JPH03102393A - Character data processing method - Google Patents

Abstract

PURPOSE:To reproduce the contour of a character which has desired line width and a desired shape by restoring the quantities of deviation by reverse Fourier transformation after varying frequency components of Fourier coefficient data, and finding positions which are the restored deviation quantities away from a reproduced skeleton line. CONSTITUTION:When the contour of the character is reproduced from data which specifies the shape of the character and consists of data including the Fourier coefficient data obtained from information specifying the skeleton line 2 of the character and the quantities of deviation from the skeleton line 2 to the contour of the character 3 through Fourier transformation, the frequency components of the Fourier coefficient data are varied properly and then the reverse Fourier transformation is carried out to restore the quantities of deviation from the skeleton line 2 to the contour 3. Then the skeleton line 2 is reproduced from the information specifying the skeleton line 2 and the positions which are the restored quantities of deviation away from the reproduced skeleton line 2 are found as points on the contour 3 of the character.

Description

[Detailed Description of the Invention] [Field of Industrial Application] The present invention relates to a character data processing method for reproducing characters from character pattern data stored in various electronic devices. The present invention relates to a character data processing method for reproducing characters with various line widths and shapes from one character pattern data that is converted into data with information on the amount of deviation to the contour.

[Prior art] Various electronic devices such as laser printers and input editing devices are
Character patterns for screen display and hard copy output are stored in a memory such as a hard disk or ROM.

These character patterns are converted into data in various ways. Examples of techniques for converting character patterns into data include the following methods.

■． Define bitmap data using run-length codes ■. Approximating the outline of characters with vectors ■. Approximating the outline of a character with a curved line [Problem to be solved by the invention] There are various methods for converting character patterns into data, but all of these methods can convert characters with different line widths or outlines from one character pattern data. There are problems such as characters that have been transformed cannot be easily reproduced. [Means for Solving the Problems] The present invention has been developed based on the above points, and when reproducing a character pattern, processing is performed to change the shape of the character, such as the line width of the character, from one character pattern data. The purpose is to provide a method for processing character data. The feature is that the character shape is composed of data that includes at least information that specifies the skeleton line of the character, and Fourier coefficient data obtained by Fourier transforming multiple deviations from the skeleton line to the outline of the character. In a character data processing method for reproducing the outline of a character from data specifying the character, the frequency components of the Fourier coefficient data are appropriately changed, and then the deviation amount from the skeleton line to the outline is restored by inverse Fourier transform. A character with a desired line width and shape can be created by reproducing a skeleton line from the information specifying the skeleton line, and determining a position separated by the restored deviation amount from the reproduced skeleton line and setting it as a point on the outline of the character. The goal is to reproduce the contours of

[Example] ■. Creation of character pattern data First, a method for creating character pattern data to which the present invention is applied will be explained. The character pattern data to which the present invention is applied is obtained by decomposing a character 1 into constituent elements (hereinafter referred to as basic strokes) la as shown in FIG. 2, and converting each basic stroke 1a into data as one unit. Create something. Below, the case where one basic stroke 1a of a character as illustrated in FIG. 3 is converted into data will be explained with reference to the processing flow diagram of FIG. 12. Note that the characters used as the basis for creating the character pattern data may be either analog original characters or character pattern data that has already been converted into data by some method.

1. Creation of skeleton line information> First, extract the skeleton line of the basic stroke and create skeleton line information that defines the skeleton line (Fig. 12, 81). There are various methods for extracting the skeleton lines, such as automatic extraction using image processing technology and manual extraction by an operator on the display screen, and any of these methods may be used. In addition, the extracted skeleton lines may be defined using any method such as straight lines, curves (functions), straight lines and curves, etc. For example, for the basic stroke 1a in FIG. 3, skeleton line information that defines the skeleton line 2 shown in FIG. 4 is created. <2. Extraction of deviation amount> Next, the deviation amount (distance) from the skeleton line to the contour is extracted (S2). The deviation amount is extracted in order from an arbitrarily determined deviation amount extraction start point. Basically, this is performed by sequentially extracting distances from division points set on the skeleton line to the contour at predetermined intervals. The interval at which dividing points are set can be set arbitrarily, and in the case of peculiar parts of the skeleton line, extraction is performed using a method suitable for that part. For example, the amount of deviation is extracted by the following method depending on the state of the skeleton line.

