JPH03126993A - Character generator - Google Patents

Abstract

PURPOSE:To generate character font data without the aid of manpower and to display high-quality characters without jaggy and flicker by subjecting the character font data to an anti-aliasing processing and edge graduation processing at the time of area filling the generated character font data. CONSTITUTION:An engineering work station EWS 28 interprets the character composing data outputted from a Japanese word processor 30, etc., and reads in the character font data stored in the form of outline font on a magnetic disk 32 etc., according to the interpreted character composing data. The work station subjects the character font data to the anti-aliasing processing and the edge blurring processing at the time of area filling the read character font data. The character font data is generated without the aid of manpower in this way and the high-quality characters free from the jaggy and flicker are displayed.

Description

[Detailed description of the invention] [Industrial application field]

The present invention relates to a character generator, particularly for producing high-definition (also referred to as high-definition television, HDTV) still image programs used as a means for database searches and information transmission in public facilities such as art galleries, libraries, and museums. The present invention relates to a character generating device suitable for use when [Conventional technology] Conventionally, in NTSC televisions and high-definition televisions, characters are displayed together with images (superimposed).
For example, when creating a high-definition still image program, characters may be displayed to explain the screen. When displaying characters on the monitor device, for example, a method can be considered in which a computer display device is connected to the monitor device and a computer display character font in this display device is displayed. Another possible method is to create character image data by reading a character block created by phototypesetting or the like with a high-definition camera or scanner, and to display characters using the character image data.

[Problems to be Solved by the Invention] However, in the method of displaying characters on the screen using the character font of a computer display device as described above, the roughness of the outline of the characters is not noticeable when displayed on a small screen. However, when displaying on a large screen, for example when enlarging and projecting on a large projection TV monitor, jaggies will become noticeable in the outlines of the characters displayed on the TV monitor.
There is a problem that the character quality deteriorates. Another problem is that flickers are likely to occur in the displayed characters, for example, when the character font is displayed on a high-definition interlaced monitor. In addition, in the method of reading the text block with a high-definition camera, etc., jaggies become less noticeable in the characters displayed on the screen, but flicker may occur, and the text block is created each time, which is time-consuming. There is a problem that it takes a lot of time. The present invention has been made to solve the above-mentioned problems, and provides a character generation device that can generate character font data without human intervention and can display high-quality characters without jaggies or flickers. The task is to do so. [Means for Achieving the Object] The present invention provides a character generating device for generating character font data having an outline vector font for a desired character, and a method for filling out the generated character font data. The above-mentioned problem is achieved by comprising means for performing anti-aliasing processing on the character font data to be filled in, and means for performing edge blurring processing on the character font data to be filled in. . [Operation] In the present invention, the character generating device generates character font data having an outline vector font for a desired character, and when filling out the generated character font data, anti-aliasing is applied to the character font data. At the same time, edge blurring processing is performed. Therefore, since the character font data for the desired character is generated without reading the character block, it can be quickly generated that corresponds to the character block, without requiring manual work such as creating the block. It becomes possible to generate character font data. Since jaggies in character outlines can be made less noticeable by anti-aliasing processing, it is possible to obtain character font data of a quality that can be used on large screens. Furthermore, since edge blurring processing is performed on the character font data at the same time, it is possible to display the text without flickering or making it inconspicuous even when displayed on the screen of a broadcast television such as a high-definition or NTSC television. Therefore, typefaces that have traditionally been avoided because they tend to cause jaggies or flicker, such as italic, italic, and Mincho fonts, can be used for NTSC television and high-definition broadcasting. It can now be used for characters displayed on commercial televisions. For example, italic and italic character fonts had many diagonal lines and jaggies were easily noticeable, so they were not often used in NTSC televisions, etc., but with the present invention, character fonts with such typefaces can also be used. become able to. Also, italic and #7 character fonts were used for characters displayed in high-definition because the 7 fonts were not noticeable when phototypesetting characters were read with a scanner. For example, by applying gradation manually using the paint system,
Since the occurrence of flicker is prevented, there is a problem in that the burden of painting work is large. On the other hand, the present invention eliminates this problem because flicker is eliminated through edge blurring processing. Additionally, Mincho typefaces are often used on non-interlaced monitors such as word processors because they are easy to read and less tiring. It wasn't used much. In contrast, according to the present invention, it becomes possible to display Mincho fonts with high quality even on broadcast television.

