JPH1056571A - Image processor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To perform the optimum edge processing to each edge of left and right in a main scanning direction of a character, line drawing, etc. SOLUTION: A left edge pixel phase signal generating part 111 generates a phase signal which shifts the dots of left edge pixels to the right, a right edge pixel phase signal generating part 112 generates a phase signal which shifts the dots of right edge pixels to the left, and non edge pixel phase signal generating part 113 generates a phase signal which indicates that the dots of non edge pixels are in the center. Output signals of phase signal generating parts 111 to 113 are selected, based on a decided result of an edge deciding part 109 by a selector 114, added to the lower bit of image data and outputted to a printer 108. The printer 108 prints dots in density which corresponds to the image data, also shifts dots to the center, left or right and prints them.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to a digital copying machine,
The present invention relates to an image processing apparatus used for a facsimile apparatus, a scanner, an image filing apparatus, and the like, and particularly to an image processing apparatus for processing an edge of an image.

[0002]

2. Description of the Related Art Conventionally, as this kind of image processing apparatus,
For example, when a mode for performing edge processing is set as disclosed in Japanese Patent Application Laid-Open No. 4-229767, a method of detecting a left edge in the main scanning direction of a character, a line drawing, or the like, and performing edge processing on several subsequent pixels. It has been known.

[0003]

However, according to the above-mentioned conventional method, only the left edge in the main scanning direction such as a character or a line drawing is detected, so that the right edge in the main scanning direction cannot be edge-processed. There is a problem. Further, in the above-described conventional method, it is only possible to switch the mode of performing the edge processing or not, so that different dot processing cannot be performed on the left and right edges,
As a result, there is a problem that optimal processing cannot be performed on the dots on the left and right edges.

[0004] In view of the above-mentioned conventional problems, the present invention has been developed for characters,
It is an object of the present invention to provide an image processing apparatus capable of performing optimal edge processing on left and right edges of a line drawing or the like in the main scanning direction.

[0005]

In order to achieve the above object, a first means determines the presence or absence of an edge in image data and, when it is determined that an edge exists, determines whether a left edge or a right edge is present. Edge determining means for determining, first gradation processing means for performing a gradation process on a pixel determined to be an edge by the edge determining means, and a first gradation processing means for determining a non-edge by the edge determining means A second gradation processing means for performing gradation processing different from the first gradation processing means, and a first gradation processing means for a pixel determined as a left edge by the edge determination means. First edge processing means for performing edge processing on the processed image data; and image data processed by the first gradation processing means for a pixel determined to be a right edge by the edge determination means. Characterized in that a second edge processing means for different edge treatment and edge processing means.

A second means is that, in the first means, the first edge processing means forms, to the right, a dot of a pixel determined to be a left edge by the edge determination means, and the second edge processing means A dot of a pixel determined to be a right edge by the edge determining means is formed on the left side.

[0007] A third means is the image processing apparatus according to the first or second means, further comprising: processing the image data processed by the second gradation processing means for a pixel determined as a non-edge by the edge determining means. A non-edge processing means for forming a dot at the center.

[0010] A fourth means is an image obtained by processing the pixel determined by the non-edge processing means as a non-edge by the edge determining means in the third means by the second gradation processing means. It is characterized in that data dots can be formed to the left or right.

[0009]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing a digital copying machine to which an embodiment of the image processing apparatus according to the present invention is applied, FIG. 2 is an explanatory diagram showing a left edge and its sampling data, and FIG. FIG. 4 is an explanatory diagram showing a reference pixel when detecting an edge pixel. FIG. 5 is a block diagram showing an edge determination unit in FIG. 1. FIG. 6 shows a reference pixel when expanding a left edge pixel. FIG. 7 is an explanatory diagram showing reference pixels when a right edge pixel is expanded, FIG. 8 is a block diagram showing a first example of a left edge pixel detection unit and a right edge pixel detection unit in FIG. 5 in detail, FIG. 9 is a block diagram showing in detail a first example of the left edge expanding section and the right edge expanding section of FIG.

