an angle formed on the complex plane with the distance from the origin, X
In the axial direction, X amplitude = √ (Xrl ² + Xim ² ) and X phase angle tan ⁻¹ (Xim / Xrl), and the same applies in the Y-axis direction.

Next, the operation of the document data processing apparatus will be described with reference to FIGS.
This will be described with reference to the flowchart shown in FIG. Here, programs for implementing the functions described in these flowcharts are stored in the storage medium 4 in the form of program codes readable by the CPU 1, and the contents thereof are loaded into the RAM 2. . The same applies to flowcharts in other embodiments described later. FIG. 5 is a flowchart showing the operation when a frequency font for one character is created. First, the character font to be processed (outline font)
Is acquired (step A1), this is stored in the frequency font storage unit 2-3 (step A2), and the first part is drawn on a bit map in the RAM 2 (step A3). Then, the contours of the drawn parts are plotted (steps A4 to A7). FIG.
Is a diagram of the operation in this case. In the figure, a set of rectangles is each drawn dot, and one rectangle corresponds to one dot. Here, scanning is performed from the upper left to the lower left with the upper left in the drawing as the origin, and when reaching the bottom, the right dot row is scanned from the upper to the lower, and the above scanning is performed until a drawn dot is found. The operation is repeated (see FIG. 6A). In addition, such a scanning direction is meaningful, and in the case of Japanese, especially a font close to handwriting, the lines that make up the character must be written from the left side, whether it is a horizontal bar, flip up, swing down This is because, by adjusting the sampling start position to the start of writing of characters, font deformation described later becomes convenient. Note that if scanning in the left-right direction (horizontal scanning) is performed in contrast to scanning in the up-down direction (vertical scanning) as described above, the start of writing of a character often does not start sampling. In this embodiment, a vertical scan is performed. In addition, since the writing start position is synonymous with the writing end position, the sampling start position may be set as the character writing end position, and sampling is performed by tracing the character outline clockwise from this position. Perform the operation.

By repeating such a scanning operation, in step A4, the first dot is searched for, the starting point is used as the starting point, the contour is traced, and the length of the periphery is counted (step A5). In this case, as shown in FIG. 6 (C), the periphery of the currently focused dot is examined, and if a dot exists in any one of the four directions, the point of interest is moved in the direction of the dot to move the point on the contour line. However, when there are a plurality of dots in the periphery, FIG.
The order shown in (B) (the order of numerals 1, 2, 3, and 4) is to be followed. Since there is no dot in the direction of "1" around the dot start point, "2"
The direction of interest is examined. Since there is a dot in that direction, the point of interest is moved rightward in the figure from the start point (FIG. 6).
(C)). Then, from the next dot, the surroundings are determined based on the direction turned 90 degrees to the left from the current direction (“1” direction if the direction is “2”, “4” direction if the direction is “1”). Then, by moving the point of interest in the same manner, it is possible to make one round while turning clockwise on the contour line. In this process, the length of the contour can be obtained by counting how many dots exist on the trajectory up to one round. The number of sampling steps is determined based on the count value (the length of the contour for one part) thus obtained (step A6). In this case, an appropriate sampling interval is determined by comparing the sampling value arbitrarily set in advance with the count value (length). FIG. 6C shows that every four dots are sampled for the entire length (32 dots), and eight samples are taken in total. Here, in the figure, a cross-hatched portion indicates a sample target, which is a 4-dot interval from a dot start point. Then, each coordinate value of the sample target point is sampled following the contour line (step A7).

