JP2839806B2 - Resolution improvement method in image forming apparatus - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for improving resolution in an image forming apparatus.

[0002]

2. Description of the Related Art RE technologies for image enhancement related to resolution enhancement (hereinafter referred to as RE) include US Pat. Nos. 4,437,641 (Xerox) and US Pat. Nos. 4,847,641.

In these patents, pixels of M columns × N rows around a bit to be subjected to RE are used as a template, and a plurality of predetermined pixel patterns of M columns × N rows are used. The direction of the shift and the presence or absence of the shift are determined by the pattern match.

FIG. 23 is a view for explaining jagged correction of a line approximating a horizontal line for a pixel.

For example, in the case of a pattern as shown in FIG. 23 (a), a plot as shown in FIG. 23 (b) is originally provided as shown in FIG. 23 (c) to prevent jaggies (jaggies). Is a method of plotting small dots on The RE based on this pattern matching was effective in preventing jaggies in lines approximating a horizontal line and in preventing jaggies in lines approximating a vertical line.

[0006]

However, the conventional pattern
The method based on matching is effective in preventing jaggies, but is less effective in correcting broken lines (font decoration, SERIF) and correcting toner intersections. When performing pattern matching to determine whether an intersection or non-intersection or a serif or a non-intersection is required in comparison with jaggies, it is necessary to increase the size of the template and increase the number of patterns. As a result, the number of gates when executed by the ASIC increased, and the number of gates and speed became a bottleneck, and the effect was small.

SUMMARY OF THE INVENTION An object of the present invention is to provide an image forming apparatus having an image output device with low and medium resolutions, by shifting a pixel, changing a pixel size, and adding a pixel with respect to an image to be processed, so as to realize an apparent resolution. It is an object of the present invention to provide a method capable of improving the above.

[0008]

According to the present invention, there is provided an image forming apparatus provided with a counting means for counting the number of pixels and a pixel control means for controlling pixels forming an image to be output. Setting a template that defines the peripheral pixels of the pixels that form the image to be formed, and counting the number of pixels in each row and each column of the set template;
Counted comparison result in the rows of each row the number of pixels and or <br/> pixel shift based on a comparison result between each column of the number of pixels in each column, the change of the pixel size, and additional pixels Controlling whether or not any one of the operations is performed by the pixel control means.

[0009]

[Action] counts the number of pixels each row and each column of setting the template that defines the peripheral pixels of pixels forming an image to be output counting means the set template, each line pixel control means is counted comparison及<br/> beauty or pixel shift based on a comparison result between each column of the number of pixels each column in between rows of the number of pixels, the change of the pixel size, and any of the additional pixel By controlling the presence or absence of one implementation, it is possible to improve the apparent resolution by performing jaggy correction, crushed serif correction, and toner accumulation without increasing the storage capacity of the bitmap memory, and clear to the image output device Can output a simple image.

When performing RE according to an embodiment of the present invention,
The embodiment of the present invention has the following features.

(1) When performing jagged correction of a line approximating a vertical line, the peripheral pixels of the processing pixel are used as a template, and whether or not the processing pixel is shifted in one direction (right direction) is determined. It can be determined whether to shift in the direction (left direction).

(2) A jagged line approximating the vertical line in (1)
When making corrections, the number of blacks in each column of the template is counted as a criterion for determining whether to shift right or not, or shift left or not, and the difference between the number of blacks in the left and right columns is calculated. And the number of black columns at the center, it is possible to determine the direction of the shift and the presence or absence of the shift.

(3) When performing jagged correction of a line approximating a horizontal line, pixels surrounding the processing pixel are used as a template, and whether or not to add a pixel (dot) and the number of dot sizes are determined. Can be determined.

(4) When performing jagged correction of a line approximating a horizontal line, the number of blacks in each row of the template is counted, and the difference between the number of blacks in the upper and lower rows and the black in the center row are counted. By comparing the numbers, it is possible to determine whether or not to add a dot and what percentage of the duty of the dot.

(5) A template for performing jagged correction of a line approximating a vertical line may be different from a template for performing jagged correction of a line approximating a horizontal line. it can.

(6) In the case of performing a corrected serif correction, it is possible to use the peripheral pixels of the processing pixel as a template to determine whether to plot the dots or to determine the size of the dots.

(7) When performing the above-mentioned crushed dialogue correction,
The number of blacks in each row of the template is counted, the size of each row is compared, the number of blacks in each column is counted, and the size of each column is compared. Or whether to change the dot size.