■． Middle partial skeleton line 2 where the skeleton line is straight or gently curved
However, as shown in FIG. 5(1), in the middle part of a straight line or a gentle curve, the dividing points are set at points where the skeleton vA2 is divided at equal intervals as shown in FIG. 5Q). The distance to the point where the normal line of the dividing point intersects with the contour 3 (hereinafter referred to as a sample point) becomes the deviation amount. This is the extraction method described above.

■． End of the Skeletal Line The end of the skeleton line 2 as shown in Figure 6 ω cannot be divided into dividing points at equal intervals as in the method ① described above. ) Divide into equal angles as shown. The distance from the end 2e of the skeleton line to the sample point that intersects with the extension line of the divided angle line is the deviation amount. ■． A skeleton line 2 adjacent to both ends 2a, 2b of a partial skeleton line 2, in which the skeleton lines adjacent to both ends of the skeleton line have an angle. ，
, 2A, are angled as shown in Figure 7 ω, the two skeleton lines 2It+ll 2a-1 on the side adjacent to both end points 2a and 2b as shown in Figure 7(2) The intersection point 4 of the two normals of the end points is found, and the distance to the sample point where the line connecting the intersection point 4 and the division points obtained by dividing the skeleton line 2 at equal intervals intersects the contour 3 is extracted as the amount of deviation.

The contour on the opposite side to the intersection point is extracted by the method (2) described above.

The method for extracting the amount of deviation is not limited to the methods (1) to (4) above, and it goes without saying that other methods may be used.

FIG. 8 is a graph showing the amount of deviation extracted as described above for one basic stroke. The vertical axis represents the amount of deviation. The horizontal axis represents information that identifies which sample point on the contour the amount of deviation corresponds to.For example, if the sample point on the contour from which the amount of deviation was extracted is
A number (hereinafter referred to as sample item number) is assigned in order, and the horizontal axis is the number. 3. Fourier Transform> The deviation amount of the basic stroke extracted as described above is subjected to Fourier transform (S3).

Since the amount of deviation is a discrete value formed by a data string as shown in FIG. 8, the Fourier transform performed here is a discrete Fourier transform.

Discrete Fourier transform is generally expressed by the following equation.

N: Number of data g(4): Fourth data Here, N: Number of sample item numbers of one basic stroke g(4): Amount of deviation in the fourth sample item number.

Therefore, here we perform the Fourier transform expressed by the above formula. If the result of Fourier transformation is expressed in a graph, it will be as shown in FIG. 9, for example. In FIG. 9, the horizontal axis is the order, and the vertical axis is the Fourier coefficient. Also, Fig. 9 (1) is a graph showing the real part, and Fig. 9 <2) is a graph showing the Fourier coefficient of the imaginary part. In FIG. 9, the real part of the Fourier coefficients is left-right even symmetric, and the imaginary part is also odd symmetric. This is not a special characteristic of the example shown in Figure 9, but rather a characteristic peculiar to discrete Fourier transform. Therefore, there is no need to memorize all Fourier coefficients,
Memorizing only half of it is enough.

4. Data Storage> Next, storage of data regarding one basic stroke, which is composed of skeleton line information, Fourier coefficients, etc. obtained as described above, will be explained.

Although the obtained Fourier coefficient data may be stored as is, a method of compressing and storing the data amount will be described here.

FIG. 10 is a diagram showing an example of the format of Fourier coefficient data. The information represented by each data in FIG. 10 is as follows.

■． total order The order of the Fourier coefficients of that elementary stroke.

■． Data type Data that distinguishes whether the Fourier coefficients that follow are the real part or the imaginary part.

■． quantization pitch Pitch when Fourier coefficients are quantized.

■． Number of sections As shown in Figure 11, all data is divided into several sections in order to change the number of bits allocated depending on the magnitude of the Fourier coefficient value. Data indicating the number of sections at that time.

