JPH1056561A - Image processor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To satisfactorily convert the resolution of an input image while suppressing degradation in picture quality after resolution conversion by synthesizing correspondent patterns coincident with two predetermined thresholds and resolution converted image data when an area suitable for the patterns set by using the respective thresholds is detected. SOLUTION: The inputted image data of resolution 400dpi, for example, are stored from an input terminal A to a line buffer 11 successively from the leading picture element of a leading line and outputted to a suitable pattern detector 12 and a pattern separator 13. When any one of 18 kinds of patterns is detected, for example, at the pattern detector 12, the value of a concerned picture element in that pattern is inverted by the pattern separator 13 and inputted to a resolution converter 14. Binary image data of which the resolution is converted to 600dpi, for example, from the resolution converter 14 and the binary pattern data of 600dpi selected by a pattern generator 15 are synthesized by a synthesizer 16 and outputted to an output terminal B.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to an image processing apparatus, and more particularly to an image processing apparatus for converting the resolution of an image.

[0002]

2. Description of the Related Art In recent years, the resolution of a print engine has been rapidly increasing with the development of xerographic technology and thermal ink jet technology. Recently 600 dpi
A print engine having a resolution as high as about (dpi = dot per inch, the number of pixels per inch) is mounted on a copying machine, a printer, a facsimile machine, and a composite machine thereof.

[0003] Images input to such a print engine have various resolutions, and do not always match the resolutions of the print engine. At this time, if the input image is printed out as it is, if the resolution of the input image is lower than the resolution of the print engine,
The image is reduced and printed out. For this reason, small parts of small characters and patterns are crushed. Conversely, if the resolution of the input image is higher than the resolution of the print engine, the image is enlarged and printed out.

[0004] For this reason, for example, there arises a problem that an image originally having an A4 size does not fit on an A4 sheet but partially protrudes. Therefore, it is necessary to convert the resolution of the input image so as to match the resolution of the print engine and print out the image. Various methods, such as a nearest neighbor method, a logical sum method, and a projection method, have been proposed as methods for performing such resolution conversion. However, in particular, when the resolution conversion ratio is not an integer, for example, when the resolution is converted from 400 dpi to 600 dpi (the resolution conversion ratio is 1.5), the image quality tends to deteriorate.

[0005]

As a method for solving the above problem, for example, there is a resolution conversion method disclosed in Japanese Patent Laid-Open No. Hei 4-240970. According to this method, an isolated black pixel (a black pixel whose peripheral pixels are all white pixels) is used by using a plurality of monitoring areas having different sizes at the time of resolution conversion.
Is detected according to the degree of isolation, and the detected isolated black pixel is divided into a plurality of black pixels according to the size of the monitoring area.

That is, isolated black pixels generated in a low-density area (highlight area) of an image are dispersed to improve graininess in the area. However, if the resolution conversion ratio is not an integer, in order to match the shading in the area before and after the resolution conversion, it is necessary to divide the same isolated black pixel into a case where it is dispersed and a case where it is not dispersed, which complicates the processing.

Another method is disclosed in, for example,
There is a resolution conversion method disclosed in Japanese Patent No. 44322. This method converts the target pixel into a plurality of pixels according to the pattern of the target pixel and its surrounding pixels. According to this method, resolution conversion can be performed while preserving isolated black pixels or one-dot fine lines.

However, if the pattern detection range is widened, the number of conversion patterns becomes enormous, resulting in an increase in circuit scale or a reduction in processing speed. On the other hand, if the range in which the pattern is detected is narrowed, the pattern is converted into the same pattern irrespective of the situation around the detected pattern, so that the image quality after conversion tends to deteriorate.

As another method, for example, Japanese Patent Laid-Open No. 2-2
There is a resolution conversion method disclosed in JP-A-39280. In this method, an input binary image is converted into a multi-valued image by a projection method, the resolution is converted, and the multi-valued input image after the resolution conversion is re-binarized by an error diffusion method. Since the multi-valued image is estimated from the input binary image and the resolution is converted and binarized again, the image quality hardly deteriorates.

A well-known document relating to the projection method is described in "A Study of Facsimile Line Density Conversion" by Arai and Yasuda (Journal of the Image Electronics Society of Japan, Vol. 7, No. 1, page 11, pages 18-18).
P., 1978), and well-known documents relating to the error diffusion method include M.S. R. "Images" by Schroeder
from Computer ”(IEEE Spee)
ctrum, vol. 6, 1969).

However, when the above technique is applied to an image which has been binarized by the error diffusion method, which is often used when transmitting a picture such as a photograph by facsimile, when the low density portion is binarized by the error diffusion method, Appearing isolated black pixels are converted into a plurality of connected black pixels to form an unsightly noise pattern, or conversely, isolated black pixels disappear, and the grayscale expression of the original image and the image after resolution conversion do not match. Further, the same phenomenon occurs for an isolated white pixel (white pixels whose peripheral pixels are all black pixels) that appear when the high-density portion is binarized by the error diffusion method.