FIG. 10 is an explanatory diagram showing a reference pixel when expanding a left edge pixel, FIG. 11 is an explanatory diagram showing a reference pixel when expanding a right edge pixel, and FIG. 12 is a left edge pixel detecting section in FIG. FIG. 13 is a block diagram illustrating in detail a second example of the right edge pixel detection unit, and FIG. 13 is a block diagram illustrating a second example of the left edge expansion unit and the right edge expansion unit in FIG. 5 in detail.
FIG. 14 is a block diagram showing in detail a third example of the left edge pixel detection unit and the right edge pixel detection unit of FIG. 5, and FIG.
FIG. 4 is an explanatory diagram showing coefficients and arithmetic processing of a smoothing filter of No. 4;
FIG. 16 is an explanatory diagram showing a process of shifting the left edge pixel to the right,
FIG. 17 is an explanatory diagram showing a process of shifting the right edge pixel to the left,
FIG. 18 is an explanatory diagram showing the processing of non-edge pixels, FIG. 19 is a block diagram showing the main part of FIG. 1 in detail, FIG. 20 is an explanatory diagram showing the process of adding the phase data of FIG. 19 to the image data, and FIG. 19 is an explanatory diagram showing a process of replacing the phase data of FIG. 19 with lower bits of image data, FIG. 22 is a flowchart showing a printing process of edge pixels and non-edge pixels, and FIGS. 23 and 24 are explanatory diagrams showing operations.

As shown in FIG. 1, a digital copying machine according to the present embodiment is a scanner 10 for reading an image from a document and outputting image information (digital signal) after A / D conversion.
1, a filter processing unit 102 that performs various types of filter processing on the image information input from the scanner 101, a scaling processing unit 103 that performs scaling of the image information, and image information from the scaling processing unit 103 that are set in advance. A gradation processing unit (1) 104 and a gradation processing unit (2) 105 for converting into multi-valued data,
An edge determination unit 109 that determines left and right edges in the main scanning direction of a character, a line drawing, or the like based on image information from the scaling unit 103, and a determination result of the edge determination unit 109 that outputs the outputs of the gradation processing units 104 and 105. Selector 106 that selects according to
And a γ correction unit 107 that performs γ correction processing on the image signal selected by the selector 106.

Here, the gradation processing unit (1) 104 includes a character,
In order to improve the image quality of the edge of the line image, the image information from the scaling processing unit 103 is output as it is, for example, as 8 bits, and the image information is selected in the pixel section determined by the edge determination unit 109 as the left or right edge. On the other hand, the gradation processing unit (2) 10
Numeral 5 performs multi-level error diffusion 2-dot processing to improve the image quality of pictures and photographs. In this processing, the pixel data after the multi-level error diffusion processing is output in units of two pixels, so that the gradation reproducibility is superior to that of one pixel. However, if this gradation processing is performed on characters and line drawings, the image quality naturally deteriorates. Therefore, the output of the gradation processing unit 105 is selected in a region other than the pixel section in which the edge determination unit 109 determines the left and right edges.

This copying machine also has a phase signal P for shifting the dot of the pixel on the left edge to the right as shown in detail in FIG.
h (= 10), a left edge pixel phase signal generator (1) 111, and a right edge pixel phase signal generator (2), which generates a phase signal Ph (= 01) for shifting the dot of the right edge pixel to the left. ) 112 and a non-edge pixel phase signal generator (3) 113 for generating a phase signal Ph (= 00) indicating that the dot of the non-edge pixel is at the center,
The output signals Ph of the phase signal generators 111 to 113 are selected by the selector 114 based on the determination result of the edge determiner 109.

The phase signal selected by the selector 114 is added to the lower bits of the image data from the gamma correction unit 107 by the adder 110, output to the printer 108, and the image is output on the screen. Here, the printer 108
Can print dots at a density corresponding to the image data, and can print the dots by shifting them to the center or right and left based on the phase signal Ph.

