JPH01160181A - Picture processor - Google Patents

Abstract

PURPOSE:To prevent unique stripe pattern in the error dispersion method and the average error minimizing method and to always obtain a reproduction picture with high quality by applying dot processing to add screen corner. CONSTITUTION:A picture data read by an input sensor section 11 is fed sequentially to an A/D converter 12 and a data of each picture element is converted into a digital data. Then a selector 15 is provided to apply dot processing by a pre-processing circuit 16 for a part where no edge exists and the block data inputted by the pre-processing circuit 16 are summed and the pseudo dot processing where the total sum of the density data in a block is replaced as one picture element or plural picture densities in a block is applied and screen corner is applied to the pseudo dot. Thus, the production of stripe pattern is prevented and a reproduced picture with high quality is obtained.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to image processing apparatuses such as digital printers and digital facsimiles that handle images as digital signals.

[Conventional technology]

Generally, images are sampled using a CCD sensor, etc.
2. Description of the Related Art With the development of digital equipment, digital copying devices that output digitized data from a digital printer such as a laser beam printer and reproduce images are becoming widely used in place of conventional analog copying devices.

In order to reproduce halftones, this digital copying apparatus generally uses a dither method or a density pattern method to reproduce gradations. However, in such a method, (1) If the original is a halftone image such as a print, a periodic striped pattern that is not present in the original appears in the copied image.

(2) If the original contains line drawings, characters, etc., the edges will be cut off by the dithering process and the image quality will deteriorate.

There were other drawbacks.

The phenomenon (1) is called a moiré phenomenon, and its occurrence is thought to be caused by: A) Beat between the halftone dot document and input sampling B) Beat between the halftone dot document and the dither threshold matrix.

In particular, phenomenon B) is caused by the fact that when the dither threshold is arranged in a dot-concentrated manner, the output image also has a pseudo-dot structure, which causes a beat between the input dot document and the moiré phenomenon. cause

On the other hand, a binarization method that has recently attracted attention is a method called an error diffusion method. This method calculates the density difference between the original image density and the output image density for each pixel, and distributes the error amount, which is the result of this calculation, by applying specific weighting to surrounding pixels. Regarding this, please refer to the literature R,
W, F1oyd and L, Steinberg “A
n Adaptive Algorithm for
5patial Gray 5cale”SID
, 75 Digest.

There is also a method called the minimum average error method, which is considered to be equivalent to the error diffusion method.

When binarization is performed using such a method, there is no periodicity in error processing, so moiré does not occur in the halftone image, and the resolution is good compared to dither methods etc. However, a unique striped pattern occurs in the --like density area, which causes a decrease in image quality.

〔the purpose〕

The present invention eliminates the drawbacks of the conventional example described above, and also prevents the occurrence of unique striped patterns in the error diffusion method or the minimum average error method, which is equivalent thereto, and enables high-quality reproduction of any image. The purpose is to obtain images.

FIG. 1 is a block diagram showing one embodiment of the present invention. The image data read by the input sensor unit 11, which has a photoelectric conversion element such as COD and a drive system that scans it, is
The signal is sequentially sent to the A/D converter 12.

Here, for example, the data of each pixel is converted into 8-bit digital data. This means that the data has been quantized to have 256 levels of gradation.

Next, in the correction circuit 13, corrections such as shading correction for correcting sensitivity unevenness of the sensor and unevenness of illuminance due to the illumination light source are performed by digital calculation processing.

Next, this corrected signal 100 is sent to the edge detection circuit 1.
4 and selector 15. At this time, data is transferred in blocks with mXn pixels as one block, but in this embodiment, m=3. The explanation will be made assuming that n==6.

The edge detection circuit 14 determines whether an edge exists within the block. As a method of judgment, 3
Find the difference between the x6 Laplacian and the minimum and maximum values in the block, and that value is the threshold T (here T, = 30)
If it is greater than or equal to T1, an edge exists, and if it is smaller than T1,
There are methods such as determining that an edge does not exist, but we will use the latter here.

