JPH08138066A - Contour line sampling method for character - Google Patents

Abstract

PURPOSE: To accurately sample line-pair information required for line width correction processing on all characters provided with various features. CONSTITUTION: A character plotting part 13 in which a line width correcting part 134 equipped with a closed area separating part 1341, a horizontal line/ vertical line sampling part 1342, a line pair judging part 1343 and a coordinate point correcting part 1344, a data readout part 131, a font storage part 132, a coordinate transformation part 133, work memory 135 and a paint-out part 136 are provided, an input part 11 and an output part 14 are provided, and they are controlled by a control part 12, and the line pair judging part 1343 is provided with a judging part 24 in which a pair flag judging part 25 for target edge, a directional condition judging part 26 and a coordinate value condition judging part 27 are provided, a reference edge setting part 21, a pair flag judging part 22, a target edge setting part 23 and a pair flag setting part 28, and it is controlled by a line pair judging control part 70.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method of outputting high-quality characters on a printer or a display by using character data (outline font) composed of character contour coordinate data.

[0002]

2. Description of the Related Art In order to display and print an outline font, the font expression must be converted into a dot pattern. Therefore, even if the outline font is originally expressed with the same line width, the resolution of the output device such as a display or a printer or the reduction of the characters causes
The line width varies and is displayed and printed.
Therefore, when converting an outline font into a dot pattern, processing (hinting) is performed to correct variations in line width. The main hinting is disclosed in Japanese Patent Application No. 2-81657. Further, at the time of hinting, it is necessary to perform line pair extraction processing such as the upper and lower ends of the contour lines that generate horizontal lines and vertical lines, and the left and right ends. Conventionally, Japanese Patent Application No. 4-18819 is used for line pair extraction.
A method such as the one disclosed in Japanese Patent No. 0 was used.

As a line pair extraction rule, this is a line pair that manages only the horizontal and vertical line directions and the (x, y) coordinate values of the representative points, sorts the representative points, and observes the directions. It was to judge whether or not. As a result, x1 and x2, like the conventional line pair 82 of FIG.
Since x7 and x8 are extracted as line pairs, the line width varies even after hinting.

[0004]

In the above conventional method, an accurate line pair can be extracted for a simple character, but a correct line pair cannot be extracted for a special character, resulting in an uneven line width. In some cases, it was standing.

It is an object of the present invention to extract horizontal and vertical line information in advance in order to perform the line width correction which is indispensable for expressing the beauty of the original outline font. The line pair is accurately extracted by using a line pair determination method applicable to all characters.

[0006]

The present invention has been made in order to achieve the above-mentioned object, and proposes a line pair extraction rule which can deal with all characters. In addition to one end point that generates a direction and an edge, for example, in the case of a horizontal line, the maximum x value, the minimum x value and the y value that generate a horizontal line, and the direction of the line are managed. Similarly, in the case of a vertical line, the maximum y value, the minimum y value and the x value for generating the vertical line, and the direction of the line are managed. By using the line direction and coordinate values, line pairs can be accurately extracted even for characters with various characteristics, and line width correction does not deteriorate the character quality, and It is possible to express the fullness.

[0007]

[Operation] When characters and size are input by the input means,
The control means reads out the outline data from the font storage section onto the work memory, and the coordinate conversion section performs device coordinate conversion and size conversion. Next, the line width correction unit corrects the line width variation due to the conversion, the filling unit generates a contour line from the intersection of the straight line approximation data and the scan line, fills the inside of the contour line, and outputs the character from the output unit. . The line width correction unit includes a closed region separation unit, a horizontal line vertical line extraction unit, a line pair determination unit, and a coordinate point correction unit.
The closed area separation unit separates the contour coordinate data of the character into closed areas. Next, in the horizontal line / vertical line extraction unit, the y coordinate value of the horizontal line outline coordinates from the coordinate-converted character outline coordinate data, the x coordinate value of the endpoints that generate the horizontal line and the line direction, and the x coordinate value of the vertical line outline coordinates. , Y-coordinate values of both end points that generate a vertical line and the direction of the line are extracted, sorted by the y-coordinate value for the horizontal line and x-coordinate value for the vertical line and stored in the table. Next, the line pair determination unit determines whether or not the extracted horizontal lines and vertical lines satisfy the conditions that can be the respective line pairs. In the coordinate point correction unit, the interval between the line pairs obtained by the line pair determination unit is obtained, and the interval is converted into an integer,
The coordinate value is corrected according to the interval between the line pairs that have been converted into integers, and the entire length is corrected.