Next, the edge judgment processing by the edge judgment unit 109 will be described. FIGS. 2 and 3 schematically show 8-bit (0 to 255) sampling data in the main scanning direction in the vicinity of the edge when the left and right edges cross in the main scanning direction, respectively. "Is white and" 255 "is black. Such an edge can be detected by referring to the target pixel Dj and the pixels Dj-1 and Dj + 1 before and after the target pixel Dj as shown in FIG. As shown in FIG. 5, the edge determination unit 109 includes a left edge pixel detection unit 501, a left edge expansion unit 502 that expands the detection result of the left edge pixel detection unit 501, a right edge pixel detection unit 503, and a right edge pixel A right edge expansion section 504 for expanding the detection result of the detection section 503.

<Edge Pixel Determination 1> Next, a first example of the edge determination unit 109 will be described. Left edge pixel detection unit 50
1 and the right edge pixel detection unit 503 refer to the target pixel Dj and the preceding and succeeding pixels Dj−1 and Dj + 1 shown in FIG. 4 and calculate Dj + 1−Dj− based on equations (1) and (2). The target pixel Dj that satisfies 1 ≧ TH1 and Dj <TH2 (1) is determined as the left edge pixel of a character or a line drawing, and Dj−1−Dj + 1 ≧ TH1 and Dj <TH2 (2) holds Is determined to be the right edge pixel of the character or line drawing.

That is, the difference | Dj-1−Dj + 1 | of the density values of the pixels before and after the target pixel Dj is the threshold T
H1 or more and the density Dj of the pixel of interest is equal to the threshold TH2
If smaller, the target pixel Dj is determined to be an edge pixel. Here, the threshold value TH1 in Expressions (1) and (2)
TH2 is set so that the pixel D1 on the white side of the boundary where the boundary changes from white to black is detected in the case of the left edge as shown in FIG. 2, and in the case of the right edge as shown in FIG. Is set such that a pixel D13 on the white side of the boundary where black changes to white is detected.

Next, the processing of the left edge expansion section 502 and the right edge expansion section 504 will be described. The left edge expansion unit 502 expands the detection result of the left edge pixel detection unit 501 to the right. That is, in FIG. 6, Elj = Cj‖Cj-1‖Cj-2 Cj-3 (3) based on the detection results of the three pixels Cj-3, Cj-2, and Cj-1 on the left side of the target pixel Cj. , ‖ Indicate a logical sum (the same applies hereinafter).

Thus, the left three pixels Cj-3, Cj-2, C
If at least one of j-1 (and the target pixel Cj) is a left edge pixel, the target pixel Cj is set as a left edge pixel.

The right edge expanding section 504 expands the detection result of the right edge pixel detecting section 503 to the left. That is, the three pixels Cj + 1 on the right side of the target pixel Cj in FIG.
, Cj + 2, Cj + 3 based on the detection results of Erj = Cj‖Cj + 1‖Cj + 2‖Cj + 3 (4), and the three right-hand pixels Cj + 1, Cj + 2, Cj + 3 If at least one of the pixels (and the target pixel Cj) is a right edge pixel, the target pixel Cj is set as a right edge pixel.

That is, in the left edge expansion section 502 and the right edge expansion section 504, the pixels detected as the left edge pixel and the right edge pixel, respectively, are white pixels near the boundary where white changes to black and black changes to white. Therefore, in order to expand a white pixel detected as an edge pixel in the direction of a black pixel, the white pixel is expanded to the right in the case of a left edge and to the left in the case of a right edge.

Next, a specific configuration and operation of the left edge pixel detection unit 501 and the right edge pixel detection unit 503 will be described with reference to FIG. The pixel data D is delayed by four pixels by flip-flops (FF) 801a to 801d,
Adder 802 adds the value of 5 pixels together with the current input pixel.
Is used for edge determination. The value smoothed by the adder 802 is F
Each pixel is delayed by two pixels by F801e and 801f, and the density values of the target pixel Dj and the preceding and following pixels Dj-1 and Dj + 1 are obtained as three pixels combined with the current input value. Here, as for the pixel data D, the right pixel of the document is input first. Next, the subtractor 803 calculates Dj + 1−Dj−1 shown in the equation (1), and the subtractor 804 calculates Dj−1−Dj + 1 shown in the equation (2).