(8) The template for correcting the above-mentioned crushed dialogue can be different from other templates.

(9) When performing toner rounding correction at an intersection,
Using the peripheral pixels of the processing pixel as a template, it is possible to determine whether or not to change the dot size.

(10) When performing toner rounding correction at the intersection, the number of blacks in each row of the template and the number of blacks in each column are counted, and the dot size is changed by comparing the sizes. Or not.

(11) The above four processes, ie, jaggies correction of a line approximating a vertical line, jaggies correction of a line approximating a horizontal line, correction of crushed serifs, and correction of a toner round at an intersection can be simultaneously performed. .

(12) When counting the number of black pixels, the closer the pixel is to the processing pixel, the greater the weight can be.

(13) When counting the number of blacks in a row because the line is thicker than one dot, take the exclusive OR with the immediately preceding row and determine the number of blacks in the adjacent column. When counting, the exclusive OR with the immediately preceding column can be taken.

Taking an electrophotographic printer as an example, according to the embodiment of the present invention, in the example of a line approximating the horizontal line in FIG. 23A, if RE is not performed, the result is as shown in FIG. 23B. Fig. 23 (c) shows the result of the addition.
Correction is made. Also, taking a line similar to the vertical line in FIG. 1 (a) as an example, if RE is not performed, the line will be as shown in FIG. 1 (b). However, by shifting the dots in the horizontal direction, a line without jaggies (FIG. It is possible to print a clean line such as 1 (c)).

In the example shown in FIG. 2A, by changing the dot size and adding a dot, it is possible to correct the broken lines (FIG. 2B) (FIG. 2C).

In the example of FIGS. 3 (a) and 4 (a), the dot size at the central intersection (FIG. 3 (b) and FIG. 4 (b) is reduced (FIGS. 3 (c) and 4 (b)). c)) Toner drift correction can be performed.

[0027]

DESCRIPTION OF THE PREFERRED EMBODIMENTS An RE according to an embodiment of the apparatus of the present invention will be described in detail for each correction mode.

First, jagged correction of a line approximating a vertical line will be described.

Using the peripheral pixels of the processing pixel as a template, the number of blacks in each column is counted, and the number of blacks in the left and right columns and the number of blacks in the center column are compared to determine the shift. Decide the direction and shift.

In FIGS. 5 and 6 (a) to 6 (e), the template 10 will be described as 3 × 7 (pixels). Pixels to be processed (hereinafter, referred to as hatched rectangles, hereinafter referred to as processed pixels) take a template of 3 × 7 and sum the sum of black pixels in each column. In FIG. 6A, the sum of “1” (black) in the left column of the column including the processing pixel is 3, and the total of “1” (black) in the column including the center processing pixel is 4. The sum of “1” (black) in the column to the right of the column including “” is zero. As described above, since the sum of black in each column is 3, 4, 0, the processing pixel is shifted to the left. This shift amount may or may not be changed depending on the difference between the number of “1” in the left column and the number of “1” in the processing pixel. In addition, the weight may be or may not be weighted toward the center of the seven pixels.

In FIG. 6B, the number of "1" (black) in the left column is 0, the number of "1" (black) in the column of the processing pixel is 4, and "1" (black) in the right column. Is 3. Comparing the sum of the three columns, the processed pixel shifts to the right because the sum of the left column is smaller than the sum of the right column.

In FIG. 6C, the number of "1" (black) in the left column is 1, the number of "1" (black) in the column of the processing pixel is 5, and "1" (black) in the right column. Is 1. Since the sum of the three columns is compared, the sum of the left column is equal to the sum of the right column, so that the processing pixel is not shifted. Even if the sum of the left column and the sum of the right column are equal, it is possible to shift by increasing the weight closer to the processing pixel. However, in this example, although it does not matter whether the weight is increased, weighting may or may not be performed.

In FIG. 6 (d), the number of "1" (black) in the left column is 2, the number of "1" (black) in the column of the processing pixel is 3, and the number of "1" (black) in the right column. Is 2. Since the sum of the three columns is compared, the sum of the left column is equal to the sum of the right column, so that the processing pixel is not shifted.

In FIG. 6 (e), the number of "1" (black) in the left column is 6, the number of "1" (black) in the column of the processing pixel is 1, and "1" (black) in the right column. Are zero. Comparing the sum of the three columns, the processed pixel shifts to the left because the sum of the left column is greater than the sum of the right column.