■． Degree of interval Data that indicates how many orders of Fourier coefficients the interval corresponds to. ■ Number of allocated bits Number of bits allocated to represent the Fourier coefficient in that interval 6 If the same number of bits is given uniformly to represent the largest Fourier coefficient, bits will be given redundantly to the part of the small Fourier coefficient. The amount of data will increase. Therefore, as shown in FIG. 11, the number of bits to be allocated is changed depending on the magnitude of the absolute value of the Fourier coefficient. In other words, when the absolute value of the Fourier coefficient is large, more bits are allocated, and when it is small, fewer bits are allocated, thereby reducing the overall amount of data.

■． Fourier coefficient The value of the Fourier coefficient for one section obtained by quantizing the Fourier coefficient at the above-mentioned quantization pitch. In other words, Fourier coefficients are stored for the number of degrees of the above-mentioned interval.

In this way, as shown in Figure 10, Fourier coefficient data consists of ■■■ data that constitutes one interval data, and the interval data is collected in the number indicated by the ■interval number to form the real part or imaginary part. It composes the data.

The Fourier coefficient data configured in the above format is stored together with the above-described skeleton line information and deviation amount extraction method data, and is used as data regarding the basic stroke (FIG. 12, 34).

One character pattern data is composed of data regarding a plurality of basic strokes.

In this way, for example, character pattern data of characters up to the JIS 1st and 2nd standards is created and used as character pattern data to be installed in various electronic devices such as laser printers and input editing devices.

■． Reproduction of Character Patterns Next, an embodiment of the present invention for reproducing characters with various line widths and characters with deformed outlines from the character pattern data created as described above will be described with reference to the processing flow diagram in FIG. . The following is a description of the reproduction of one basic stroke.

First, Fourier coefficient data as shown in FIG. 9a>(2) is restored from data stored in the format shown in FIG. 10 (SL in FIG. 1).

Next, a process of appropriately changing the restored Fourier coefficient data is performed (S2).

In this way, the present invention modifies the line width and contour shape of the reproduced character by appropriately changing the Fourier coefficient data.

For example, the specific changes and changes in character shape are as follows.

■． Change to increase or decrease the low frequency component value of the Fourier coefficient By performing this change, the line width of the character can be changed.

■． By adding a low frequency component or a high frequency component, especially to the Fourier coefficients of a simple and smooth contour, it is possible to reproduce characters with uneven contours.

■． Modification to remove a certain amount of high frequency If there is noise in the outline that occurred during the creation of character pattern data, this change to the Fourier coefficients will remove the noise and reproduce a character with a smooth outline.

As described above, you can change the line width and outline shape of characters. Details of the processing results will be described later.

Next, the Fourier coefficient data appropriately changed as described above is subjected to inverse Fourier transform (S3).

Inverse Fourier transform is generally expressed by the following equation.

N: Number of sample item numbers of one basic stroke g(4): Amount of deviation in the fourth sample item number Next, the skeleton line of the basic stroke is restored (S4).

The skeleton line is restored from the skeleton line information stored together with the Fourier coefficient data.4 From the data of the deviation amount, skeleton line, and deviation extraction method restored as above, the basic. The outline of the stroke is reproduced (S5).

FIG. 13 is a diagram illustrating an example of processing in which the line width of a character is changed by making the above-mentioned change. Figure 13 (1) shows the basic stroke, deviation amount, and
FIG. 13(2) shows a state in which the line width of a character is changed by reducing a certain low frequency component of the Fourier coefficient.

FIG. 14 is a diagram illustrating an example of processing in which a character with an uneven outline is reproduced by changing the above-mentioned ■. Figure 14 (1) shows the basic strokes, deviation amounts, and Fourier coefficients represented by the original character pattern data, and Figure 14 (2) shows changes that add a certain amount of low frequency and a certain amount of high frequency to the Fourier coefficient as appropriate. This shows the state in which characters with uneven outlines are reproduced.

FIG. 15 is a diagram illustrating an example of processing for removing contour noise by performing the above-mentioned change. Figure 15 (1) shows the basic strokes, deviation amounts, and Fourier coefficients represented by the original character pattern data, and Figure 15 (2) shows the changes that remove high-frequency components of the Fourier coefficients to create a smooth, noise-free image. This shows the state in which characters with a sharp outline are reproduced.