Next, in the comparator 805, equation (1)
Is calculated, and the comparator 806 calculates Dj + 1−Dj−1 ≧ TH1 shown in Expression (1). Then, the results of the comparators 805 and 806 are applied to an AND circuit 809, and a left edge is detected by a logical product (a left edge detection signal El). Similarly,
The comparators 807 and 808 calculate and compare the expressions (1) and (2), and the AND result of the comparison is calculated by the AND circuit 810 to detect the right edge (right edge detection signal Er).

Next, referring to FIG.
2 and the specific configuration and operation of the right edge expansion section 504 will be described. The left edge pixel detection signal El is delayed by three pixels by flip-flops 901a to 901c, and the target pixel Cj and the left three pixels Cj-3, Cj-2, Cj shown in FIG.
The left edge pixel detection signal El for j-1 is obtained. Next, a left edge pixel detection signal ED1 in which the left edge pixel detection signal L expands rightward is obtained by the OR gate 902a as shown in the following equation (5).

Further, the right edge pixel detection signal Er is delayed by three pixels by flip-flops 901d to 901f, and the target pixel Cj and the right three pixels Cj + 1,
Right edge pixel detection signals Er for Cj + 2 and Cj + 3 are obtained. Next, a right edge pixel detection signal EDr obtained by expanding the right edge pixel detection signal Er to the right is obtained by the OR gate 902b as shown in the following equation (6).

EDlj + 3 = Elj‖Elj + 1‖Elj + 2‖Elj + 3 (5) EDrj = Erj‖Erj + 1‖Erj + 2‖Erj + 3 (6) <Edge pixel judgment 2> Next In addition, the second
Will be described. Similarly, the edge determination unit 109 also includes a left edge pixel detection unit 501 shown in FIG. 12, a left edge expansion unit 502 that expands the detection result of the left edge pixel detection unit 501 as shown in FIG. A right edge pixel detection unit 503 and a right edge expansion unit 504 that expands the detection result of the right edge pixel detection unit 503 as shown in FIG.

The left edge pixel detection unit 501 and the right edge pixel detection unit 503 shown in FIG. 12 do not perform the smoothing as in the first example, and perform the smoothing operation as shown in FIG.
1 and Dj + 1, the pixel Dj that satisfies Dj + 1−Dj−1 ≧ TH1 and Dj <TH2 (7) based on equations (7) and (8) is defined as the left edge pixel of a character or line drawing. The pixel Dj in which Dj-1−Dj + 1 ≧ TH1 and Dj <TH2 (8) holds is determined as a right edge pixel of a character or a line drawing.

That is, as in the first example, the difference | Dj-1 between the density values of the pixels before and after the pixel of interest Dj on both the left and right sides.
−Dj + 1 | is equal to or greater than the threshold value TH1, and when the density Dj of the target pixel is smaller than the threshold value TH2, the target pixel Dj is determined to be an edge pixel. Further, the threshold values TH1 and TH2 in Expressions (7) and (8) are set such that the pixel D1 on the white side of the boundary where the boundary changes from white to black is detected in the case of the left edge as shown in FIG. Also, as shown in FIG. 3, in the case of the right edge, the pixel D on the white side of the boundary where the boundary changes from black to white
13 is set to be detected.

Next, the left edge expansion section 502 and the right edge expansion section 504 will be described. Similarly to the first example, the left edge expansion unit 502 expands the detection result of the left edge pixel detection unit 501 to the right. That is, based on the detection results of the three pixels Lj-3, Lj-2, Lj-1 on the left side of the target pixel Lj in FIG. 10, Elj = LjjLj-1‖Lj-2‖Lj-3 (9) If at least one of the left three pixels Lj-3, Lj-2, Lj-1 (and the pixel of interest Lj) is a left edge pixel, the pixel of interest Lj is determined to be a left edge pixel.