The above processing is performed on a scanning line as shown in FIG. The uppermost column has no upper part of the processing pixel, so that the second column from the top is as shown in FIG. 8B, and the third column from the top is the one shown in FIG. As shown in c), the fourth row from the top takes a template as shown in FIG. The template processing is performed in the same manner as approaching downward. Also,
When scanning on a scanning line, not only jagging correction of a line near the vertical line but also the following processing is performed at the same time.

(A) Jagi correction of a line close to a horizontal line (B) Collapsed serif correction (C) Toner dull correction As described above, jagg correction of a line approximating a vertical line is as shown in FIG. 9 (a). FIG. 9 (b) shows a horizontal template, and FIG. 9 (c) shows a horizontal template, and FIG. 9 (c) shows a jaw correction for a line near a horizontal line. A template close to a square is used. For jagged correction of a line near the horizontal line, count the number of blacks in each row, and for the other two processes, crushed serif correction and toner stagnation, count the number of blacks in both columns and rows. .

The jagged correction of near-vertical lines will be described in further detail, and the advantages of the present invention will be briefly described as compared with a general apparatus method.

FIG. 10 shows an example in which the template is 3 × 7. P13 in FIG. 10 is the processing pixel output by the laser and P0
21 pixels from 0 to P26 are templates. As shown in FIG. 12, the number of blacks P00 to P06, P10 to P16, and P20 to P26 is counted by the counting means 11, respectively. Each count result x00 to x02 shown in the black number counting circuit described later
, X10 to x12 and x20 to x22 are determined by the control signal generator 13.
The comparison is made by the size comparison unit 12 constituting the pixel control means. X00 to x02, which is the number of blacks of P00 to P06, are P20 to P
The control signal generator 13 generates a signal that shifts to the left if it is larger than x20 to x22, which is the number of blacks of 26, and generates a signal that shifts to the right if it is smaller. If they are the same, do not shift.

FIG. 11 shows a circuit example of the counting means. Sign
15 to 22 are exclusive OR circuits, 23 to 33 are OR circuits, and 34 to 42 are AND circuits, and signals corresponding to respective pixels are input from the left side of the drawing (P00 to P06), Number of black x00
~ X02 is output.

According to the general pattern matching method, there is a drawback that the circuit scale increases as the number of patterns increases. However, this black counting method can provide a great effect with a very small circuit scale as shown in FIG.

Although the example in which the line thickness is one dot has been described above, it is often the case that the line thickness is not actually one dot. This is because when the number of blacks in a row is counted because the line thickness is thicker than one dot, the exclusive OR of the previous row is taken and the number of blacks in the adjacent column is counted. Takes an exclusive OR with the immediately preceding column. Fig. 13 (a)-(c), Fig. 14
As shown in (a) to (c), bitmap data one column larger than the template for counting the number of blacks shown in FIG. 13 (a) is processed. When counting is performed for each column, an exclusive OR of two pixels in adjacent columns of the bitmap data to be printed is obtained.

The exclusive OR of P00 and P10 in FIG.
The exclusive OR P011 of P01 and P11 and the exclusive OR of P10 and P20 are designated as P120, and the exclusive OR is taken, and the resulting P010 to P236 is taken as the sum of the above black. In this case, P010 to P236 do not need to be latched, so the circuit to be added is only an exclusive OR circuit. In this case, the circuit to be added is equivalent to one pattern of a method using pattern matching of a general device. Only. Figures 13 (a), (b), (c),
When the exclusive OR of the bitmap data shown on the left side of each of FIGS. 14 (b) and (c) is taken, FIGS. 13 (a), (b), (c) and FIG.
(b) and (c) are shown on the right side of FIG.
(a) to (e). Since FIGS. 6 (a) to 6 (e) have already been described, even when the thickness is not 1 dot, it is possible to easily cope with the circuit without increasing the circuit scale by using the exclusive OR circuit shown in FIG. it can.

In the case of exclusive OR as shown in FIG. 15 (a), there are two cases shown in FIGS. 15 (b) and (c). FIG. 15B shows a case where the color changes from black to white, and FIG. 15C shows a case where the color changes from white to black. In FIG. 15 (b), the processing pixel is
The pixel of FIG. 15 (c) is the pixel of P13 of FIG.
P23 pixel. To make this distinction, the AND circuit shown in FIG. 17 may be used. In FIG. 14 (a), P123 is "1".
When P13 is “1”, it is the change from black to white shown in FIG. 15 (b), and when P123 is “1” and P13 is “0”, FIG.
This is the change from white to black as shown in (c). This can be performed in parallel, no latch is required, and the circuit scale does not increase so much.