SUMMARY OF THE INVENTION The present invention has been made to solve the following problems, and an image processing apparatus capable of satisfactorily converting the resolution of an input image while suppressing deterioration in image quality after resolution conversion is provided. The purpose is to provide.

[0013]

In order to achieve the above object, an image processing apparatus according to a first aspect of the present invention comprises a resolution conversion means for converting the resolution of input image data, and further comprises: Pattern detection means for detecting an area matching a predetermined specific pattern from the data, corresponding pattern generation means for generating a corresponding pattern defined corresponding to the specific pattern, and detection of the area by the pattern detection means In this case, there is provided a synthesizing means for synthesizing the corresponding pattern corresponding to the matched specific pattern and the image data whose resolution has been converted by the resolution converting means.

According to this, an area in which the image quality may be degraded due to conversion into a noise pattern or loss of pixels, etc., which is generated in the prior art, is detected in advance as a specific pattern, and a corresponding countermeasure is taken. By preparing a pattern and reflecting it on the image after resolution conversion, an image with good image quality can be obtained. In addition, since a corresponding pattern is not generated or combined for an area that does not correspond to a specific pattern, an increase in the circuit scale of the apparatus and a decrease in processing speed do not occur.

[0015] The image processing apparatus according to claim 2 is
When the area is detected by the pattern detection means, the image processing apparatus further includes a pattern separation unit that replaces a specific pixel in the area with an alternative pixel, and the resolution conversion unit uses the alternative pixel to replace the specific pixel by the pattern separation unit. When the image data is replaced, resolution conversion is performed on the replaced image data.

According to this, when a specific pattern is detected, the pixel of interest is replaced with a substitute pixel, resolution conversion is performed, and then synthesis with the corresponding pattern is performed. Obtainable.

Further, in the configuration according to the third aspect,
3. The image processing apparatus according to claim 1, wherein the specific pattern is a pattern determined by two thresholds, and the second threshold is set to a value equal to or less than the first threshold. In the case where the value of the pixel of interest in the region is larger than the first threshold, the value of the pixels around the pixel of interest is a pattern smaller than the first threshold, When the value of the pixel of interest is smaller than the second threshold value, it is determined that the pattern of the pixels around the pixel of interest is larger than the second threshold value.

According to this, since the specific pattern is set using the two thresholds, the specific pattern can be easily set in both the binary image and the multi-value image.

Further, in the structure according to the fourth aspect,
3. The image processing device according to claim 1, wherein the specific pattern is determined by two threshold values, and
The second threshold value is a pattern set to a value equal to or less than the first threshold value, and when the value of the pixel of interest in the area is larger than the first threshold value, Is a pattern in which the value of pixels around is smaller than the first threshold value except for one pixel, and when the value of the target pixel is smaller than the second threshold value, Are determined so as to be a pattern larger than the second threshold value except for one pixel.

According to this, since the specific pattern is set by using two threshold values, the specific pattern can be set easily in both the binary image and the multi-valued image. Since one of the peripheral pixels is set to a threshold value different from that of the other peripheral pixels, a detailed specific pattern can be set.

[0021]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, an embodiment of an image processing apparatus according to the present invention will be described with reference to FIGS. 1 to 15 in which a binary image having a resolution of 400 dpi is converted to a resolution of 600 dpi.
This will be described with reference to FIG. The white in the binary image is'
0 and black are set to '1'. (Embodiment 1)

First, an image processing apparatus according to a first embodiment of the present invention will be described with reference to FIGS. FIG.
1 shows a configuration example of the first embodiment of the image processing apparatus according to the present invention. 11 is a line buffer, 12 is a pattern detector as pattern detection means, 13 is a pattern separator as pattern separation means, 14 is a resolution converter as resolution conversion means, 15 is a pattern generator as pattern generation means, 16 Is a synthesizer as synthesis means.

ID is an image bus between the input terminal A and the line buffer 11, BD is an image bus between the line buffer 11 and the pattern detector 12, and BID is a line buffer 1.
NPD is an image bus between the pattern separator 13 and the resolution converter 14, CPD is an image bus between the pattern generator 15 and the synthesizer 16, and CNPD is an image bus between the pattern separator 13 and the resolution converter 14. The image bus between the resolution converter 14 and the synthesizer 16, the CID is the image bus between the synthesizer 16 and the output end B, and the PTS is the pattern separator 12, the pattern generator 15, Synthesizer 1
6 is a control signal bus output to the control signal bus 6.