The right edge expansion section 504 also expands the detection result of the right edge pixel detection section 503 to the left similarly to the first example. That is, based on the detection results of the three pixels Rj + 1, Rj + 2, and Rj + 3 on the right side of the target pixel Rj in FIG. 11, Erj = Rj + Rj + 1 か ら Rj + 2‖Rj + 3 (10) If at least one of the right three pixels Rj + 1, Rj + 2, Rj + 3 (and the pixel of interest Rj) is a right edge pixel, the pixel of interest Rj is set as a right edge pixel.

That is, in the left edge expansion section 502 and the right edge expansion section 504, the pixels detected as the left edge pixel and the right edge pixel, respectively, are white pixels near the boundary where white changes to black and black changes to white. Therefore, in order to expand white pixels detected as edge pixels in the direction of black pixels,
In the case of the left edge, it expands to the right, and in the case of the right edge, it expands to the left.

Next, a specific configuration and operation of the left edge pixel detection unit 501 and the right edge pixel detection unit 503 of the second example will be described with reference to FIG. The pixel data D is delayed by two pixels by flip-flops (FF) 1301 and 1302, and the respective density values of the target pixel Dj and the pixels Dj-1 and Dj + 1 before and after it are obtained. Here, the pixel data D is input first to the right pixel of the document. Then the subtractor 130
In (3), Dj + 1−Dj−1 = a shown in the equation (7) is calculated, and in the subtractor 1304, Dj−1−Dj + 1 = b shown in the equation (8) is calculated.

Next, the comparator 1305 calculates Dj <TH2 shown in the equations (7) and (8), and the comparator 1306 calculates the equation (7).
Dj + 1−Dj−1 ≧ TH1 is calculated as shown in FIG.
Dj−1−Dj + 1 ≧ TH1 shown in FIG. Next, the AND gate 1308 outputs the left edge pixel detection signal L according to the two determination expressions shown in Expression (7), and the AND gate 1309 outputs the right edge pixel detection signal R according to the two determination expressions shown in Expression (8). Is output.

Next, a specific configuration and operation of the left edge expansion section 502 and the right edge expansion section 504 of the second example will be described with reference to FIG. Similarly to the first example, the left edge pixel detection signal L is output from the flip-flops 1041a to 1041.
c, the target pixel Lj shown in FIG.
And a left edge pixel detection signal L for three pixels Lj-3, Lj-2, and Lj-1 on the left. Next, an OR gate 1402
a, as shown in equation (9), the left edge pixel detection signal L
Is expanded to the right to obtain a left edge pixel detection signal El.

Similarly, the right edge pixel detection signal R is delayed by three pixels by the flip-flops 1041d to 1041f, and the target pixel Rj and the right three pixels Rj + 1, Rj + 2, Rj + 3 shown in FIG. Is obtained. Then, the expression (1)
As shown in (0), a left edge pixel detection signal Er obtained by expanding the right edge pixel detection signal R to the right is obtained.

<Edge Pixel Judgment 3> Next, a third example of the edge judgment unit 109 will be described. FIG. 14 is a block diagram showing the left edge pixel detection unit 501a and the right edge pixel detection unit 503a in detail, and FIG. 15 is an explanatory diagram showing the configuration of the smoothing filter in FIG. Modifications 501a, 503a
In the configuration shown in FIG.
01 to 1504 and a smoothing filter 1505 are added, the density values a to e for five pixels in the main scanning direction are smoothed, and a left edge pixel L and a right edge pixel R are detected. Here, FIG. 15A shows a configuration which can be applied to either of the left and right edges as an example of the smoothing filter 1505, and ￣Dj−1 = ( a + b + c + d + e) / 5 (11) is calculated. However, "￣" before Dj-1 in equation (11)
Indicates an overline in the figure (the same applies hereinafter). FIG.
(B) shows, as another example, a smoothing filter 1505 which is applied independently to the left edge and the right edge, and based on the density values a to d of four pixels, c (left) ← (2a + 3b + 2c + d) × 2/8. b (right) ← (a + 2b + 3c + 2d) × 2/8 That is, the coefficients of the left and right smoothing filters are reversed left and right.