The coordinate values of each point sampled in this way are subjected to FFT processing (steps A8 to A15). In this case, since the FFT calculation is performed separately for the X coordinate component and the Y coordinate component, the X component of the sampled coordinates is first stored in the frequency conversion work unit 2-2 (step A).
8). Here, as shown in FIG.
-2 is the Real part, the Imag part is F0, F1,
Since the sampled X coordinate is real number data, the X coordinate value is stored in the Real part in the frequency conversion work unit 2-2 (step A8). Im in the conversion work part 2-2
"0" is substituted for all the ag parts (step A9).
Then, based on the contents of the frequency conversion work unit 2-2, the same FFT calculation as usual is performed, and all the conversion results are set as F0, Fmag as Real part and Imag part values.
1, F2... Temporarily stored in the frequency conversion work unit 2-2 (step A10), and the contents of the frequency conversion work unit 2-2 are stored in the frequency font storage unit 2-3.
Is stored as the value of the X axis real number axis part and the value of the imaginary axis part (step A11). After the conversion and storage of the X coordinate component is completed, the same conversion and storage is performed for the Y coordinate component. That is, the sampled Y coordinate is stored in the Real part of the frequency conversion work unit 2-2 (step A12), and the frequency conversion work unit 2-2 is stored.
Substitute "0" for the Imag part (step A
13). Then, an FFT calculation is performed based on the contents of the frequency conversion work section 2-2, and the conversion result is stored in the frequency conversion work section 2-2 (step A1).
4) The contents of the frequency conversion work unit 2-2 are stored in the frequency font storage unit 2-3 in the value Yr of the real axis portion in the Y-axis direction.
1, stored as a value Yim of the imaginary axis part (step A
15).

When the font conversion processing for one part is completed, the process proceeds to step A16 to check whether all parts for one character have been processed.
Returning to 3, the next part is drawn, and the above operation is repeated for each part. FIG. 7 shows a state in which the character "wa" is decomposed for each part and subjected to the FFT processing. The vertically long parts forming the "nogi bias" and the large bends forming the "shi" are shown. A table is shown in which the values of the real part and the values of the imaginary part obtained by performing the FFT processing for each part are discretely distributed on the frequency axis. In this frequency distribution, the characteristics of the vertically long parts are exhibited at the fundamental frequency F1, and the characteristics of the parts which are greatly bent are exhibited at the next frequency F2 in addition to the fundamental frequency F1. FIG. 8 shows specific numerical values of a real axis portion, an imaginary axis portion, a real axis portion, and an imaginary axis portion in the X-axis direction obtained by performing the FFT processing on the character "sum". FIG. 8 is a diagram showing a correspondence relationship between each part constituting the “sum” and frequency data of each part. Here, as shown by "1" to "7" in the figure, since this character "Sum" is composed of seven parts, a set of the corresponding spatial frequency components is stored in the frequency font storage unit 2-3 for seven parts. Is stored in As described above, the imaginary number of F0 is always “zero”, and FIG.
Although the values up to F9 have been exemplified, the value actually continues up to the maximum order specified by the sampled number.

On the other hand, in this embodiment, a character is disassembled as a part for each stroke as described above. However, in the case of a character such as the character "Otsu" which is long and largely bent, for example, Decomposing the entire character into three line segments close to a straight line is better for later-described deformation processing (particularly, thickness control).
Becomes reliable and easy. In other words, when the original font (bitmap font or outline font) is decomposed for each line segment of a continuous stroke portion based on a straight line, this character is sampled for each part and subjected to FFT processing. By controlling the thickness of each part, even a character such as "Otsu" can be reproduced with its thickness changed without changing the overall shape.

Thus, the frequency font storage unit 2-
In a state where frequency fonts corresponding to various characters are stored in the character string 3 and character printing is performed using this frequency font, the procedure shown in FIG. FIG.
0 is a flowchart showing the operation at the time of character printing in this case. That is, the frequency font storage unit 2-3 is searched based on a character arbitrarily designated as an output target,
The corresponding frequency font is read out and stored in the work memory in the RAM 2 (step B1). And
The frequency components are read out from the work memory for each part and subjected to Fourier inverse transform (IFFT) to return to the coordinates at the time of plotting (step B2). This is drawn in the print buffer 2-5 as a short vector (step B3). That is, an outline font having the coordinates generated by the IFFT as a short vector is generated, and is expanded in the print buffer 2-5. here,
The plot interval when outputting characters is not related to the number of plots when generating a frequency font, and can be arbitrarily controlled. That is, at the time of FFT conversion, the number of data (order) is related to the number of plots.
Is performed, the plot interval becomes narrow, and conversely, if the number of data (order) is reduced, the plot interval can be widened.