Note that in FIGS. 13, 14, and 15 described above, only the real parts of the Fourier coefficients are illustrated, and the imaginary parts are omitted.

The contour of the basic stroke reproduced in this way is obtained in the form of points on the contour, as shown in FIG. Therefore, when outputting characters using an electronic device or the like equipped with character pattern data according to the present invention, interpolation is performed between points on the contours, and processing is performed to fill in the inside of the contours. Interpolation and filling processing may be performed using conventionally known methods.

By playing multiple basic strokes in this way,
Can play one character.

■． Example of an electronic device equipped with character pattern data according to the present invention FIG. 17 is a block diagram showing an example of a laser printer in which character pattern data is stored in a memory by the method described above.

In Fig. 17, 5 is the character pattern made by the method described above.
6 is a CPU, 7 is an input unit for issuing various commands such as character type, line width, deformation type, etc., and 8 is a character read out from memory 5. A decoder for reproducing a character pattern from pattern data; 9 a bitmap memory for bit-expanding the characters reproduced by the decoder 8; 10 an output control unit; 11 an output for outputting a hard copy of the data expanded in the bitmap memory 9. Department. The output section l1 is, for example, a well-known laser output unit that outputs characters on plain paper or photosensitive material by scanning a laser beam with a polygon mirror. To output a character, the CPU 6 first reads the character pattern data of the character to be output from the memory 5 and sends it to the decoder 8. The decoder 8 performs the character pattern reproduction process described above to reproduce the character pattern with predetermined characters, line widths, and shapes.
The information is related to the bitmap memory 9.

After the data of a plurality of character patterns is expanded into the bitmap memory 9 by repeating such processing, the output unit 11 outputs the characters under the control of the output control unit 10. The embodiments of the present invention have been described above. In the above embodiments, character pattern data is created by decomposing characters into basic strokes. However, in the present invention, character pattern data may be created without decomposing characters into basic strokes. Of course, [Effects of the Invention] The present invention can reproduce the results shown in FIGS. 13 and 14 by simply performing a simple process of appropriately changing the frequency of the Fourier coefficient data of the real part when reproducing characters. Since the line width and character shape can be easily changed as illustrated in FIGS. 15 and 15, there is an effect that characters with various line widths and shapes can be reproduced from one character pattern data.

[Brief explanation of drawings]

Figure 1 is a process flow diagram showing the processing procedure for character reproduction according to the present invention, Figure 2 is a diagram explaining the division of characters into basic strokes, Figure 3 is a diagram showing an example of basic strokes, and Figure 4. is a diagram explaining the skeleton of a character, Figures 5, 6, and 7 are diagrams explaining the method of extracting the amount of deviation, Figure 8 is a graph of the amount of deviation, and Figure 9 is a graph of Fourier coefficient data. Graphs, FIG. 10 is a diagram showing the format of Fourier coefficient data, FIG. 11 is a diagram explaining the sections of Fourier coefficient data, FIG. 12 is a processing flow diagram showing the processing procedure for creating character pattern data,
Figures 13, 14, and 15 are diagrams showing processing examples of the present invention, Figure 16 is a diagram showing points on the contour obtained by reproduction processing, and Figure 17 is a block diagram of an example of a laser printer. It is. 5...Memory 6...CPU 7...Input section 8...Decoder 9...Bitmap memory 10...Output control section E1...Output section

Claims (1)

[Claims] (1) Identifying the shape of a character made up of data that includes at least information that specifies the skeleton line of the character and Fourier coefficient data obtained by Fourier-transforming multiple deviations from the skeleton line to the outline of the character. In a character data processing method for reproducing character contours from data, after appropriately changing the frequency components of the Fourier coefficient data, inverse Fourier transform is performed to restore the deviation amount from the skeleton line to the contour, and the skeleton line is The outline of the character with the desired line width and shape can be created by regenerating the skeleton line from the specified information, finding a position that is separated by the restored deviation amount from the regenerated skeleton line, and setting it as a point on the outline of the character. A character data processing method characterized by reproduction.