<Phase Switching> Next, the process of switching the phase based on the edge determination result will be described. Here, in the printer 108 capable of printing multi-valued data, dots are formed by modulating the power of the laser beam for writing an image in accordance with the magnitude of the density value. With power, large dots are printed. Also, by shifting the dots printed by the printer 108 rightward or leftward, a boundary (left edge) where white pixels change to black pixels in the main scanning direction, and a boundary where black pixels change to white pixels (in the main scanning direction). Right edge)
Can improve the image quality of the edge portion. Therefore,
In this specification, shifting a dot printed by the printer 108 to the right or to the left is referred to as “phase”.

FIG. 16 shows 3 × 3 pixels at the left edge. FIG. 16A shows each value (8 bits) of 3 × 3 pixels. Since the spot light of the laser beam is a circle or an ellipse, the dots to be printed are indicated by a circle or an ellipse. Further, since dots are not printed on pixels of data "0", the positions of the pixels are indicated by rectangles for convenience. The size of the circle or ellipse is proportional to the value of the data.
In the case of the left edge shown in (a), the data value of column j1 which is the edge is small, and the data value of column j2 on the right side is large. Therefore, if dots are formed without changing the phase of the column j2 as shown in FIG. 16B, a gap is generated between the dots of the right column j3, so that the reproducibility of the left edge is poor and the image quality is deteriorated. Accordingly, as shown in FIG. 16C, the image quality can be improved by shifting the dots in column j2, which is the left edge, to the right and touching or partially overlapping the dots in column j3 on the right.

FIG. 17 similarly shows 3 × 3 pixels at the right edge, and FIGS. 17 (a) to 17 (c) show FIGS.
(C) is supported. In the case of this right edge, FIG.
As shown in FIG. 7B, if dots are formed without changing the phase of the column j2 which is the right edge, a gap is generated between the dots of the column j1 on the left side, so that the reproducibility of the right edge is poor and the image quality is deteriorated. . Therefore, as shown in FIG. 17C, the image quality can be improved by moving the dots in the column j2, which is the right edge, to the left and touching or partially overlapping the dots in the left column j1.

FIG. 18A shows a 6 × 3 pixel having a left edge column j2 and a right edge column j5 and non-edge columns j3 and j4.
It is shifted to the left. FIG. 18B shows a 6 × having a left edge column j1 and a right edge column j6 and non-edge columns j2 to j5.
The left edge column j1 and the right edge column j6 are respectively shifted right and left, and the non-edge column j2 is shifted left, j3 is shifted right, j4 is shifted left, and j5 is shifted right. ing. By the processing shown in FIG. 18B, it is possible to cover defects such as jitter generated by mechanical factors of the copying machine. Note that the left edge column j1 and the right edge column j6 and the non-edge columns j2 to j5 have different gradation processes, so that even if all of them are shifted to the left and right, the image expression is different.

The phase signal Ph shifted to the left and right as described above is converted into phase signal generators 111 to 113 as shown in detail in FIG.
Are selected by the selector 114 based on the edge determination signal of the edge determination unit 109, and are added to the lower bits (LSB side) of the image signal as shown in FIG. Here, the number of bits of the phase signal Ph is 1 bit as shown in FIGS. 20 (a) and 20 (c) if it has only the left and right phases, and if the phase signal includes other phases such as the center in addition to the left and right sides. 2
It is 2 bits as shown in 0 (b) and (d).

When expressing three types of left edge, right edge and non-edge, edge data can be used as it is as phase data. Instead of adding to the image data, the phase data Ph may be replaced with the lower one bit or two bits of the image data as shown in FIG. Further, according to the configuration shown in FIG. 19, the phase data Ph can be obtained from the edge data, so that it is not necessary to separately perform a calculation for determining the phase.