Here, the reason for determining whether or not edges exist in a block is to avoid a decrease in resolution that would occur if a block with edges is halftone-formed. Execute processing.

The edge detection circuit 14 determines the presence of an edge,
“l” if the edge exists, “0” if it does not exist,
It is output as a signal 200. This signal 200 is input to the selector 15, and if the signal 200 is "0", the signal 100 input to the selector 15 is transferred to the preprocessing circuit 16 block by block as a signal 300.

Also, if the signal 200 is "1", the signal 100 is changed to the signal 40.
0 and transferred to the memory 17 block by block.

Here, the preprocessing circuit 1 is applied only to the part where there is no edge.
A selector 15 is provided to perform halftone dot formation at 6.

In the preprocessing circuit 16, the sum of the block data input as the signal 300 is calculated, and the sum of the density data in the block is taken as the density of one pixel or a plurality of pixels in the block.
Pseudo halftone dot conversion is performed. At this time, a screen angle is attached to the pseudo halftone dots, and how to attach the screen angle will be explained in detail below in the section explaining the halftone dot forming circuit. By setting this screen angle, the pitch between halftone dots can be shortened and the spatial frequency can be increased. The data 500 output from the preprocessing circuit 16 is input to the memory 17. The data 600 read out pixel by pixel from the memory 17 enters the binarization circuit 18, where it is binarized. This result is input to the printer 19 as a signal 700, and the printer 19 turns on/off the dots.
The image is reproduced by off control.

FIG. 2 is a block diagram showing details of the preprocessing circuit 16.

The signal 300 output from the selector 15 enters the summation calculation circuit 21, where 18 pieces of data (3×6
The sum of the densities of pixels) is calculated. The calculated result is sent to the halftone dot formation circuit 23 or the halftone dot formation circuit 2 by the selector 22.
4 is input. The selector 25 outputs the data output from the halftone dotting circuit 23 or 24 as a signal 500. Regarding switching of selector 22 and selector 25,
This will be explained using FIG. As shown in FIG. 3, the data is sequentially processed for three lines (3×6 pixels).

At this time, the three lines are regarded as one area, and this is called a block line. Block being processed (3×
9 pixels) is within block line A, selector 2
2, the data output from the sum calculation circuit 21 is input to the halftone dot formation circuit 23, and the selector 25 selects the output from the dot formation circuit 23 and outputs it as a signal 500. Further, if the data from the summation calculation circuit 21 is within block line B, the selector 22 inputs the data output from the summation calculation circuit 21 to the halftone dot formation circuit 24, and the selector 25 inputs the data output from the summation calculation circuit 21 to the dot formation circuit 24. Select the output and output it as signal 500. That is, the selector 22 and the selector 25 alternately switch between the output from the halftone dot forming circuit 23 and the output from the halftone dot form circuit 24 for each block line.

4-a and 4-b are diagrams for explaining the operation of the halftone dotting circuit 23. Let S be the signal output from the sum calculation circuit 21. If the block being processed is within the block line A, the signal S is input by the selector 22 to the halftone dotting circuit 23. At this time, S≦D in a relatively low concentration area
, , , (however, D max is the output density of one dot in the printer, and in this example, D = 255), then the density of the (2, 2) pixel is set to S as shown in Figure 4-a. year,
Set all the rest to 0.

Also, when S > D + nax in a block with high density, the density of the (2, 2) pixel is set to D as shown in Figure 4-b.
max, (1, 2).