[0008]

EXAMPLE An example of the present invention will be described below. FIG. 1 is a block diagram of a system showing an embodiment of the present invention.

Reference numeral 11 is an input unit for inputting characters and sizes, 12 is a control unit for controlling a series of processes, and 13
Reference numeral 2 is a font storage unit for storing font data, 131 is a data reading means for reading the font data from the font storage unit, 133 is a coordinate conversion unit for converting the font data into a predetermined character size, and 134 is a conversion unit. A line width correction unit for correcting the line width from the character contour coordinate data, 135 is a work memory for calculating the read character contour coordinate data, and 136 is a character contour for generating a character image. Reference numeral 14 is a filling unit for filling the inside, and 14 is an output unit for outputting the generated character image. Reference numeral 1341 denotes a closed area separation unit that separates the character contour coordinate data for each closed area of the character, and 1342 extracts a horizontal line and a vertical line for extracting the horizontal and vertical lines of the character for each closed area separated by the closed area separation unit. Part, 13
Reference numeral 43 is a line pair determination unit that determines whether or not each of the horizontal lines and vertical lines extracted by the horizontal line vertical line extraction unit can be a line pair, and 1344 is the line pair information based on the line pair information generated by the line pair determination unit. This is a coordinate point correction unit that determines the line width and corrects the coordinate points.

The operation of each unit will be described.

When the input unit 11 transfers the print command to the control unit 12, the control unit 12 generates a character image from the print command and transfers the character image to the output unit 14, thus controlling a series of processes.
The output means is a printer, a display, etc., and prints or displays a character image. Data reading unit 131
Then, the typeface information and the character code sent from the input unit 11 are decoded, and the designated font data is read from the font storage unit 132 to the work memory 135. The coordinate conversion unit 133 receives the character size information from the input unit 11,
The font data read to the work memory 135 is converted to have a predetermined resolution and size. At this time, when the character contour coordinate data includes a curved line expression such as a Bezier curve, the coordinate conversion unit 133 performs a linear approximation process. The converted character contour coordinate data is corrected by the line width correction unit 134. Filling section 136
Generates a character image by filling the inside of the outline based on the corrected character outline coordinate data and transfers it to the output unit 14.

In the line width correction unit 134, first, the closed area separation unit 1341 separates the character contour coordinate data for each closed area of the character, and the horizontal line vertical line extraction unit 1342 separates the coordinate conversion unit.
A horizontal line and a vertical line are extracted from the character outline coordinate data converted by 133, and the horizontal line is sorted for the y coordinate value and the vertical line is sorted for the x coordinate value and stored in the table. The line pair determination unit 1343 determines whether or not the horizontal line and the vertical line extracted by the horizontal line / vertical line extraction unit can be a line pair, and extracts line pair information. The coordinate point correction unit 1344 obtains the line width between the line pairs based on the line pair information calculated by the line pair determination unit, corrects the coordinate points to make the line widths uniform, and also corrects the overall length.

FIG. 2 is a block diagram showing the operation of the line pair determination unit of the present invention.

Reference numeral 20 is a line pair determination unit that controls a series of line pair determination processing, and reference numeral 21 is a reference edge setting unit that sets a reference edge (horizontal line or vertical line) in the line pair determination. In this embodiment, a pair flag is used to determine whether each edge has a line pair and is on or off. Therefore, the pair flag determination unit 2
At 2, it is determined whether the reference edge has a line pair. Reference numeral 23 is a target edge setting unit that sets an edge that is a target for determining whether the target edge is a line pair, and reference numeral 24 is a determination unit that determines whether the target edge satisfies a condition of being a line pair. The determination unit 24 is
Target edge pair flag determination unit 25, orientation condition determination unit 2
6 and the coordinate value condition determination unit 27. The target edge pair flag determination unit 25 is a unit that determines whether or not the target edge pair flag is off, and the orientation condition determination unit 2
Reference numeral 6 is a portion for determining whether the directions of the reference edge and the target edge are opposite to each other. The coordinate value condition determination unit 27 determines using the x coordinate range for horizontal lines and the y coordinate value range for vertical lines. It is a part.

The operation of each unit will be described.