The process will be described with reference to FIG. 22. If it is not an edge, a dot is formed at the center (Ph = 00) (step S1 → S2). On the other hand, in the case of an edge, a dot is formed rightward (Ph = 10) at the left edge (Step S1 → S3 → S4), and a dot is formed leftward (Ph = 01) at the right edge (Step S1 → S3 → S4). Step S
1 → S3 → S5).

Next, the above operation will be specifically described with reference to FIG. Here, FIG. 23 shows a case where image data is input to each circuit in order of time from the right side of the document, and the signal on the left side is temporally new data and on the left side on the document. The data of one pixel is 8 bits (= 0 to 255), and the numerical value “0” = white and the numerical value “255” = black. The numerical value “50” indicates an intermediate density at the boundary (edge) between white and black.

The data Dj-1, D and Dj + 1 are respectively shown in FIG.
2, the input data of the left edge pixel detection unit 501 and the right edge pixel detection unit 503, the flip-flop 1301,
1302, the data a and b have positive and negative signs because they are the operation results of the subtracters 1303 and 1304, respectively. Here, in the comparators 1306 and 1307, the threshold value TH1 of the positive sign and the subtractors 1303 and 13 respectively
Since the result of the operation of the subtractor 1303 and the result of the operation of the subtractor 1303 are negative in order to simplify the circuit, the result may be clamped to “0” to eliminate the negative value.

The signals L and R are respectively AND signals shown in FIG.
FIG. 13 shows output signals of the circuits 1308 and 1309, and
The left edge expansion section 502 and the right edge expansion section 504 shown in FIG.
Is the input signal. Here, for example, if the threshold values TH1 = 40 and TH2 = 80 in Expressions (7) and (8), the time “6”,
At “7”, the left edge (L = 1), and at times “14” and “15”, the right edge (R = 1).

Next, the left edge expansion signal El and the right edge expansion signal Er are OR circuits 1402a and 1402b, respectively.
Is the output signal. In this example, the left edge expansion signal El
Is the left edge (El = 1) at times "6" to "10"
And the right edge expansion signal Er changes from time “11” to
At "15", the right edge (Er = 1) is set. Finally, the 2-bit phase signal Ph is determined by a combination of the left edge expansion signal El and the right edge expansion signal Er, and is shifted rightward (Ph = 10) at times "6" to "10" and at times "11" to "15". To the left (Ph = 01),
At other times, it is at the center (Ph = 00).

FIG. 24 shows an operation example in the left edge pixel detecting section 501a and the right edge pixel detecting section 503a to which the flip-flops 1501 to 1504 and the smoothing filter 1505 are added as shown in FIG. 14 are input signals of the flip-flops 1501 to 1504 and input signals of the smoothing filter 1505, and ￣Dj-1 is an output signal of the smoothing filter 1505. In this example, the smoothing filter 1505 calculates Expression (11). Also, the data $ D,
￣Dj + 1 are flip-flops 1301 and 130, respectively.
2 and the data f and g are subtractors 130
3, 1304. Also, the signals L, R, E
l, Er, and ED are the same as those in FIG.

[0049]

As described above, according to the first aspect of the present invention, the presence or absence of an edge of image data is determined, and in the case of an edge, whether a left edge or a right edge is determined is determined. Since different edge processing is performed depending on whether the edge is an edge or not, optimal edge processing can be performed on left and right edges in the main scanning direction of a character, a line drawing, and the like.

According to the second aspect of the invention, the dots of the pixels determined to be the left edge are formed closer to the right, and the dots of the pixels determined to be the right edge are formed closer to the left. It can be reproduced as follows.

According to the third aspect of the present invention, since the dots of the pixels determined to be non-edges are formed at the center, the reproducibility of both the left and right edges and the non-edge portions can be improved.

According to the fourth aspect of the present invention, since the dots of the pixels determined to be non-edge can be formed on the left side or on the right side, it is possible to cover defects such as jitter generated from mechanical factors of the copying machine. Can be.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a digital copying machine to which an embodiment of an image processing apparatus according to the present invention is applied.

FIG. 2 is an explanatory diagram showing a left edge and its sampling data.

FIG. 3 is an explanatory diagram showing a right edge and its sampling data.