The horizontal line jagging correction is basically the same as the vertical line jagging correction. Differences are as follows: (a) In the jagged correction of a line close to the vertical, the number of blacks in each column is counted, but in the jagged correction of a horizontal line, the number of blacks in each row is counted. (B) In jagging correction of a line that is close to vertical, the processing content is to shift right, left, or not, but jagging of a horizontal line is performed by changing the dot size and adding dots. is there.

FIGS. 18 (a) and 18 (b) show examples of jagged correction of a line approximating a horizontal line. The template is shown as 7 × 3. The number of black pixels in the top row is 4, the number of black pixels in the row including the processing pixel is 3, and the number of black pixels in the bottom row is 0. If the difference in the number of blacks between the upper row and the lower row is greater than a certain size, the dot size is changed. Fig. 18 (a)
Since the processing pixel is 0 (white), processing is performed to reduce the dot size.

In FIG. 18 (b), the number of black pixels in the top row is 3, the number of black pixels in the row including the processing pixel is 4, and the number of black pixels in the bottom row is 0. It is. Also in this case, the dot size is changed. In FIG. 18A, the processing pixel is white, but in FIG. 18B, the processing pixel is black.
In (a), the size is made larger than the size of the originally added dot and smaller than the normal dot size. Jaggi correction of a line near a horizontal line when the thickness is not 1 dot can be performed by performing an exclusive OR operation in the same manner as jaggies correction of a line near a vertical line. However, it goes without saying that the rows are not adjacent columns but are adjacent rows.

Next, the correction of a broken line will be described. The correction of the crushed dialogue means that the crushed dialogue is sharpened by reducing the pixel (dot) diameter because the resolution of the portion surrounded by the broken line is poor as shown in FIG. (b)).

In the processing described so far, either the x direction or the y direction is used to count the number of black dots. However, the number of black dots in the x direction and the y direction is counted for correction of the crushed dialogue. Although the template has been rectangular in the past, a shape close to a square is suitable for correcting a crushed dialogue. Count the number of black dots in each row of the template and the number of black dots in each column of the template, compare the size of each row and each column, and add a dot based on the result. Decide if you want to resize.

Next, an example will be described. Figure from Figure 20 (a)
Assuming that the oblique-lined rectangular portion of FIG. 20 (d) is a processing pixel, the four templates of FIGS. 20 (a) to 20 (d) are used. Fig. 20 (a)
20 (b) is facing left, FIG. 20 (c) is facing downward, and FIG.
(d) is a template for the right-pointing dialogue.

First, the upward dialogue will be described. The template performs processing on the scanning lines using FIG. 20 (a). If there is an acute angle pattern in the upward direction, processing with the template of FIG. 20 (a) will result in black dots appearing on the template as shown in FIG. 21 (e). Black dots appear on the template as shown in FIG. 21 (d). Fig. 18 (d) also shows Fig. 18.
In (e) also, no processing is performed because the number of black dots in the center row is zero.

When processing the line below, the template is such that the sum of the number of black dots in each line is from the top as shown in FIG.
(0,0,1,3,5), and the sum of the number of black dots in each column is from the left
(1,2,3,2,1). The condition of the acute angle in the upward direction is as follows: the number of blacks in the top row ≤ the number of blacks in the second row ≤ the number of blacks in the third row ≤ the number of blacks in the fourth row ≤ the number of blacks in the fifth row Since it is the number of blacks, among the above conditions, if it is more beautiful to add a small dot, add a small dot. The sum of the numbers of black dots in each row is (1,2,3,2,1), and the number in the center row is large. As described above, when the number of black dots in the center row is large, a small number of dots is added.
Thus, there is a row condition and a column condition, but only when both are true, a small dot is additionally plotted. The size of the dot diameter at this time is determined by the sum of the numbers of black in each row. In this example, (0,0,1,3,5) is smaller than (0,0,1,2,3) because this is an obtuse angle compared to (0,0,1,2,3). Add a dot.