On the other hand, the outline of the character can be changed by processing the data constituting the frequency font. In this case, the original coordinate data may be generated by performing FFT processing on a deformed font for deforming the outline of the character based on the frequency font, and the character font may be developed and output based on the coordinate data. Here, when changing the size of the character, it is possible to enlarge or reduce the size of the character by multiplying all the numerical values by a coefficient, as in a normal outline font, and the size of the character can be freely changed. . Furthermore, if high frequency components are cut by a digital filter, round characters can be output,
Furthermore, if low frequency and specific high frequency are combined,
A fine change (vibration) can be added to the outline of the character, and when controlling the thickness of the outline, F1 and F2
Various modifications are possible, for example, by changing the predetermined value of the low frequency component (constant times).

As described above, since the character font is represented by a set of discrete spatial frequency components in the outline constituting the character, various deformations can be easily performed, and the deformed character is pasted on the background image. If attached, an effective guide letter or the like can be created. In particular, even if complicated outline shapes that were practically impossible to implement with conventional outline fonts were not implemented as fonts, just by processing the data to draw frequency fonts, Can also be easily deformed. Therefore, the data amount can be significantly reduced as compared with a case where a complicated shape is implemented as an outline font. In addition, the frequency font is divided into a real axis part and an imaginary axis part on a complex plane for each frequency component and represents the size, so that it is suitable for high-speed processing. Also, if the value is represented as an amplitude and a phase component for each frequency component,
This makes it easy to conceptually understand the low frequency range, and is effective when arbitrarily specifying a deformation. In addition, since the frequency font is generated by sampling and FFT processing following the outline of the normal font, the frequency font does not need to be mounted in advance, and the frequency font generation function can be simply incorporated into a word processor or the like. I just need. The unit of sampling is 1 that constitutes a character.
Since each image is converted to a set of frequency components for each image, if the font has only low-frequency components that govern the general shape of the character, the deformation is straightforward and the basic character shape is preserved. Dripping. In addition, even if the element that constitutes a character is one stroke, such as “Otsu”, the basic character shape can be maintained by sampling it by dividing it into three for each continuous stroke based on a straight line. By setting the sampling start position at the start or end of writing of a character, when performing a deformation that changes the thickness or basic character shape of the character, the deformation can be easily controlled due to the characteristics of Japanese.

(Second Embodiment) A second embodiment of the present invention will be described below with reference to FIGS. In the above-described first embodiment, all the spatial frequency components obtained by the FFT processing are stored up to the maximum order specified by the sampling number. In the second embodiment, the spatial frequency components obtained by the FFT processing are obtained. Only the low frequency components of the frequency components are stored, so that a rounded character suitable for a play application can be output. FIG. 11A shows F3.
The data structure of the frequency font when all the subsequent frequency components are set to zero is shown.
It consists of only data of 0, F1, and F2. FIG. 11B shows an output shape when a frequency font having such a data structure is printed out. In the drawing, the output shape of the frequency font of the first embodiment is shown on the left side, and the output shape of the frequency font of the second embodiment is shown on the right side. Compared with the output example of the font having the data of the maximum order, the second embodiment has a round shape as if the character were drawn with a paintbrush. When a character having such a shape is represented by a normal outline font, assuming that one outline is represented by four Bezier curves, 13 coordinate points (X, Y) and at least five coordinate point identifiers are used. 26 bytes are required even if 13 coordinate points are represented by 8 bits. On the other hand, if all data after F3 is set to zero as in the second embodiment, F0
Since the imaginary number data of is zero, the outline of one part can be represented by ten pieces of data, and even if it is represented by 16 bits, 20 bytes is sufficient.

