JPS6057264B2 - Binary pattern expansion method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION This invention aims to perform enlargement processing at an arbitrary magnification at high speed with a simple configuration and to maintain quality in an apparatus for processing character and graphic patterns represented by dot patterns.

Most conventional methods for enlarging image information represented by binary dot patterns are methods aimed at character-by-character processing; (i) all dots in the original pattern are corrected by referring to surrounding dot information; In addition, the method of enlarging the dots by replacing them with dots that are multiples of an integer; (ii) the method of detecting line segment patterns in the original pattern and expanding the line segments according to the enlargement rate; (iii) The original pattern is analogously approximated by a quadratic surface, the intervals between sample points are made finer, and the quadratic surface values at the finer sample points are binarized to create the pattern. There is a method that performs enlargement processing by determining the sample point interval and restoring the sample point interval to the original value.

These methods are intended for enlarging processing when displaying or recording character fonts, and method (i) makes it difficult to enlarge at any magnification and requires a large number of processing times, while method (ii) is based on the characteristics of kanji. None of the methods used had much meaning in the process of enlarging graphic patterns, and both methods (ii) and (ii) required a storage unit for processing and were complex in structure. Therefore, in order to apply enlargement processing at an arbitrary magnification in devices such as facsimile machines that process character strings and graphic patterns and input and output image information sequentially, it is necessary to divide them into character units or graphic block units. In addition, it required a large-capacity storage device, and there were problems in practical terms, such as a long processing time.

The present invention aims to perform high-speed enlargement processing of image information represented by a binary dot pattern at an arbitrary magnification using a small-capacity storage device while maintaining image quality, and is intended to be applied to facsimile machines and the like. It is.

An embodiment of the present invention will be described below based on the drawings.

The present invention provides an arbitrary magnification m/n (1<m/n x 2, mn positive integer) of characters and graphic patterns represented by binary dots.
In order to perform enlargement processing at high speed and maintain balance as characters or figures in the pattern after processing,
This is done by repeatedly inserting patterns with a total of m-n dot rows and m-n dot columns at equal intervals every n dots in the dot rows and dot columns that make up the original pattern. . However, when enlarging processing by inserting like this, especially with fine line width, 45
Step-like disturbances occur in the diagonal line dot pattern having 2 or 135 digits, significantly degrading the image quality.

FIGS. 1a to 1c show step-like disturbances that occur when a diagonal dot pattern having an inclination of 1 to 3 dots with a line width of 45 is enlarged by a factor of 413 by the above enlargement process.

In this case, in order to insert a pattern of 1 dot row and 1 dot column every 3 dot rows and 3 dot columns, the insertion position is relative to the diagonal line pattern. Three cases occur, and three disturbance patterns occur for each line width. In order to correct the disturbance of the diagonal dot pattern with a 1 to 3 dot line width of 45 to 135 due to such insertion enlargement processing, the enlargement processing is performed on the original pattern. 1 to 3 that cause disturbances
Check whether a diagonal dot pattern of dots exists or not, as shown in Figure 2. AO, al, a2, a3, a4, a5, a6,
A2, al whose vertices are the 3×3 dot square area consisting of a7 and a8 and the dot at the intersection.

, all, al2, al3, a39aO? Al9a99
a49a39al39al49al59al69al7
9a59aO9a69a79aO9a59al79a1
89a199a20, a21, a8, a24, a9, a
1 to 20 in FIG. 3 indicate patterns serving as determination criteria, which are determined based on patterns within four 3×3 dot square areas consisting of 1, a0, a7, a21, a22, and a23. Here, Figure 3 A is AO centered on Momo? Al9a
The pattern in the 29a39a49a59a69a7, a8 area is shown, and FIG. 3B shows the pattern in another square area area whose vertices are ~. A pattern that matches the determination pattern shown in FIG. 3 exists in the area shown in FIG. 2 near the intersection of insertion dot row 1 and insertion dot column j, which are to be L-inserted after the original dot pattern. In this case, the position of the 3×3 dot square area and the pattern types 1 to 20 are stored, assuming that the disturbance occurs after the insertion and enlargement processing. After execution of the insertion enlargement process, regarding the stored 3 x 3 dot square area, in the case of 1 to 4, insert dot row 1 and insert dot column j, and each of the subsequent inserted dot rows and dot columns. In the 4 x 4 dot area (Figure 4) centered on the 2 x 2 dot intersection area of the intersection, 5 to 2
In the case of 0, 4 with the intersection area of 2 × 2 dots as the vertex
In the ×4 dot area (Fig. 4), the Kg shown in Fig. 4
〜B3, Cl〜Diagonal line Bl of the square area shown in C3,
By replacing the 4-dot diagonal line pattern on Cl and the 3-dot diagonal pattern on each of the diagonal lines B2, b3, C2, and C3 on both sides with correction patterns in accordance with the detected pattern type. Perform correction processing. FIG. 5 shows the correspondence between the pattern after insertion and enlargement processing and the correction pattern to be replaced, and shows the insertion dot rows 1 and 2 corresponding to the judgment patterns 1 to 20 before insertion and enlargement shown in FIGS. In the pattern enlarged by insertion after insertion dot row j, the intersection area of 2 × 2 dots of the intersections of insertion dot row 1 and insertion dot row j, and the dot row and dot row inserted after that is 4x4 dots with center or apex
The enlarged pattern when viewed diagonally as shown in the figure is associated with the correction pattern to be replaced.