FIG. 2 shows a configuration example of the pattern detector 12. 201 is a block buffer having a capacity of 5 pixels in the main scanning direction and 5 pixels in the sub-scanning direction, having the contents shown in FIG.
Reference numerals 219 to 219 denote pattern detection circuits for detecting 400 dpi patterns different from each other. BBD is a bus for supplying pixel data from the block buffer 201 to the pattern detection circuits 202 to 219, and PTS0 to PTS1.
Reference numeral 7 denotes a signal which becomes active when the pattern detection circuits 202 to 219 respectively detect 400 dpi patterns to be detected. PTS is a bus composed of PTS0 to PTS17. Pattern detection circuits 202-21
The 400 dpi pattern detected by 9 is any of the 18 types of patterns shown in FIG. Table 1 shows the correspondence between the pattern detection circuit and the detection pattern.

[0025]

[Table 1]

The configuration of the pattern detection circuit is such that when the pattern detection signal is positive logic, the pattern detection circuits 202 to 21
If 0, 400d to be detected by each pattern detection circuit
The target pixel position of the pi pattern is the block buffer 201
Is the central pixel (positions 2 and 2 in FIG. 4), and the relative positional relationship of each pixel in the pattern is
An input line is configured to match the relative positional relationship of each pixel stored in the pattern No. 01, and a logical negation of an input at a position of a black pixel in the pattern and a NOR (negation of a logical sum) with another input. May be used as the pattern detection signal.

In the case of the pattern detection circuits 211 to 219, 400 dpi to be detected by each pattern detection circuit is used.
The input line is set so that the target pixel position of the pattern becomes the center pixel of the block buffer 201, and the relative positional relationship of each pixel in the pattern matches the relative positional relationship of each pixel stored in the block buffer 201. Constitute
The AND of the logical negation of the input at the position of the white pixel in the pattern and another input may be used as the pattern detection signal. Pattern detection circuit 203 and pattern detection circuit 2
12 (a) and 3 (b) show an example of the configuration of the twelfth embodiment.

In FIG. 3, 31, 32, 34 and 3
5 is a logical NOT element, 33 is a 14-input NOR circuit that takes NOR of all inputs, and 36 is a 14-input A that takes AND of all inputs.
ND circuit, BBD _{m, n} (m and n are 0 ≦ m ≦ 4, 0 ≦ n ≦
(An integer of 4) is a signal line for transmitting the value of one pixel among the data of 5 pixels × 5 pixels stored in the block buffer 201.

FIG. 4 shows the values of m and n and the block buffer 2
The relationship with the position of the pixel stored in 01 is shown. In the case where a window of 5 × 5 pixels centering on the target pixel is set on the original image and there is a portion that does not exist in the window, the block buffer 201 uses 400 dpi in the pattern detection circuits 202 to 210. When detecting a pattern, the pixel value of the portion is treated as “1”, and the pattern detection circuits 211 to 219 treat the pixel value as “1”.
When detecting a 00 dpi pattern, the pixel value of the portion is treated as '0'.

FIG. 6 shows a configuration example of the pattern separator 13. 61 is a pixel value conversion circuit for inverting the pixel value, 62
Is a multiplexer. When any of the 400 dpi patterns a to r is detected by the pattern detector 12 and any of the control signals PTS0 to PTS17 becomes active, the multiplexer 62 outputs the output of the pixel value conversion circuit 61 to the NPD, and outputs the PTS0 to PTS17. If neither is active, the input ID is output to the NPD as it is. That is, 400 in the pattern detector 12
When any one of the dpi patterns a to r is detected, the value of the pixel of interest of the pattern is inverted and input to the resolution converter 14. If no pattern is detected, the value of the pixel of interest is directly input to the resolution converter 14. Is input to

FIG. 7 shows an example of the configuration of the resolution converter 14, wherein 71 is a line buffer, and 72 is a multi-valued / multi-level
A resolution conversion circuit, 73 is a binarization circuit, 74 is a quantization error buffer, BNPD is an output of the line buffer 71, and PMD is an image output converted into a multi-level image by the multi-level / resolution conversion circuit 72 and converted in resolution. , ERD are bidirectional data buses for the binarization circuit 73 to read and write quantization error data in the quantization error buffer 74.

The multi-level / resolution converting circuit 72 multi-levels the image input from the pattern separator 13 through the bus NPD and the line buffer 71 based on the projection method, and increases the resolution from 400 dpi to 600.
The image is converted to dpi and output, and the binarization circuit 73 re-binarizes the image whose resolution has been converted by the multi-value / resolution conversion circuit 72 based on the error diffusion method.