The line pair determination control unit 20 receives the horizontal line vertical line information extracted by the horizontal line vertical line extraction unit 1342, finds a line pair of each horizontal line vertical line, and performs a coordinate point correction by the coordinate point correction unit 1344. Control the processing of. First, the reference edge setting unit 21 sets the reference edge, and the pair flag determination unit 22 determines from the pair flag of the reference edge whether or not the line pair has already existed.
Next, the target edge setting unit 23 sets an edge which is a target of line pair determination of the reference edge. Next, the determination unit 24 determines whether the target edge can be a line pair of the reference edge. First, the target edge pair flag determination unit 25 determines from the target edge pair flag whether the target edge is off. The orientation condition determination unit 26 determines whether or not a condition that the orientations of the reference edge and the target edge are opposite to each other is satisfied. Next, it is determined from the coordinate values of the reference edge and the target edge whether or not a condition that can be a line pair is satisfied. When the determination unit 24 determines that there is a line pair, the pair flag setting unit 28 sets the pair flag of the reference edge and the target edge to ON (has a line pair). In this embodiment, for each edge of the horizontal line and the vertical line, it is managed whether or not there is a line pair by using a pair flag, but it may be managed by a method other than this.

FIGS. 3 and 4 are flow charts showing horizontal line extraction and vertical line extraction processing, which are operations of the horizontal line vertical line extraction unit 1342, respectively. FIG. 5 shows a horizontal line / vertical line extraction unit 13
The horizontal line and the vertical line extracted in 42 are sorted, and the duplicate coordinate values are deleted and stored in the table. 6 and 7 are flowcharts showing the line pair determination unit 1343. FIG. 8 shows the vertical line pair extraction result for the character "out." An example of the horizontal line extraction process will be described with reference to FIG.

First, in step ST300, a starting point for horizon search is set. Step ST301
Then, the previous coordinate point (x _i-1 , y _i-1 ) and the target coordinate point (x _i ,
The difference (xw, yw) of y _i ) is calculated. In ST302, it is checked whether the condition that the difference (xw) between the x coordinate values is other than 0 and the difference (yw) between the y coordinate values is 0 is satisfied. If the condition of ST302 is not satisfied, the next coordinate point is set as the target coordinate point in ST311, and the process of ST301 is similarly performed. Also ST3
When the condition of 02 is satisfied, the target x coordinate value (x _i ) and the previous x coordinate value (x _i-1 ) are compared in ST303. If the target x-coordinate value (x _i ) is greater than or equal to the previous x-coordinate value (x _i-1 ), the maximum x value is x _i , the minimum x value is x _i-1 , and the y value is y _i (= y _i ) in ST304. _-1 ), and set the direction to right.
Also, as a result of the comparison in ST303, the x coordinate value of interest
If (x _i ) is smaller than the previous x coordinate value (x _i-1 ), ST305
, The maximum x value is x _i-1 , the minimum x value is x _i , and the y value is y _i (= y
_i-1 ), and set the direction to left. In ST306, it is checked from the edge information obtained in ST304 and ST305 whether or not edges having the same y value and the same direction are registered in the horizontal line extraction table. ST30 if not registered
At 9, the edge is registered in the horizontal line extraction table. If registered, the size of the maximum y value is compared with the registered maximum y value in ST307. If the maximum y value is larger, the registered edge is deleted from the horizontal line table in ST308, and the edge is registered in the horizontal line extraction table in ST309. As a result of the comparison in ST307, if the maximum y value of the registered edge is larger, the edge is not registered and S
At T311, the next edge is extracted. If the edge number is greater than or equal to the number of coordinate points in ST310, the y coordinate values are sorted in ST312 and the process ends. On the other hand, if the edge number is smaller than the number of coordinate points, the next edge is extracted in ST311. As described above, in the case of horizontal lines having the same y value, a fixed rule is set to eliminate duplicates and the result of sorting the y coordinate values is registered. In this embodiment, the rule with the larger maximum x value is registered as the deduplication rule, but it goes without saying that another rule may be registered. Similarly, an example of the vertical line extraction processing will be described with reference to FIG.