When the processing of the lower row is performed, the template is as shown in FIG. Also in this case, the above-mentioned condition is applied, so that additional plotting is performed. In addition, the sum of the black dots in each row is (0,1,3,5,5), which is closer to the processing pixel than (0,0,1,3,5). Becomes larger. The actual plotted dots are shown in Figure 20
(h). Further, in the processing in the lower row, the template
The result is as shown in FIG. Here, since the processing pixel is black, it is not necessary to add a dot. Therefore, it is plotted normally and plotted as shown in FIG. Similarly, the plot is as shown in FIG. 21 (c), and the collapsed dialogue can be corrected by adding the above two dots. In this way, the number of black dots in each row of the template is counted, the size is compared, and the same processing is performed on each column, so that the broken lines can be corrected.
Although the description has been made using the upward serif as an example, the same applies to other directions.

Next, the toner pool will be described. Toner accumulation correction is a process for preventing toner from accumulating at the intersection of lines, and is handled by reducing the dot at the intersection. This is also possible by counting the sum of the number of black dots in each column and the sum of the number of black dots in each row, and judging whether or not to reduce the dot diameter by comparing the sizes.

This will be described with an example. First, the template determines peripheral pixels of the processing pixel so that the processing pixel is located at the center as shown in FIG. In the case where toner stagnation occurs, the sum of the number of black dots in each row and the sum of the number of black dots in each column both increase in the center as shown in FIGS. 22 (a) to (c).
As shown in FIG. 22 (d), the number of black dots in each column and the number of black dots in each row are small at the center. Therefore, counting the number of black dots in each row and the number of black dots in each column,
If the above conditions are satisfied, the size of the dot may be reduced.

[0055]

According to the present invention, the counting means sets a template which defines the peripheral pixels of the pixels forming the image to be output, counts the number of pixels in each row and each column of the set template, and the pixel control means counts. Comparison of the number of pixels in each row between rows
The bitmap memory controls whether or not to perform any one of the pixel shift, the pixel size change, and the pixel addition based on the result and / or the comparison result between the columns of the number of pixels in each column. , The apparent resolution can be improved, and a clear image can be output to the image output device.

[Brief description of the drawings]

FIG. 1 is a diagram illustrating jagged correction of a line approximating a vertical line for a pixel.

FIG. 2 is a diagram for explaining dialogue correction for pixels.

FIG. 3 is a diagram illustrating a first toner rounding correction for a pixel.

FIG. 4 is a diagram illustrating a second toner accumulation correction for pixels.

FIG. 5 is a diagram illustrating a template.

FIG. 6 is a diagram illustrating pixel shift.

FIG. 7 is a diagram illustrating horizontal scanning of a pixel.

FIG. 8 is a diagram illustrating setting of a template.

FIG. 9 is a diagram illustrating one embodiment of a template.

FIG. 10 is a diagram illustrating details of an embodiment of the template.

FIG. 11 is a diagram illustrating a circuit diagram of a counting unit.

FIG. 12 is a diagram illustrating a counting unit and a pixel control unit.

FIG. 13 is a diagram illustrating first bitmap data.

FIG. 14 is a diagram illustrating second bitmap data.

FIG. 15 is a diagram illustrating third bitmap data.

FIG. 16 is a diagram illustrating an exclusive OR circuit.

FIG. 17 is a diagram illustrating an AND circuit.

FIG. 18 is a diagram illustrating a first jagged correction of a line approximating a horizontal line.

FIG. 19 is a diagram illustrating a first dialogue correction.

FIG. 20 is a diagram illustrating a second dialogue correction.

FIG. 21 is a diagram illustrating a third dialogue correction.

FIG. 22 is a diagram illustrating a toner pool.

FIG. 23 is a diagram illustrating a second jagged correction of a line approximating a horizontal line.

[Explanation of symbols]

 10 Template 11 Counting means 12, 13 Pixel control means

──────────────────────────────────────────────────続 き Continued on the front page (58) Field surveyed (Int.Cl. ⁶ , DB name) G06T 5/20-5/30 B41J 2/485 H04N 1/409

Claims (1)

(57) [Claims] 1. An image forming apparatus comprising: a counting means for counting the number of pixels; and a pixel control means for controlling a pixel forming an image to be output, around a pixel forming an image to be output by the counting means. set the template which defines the pixel, the steps of counting the number of pixels each row and each column of the set template, the number of pixels counted each row
Of the comparison result between each row and / or the number of pixels in each column
Controlling the presence or absence of any one of pixel shift, pixel size change, and pixel addition based on the comparison result by the pixel control means. How to improve.