Here, the projection method and the error diffusion method will be described. First, the projection method will be described. The projection method is
The value of the pixel of the image after resolution conversion to be obtained (hereinafter referred to as a conversion pixel) is determined by the value of the pixel of the image before resolution conversion (hereinafter referred to as an original pixel) in the vicinity of the pixel and the value of the neighboring pixel. This is a method of obtaining an average value weighted according to the distance from the pixel whose value is to be obtained. The converted pixel value is obtained by the following procedure.

1. 1. Divide the original image into squares of the same size, centered on each pixel position 2. Divide the converted image into squares of the same size centered on each pixel position 3. Center the squares in the upper left corner of both images. 4. Set a window centered on the center of the square on the converted image and the same size as the square on the original image Multiply the ratio of the area occupied by the square in the original image in the above window to the pixel value in each square, and take the sum of the values to obtain the converted pixel value

When the resolution is converted from 400 dpi to 600 dpi, the positional relationship between the original pixel and the converted pixel is as shown in FIG. In FIG. 8, P
₁ to P ₉ are converted pixel, P _a to P _i is the original pixel. In order to determine the value of the converted pixel P5, centered on the center of the P _5, is set the same size of the window as the original pixel (square shaded in FIG. 8) (4) above. The original pixels within the window is P _a, P _b, P _d and P _e, the percentage of each original pixel occupies in the window P _a: P _b:
Since P _d : P _e = 1: 2: 2: 4, the value of the converted pixel P ₅ can be obtained by equation (5) (5 above).

P ₁ = P _a (1) P ₂ = (P _a + 2P _b ) / 3 (2) P ₃ = (P _b + 2P _c ) / 3 (3) P ₄ = (P _a + 2P _d ) / 3 _{(4) P 5 = (P} a + 2P b + 2P d + 4P e) / 9 (5) P 6 = (2P b + P c + 4P e + 2P f) / 9 (6) P 7 = (2P d + P f) / 3 _{(7) P 8 = (4P} e + 2P f + 2P h + P 1) / 9 (8) P 9 = (2P d + 4P e + P g + 2P h) / 9 (9)

The other original pixels P ₁ to P ₄ and P ₆ to P ₉ are similarly obtained by equations (1) to (4) and (6) to (9). Although only some of the pixels are shown in FIG. 8, the values of the other pixels can be obtained by any one of Expressions (1) to (9).

[0038]

(Equation 1)

Next, the error diffusion method will be described. The error diffusion method is a gradation number conversion method that refers to a quantization error of a pixel that has already been quantized when a certain pixel is quantized. Expressions for calculating each pixel value are Expression (10) and Expression (10). 11)
Can be represented by

In equations (10) and (11), P _{i, j} is the pixel of interest, C _{i + m, j + n} is a weighting factor, and e
_{i + m, j + n} represent the quantization error, [•] represents the quantization, I is the pixel number in the main scanning direction, j is the pixel number in the sub-scanning direction, and m is the offset in the main scanning direction from the pixel of interest. The value n represents an offset value in the sub-scanning direction from the pixel of interest, i and j
Is an integer of 0 or more, m is an integer, and n is an integer of 0 or less.

The range in which m and n change in equation (10) is determined by the range in which the quantization error is referenced. For example, when referring to the quantization error of the pixel at the upper left, the upper, the upper right, and the left of the target pixel, combinations of m and n are as shown in Table 2, and Expression (10) is expressed by Expression (1).
It can be rewritten as in 2).

[0042]

[Table 2]

In the present embodiment, the resolution converter 14 converts "0" into "0" and "1" into "255" in the input image to obtain an 8-bit / pixel binary image, and then obtains the following expression ( 1) to 600 dpi, 8 bi by equation (9)
It is converted into a multi-valued image of t / pixel, and the quantization threshold value is set to “128” in Expression (12). _{i, j} '
If "1" is smaller than "128", "0" is substituted and binarized again.

The quantization error in equation (11) is assumed to be signed 9-bit data ('-255' to '255). P ' _{i, j} = [P _{i, j} + (1/8) e _{ij, j-1} + (3/8) e _{i, j-1} + (3/8) e _{i + j, j-1} + (1/8) e _{i-1, j} ] (12)

FIG. 9 shows an example of the configuration of the pattern generator 15. Numerals 901 to 918 denote pattern generating circuits for generating mutually different 600 dpi patterns.
919 is a multiplexer, and PD0 to PD17 are buses for outputting 600 dpi pattern data output from the pattern generation circuits 901 to 918, respectively. 600 generated by the pattern generation circuits 901 to 918
The dpi pattern is one of a total of 18 types of patterns shown in FIG.
The correspondence with the dpi pattern is as shown in Table 3.

[0046]

[Table 3]

FIG. 11 shows an example of the configuration of the synthesizer 16, where 111 is a control circuit, 112 is a line buffer, and BD is the control circuit 111 and the line buffer 112.
CTRL is a line buffer control signal bus. The image data CNPD output from the resolution converter 14 is output without any processing in the control circuit 111 and stored in the line buffer 112.