FIG. 12A shows a frequency font having a data structure in which all data after F2 are zero. Since the frequency font is composed of only F0 and F1, it can be represented by six data per contour, and even if it is represented by 16 bits. 12 bytes is enough.
FIG. 12B shows the lowest frequency component (fundamental frequency component F
This figure shows an output shape when the data capacity is greatly reduced while only 1) is shown. Although the character shape is considerably distorted, a rounded character suitable for use in play can be obtained. FIG. 13 shows FIG.
2 is a flowchart showing an operation when generating and registering a frequency font having the data structure shown in FIG.
This is basically the same as the flowchart of FIG. 5 in the first embodiment. However, in the second embodiment, it is necessary to store the contents of the frequency conversion work unit 2-2 in the frequency font storage unit 2-3. Only data is extracted and stored. That is, after performing the process of storing the number of parts constituting the character to be processed in the frequency font storage unit 2-3 (steps C1 and C2), one part is drawn, and sampling is performed based on the length of the outline. The number of steps is determined (steps C3 to C6), and each coordinate value of the sample target point is sampled following the contour (step C7). Then, the X component of the sampled coordinate values is substituted into the frequency transform work unit 2-2 to perform the FFT calculation (steps C8 to C10). In the conversion result obtained as described above, except for the high frequency component, the DC component F0
, And Real and Imag data of the fundamental frequency components are extracted from the frequency conversion work unit 2-2 and stored in the frequency font storage unit 2-3 (step C11). Next, the same processing is performed for the Y component, and the FFT calculation is performed by substituting the Y component into the frequency conversion work unit 2-2 (steps C12 to C14). And Real of F0 and F
1 Real and Imag only in frequency font storage 2
-3 (step C15). Step C is repeated to repeat such an operation for each part constituting one character.
At 16 it is determined whether all parts have been completed. If not, the process returns to step C3. As a result, as shown in FIG. 12A, a set of spatial frequency components constituting the frequency font is
Since the data structure includes only low-frequency components, the data capacity can be significantly reduced.

(Third Embodiment) Hereinafter, a third embodiment of the present invention will be described with reference to FIGS. In the second embodiment described above, in order to reduce the data capacity, the aggregate of the spatial frequency components has a data structure consisting of only the low frequency components. However, in the third embodiment, the specific high frequency components, For example, by adding F56 and setting the value to a constant value, it is possible to deform the outline by vibrating finely. FIG. 14B is a font output example showing the deformed state. If a character having such a shape is to be realized by a normal outline font, the data capacity becomes too large to be practical. In the embodiment, F56 is a fixed value (0.7
81), the modified character as described above is realized only by adding data. FIG. 14A shows the data structure in this case, and a set of spatial frequency components for each part is composed of F0, F1, F2, and F56.
Is a fixed value, and the outline of a character having the shape specified by F0, F1, and F2 is vibrated by the high-frequency component of F56. Can be obtained by increasing its value.
In order to obtain fine vibration, the order of the high frequency component may be increased. FIG. 14B schematically shows the vibration state of the character outline. If such characters are to be realized with a normal outline font, the number of characters increases from 26 bytes to nearly 800 bytes. In contrast, in the frequency font of the third embodiment, F56 is used for each part.
Can be realized only by adding a constant value (0.781), so that only 56 bytes are required, and the capacity of the entire font is greatly reduced.

By the way, in the modification shown in FIG. 14B, the cycle of vibration differs for each part. In this case, it is good if the appearance cycle is almost the same or the appearance cycle may be different. However, in order to make the oscillation cycle constant for each part, as shown in FIG. A set of frequency components for each is a high frequency component Fn determined according to the length of the part in addition to F0 and F1.
12 in the second embodiment.
The configuration is such that a high frequency component Fn is added to the data structure of FIG. An n column for determining the order n of the high frequency component to be added for each part is provided. This order n is determined according to the length of the contour of the part.
By setting the value of the high frequency component indicated by n to a constant value, the vibration cycle is made constant for each part.