FIG. 4 is an explanatory diagram showing reference pixels when detecting edge pixels.

FIG. 5 is a block diagram illustrating an edge determination unit of FIG. 1;

FIG. 6 is an explanatory diagram showing reference pixels when a left edge pixel is expanded.

FIG. 7 is an explanatory diagram showing reference pixels when a right edge pixel is expanded.

8 is a block diagram illustrating a first example of a left edge pixel detection unit and a right edge pixel detection unit in FIG. 5 in detail.

9 is a block diagram showing a first example of a left edge expansion section and a right edge expansion section in FIG. 5 in detail.

FIG. 10 is an explanatory diagram showing reference pixels when a left edge pixel is expanded.

FIG. 11 is an explanatory diagram showing reference pixels when a right edge pixel is expanded.

12 is a block diagram illustrating in detail a second example of a left edge pixel detection unit and a right edge pixel detection unit in FIG. 5;

FIG. 13 is a block diagram showing in detail a second example of a left edge expansion section and a right edge expansion section in FIG. 5;

14 is a block diagram illustrating a third example of a left edge pixel detection unit and a right edge pixel detection unit in FIG. 5 in detail.

FIG. 15 is an explanatory diagram showing coefficients and arithmetic processing of the smoothing filter in FIG. 14;

FIG. 16 is an explanatory diagram showing a process of shifting a left edge pixel to the right.

FIG. 17 is an explanatory diagram showing a process of shifting the right edge pixel to the left.

FIG. 18 is an explanatory diagram showing processing of a non-edge pixel.

FIG. 19 is a block diagram showing a main part of FIG. 1 in detail.

FIG. 20 is an explanatory diagram showing a process of adding the phase data of FIG. 19 to image data.

FIG. 21 is an explanatory diagram showing a process of replacing the phase data of FIG. 19 with lower bits of image data.

FIG. 22 is a flowchart illustrating a printing process of an edge pixel and a non-edge pixel.

FIG. 23 is a timing chart for explaining a specific operation when the circuits of FIGS. 12 and 13 are used;

FIG. 24 is a timing chart for explaining a specific operation when the circuit in FIG. 14 is used.

[Explanation of symbols]

 104 gradation processing unit (1) 105 gradation processing unit (2) 106, 114 selector 108 printer 109 edge determination unit 110 adder 111 left edge pixel phase signal generation unit 112 right edge pixel phase signal generation unit 113 non-edge pixel phase Signal generating unit 501, 501a Left edge pixel detecting unit 502 Left edge pixel expanding unit 503, 503a Right edge pixel detecting unit 504 Right edge pixel expanding unit 1505 Smoothing filter

Claims (4)

[Claims] 1. An edge determining means for determining the presence or absence of an edge of image data and determining, when it is determined that an edge is present, a left edge or a right edge; A first gradation processing means for performing gradation processing on the pixel, and a second gradation processing means for performing a gradation processing different from the first gradation processing means on the pixel determined to be non-edge by the edge determination means. Gradation processing means; first edge processing means for performing edge processing on image data processed by the first gradation processing means for a pixel determined to be a left edge by the edge determination means; Second edge processing means for processing image data processed by the first gradation processing means on a pixel determined to be a right edge by the means different from the first edge processing means An image processing apparatus having a. 2. The first edge processing means forms a dot of a pixel determined as a left edge by the edge determination means on the right side, and the second edge processing means determines a right edge by the edge determination means. 2. The image processing apparatus according to claim 1, wherein the dots of the selected pixels are formed on the left side. 3. The image processing apparatus according to claim 1, further comprising a non-edge processing unit for forming, in the center, a dot of the image data processed by the second gradation processing unit for a pixel determined as a non-edge by the edge determination unit. The image processing apparatus according to claim 1 or 2, wherein: 4. The non-edge processing means forms dots of image data processed by the second gradation processing means to the left or right with respect to pixels determined as non-edges by the edge determination means. The image processing apparatus according to claim 3, wherein the image processing apparatus is capable of performing the processing.