On the other hand, the image data CPD output from the pattern generator 15 is ORed with the output image data of the resolution converter 14 stored in the line buffer 112 when the generated patterns are patterns A to I. If the generated pattern is pattern J to pattern R, AND (logical product) with the output image data of the resolution converter 14 stored in the line buffer 112 is calculated and stored in the line buffer 112 again. I do.

Hereinafter, an operation when a 400 dpi resolution image is input to the image processing apparatus of this embodiment will be described. First, the resolution 40 input to the image processing apparatus of the present embodiment is
The image data of 0 dpi is temporarily stored in the line buffer 11 from the input terminal A in order from the head pixel of the head line,
It is output to the pattern detector 12 and the pattern separator 13 as appropriate.

The pattern detector 12 sends block data of 5 pixels × 5 pixels centered on the target pixel from the line buffer 11 to the block buffer 2 according to the position of the target pixel.
01, the block data is stored in the block buffer 201, and the pattern detection circuits 202 to 219 use the 18 types of patterns a to r shown in FIG.
Is detected in the block data, one of PTS0 to PTS17 is asserted according to the type of the detected pattern.

The pattern separator 13 includes PTS0 to PTS
17 is asserted, the line buffer 1
Invert the pixel value of the image data input from
If none of S0 to PTS17 is asserted, the image data input from the line buffer 11 is input to the resolution converter 14 as it is.

The resolution converter 14 is calculated by the equations (1) to (9).
The image data input from the pattern separator 13 is once multi-valued by the projection method of calculating the pixel value after the resolution conversion by using the formula (1), and the resolution is converted to 600 dpi.
Binarization is performed again by 1) and equation (12).

The pattern generator 15 includes patterns A to R generated by the pattern generation circuits 901 to 918.
Any of the 600 dpi pattern data in PTS0
One of the pattern data is selected by the multiplexer 919 in accordance with whether any one of .about.PTS 17 is asserted and output to the synthesizer 16.

The synthesizer 16 outputs the binary image data whose resolution has been converted to 600 dpi, which is output from the resolution converter 14, and the 600 dpi which is output from the pattern generator 15.
The binary pattern data of i is synthesized according to which of PTS0 to PTS17 is asserted and output to the output terminal B.

At this time, when PTS0 to PTS8 are asserted, the image data output from the resolution converter 14 and the pattern data output from the pattern generator 15 are combined by logical OR, and PTS9 to PTS17 are asserted. And the image data output from the resolution converter 14 and the pattern data output from the pattern generator 15 are combined by AND, and when none of PTS0 to PTS17 is asserted, the image data output from the resolution converter 14 Does nothing. As for the image data stored in the line buffer 112, the portion where the synthesis processing is completed is read out from the line buffer 112 and output from the output terminal B to the outside of the present embodiment.

For example, when the pattern a is detected by the pattern detector 12, the output PTS0 of the pattern detection circuit 202 is asserted. When the pattern separator 13 detects that PTS0 is asserted,
The output of the pixel value conversion circuit 61 is selected and output to the output image bus NPD of the multiplexer 62, and the resolution converter 1
4 is input.

That is, the value of the target pixel, which is a black pixel, is inverted (converted to a white pixel) and input to the resolution converter 14. In the resolution converter 14, the pixel value input from the pattern separator 13 is once multi-valued based on the equations (1) to (9) and the resolution is set to 600 dpi.
And based on equations (11) and (12),
The value is output to the synthesizer 16.

On the other hand, in the pattern generator 15, P
When detecting that TS0 is asserted, the output PD0 of the pattern generation circuit 901, that is, the pattern A, is selected by the multiplexer 919 and output to the synthesizer 16. When the synthesizer 16 detects that PTS0 is asserted, the pattern A is synthesized with the image stored in the line buffer 112 and output from the resolution converter 14 by logical OR.

(Embodiment 2) Next, a second embodiment different from the above will be described with reference to FIGS.
FIG. 12 is a diagram illustrating a configuration example of the present embodiment.
1 is a line buffer, 122 is a pattern detector, 123
Is a resolution converter, 124 is a pattern generator, 125 is a synthesizer, ID is an image bus between the input terminal A and the line buffer 121, and BD is an image bus between the line buffer 121 and the pattern detector 122. , NPD are the image bus between the line buffer 121 and the resolution converter 123, C
PD is an image bus between the pattern generator 124 and the synthesizer 125, and CNPD is a resolution converter 123 and the synthesizer 125.
CID is an image bus between the synthesizer 125 and the output terminal B, and PTS is an image bus between the synthesizer 125 and the output terminal B.
This is a control signal bus output to the pattern generator 124 and the synthesizer 125.