FIG. 16 is a flowchart showing the font generation processing in this case. Here, steps D1 to D
Reference numeral 15 is the same as steps C1 to C15 in FIG. 13 in the above-described second embodiment, and stores only the components F0 and F1 in the frequency font storage unit 2-3. When the processing for one part is completed in this way, the length of the contour line obtained in step D5 is recorded and held in the work memory (step D16). Then, the process returns to step D3 until the end of processing for all parts is detected in step D17, and the above operation is repeated, so that a frequency font consisting of only F0 and F1 is generated, and the outline of The length is recorded. Then, by comparing the length of the outline of each part, the maximum value is extracted (step D18), and the length of the outline of the first part is extracted (step D19). Then, calculation for obtaining the order n of the high frequency component Fn to be added is performed (step D2).
0). That is, the ratio of the length of the corresponding part to the maximum value is determined based on the length of the contour of the relevant part extracted in step D19 and the maximum value of the length extracted in step D18, and the value is calculated as the fundamental frequency component F1. To obtain the order n of the high frequency component to be added. In this case, when a fraction is generated in the calculation result as described above, a fraction process such as rounding, rounding up, and rounding down is performed. Then, proceed to Step D21,
A constant value Q is substituted for the value of the high frequency component Fn represented by the order n. Here, “−Q” is applied to the real part XrlFn of the X axis of the high frequency component Fn, and “Q” is applied to the imaginary part XimFn.
And further, the real part YrlFn and the imaginary part Yi of the Y axis
“Q” is substituted for mFn. In this case, the real part X of the X axis
The reason why “−Q” is substituted for rlFn is that the sampling operation is performed in the clockwise direction as shown in FIG. 6, and the constant addition value is “0.781” or the like. It is optional depending on the strength of the vibration. The high frequency component Fn generated in this manner is additionally registered in the frequency font storage unit 2-3 together with the order n in association with the corresponding part (step D22). Then, it is checked whether or not all parts have been processed (step D23). If not, the process proceeds to step D24, where the length of the contour of the next part is extracted, and the process returns to step D20. Then, the order of the high-frequency component corresponding to the length is determined, and a process of substituting a constant value Q for additional registration is performed for all parts. By generating a frequency font in this way, the vibration cycle can be made constant for each part.

In each of the above-described embodiments, an outline font is used as an example of a normal font when the normal font is converted to a frequency font by FFT processing. However, a bitmap font may be used. , Image data read by an image reader. Also, the normal font is subjected to FFT processing and converted into a frequency font, and the frequency font is stored in the frequency font storage unit 2-
3, but when outputting a character string, 1
After performing FFT processing for each character in real time, IFFT
You may make it process and output. Further, the frequency font may be generated according to another frequency conversion method without being limited to the FFT processing. The contents of the frequency font storage section 2-3 may be externally supplied via a recording medium such as a floppy disk or a communication line, or may be a document data processing apparatus having no frequency conversion function. . The characters are not limited to Chinese characters, but may be kana characters, symbols, alphabets, figures, and the like.

[0026]

According to the present invention, by expressing the outline of a character by a set of discrete spatial frequency components, the basic outline and thickness of the character can be significantly changed while suppressing an increase in the amount of data. For example, conventional outline fonts do not have a degree of freedom of deformation, such as being able to add various expressions, and have a wide range of applications.

[Brief description of the drawings]

FIG. 1A is a block diagram showing an overall configuration of a document data processing apparatus, and FIG. 1B is a diagram showing a main configuration of a RAM 2.

FIG. 2 is a diagram illustrating a configuration of a frequency font storage unit 2-3 and a data structure in which a size is divided into a real part and an imaginary part on a complex plane for each frequency component constituting the frequency font. .

FIG. 3 is a diagram illustrating a process of sampling a contour of a normal font for each part, performing FFT processing, generating a frequency font, and registering the frequency font in a frequency font storage unit 2-3.

FIG. 4 is a diagram for each frequency component constituting a frequency font.
The figure which showed the other data structure which represented the value by the amplitude and the phase component.

FIG. 5 is a flowchart showing an operation from generation of a frequency font from a normal font to registration in a frequency font storage unit 2-3.

FIGS. 6A to 6C are diagrams for explaining an operation when sampling is performed by tracing the contour on a bitmap drawn for one part.

FIG. 7 shows a state in which one character is decomposed into parts and subjected to FFT processing, and values of a real part and an imaginary part obtained by performing FFT processing are discretely distributed on a frequency axis. The figure which showed the frequency distribution diagram.

FIG. 8 is a diagram showing specific numerical values of each part constituting a frequency font and showing a correspondence relationship between which part corresponds to which data.

FIG. 9 is a diagram illustrating a process from IFFT processing of a frequency font to character printing.

FIG. 10 is a flowchart showing an operation when characters are printed using a frequency font.

FIGS. 11A and 11B are diagrams for explaining the second embodiment, and FIG.
Is a diagram for explaining the data structure of the frequency font,
FIG. 3B is a diagram showing an output shape of a character represented by a frequency font having such a data structure.