(2,1), (2,3), (3,2) pixel density to 01
111, and the remaining pixel density is set to Dav2. Davl
+Dav2 is given by the following formula depending on the density within the block.

i) When D max < S≦5 D m, oav+ = (S D max )
/ 4Dav2 = O ii ) S > 5Dmax oav+ = D max oavz = (s s Dmax ) / 13 FIG. 5 is a diagram illustrating the operation of the halftone dotting circuit 24. Let S be the signal output from the sum calculation circuit 21. If the block being processed is within block line B, the signal S is input by the selector 22 to the halftone dotting circuit 24. At this time, S≦D, (however, D
If man is the output density of one dot in the printer (Dmax = 255 in this example), then the density of the (2,5) pixel is S as shown in Figure 5-a, and the rest are all 0. do.

Also, when S > D max, the density of the (2°5) pixel is Dmax, (1, 5), (2
, 4), (2, 6), (3°5) pixel density as Davl
, the remaining pixel density is Dav2.

Davl and Dav2 are given by the following formula depending on the density within the block.

) When D max < S ≦ 5 D ma-, Davl = (SD max ) / 4Dav
2 ” 0 ii) S>5Dmax Davl ”D max Dav2 ” (S 5 D max) / l 3
The selector 25 in FIG. 2 selects every three lines of data output from the halftone processing circuit 23 or 24 that performs different halftone processing to give a screen angle, and outputs a signal 500.
It is output to the memory 17 as .

FIG. 6 is a block diagram showing details of the binarization circuit 18 of FIG. 1.

The image data 600 (xi) output from the memory 17 has an error ε+i stored in the error buffer memory 63.
(Previously generated correction data X'11 and output data 1Y
The adder 61 adds the value obtained by multiplying the difference from ii) by the weighting coefficient α1□ specified by the weighting generator 62. Writing this as a formula is as follows.

An example of weighting coefficients is shown in FIG. * in FIG. 7 indicates the pixel position currently being processed.

Next, the correction data X',i is processed at a threshold value T
(Here, = 255.T = 127)
and outputs data Yij. Here, Yi+ is binary data such as DITlax or 0.

The binarized data is stored in the output buffer 67,
Output data 700.

On the other hand, the arithmetic unit 64 calculates the difference ε1. between the correction data X'ii and the output data Yii. is calculated, and the result is stored in the error buffer memory 63 at a location corresponding to pixel position 166. By repeating this operation, binarization is performed using the error diffusion method.

As described above, according to the above-mentioned embodiment, areas where edges do not exist, for example, uniform density areas such as highlight areas and shadow areas of an image, are halftone-ized as preprocessing and binarized using the error diffusion method. It is carried out. Therefore, it is possible to align the dots in the highlight portion or the shadow portion, thereby reducing the unique striped pattern that occurs when the error diffusion method is performed.

Moreover, by arranging the dots, regularity without the appearance of noise can be created, so that it is possible to prevent the generation of grainy noise that is felt in highlight areas or shadow areas.

Furthermore, in this embodiment, the halftone processing is changed for each block to form an image with a screen angle of 45 degrees.
It is possible to prevent a decrease in spatial frequency (the gap between dots becomes wider) during halftone processing, and to obtain a smooth reproduced image.

Also, since halftone processing is not performed on the edges, characters,
The resolution of line drawings, etc. can be maintained.

FIG. 8 is a block diagram showing an embodiment in which this embodiment is used for color image processing.

A red signal, a green signal, and a blue signal separated into three colors are output from the color image human motion sensor 81, and the A/D
A converter 82 converts each color into an 8-bit digital signal.

The correction circuit 83 performs shading correction, complementary color conversion from RGB signals to YMC signals, and masking processing.
Yellow signal, M agenta signal,
A Cyan signal is output. Here, the Y, M, and C signals are represented as a signal 801. Also, the data is Y,
M.

Block transfer is performed on the C signal 801 in which each block is made up of 3×6 pixels.

Further, the edge detection circuit 84, selector 85, preprocessing circuit 86, memory 87, and binarization circuit 88 are respectively the edge detection circuit 14, selector 15, preprocessing circuit 16, and
This can be realized by having the memory 17 and the binarization circuit 18 for three colors, but in order to prevent color moiré due to overprinting of the three MMC colors, the configuration of the preprocessing circuit 86 is changed as shown in FIG. do.