In this embodiment, the line buffer 121 is the line buffer 11 of FIG. 1, the pattern detector 122 is the pattern detector 12 of FIG.
23 is a resolution converter 14 and a pattern generator 124 of FIG.
Operate in the same manner as the pattern generator 15 of FIG. 1, and a detailed description thereof will be omitted.

FIG. 13 is a diagram showing an example of the configuration of the synthesizer 125 in FIG. 12. Reference numeral 131 denotes a control circuit, 132 denotes a line buffer, and BD denotes an image bus between the control circuit 131 and the line buffer 132. , CTRL are line buffer control signal buses. The image data CNPD output from the resolution converter 123 is output without any processing in the control circuit 131 and stored in the line buffer 132. On the other hand, the image data CPD output from the pattern generator 124 is overwritten on the image data output from the resolution converter 123 and stored in the line buffer 132.

The operation when an image having a resolution of 400 dpi is input to the image processing apparatus according to the present embodiment will be described below.
Resolution 400d input to the image processing apparatus of the present embodiment
pi image data, in order from the first pixel of the first line,
From the input terminal A, the data is temporarily stored in the line buffer 121 and output to the pattern detector 122 and the resolution converter 123 as appropriate. The pattern detector 122 uses the 18 shown in FIG.
When any of patterns a to r is detected in the block data, one of PTS0 to PTS17 is asserted according to the type of the detected pattern.

The resolution converter 123 temporarily multi-values the image data input from the line buffer 121 and adjusts the resolution by the projection method of calculating the pixel value after the resolution conversion by the equations (1) to (9). It is converted to 600 dpi and binarized again according to the equations (11) and (12).

The pattern generator 124 converts the 600 dpi pattern data of any of the patterns A to R generated by the pattern generation circuits 901 to 918 into PTS data.
One pattern data is selected by the multiplexer 919 according to which one of 0 to PTS 17 is asserted and output to the combiner 125.

The synthesizer 125 outputs the binary image data whose resolution has been converted to 600 dpi output from the resolution converter 123 and the binary image data 60 which is output from the pattern generator 124.
0 dpi binary pattern data is converted to PTS0 to PTS1.
7 are synthesized according to which one of them is asserted, and output to the output terminal B. At this time, when PTS0 to PTS8 are asserted, the image data output from the resolution converter 123 and the pattern data output from the pattern generator 124 are combined by OR, and when PTS9 to PTS17 are asserted, the resolution is increased. The image data output from the converter 123 and the pattern data output from the pattern generator 124 are combined by a logical product.

On the other hand, when none of PTS0 to PTS17 is asserted, no processing is performed on the image data output from the resolution converter 123. As for the image data stored in the line buffer 112, the portion where the synthesis processing is completed is read out from the line buffer 112,
The signal is output from the output terminal B to the outside of the present embodiment.

For example, when the pattern a is detected by the pattern detector 122, PTS0 is asserted. In the resolution converter 123, the line buffer 1
The pixel value input from 21 is once multi-valued based on equations (1) to (9), the resolution is converted to 600 dpi, and re-binarized based on equations (11) and (12). Output to the synthesizer 125.

On the other hand, in the pattern generator 124,
When detecting that PTS0 is asserted, the pattern A is output to the synthesizer 125. When the synthesizer 125 detects that PTS0 is asserted, it synthesizes the image output from the resolution converter 123 by overwriting the pattern A.

As described above, by the error diffusion method, the two signals input to the first and second embodiments from the input terminal A are used.
The valued 400 dpi image is converted into a 600 dpi image while the isolated black pixels in the low density portion and the isolated white pixels in the high density portion are preserved by the pattern conversion. Or isolated white pixels are converted into an unsightly noise pattern, or conversely, disappear, and the original pixel and the grayscale expression of the image after resolution conversion do not match, thereby suppressing image quality deterioration, and achieving a good resolution of 600 dpi.
Can be converted to

In the first and second embodiments, the case where the resolution is converted from 400 dpi to 600 dpi is taken as an example. However, the gist of the present invention is that it can be applied to other resolution conversions having the same conversion ratio. It is obvious to those skilled in the art that the present invention can be applied to resolution conversion of other conversion ratios.

Further, in the first and second embodiments, the image binarized by the error diffusion method is targeted, but the gist of the present invention is that the image is binarized by another binarization method such as the dither method. Obviously, the present invention can be applied to a binarized image.

Further, in the first and second embodiments, since the binarized image is targeted, if the target pixel is a black pixel (ie, '1'), all of its peripheral pixels If the target pixel is a white pixel (that is, '0') or only one of its peripheral pixels is a black pixel and the other peripheral pixels are white pixels, The pattern detector 12 or 122 detects a pattern in which all peripheral pixels are black pixels, or only one of the peripheral pixels is a white pixel and the other peripheral pixels are black pixels. , A first threshold and a second threshold that is less than or equal to the first threshold.
Are set to values larger than '0' and smaller than '1', for example, '0.5'.