【Example】

Embodiments of the present invention will be described in detail below with reference to the drawings. This embodiment is a character generating device for generating characters used in a still image program, which is provided in a high-definition still image program producing apparatus as shown in FIG. The structure of this character generator is shown in FIG. The high-definition still image program production device shown in FIG. A high-definition still image editing unit 12 performs editing processing such as inputting character data and various effects on the converted high-definition still image data, and outputs editing instructions as programming data. and a high-definition still image reproducing unit 16 for displaying high-definition still image data in the still image editing unit 12 and reproducing a high-definition still image program according to the output programming data. It will be done. The high-definition still image editing section 12 includes a magnetic tape 18 and a magnetic disk 20 for storing image data read from a scanner and data of a high-definition still image program, and a first magnetic disk for monitoring the high-definition image to be edited.
The computer has a high-definition monitor 22, and a keyboard and stylus pen 24 for inputting editing commands. The fi production unit 14 outputs an editing command for high-visibility still image data to an image processing unit 36 (described later), creates programming data for the command, and also edits character font data stored in the magnetic disk 32. an engineering workstation for generating necessary character font data by reading out the character code and typesetting command (hereinafter referred to as character typesetting data) inputted from the Japanese word processor 30, and performing various processing on the character font data. (EWS) 28, a magnetic tape 29 for storing still image data before and after processing, and a frame memory 34 for storing image data in order to superimpose processed character font data on image data. An image processing unit 36 that processes and edits image data according to output editing commands, a non-interlace monitor 38 that displays still image data and character data before and after processing, and a non-interlace monitor 38 that displays still image data for high-definition still image programs. It has a high-definition converter for displaying at pixel density and three second high-definition monitors. A high-definition processor (not shown) that displays images, a third high-definition monitor 40 for displaying the images, and an optical disk 41 for storing high-definition still image data are provided. A processing system before and after an EWS 28 is shown in Fig. 2, and this EWS 28 has a detailed configuration shown in Fig. 3. As shown in Fig. 3, this EWS 28 includes the A decoding unit 42 decodes the character typesetting data outputted from the word processor 30, and an outline font (
A character font reading unit 44 that reads character font data stored in the form of vector data), and an affine transformation unit 4 that performs affine transformation of the read character font data.
6, an arrangement section 48 that arranges the affine-converted character font data in four ways, for example, horizontal writing, vertical writing, Roman (kana) half-width writing, and European (kana) pitched writing, and a character font by filling in the character font data. A character font filling section 5o that converts data from vector data to raster data, an anti-aliasing processing section 52 that performs anti-aliasing processing on the filled character font data (raster data), and an anti-aliasing processing section 52 that performs anti-aliasing processing on the filled character font data (raster data); Edge blurring processing unit 54 that performs edge blurring processing
and are stored in the frame memory 34 as "C;E,D
raster data of character fonts (hereinafter referred to as character raster data) generated in a frame memory reading section 56 for reading the original image data and a character raster development section 58 from the character font filling section 50 to the edge blurring processing section 54; A background superimposition processing section 60 that superimposes the read background image data and a frame memory writing section 62 that writes still image data on which characters are superimposed into the high-definition frame memory 34 are provided. Next, the operation of the embodiment will be explained. In this embodiment, the process of generating characters and writing them into the background image is performed on an engineering workstation (E
WS) 28, the process is carried out in accordance with the steps indicated by the arrows shown in FIG. That is, first, the character typesetting data decoding unit 42 reads the character code (for example, JIS code, etc.) input from the Japanese word processor 30 and the typesetting command (typeface specification, etc.) inputted from the host processor of the word processor 30, for example, a personal computer. The character font file in the magnetic disk 32 is searched. Next, the character font reading unit 44 reads outline vector font data (vector data) of the character font from the retrieved character font file based on the input character code and typesetting command. Next, the character font affine conversion unit 46 performs character enlargement, reduction, rotation, and transformation (half-length, oblong, italic) processing on the read character font data (outline vector font). For example, as shown in FIG. 4, a process of converting a character font input in its original form to italic is performed. Next, the character font placement unit 48 performs placement processing on each affine-converted character font data to form a predetermined wording. For example, in horizontal writing, use Figure 5 (A), in vertical writing, use Figure 5 (B), and in Roman half-width writing, use Figure 5 (C). ), and in the case of Roman letters, the fonts are arranged as shown in FIG.
Converts character font data to raster data by filling in the outline of vector data). This filling is done by sampling each pixel into a raster and filling in the outline of the outline vector font. For example, as shown in Figure 7 (
If there is a relationship between the pixels and the outline vector font as shown in A), if the pixels are simply filled in, jaggies will occur in the outline portion as shown in FIG. 7(B). Therefore, the anti-aliasing processing unit 52 first virtualizes the pixel density to N@, for example four times as finely as shown in FIG. Then, for the raster data of the actual pixel, give the average value of the finely divided pixel values corresponding to the actual pixel.For example, as shown in Figure 7(C), 4@finely sampled. When the average value of the pixel density data is taken, as shown in FIG. 7(D), pixels at the boundary between the outline and in-font are painted with halftone. Note that the strength of this anti-aliasing processing becomes stronger as the pixel density becomes finer, and the setting value can be changed depending on the size of the monitor screen, for example. Next, when filling in the character font, the edge blurring processing unit 54 performs edge blurring processing by giving each pixel a width and filling the periphery of the pixel with a smaller weight than the center. 8(A), but in this embodiment, in order to perform edge blurring, the distribution of surrounding pixels is changed as shown in FIG. 8(B). The weights of each pixel before and after filling with this distribution, for example, are shown in Fig. 8 (A) and (B).