First, in step ST400, a starting point for vertical line search is set. Step ST401
Then, the previous coordinate point (x _i-1 , y _i-1 ) and the target coordinate point (x _i-1
The difference (xw, yw) between _i , y _i ) is calculated. In ST402, the difference between the x coordinate values (xw) is 0 and the difference between the y coordinate values (y
Check whether the condition that w) is other than 0 is satisfied. If the condition of ST402 is not satisfied, the next coordinate point is set as the target coordinate point in ST411, and the process of ST401 is performed in the same manner. If the condition of ST402 is satisfied, ST4
At 03, the target y coordinate value (y _i ) and the previous y coordinate value (y _i-1 ) are compared. The target y-coordinate value (y _i ) is the previous y-coordinate value (y _i-1 )
If it is greater than or equal to ST404, the maximum y value is y _i and the minimum y value is y.
_{The i-1} and x values are set to x _i (= x _i-1 ) and the direction is set to up. In addition, as a result of the comparison in ST403, the y-coordinate value of interest
If (y _i ) is smaller than the previous y coordinate value (y _i-1 ), ST405
, The maximum y value is y _i-1 , the minimum y value is y _i , and the x value is x _i (= x
_i-1 ), and set the direction to down. In ST406, it is checked from the edge information obtained in ST404 and ST405 whether or not edges having the same x value and the same direction are registered in the vertical line extraction table. ST if not registered
At 409, the edge is registered in the vertical line extraction table. If registered, the maximum y value is compared with the already registered maximum y value in ST407. If the maximum y value is larger, the registered edge is deleted from the vertical line straight line table in ST408, and then the edge is registered in the vertical line extraction table in ST409. As a result of the comparison in ST407, if the maximum y value of the registered edge is larger, the edge is not registered and the next edge is extracted in ST411. If the edge number is greater than or equal to the number of coordinate points in ST410, the x coordinate values are sorted in ST412, and the process ends. On the other hand, if the edge number is smaller than the number of coordinate points, the next edge is extracted in ST411. In this way, in the case of vertical lines having the same x value, a certain rule is set to eliminate duplicates and register information after sorting. In this embodiment, the rule with the largest maximum y value is registered as the deduplication rule, but it goes without saying that another rule may be registered. Further, in the present embodiment, an accurate line pair can be recognized by managing the x-direction end point coordinates representing the horizon if the line is a horizontal line and the y direction end point coordinates representing the vertical line if the line is a vertical line. I chose FIG. 5 shows an example of each piece of edge information. The horizontal line extraction table 51 stores information about edges extracted as horizontal lines. In this embodiment, the y value, the maximum x value, the minimum x value, and the direction are stored in the table as horizontal line edge information. Similarly, the vertical line extraction table 52 stores information on edges extracted as vertical lines. Horizontal line extraction table 5
1. By storing information in the vertical line extraction table 52, it is possible to extract a line pair that has been difficult to extract conventionally. Next, an example of the line pair determination processing in the case of vertical lines will be described with reference to FIG.

First, in ST60, a vertical line for starting the line pair determination is set. In ST61, it is checked whether the pair flag of the vertical line n of interest is off. If on,
Since the vertical line n has already been extracted as a pair flag of another vertical line, it is compared in ST66 whether the vertical line number n is the vertical line number (M) or more, and if it is the vertical line number or more, the processing is performed. To finish. If the vertical line number n is smaller than the number of vertical lines, the vertical line number is incremented and the same processing is performed.
If the pair flag is off in ST61, a line pair of vertical line n (vertical line m) is extracted in ST62. Then ST63
Then, the pair flag of the vertical line n and the vertical line m extracted in ST62 is turned on. In ST64, it is determined whether the pair flags of all the vertical lines are on, and if all are on, the process ends. On the other hand, if the pair flags of all vertical lines are not turned on in ST64, it is compared whether or not the vertical line number n is equal to or more than the vertical line number.
The vertical line number is incremented at 65 and the same processing is repeated. Here, when the vertical line number is equal to or larger than the number of vertical lines, the processing is ended. Next, the line pair extraction processing in ST62 will be described with reference to FIG. Here, an embodiment will be described in which the vertical line m is focused in order to search for a line pair of the vertical line n.

First, in ST70, attention is paid to the vertical line m (m = 1) having the vertical line number 1 as a line pair candidate for the vertical line n. If the reference vertical line n is different from the vertical line m of interest in ST71, it is checked in ST72 whether the pair flag of the vertical line m is off. If it is off, it is checked in ST73 whether the directions of the vertical line n and the vertical line m are opposite. If the conditions of opposite directions are met in ST73, in ST74

[0022]

(1) (minym ≦ maxyn ≦ maxym) or

[0023]

