JPS6049314B2 - Character pattern correction method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION (1) Field of Application of the Invention The present invention is intended to solve the problem of missing character patterns in character pattern data of approximately 16 rows and 64 columns, which is expressed using a character pattern matrix. This relates to a method for correcting dot protrusion.

(2) Prior art In the past, several methods have been reported for correcting the roughness of diagonal lines when enlarging a character pattern (Japanese Patent Application Laid-Open No.
0-10925 character enlargement method, patent application 1984-12436
1 enlarged character correction method, etc.), but this is because it is known in advance that dots will be missing during enlargement, and there is no method for general character patterns like this method.

Therefore, in the past, corrections were specified in units of one dot by a human visually checking, resulting in a huge amount of work. (3) Purpose of the Invention The present invention provides a dot-type character pattern created using an optical scanner (facsimile, etc.), a character pattern created by enlarging or reducing an existing dot-type character pattern, etc. An object of the present invention is to provide a method for correcting missing dots and protruding dots.

(4) General description of the invention In order to achieve the above object, it is necessary to distinguish missing dots from intersections of straight lines, and protruding dots from corners of straight lines.

Therefore, in the present invention, two orthogonal sides around the dot are both dots of a different type from the dot, and the no(
If the dot is black, it is white; if it is white, it is black) If there are 3 or more dots (but not including intersections), it is considered to be an intersection of straight lines or a corner of a straight line, and the dots are not corrected, and at least one of the dots is If it is 2 dots, there is a missing dot or
A method is used in which the dots are treated as protruding dots and the dots are corrected. Figure 1 shows white dots that need to be corrected due to missing dots (indicated by ●), black dots that need to be corrected due to protruding dots (indicated by O), and white dots that need to be corrected due to straight line intersections. No white dot (indicated by an x)
, indicates a black dot (indicated by 2) that does not need to be corrected because it is a corner of a straight line.

Note that O indicates a black dot, and blank space indicates a white dot. According to the present invention, a(5b, c(5d) in FIG. 1 above are corrected separately. (5) Embodiments The present invention will be described in detail below with reference to embodiments.

FIG. 2 shows a circuit configuration of a character generator according to an embodiment of the present invention.

First, the dot data read out from the storage device for 1 hour is
The data enters the shift register 2 one row at a time, and is shifted one dot at a time by a shift pulse.

The data in the shift register is sequentially transferred to the shift registers 3 to 6 and the 5-bit 5-stage register of the pattern detection circuit 7 by a shift pulse. 1
When the row has been shifted, the next row is placed in shift register 2.

The 5-bit 5-stage register detects the dot patterns shown in a-h in FIG. Send out.

If there is no match, the circuit 8 sends out the original dot. Third
The diagram shows the pattern detection circuit 7 in FIG. ○ is a black dot, × is a white dot, Δ can be black or white, but both dots must not be black at the same time.
The mouth can be black or white, but it cannot be white at the same time; the blank space indicates a dot that can be black or white.

An example of only the detection circuit shown in FIG. 5 is shown in FIG.

Here, R,,3 indicates the j-th bit from the left of the i-th stage of the 5-bit, 5-stage register. P. ．． is ``1'' when the pattern in Figure 3 a is detected, and ``0'' otherwise.
It is a signal that becomes . The dot correction circuit 8 is a
This is a circuit for correcting dots when a -h pattern is detected, as shown in FIG. If the third
If even one of the patterns a to h in the figure is detected, 1 is applied to one terminal of the OR gate 11, and therefore, 1 is applied to one of the AND gates 12 and the OR gate 13 is applied.
1 is applied to one of the .

Therefore, the output of OR gate 13 is 1, and
1 is applied to one of the gates 15, and the input at the other terminal is output as is. On the other hand, since 1 is applied to one terminal of the AND gate 12, R2
, 3, the output of the AND gate 12 is 1
,0. Next, since it passes through the inverter 14, R3
, 3, 011 is applied to the AND circuit 15. Therefore, as mentioned above, and gate 1
Since 1 was applied to one terminal of 5, it is output as is. From the above, at least 1 of Pa-Ph
If it becomes 1, bit inversion of R3,3 occurs. Furthermore, if none of the patterns a to h in FIG. applied.

Therefore, the output of the AND gate 12 becomes O and is applied as 1 to the AND gate 15 through the inverter 14. On the other hand, since one terminal of the OR gate 13 was 0, the output of 1, 0 corresponds to 1, 0 of R3, 3.
is output, and similarly, 1, 0 are output from the AND gate 15.
is output. From the above, all of Pa-Ph are O
If so, output R3,3 as is. Next, the operation of the apparatus according to the embodiment of the present invention will be explained using the example shown in FIG.

FIG. 4A shows a pattern to be corrected, in which numbers 1 to 25 are numbers for distinguishing dots; those with 0 represent black dots, and those without 0 represent white dots.

The 5×5 matrix on the left side of B-R in the same figure is a simplified representation of shift registers 2 to 6 in FIG.
The 5×5 matrix on the right is the 5 of pattern detection circuit 7.
A staged 5-bit register is shown. In this device, the central bit (
That is, since the processing is performed by focusing on the third bit from the left in the third row, information for three lines is first stored in registers 2, 3, and 4.

If the contents of each register 2 to 6 are shifted by 2 bits from this state, the result will be as shown on the left side of FIG. 4B, and the contents of the detection circuit 7 will be as shown on the right side. FIG. 4B shows a state in which the center dot of the detection circuit 7 is 1, so processing is being performed on the 1 dot in FIG. 4A, but the pattern on the right side of FIG. Since both dots do not match, the 1 dot (that is, white) is output as is. As the pattern is further shifted one bit at a time, the pattern changes as shown in C, D, and E in the figure, and processing is performed on dots corresponding to 2, 3, and 4 of the original pattern, respectively.

When the shift register 2 becomes empty as shown in FIG. 4E, the leftmost column of the 5-stage 5-bit register is reset in order to prohibit connection between rows in the pattern detection circuit 7.

Furthermore, when information for I rows is to be input into the shift register 2 next from the state shown in FIG. As a result, pattern G in the figure is obtained. After one line of information has been put into the shift register 2, a blank shift is performed until the next dot 6 to be processed comes to the center position of the 5-stage 5-bit register.

As a result, a state like 1 is reached, and processing for dot 6 is performed. K in the same figure shows a state in which processing is performed on dot 8. In this state, since the pattern matches the pattern in FIG. Reversed to dots. In other words, the protruding portion of the dots in the original pattern of FIG. 4A is corrected correctly. The above-mentioned processing is continued even if there are no more lines to input until the last dot of the last line reaches the center position of the 5-stage 5-bit register.

(6) Summary As explained above, according to the present invention, it is possible to automatically correct missing dots and protruding dots in characters output by a kanji printer, etc., thereby improving character quality when printed and displayed. improves.

[Brief explanation of drawings]

Figure 1 is a detailed explanatory diagram of the present invention, Figure 2 is a first aspect of the present invention.
FIG. 3 is a schematic diagram showing a detection pattern, FIGS. 4A and 4B are operation explanatory diagrams, and FIGS. 5 and 6 are detailed diagrams of a part of the block in FIG. 2.

Claims (1)

[Claims] 1. In a method for correcting a character pattern expressed in a dot matrix, each white dot or black dot forming the character pattern has 3 or more black dots, or If it is a white dot, leave it as a white dot or black dot, one of which is 2 black dots or white dot and the other is 3 dots.
Black dots or white dots, or 2 on both sides
A character pattern correction method characterized in that if the dot is a black dot or a white dot, the white dot or black dot is corrected to a black dot or a white dot.