FIG. 6 shows the correction procedure in the case shown in 1 of FIG.

In this case, since there is a noise component in the C1 direction, a pattern is detected in the C1 direction, and when an enlarged pattern as shown on the left in 1 of Fig. 5 is obtained, it is replaced with a correction pattern as shown on the right. The correction is completed. The types of correction patterns shown in FIG. 5 are limited to specific ones as shown in FIG.

FIG. 8 shows an undisturbed pattern obtained when the correction process shown in FIG. 5 is performed on the enlargement process shown in FIG. 1.

A simple processing configuration of the present invention will be explained with reference to FIG.

The input section includes a reading section 21 for inputting the original pattern, and an input register 22 having a capacity of one dot row of the original pattern. The correction/enlargement processing section uses an input memory 23 having a capacity of 5 dot rows to determine whether the reference pattern shown in FIG. 3 exists for the square area shown in FIG. 2, and if so, its position and information. A detection section 24 that detects information on the pattern type, an insertion section 25 that inserts dot rows and dot columns, an output storage section 26 that stores the enlarged pattern and has a capacity for 6 dot rows of the enlarged pattern, and a 4×4 dot a correction register 31 that performs correction processing on the square area;
In response to an instruction from the detection unit 24, the output storage unit 26 outputs data from the output storage unit 26 to the correction register 31, and from the correction register 31 to the output storage unit 2.
Gate section 2 that transfers a 4x4 dot square area to 6
7. Correction pattern register 29 that outputs a correction pattern to correction register 31 according to instructions from detection unit 24
, 30. The output section includes an output register 28 having a capacity of one dot line of the enlarged pattern, and an output section 32 for outputting the enlarged pattern. Next, its operation will be explained.

The original pattern is input from the reading section 21, transferred to the input storage section 23 one dot row at a time via the input register 22, and the pattern detection section 24 detects the square shown in FIG. 2 near the intersection of insertion rows j in insertion example 11. It is determined whether the reference pattern shown in FIG. 3 exists in the area, and if there is a reference pattern, its position and pattern types 1 to 20 are detected and stored. When the detection is completed, one dot row at a time is inserted through the insertion section 25 for enlargement and transferred to the output storage section 26, and the enlarged pattern to be corrected by the gate section 27 is transferred to the correction register 31. , the correction pattern is replaced with the correction pattern in the correction pattern registers 29 and 30 according to an instruction from the detection unit 24, and the correction is completed. After the correction, the gate section 27 transfers the pattern back to the output storage section 26, and the corrected enlarged pattern is outputted to the output section 32 via the output register 28. As described above, according to the present invention, enlargement processing at an arbitrary magnification of characters and graphic patterns represented by binary dots is performed by mechanically repeating insertion into dot rows and dot columns that constitute the original pattern. Detecting the disturbance of the diagonal dot pattern caused by the above enlargement process in a 3 x 3 dot square area centered on the intersection of the insertion dot row and insertion dot column, and detecting the 4 x 4 dot square area after the above enlargement process. By performing correction by area correction processing, it is possible to provide correction and enlargement processing at an arbitrary magnification that maintains quality at high speed with a small storage capacity and a simple configuration.

[Brief explanation of drawings]

Figure 1a is an explanatory diagram showing the disturbance of the diagonal line pattern with a 1-dot line width when 41S is inserted and enlarged. The figure shows correction information detected in the original pattern 3
An explanatory diagram of the position of the square area of ×3 dots, Figure 3 A, B
is an explanatory diagram showing the correction criteria pattern for the 3 x 3 dot square area shown in Fig. 2, and Fig. 4 is an explanation of the correction position for the 4 x 4 dot square area which is an enlarged version of the square area shown in Fig. 3. 5 is an explanatory diagram showing the correspondence between the judgment reference pattern shown in FIG. 3 and the correction pattern, and FIG.
The figure is an explanatory diagram showing the correction processing procedure in case 1 of Fig. 5, Fig. 7 is an explanatory diagram showing the types of correction patterns in Fig. 5, and Fig. 8 a-c are shown in Fig. 1 a-c. FIG. 9 is an explanatory diagram showing a pattern after correction processing has been performed on the enlargement processing. FIG. 9 is a processing configuration diagram showing an embodiment of the present invention.

Claims (1)

[Claims] 1 Enlarge a character or figure pattern represented by binary dots by an arbitrary magnification m/n (1<m/n<2, mn a positive integer) at intervals of n dot rows and n dot columns, m - A processing method that is performed by repeatedly inserting patterns of n dot rows and m-n dot columns at equal intervals, and whether a diagonal dot pattern exists on the original pattern to correct the disturbance that occurs in the diagonal dot pattern. At all intersections of dot rows and dot columns that are the insertion positions of m-n dot rows and columns between the above n dot rows and columns, a 3×3 dot area with the intersection as the center or the intersection as the vertex. After performing the enlargement process by inserting the area, a 2 x 2 dot area is centered or For the 4 x 4 dot area as the vertex, 4 on the diagonal
A binary pattern characterized in that the dot pattern and each 3-dot pattern on both sides of the diagonal line are replaced with a fixed pattern based on the determination result to correct the disturbance of the diagonal line pattern in enlargement processing at an arbitrary magnification. Expansion method.