## EQU2 ## It is checked whether (minyn ≦ maxym ≦ maxyn) is satisfied. If the condition of ST74 is satisfied, the vertical line m is extracted as a line pair of the vertical line n. ST7
The vertical line m cannot be a line pair of the vertical line n unless the conditions of 1, ST72, ST73, and ST74 are satisfied. Therefore, in ST76, the vertical line number n is incremented and the process is repeated. ST71, ST72, ST7
Even if there are a plurality of vertical lines that satisfy the conditions of ST3 and ST74, the horizontal line vertical line extraction unit sorts the y coordinate values for horizontal lines and the x coordinate values for vertical lines, and then registers them in the table. Therefore, it is obvious that the vertical line that first satisfies the condition may be determined as the line pair. 6 and 7 show the line pair extraction processing in the vertical line, but in the case of the horizontal line, the condition of ST74 is

[0024]

(3) (minxm ≦ maxxn ≦ maxxm) or

[0025]

## EQU4 ## The same processing is performed except that (minxn≤maxxm≤maxxn).

FIG. 8 shows a line pair extracted with respect to a vertical line by using the above-mentioned line pair determination processing. Vertical line pair 81 shows line pair information. A vertical line pair 81 indicates that x0 and x2, x1 and x3, x4 and x5, x6 and x8, and x7 and x9 are line pairs, respectively. According to the present invention, a correct line pair can be extracted even if it has a complicated character structure as shown in FIG.

[0027]

According to the present invention, for all horizontal and vertical lines that make up a character, the maximum x for generating horizontal lines is x.
By controlling the values, minimum x value, y value, line direction, and maximum y value, minimum y value, x value, and line direction that generate vertical lines, the correct line for characters with various characteristics Pairs can be extracted. As a result, by performing the line width correction, the original beauty of the character can be sufficiently expressed.

[Brief description of drawings]

FIG. 1 is a block diagram showing a system of the present invention.

FIG. 2 is a block diagram showing a line pair determination unit.

FIG. 3 is a flowchart showing an example of extracting a horizontal line.

FIG. 4 is a flowchart showing an example of extracting a vertical line.

FIG. 5 is an explanatory diagram of a table that manages extracted horizontal lines and vertical lines.

FIG. 6 is a flowchart showing an example of line pair determination processing for vertical lines.

FIG. 7 is a flowchart showing an example of line pair determination processing for a vertical line.

FIG. 8 is an explanatory diagram showing line pair information of “out” vertical lines.

[Explanation of symbols]

11 ... Input part, 12 ... Control part, 13 ... Character drawing part, 14
Output unit 131 Data read unit 132 Font storage unit 133 Coordinate conversion unit 135 Work memory 136 Fill unit 1341 Closed region separation unit
1342 ... Horizontal line / vertical line extraction unit, 1343 ... Line pair determination unit, 1344 ... Coordinate point correction unit.

Claims (5)

[Claims] 1. Input means for inputting an arbitrary character and size, output means for outputting a character, font storage means for storing character contour coordinate data forming an outline font, and said input means. Data reading means for reading the character contour coordinate data corresponding to the character code, coordinate conversion means for converting the character contour coordinate data to a desired resolution and size, and filling means for filling the inside of the generated contour coordinate points. In the character output unit including the line width correction unit that corrects the line width using the character contour coordinate data converted by the coordinate conversion unit, when extracting line pair information of the horizontal line and the vertical line, the maximum horizontal line is formed. A horizontal line vertical line extracting unit for extracting information on the value and the minimum value and information on the maximum value and the minimum value forming the vertical line; Using the extracted information in a vertical line extraction unit, each of the horizontal lines, contour extraction method characterized by comprising determining line pair determining section whether a line pair in a vertical line. 2. The orientation condition determining means for determining whether a line pair is a horizontal line or a vertical line when determining whether or not it is a line pair, and in the case of a horizontal line, a horizontal line contour coordinate y. A contour line extraction method having a coordinate value condition judging means using a range of coordinate values and x coordinate values, and in the case of a vertical line, a range of x coordinate values and y coordinate values of vertical line contour coordinates. 3. A contour line extraction method according to claim 2, wherein a line pair is judged to be a line pair when both conditions are satisfied. 4. The contour line extracting method according to claim 3, wherein the judgment condition is whether the directions of the contour lines are opposite to each other. 5. The method according to claim 3, wherein the x-coordinate value range of one horizontal line and the x-coordinate value range of the other have a common portion, and similarly, the y-coordinate value range of one vertical line and the y-coordinate value of the other vertical line. A contour line extraction method in which the judgment condition is that there is a common part in the range.