It will also be shown inside. Note that in FIG. 8(B), the coefficient a is, for example, 1/4. This weighted edge blurring process weights the filling of each pixel with a filter matrix.For example, if the attenuation coefficient is ε(>1) in a (2M+1) The element Pi J, that is, the weight of the pixel is expressed by the following equation (1). Pij=aε-1·ε-IJ (-M≦i, J≦M) (1) Here, the coefficient a in equation (1) is expressed by the following equation (2). In the example shown in FIG. 8, the damping coefficient ε=2.0, M=
1, and the coefficient a=1/4 as described above. Therefore, the surrounding pixels of the character font to be filled will be filled with a distribution as shown in Figure 9 (A), and when filled, the surrounding pixels will be filled as shown in Figure 9 (B). The edges of the character font are blurred using a gradation.The strength of the edge blurring process can be changed, for example, depending on the size of the monitor screen.From the above, , in the character raster development section 58 from the character font filling section 50 to the edge blurring processing section 54,
Create raster data with gradation from line vector font data. Next, the frame memory reading unit 56 reads the image currently displayed on the high-definition monitor (background image @). However, if superimposition of characters on the background image is not performed, this reading process is not performed. Give a certain background color as the base color. Next, in the background superimposition processing section 60, the character raster data (gradation data) generated by the character raster development section 58 and the background image read by the reading section 56 are superimposed (2), for example, as shown in FIG. A background image as shown in FIG. 10(B) is superimposed on character raster data as shown in FIG. 10(C) to obtain an output image as shown in FIG. 10(C). This character raster data (for example, 16-bit data) and background image data (8-bit data for each of red R, green G, and blue B)
For example, the gradation S of character raster data can be set from 0 to
As a range of 327'67, set the text color to red RO, green Go
, when the colors of the blue BO1 background image are red Rb, green Gb, and blue Bb, data for each color of the output image (red R1 green G, blue B) can be obtained using the following equations (3) to (5). This is done by R= (1-f) Rb 10f Ro -<3)
G=(1-f)Gb+fGo-(4)B=(1
-f ) Bb +f Bo - (5) However, the coefficient f is determined by f = S/32767 and takes a value between 0 and 1. In the case of FIG. 10, in the character raster data, the gradation of the line drawing part is 32767, and the background image except the line drawing part has a gradation of O1. Red R1 green G and blue B are given as 8-bit data. Next, the output image data in which the background and character data have been superimposed as described above is written into the frame memory 34,
It is displayed on the high-definition monitor 38. In addition, if necessary, the image data is written to an external storage device such as the magnetic disk 32 or an optical disk and stored as still image data containing characters. Then, the video production section 14 produces and edits the still image program, and the high-definition processor in the high-definition still image reproduction section 16 reproduces the high-definition still image data according to the programming data used during editing. The image data is corrected and edited in the high-definition still image editing department and created as material data for the final high-definition still image program. In the above embodiment, the present invention is adopted in the high-definition still image program production device. was generating characters, but
The device to which the present invention is applied is not limited to such a high-definition device, and the present invention can also be applied to other devices, such as those that generate characters to be displayed on an NTSC television. [Effects of the Invention] As explained above, according to the present invention, character font data can be generated without requiring any manual work such as creating a block when creating character data, and it is possible to generate character font data without jaggies or flickers. Can display high-quality characters. Therefore, it is possible to use fonts that have traditionally been avoided due to their tendency to cause jaggies or flicker by overlaying them on images6.For example, italics and slanted fonts have many diagonal lines that make jaggies more noticeable. Although it has not been used much for various reasons, it can now be used with the present invention. Mincho fonts have not been widely used in broadcast televisions such as NTSC televisions and high-definition televisions because they have thin horizontal lines and tend to flicker; however, the present invention allows them to be used in broadcast televisions. Excellent effects such as being able to do this can be obtained.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing the overall configuration of a high-definition still image program production device according to an embodiment of the present invention, FIG. 2 is a block diagram showing the configuration of a character generation system in the embodiment, and FIG. FIG. 4 is a block diagram showing a detailed configuration of the system; FIG. 4 is a plan view showing an example of character font affine conversion for explaining the operation of the embodiment; FIG. 5 is a diagram showing an example of character font placement processing. FIG. 6 is a plan view showing an example of the character font filling process; FIG. 7 is a plan view illustrating the anti-aliasing process; FIG. Similarly, FIG. 9 is a schematic diagram and line diagram for explaining the edge blurring process, FIG. 9 is a plan view for explaining the process, and FIG. 10 is a plan view for explaining the background superimposition process. be. 10...Scanner, 12...High-definition still image editing department, 14...Video production department, 16...High-definition still image playback section, 28...Engineering workstation (EWS), 30...Japan word processor, 34... frame memory, 36... image processing section, 42... character typesetting data reading section, 44... character font reading section, 46... character font affine conversion section, 48... - Character font placement section, 50... Character font filling section, 52... Anti-aliasing processing section, 54... Edge blurring processing section, 56... Frame memory reading section, 58... Character raster expansion 60...Background superimposition processing unit, 62...Frame memory writing unit.

Claims (1)

[Claims] (1) Character generation means for generating character font data having an outline vector font for a desired character, and performing anti-aliasing processing on the character font data to be filled in when filling out the generated character font data. A character generating device comprising: means for performing edge blurring processing on the character font data to be filled in.