Reference numeral 91 in FIG. 9 is a selector for separating the YMC signal into a Y signal, an M signal, and a C signal. 92a to 92c are Y,
This is a sum calculation circuit that calculates the sum of densities within a block from each data of M and C. The sum of the calculated densities is sent to the halftone dotting circuit 93 as signals 901, 902, 903.
a, 93b, -, 95a, 95b.

These halftone dot forming circuits 93 a , 93 b , . . .
The processing in 95a and 95b will be explained using FIG. 1O.

In the halftone forming circuit 93a, i) (signal 901)≦D max (D max=
255) when A21=(signal 901) Other pixel density=O ii ) D max < (signal 901)≦4 D
m5-A21=D max AII=A22=A31=((signal 901)-D
,,,a,) /3Other pixel density=Oiii) (signal 901)>4DmaxA 11 =
A 21 = A n = A 31 = D ma
w Other pixel density = ((signal 901) 4D-a
A pseudo halftone dot is formed as follows: x)/14. Here, (signal 901) is the sum of the concentrations output from the sum calculation circuit 92a, and AiI(j" 1. . . , 3,
j=1. ..., 6) represents the density of the (i,D pixel) in the block.

In the halftone forming circuit 93b, i) When (signal 901)≦D max, A 24
” (Signal 901) Other pixel density = O ii ) D max < (Signal 901)≦5 D
When ff1ax, A24 = D max Al1 = A23 = A25 = A34 = ((signal 901)
Dma-) / 4 Other pixel density = O iii) (signal 901) A 14 when > 50m@x
” A 23 = A 24 = A y, = A
34 = D max other pixel density = ((signal 90
1) A pseudo halftone dot is formed as follows: 5D mIX)/13. Here, (signal 901) is the sum of the concentrations output from the sum calculation circuit 92a, and A++(
i=1. ..., 3. j=1. ..., 6) represents the density of the (i,D pixel) in the block.

In the halftone forming circuit 94a, i) When (signal 902)≦D maw, A
= (signal 902) Other pixel density = O ii ) D + nax < (signal 902)≦4 D
lrl,.

A13=D mat A, □=A, 4=A23=((signal 902) −D
max) / 3 Other pixel density = O 1ii) (signal 902) > 4 D mawA
12 = A 13 = A 14
A pseudo halftone dot is formed as follows: = A 23 = D max and other pixel densities = ((signal 902) 4Dmax)/14. Here, (signal 902)
is the sum of the concentrations output from the sum calculation circuit 92b, and Atj (1=1...., 3, j=x,...,
6) represents the density of the (i, D pixel in the block).

In the halftone forming circuit 94b, i) When (signal 902)≦D max, A +s
= (signal 902) Other pixel density 20ii) D max< (signal 9o2)≦4 D,
,,.

A16 = D max Ass = A2S = A211 = ((signal 902) -
Dm,, ) /3 Other pixel density = O iii) (signal 902) > 4Dmax A 16 = A 16 = A 25 = A 26
= D max Other pixel density = ((signal 902)
A pseudo halftone dot is formed as follows: 4Dmax)/14. Here, (signal 903) is the total sum calculation circuit 92
It is the sum of the densities output from Air(i=1.b).
..., 3, j=1. ..., 6) is (i
, j) represents the density of the pixel.

In the halftone forming circuit 95a, 1) When (signal 903)≦D mmX, A33=(signal 903) Other pixel density
axA33 = D ma engineering A23 = A32 = A34 = ((signal 903
) Dr, l-*) / 3 Other pixel density 20; 1i) (signal 903) > 4 D m1xA 2
3 = A 32 = A aa = A 34 = Dmax
Other pixel density = ((signal 903) 4D ma
A pseudo halftone dot is formed as follows: x)/14. Here, (signal 903) is the sum of concentrations output from the sum calculation circuit 92c, and A tH(r ≦1. . . ,
3, j=t, . . . , 6) represents the density of the (i, j) pixel in the block.