Here, when a multivalued image is targeted,
A first threshold value and a second threshold value that is equal to or less than the first threshold value are set, and when the value of the target pixel is larger than the first threshold value, the vicinity of the target pixel Pixel values are smaller than the first threshold value, or 1
The pattern in which only one pixel is larger than the first threshold value and the other peripheral pixels are smaller than the first threshold value, and when the value of the target pixel is smaller than the second threshold value, Are all larger than the second threshold value, or only one of the peripheral pixels is smaller than the second threshold value and the other peripheral pixels are larger than the second threshold value. The pattern detector 12 or the pattern detector 122 may detect a large pattern.

In the processing of the input image in the resolution converter 14 or the resolution converter 123, the number of gradations of the multi-valued conversion by the projection method is set to be equal to or larger than the number of gradations of the original image. The processing may be performed by setting the quantization thresholds having different values from each other and within the range of pixel values by the number obtained by subtracting 1 from the number of gradations of the original image. For the pattern generated by, a pattern having the same number of gradations as that of the original image may be generated.

In the first and second embodiments, the pattern detected by the pattern detector 12 and the pattern detector 122 is the pattern shown in FIG. 2, but it is obvious that the present invention is not limited to this. .

In the first and second embodiments, the correspondence between the control signals PTS0 to PTS17 output from the pattern detector 12 and the pattern detector 122 and each detection pattern is assigned as shown in Table 1. However, it is clear that the present invention is not limited to this.

In the first and second embodiments, the output from the pattern generator 15 and the pattern generator 124 is performed according to the value of the control signal PTS output from the pattern detector 12 and the pattern detector 122. 6
The 00 dpi pattern is the pattern shown in FIG. 3, but it is clear that other patterns may be used without departing from the spirit of the present invention.

Further, in the first and second embodiments, the binarization by the error diffusion method is performed by the equations (11) and (1).
Although it is performed according to 2), it is clear from Expression (10) that the binarization expression by the error diffusion method is not limited to Expression (12).

In the first and second embodiments, the quantization threshold value for binarization by the error diffusion method is
Although it is fixed at 128 ', the present invention is not limited to this, and it is apparent that changing the quantization threshold for each pixel does not depart from the gist of the present invention.

In the first and second embodiments, the synthesizer 16 is configured as shown in FIG.
Another configuration may be used as long as it performs the same function.
The configuration shown in FIG. In FIG. 14, 141 is a line buffer, 142 is a control circuit, 14
Reference numeral 3 denotes a synthesizing circuit, LBD denotes an output image data bus of the line buffer 141, SEL denotes a control signal output from the control circuit 142 to the synthesizing circuit 143, and CTRL denotes a control signal output from the control circuit 142 to the line buffer 141. is there. FIG. 15 shows a configuration example of the synthesis circuit 143, where 151 is an AND circuit, 152 is an OR circuit,
153 is a multiplexer.

14 and 15, 600 dpi pattern data output from the pattern generator 15 is temporarily stored in the line buffer 141, and
43. In the synthesizing circuit 143, the multiplexer 153 uses the AND circuit 151 and the OR circuit 152 for each pixel of the image data input from the resolution converter 14 obtained by the AND circuit 151 and the pattern data input from the line buffer 141. Either of the obtained image data input from the resolution converter 14 and the logical sum of the pattern data input from the line buffer 141 for each pixel and the image data input from the resolution converter 14 are input from the control circuit 142. Selected according to the value of the signal SEL to be output.

At this time, the signal SEL has a pattern to be synthesized with the image data input from the resolution converter 14, as shown in Table 2.
In the case of the pattern I shown in FIG.
When the pattern is J to R, the output of the AND circuit 151 is controlled to the bus CID, and when the pattern is not synthesized, the image input from the resolution converter 14 is directly output to the bus CID.

In the first and second embodiments, the configuration of the image processing apparatus according to the present invention is the configuration shown in FIG. 1. However, any other configuration for realizing the gist of the present invention can be used. It may be a configuration.

[0084]

As will be apparent from the following description,
According to the image processing apparatus according to the first aspect of the present invention, a region corresponding to a specific pattern is detected, a corresponding pattern corresponding to the detected region is prepared, and the corresponding pattern is reflected on the image after the resolution conversion, thereby obtaining a low density image. The isolated black pixel and the isolated pixel in the high-density part are converted into an unsightly noise pattern, or the isolated black pixel and the isolated white pixel disappear, and the grayscale expression of the original image and the image after resolution conversion do not match. It is possible to convert the resolution of the image satisfactorily while suppressing image quality deterioration,
In particular, the effect is remarkable when the resolution conversion ratio is not an integer.