FIG. 12 is a view for explaining the second embodiment, and FIG.
FIG. 4 is a diagram for explaining another data structure of the frequency font, and FIG. 4B is a diagram showing an output shape of a character represented by the frequency font having such a data structure.

FIG. 13 is a diagram showing an operation when generating and registering a frequency font having the data structure shown in FIG.

FIGS. 14A and 14B are diagrams for explaining the third embodiment, in which FIG.
Is a diagram for explaining the data structure of the frequency font,
FIG. 3B is a diagram showing an output shape of a character represented by a frequency font having such a data structure.

FIG. 15 is a diagram for explaining a configuration of a frequency font storage unit 2-3 and another data structure of a frequency font stored therein in the third embodiment.

FIG. 16 is a flowchart showing an operation when generating and registering a frequency font having the data structure shown in FIG. 15;

[Explanation of symbols]

 Reference Signs List 1 CPU 2 RAM 2-1 Character string storage unit 2-2 Frequency conversion work unit 2-3 Frequency font storage unit 2-4 Modified font storage unit 2-5 Print buffer 3 Storage device 3-1 Normal font storage unit 4 Storage medium 5 input device 6 display device 7 printing device

Claims (14)

[Claims] 1. A character font storage device used for a character font output device for drawing and outputting a character font, wherein the character font storage device stores a character font to be output. A character font storage device which stores a frequency font having a data structure represented by: 2. The character font storage according to claim 1, wherein the frequency font is formed by expressing the outline of a character by a set of discrete spatial frequency components in accordance with a discrete Fourier series expansion method. apparatus. 3. The set of spatial frequency components is numerical information representing the magnitude of a real axis part and an imaginary axis part on a complex plane for each frequency component. 1. The character font storage device according to 1. 4. The character font storage device according to claim 1, wherein the set of spatial frequency components is numerical information representing the value of each frequency component as amplitude and phase components. 5. The character font storage according to claim 1, wherein the aggregate of the spatial frequency components is sampled by tracing on the contour of the bitmap data and is generated by discrete frequency decomposition. apparatus. 6. The method according to claim 1, wherein the set of spatial frequency components is sampled by tracing the contour of a character developed based on an outline font, and is sampled by discrete frequency decomposition. 1. The character font storage device according to 1. 7. The character font storage according to claim 5, wherein, when sampling while tracing the outline of the character, sampling is performed for each stroke in units of one element constituting the character. apparatus. 8. A unit for performing sampling by following the contour of a character is, for each line segment of a continuous stroke portion based on a straight line, even if the element constituting the character is one stroke. 7. The character font storage device according to claim 5, wherein: 9. A position at which sampling is started by tracing the outline of a character is a position at which writing of a character starts or ends, and a sampling operation is performed by tracing the outline of a character from that position. 7. The character font storage device according to claim 5, wherein: 10. The set of spatial frequency components comprises only low-frequency components that govern the general shape of a character, and the amount of font data is reduced by deleting all high-frequency components. The character font storage device according to claim 1. 11. The set of spatial frequency components comprises a combination of a low frequency component that governs the general shape of a character and a specific high frequency component, and a contour is formed by adding a specific high frequency component. 2. The character font storage device according to claim 1, wherein the character font storage device has a fluctuation element. 12. A font storage means for storing a frequency font having a data structure in which a contour constituting a character is represented by a set of discrete spatial frequency components, and a frequency font in the font storage means is read out. A character font output device comprising: a conversion means for converting the frequency font into coordinate data; and an output means for developing and outputting a character font based on the coordinate data converted by the conversion means. 13. A font generating means for reading an existing ordinary character font, sampling the contour along its outline, and generating a frequency font by discrete frequency decomposition, and providing a font generated by the font generating means. 13. The character font output device according to claim 12, wherein said frequency font is written in said font storage means. 14. A function for reading out a frequency font in which a contour constituting a character is represented by a set of discrete spatial frequency components to a computer and converting the frequency font into coordinate data; A recording medium on which a program for realizing a function of developing and outputting a character font based on a program is recorded.