In the halftone conversion circuit 95h, i) (signal 903)≦Dmax, A = (signal 903), other pixel density 20ii) D mat < (signal 903)≦4 D
ma-A 36'' D max A26 = A28 = A3S = ((signal 903)
-D,,,,,)/3 Other pixel density=O iii) (signal 903)>4Dm,l! A25 :A
26 = A 35 = A 311 ” D ma
x Other pixel density = ((signal 903) -4Dmax
)/14 pseudo halftone dots are formed. here,
(Signal 903) is the sum of the concentrations output from the sum calculation circuit 92c, A i: (1=1,...,3
, j=1. ..., 6) represents the density of the (i, j) pixel in the block.

As described above, the halftone forming circuits 93a, 93b,
By configuring 95a and 95b, it is possible to set a screen angle for each color, and as mentioned above, it is possible to improve the striped pattern, and by changing the formation position of halftone dots for each color. It is also possible to prevent color moire due to registration misalignment, etc.In other words, rather than having dots of each color printed without periodicity, it is possible to set a screen angle for each color in advance,
By forming dots periodically, an image that appears to have a uniform density can be formed due to the integral effect of the eye, and color moiré can be prevented from occurring.

Furthermore, here the block size for all three colors is 3 x 6 pixels, but by changing the block size for each color and changing the number of halftone dots in each block, the screen angle can be adjusted for each color. You can also set

96 to 98 are switchers, and the switch 96 has an input signal 9.
00, the outputs of the halftone dot forming circuit 93a and the halftone dot forming circuit 93b are switched. Switches 97 and 98 similarly switch the outputs of 94a and 94b, and 95a and 95b. The selector 99 converts the data from the switches 96 to 97 into the MMC signal 8.
Output as 03.

As explained above (according to this embodiment, by performing halftone processing that adds screen angle), it is possible to prevent the occurrence of striped patterns and grainy noise in the highlight and shadow areas of an image where there are no edges. .

Furthermore, for color images, by setting a screen angle for each color, it is possible to prevent striped patterns as well as color misalignment (color moiré) caused by registration misalignment and the like.

Note that the halftone method for adding screen angles is not limited to this embodiment (it can also be realized by changing the formation position of blocks or dots other than 3×6 pixels).

〔effect〕

As described above, according to the image processing apparatus of the present invention, it is possible to prevent the unique striped pattern caused by the error diffusion method and the minimum average error method, and it is possible to obtain a high-quality reproduced image for any image.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing one embodiment of the present invention, and FIG.
3 is a diagram for explaining the operation of the selector 22, FIG. 4-a, FIG. 4-b, FIG. 5-a, and FIG. 5-b. teeth,
FIG. 6 is a block diagram of the binarization circuit 18, FIG. 7 is a diagram showing an example of weighting coefficients, and FIG. A block diagram when this embodiment is partially modified, FIG. 9 is a block diagram of the preprocessing circuit 86, and FIG.
FIG. 3 is a diagram for explaining processing of a halftone dot formation circuit. 11 is an input sensor, 12 is an A/D converter, 13 is a correction circuit, 14 is an edge detection circuit, 15 is a selector, 16 is a preprocessing circuit, 17 is a memory, 18 is a binarization circuit, and 19 is a printer. .

Claims (2)

[Claims] (1) Processing means for halftone processing image data, quantization means for quantizing the halftone image data, and correction for correcting error information generated during quantization in the quantization means. 1. An image processing apparatus comprising: means for performing halftone dot processing such that a screen angle is formed when the halftone dot processing is performed by the processing means. (2) The image processing apparatus according to claim 1, wherein a screen angle is set for each color of the image data when the processing means performs halftone processing.