According to the image processing apparatus of the second aspect, when a specific pattern is detected, the pixel of interest is replaced with a substitute pixel, resolution conversion is performed, and then synthesis with the corresponding pattern is performed. Since the resolution conversion is performed, a better resolution conversion result can be obtained.

Further, according to the image processing apparatus of the present invention, since the specific pattern is set using two threshold values, the specific pattern can be easily specified in both a binary image and a multi-value image. The pattern can be set.

According to the image processing apparatus of the present invention, since the specific pattern is set using two threshold values, the specific pattern can be easily specified in both a binary image and a multi-value image. Since a pattern can be set and one of the peripheral pixels is set to a threshold value different from that of the other peripheral pixels, a detailed specific pattern can be set.

[Brief description of the drawings]

FIG. 1 is a diagram illustrating a configuration example of a first embodiment of an image processing apparatus according to the present invention.

FIG. 2 is a diagram showing a configuration example of a pattern detector 12 in FIG.

3A and 3B are diagrams illustrating a configuration example of a pattern detection circuit in FIG. 2; FIG.
3 is a diagram illustrating a configuration example of a pattern detection circuit 212. FIG.

FIG. 4 is a diagram showing the contents of a block buffer 201 in FIG. 2;

FIG. 5 is a diagram illustrating pattern detection circuits 202 to 21 in FIG. 2;
FIG. 9 is a diagram showing a pattern detected in 9.

FIG. 6 is a diagram showing a configuration example of a pattern separator 13 in FIG.

FIG. 7 is a diagram showing a configuration example of a resolution converter 14 in FIG.

FIG. 8 is a diagram for explaining the principle of the projection method.

9 is a diagram showing a configuration example of a pattern generator 15 in FIG.

FIG. 10 shows a pattern generation circuit 901-9 in FIG.
FIG. 8 is a diagram showing a pattern generated in 18.

11 is a diagram illustrating a configuration example of a combiner 16 in FIG.

FIG. 12 is a diagram illustrating a configuration example of an image processing apparatus according to a second embodiment of the present invention.

13 is a diagram illustrating a configuration example of a combiner 125 in FIG.

FIG. 14 is a diagram illustrating a configuration example of the combiner 16 in FIG. 1 that is different from the configuration example illustrated in FIG. 11;

15 is a diagram illustrating a configuration example of a synthesis circuit 143 in FIG.

[Explanation of symbols]

 12, 122 pattern detector 13 pattern separator 14, 123 resolution converter 15, 124 pattern generator 16, 125 synthesizer 61 pixel value conversion circuit 62, 153, 919 multiplexer 71, 112, 132, 141 line buffer 72 multi-value Conversion / resolution conversion circuit 73 binarization circuit 74 quantization error buffer 111, 131, 142 control circuit 143 synthesis circuit 151 logical product circuit 152 logical sum circuit 201 block buffer 202-219 pattern detection circuit 901-918 pattern generation circuit

Claims (4)

[Claims] 1. An image processing apparatus comprising: resolution conversion means for converting the resolution of image data; a pattern detection means for detecting, from the image data, an area that matches a predetermined specific pattern; A corresponding pattern generating means for generating a corresponding pattern determined as follows; and a synthesizing means for synthesizing the corresponding pattern and image data whose resolution has been converted by the resolution converting means when the area is detected by the pattern detecting means. An image processing apparatus comprising: 2. The image processing apparatus according to claim 1, further comprising: a pattern separating unit that replaces a specific pixel in the region with a substitute pixel when the region is detected by the pattern detecting unit. The image processing apparatus according to claim 1, further comprising a unit configured to perform resolution conversion on image data in which a pixel is replaced with the substitute pixel. 3. The pattern in which the specific pattern is determined by two thresholds, and the second threshold is set to a value equal to or less than a first threshold. When the value of the pixel is larger than the first threshold, the value of the pixel around the pixel of interest is a pattern smaller than the first threshold, and the value of the pixel of interest is the 3. The image processing according to claim 1, wherein when the pixel value is smaller than the second threshold value, a value of a pixel around the target pixel is a pattern larger than the second threshold value. 4. apparatus. 4. The specific pattern is a pattern determined by two thresholds, and a second threshold is set to a value equal to or less than a first threshold. When the value of the pixel is larger than the first threshold value, the pattern of the pixels around the target pixel is smaller than the first threshold value except for one pixel. When the value of the pixel is smaller than the second threshold value, it is determined that the value of the pixel around the target pixel is a pattern larger than the second threshold value except for one pixel. The image processing device according to claim 1 or 2, wherein