JPH04265072A - Picture processor - Google Patents

Abstract

PURPOSE:To obtain a reproduced picture of high quality by performing gamma correction without degrading the gradient of a low-density area in a printer having any gradation characteristic. CONSTITUTION:An average density calculating circuit 2 calculates an average density based on plural binary data, and an average density correcting circuit 3 corrects this calculated average density based on a prescribed gradation characteristic corresponding to density data of a noticed picture element, and a binarizing circuit 1 calculates the error based on the average density corrected by the average density correcting circuit 3 and distributes the calculated error to density data of unbinarized picture elements following the noticed picture element and binarizes data, where the error obtained from picture elements preceding the noticed picture element and density data of the noticed picture element are added, with the average density as the threshold.

Description

[Detailed description of the invention]

0001

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing apparatus, and, for example, to an image processing apparatus that binarizes and outputs multivalued (gradation) image data.

0002

2. Description of the Related Art Conventionally, in this type of apparatus, a dither method has been used as a means for binarizing multivalued image data. Normally, in the dither method, an m×n dither matrix (m and n are natural numbers) is prepared, and multivalued data is compared with a threshold value in the corresponding matrix element to perform binary judgment, resulting in m×n binarization. This method forms blocks to reproduce a pseudo-halftone image. However, in the dither method, the number of gradations that can be expressed is limited to m×n+1, so the resolution cannot be said to be good.

[0003] Regarding this dither method, in 1975, Floyd and Steinberg proposed “An Adaptive
e Algorithm for Spatial G
The error diffusion method proposed in the paper ``rayscale ``SID DIGEST'' is a method that is superior to the dither method in terms of resolution and gradation, and is a method that has recently attracted particular attention.This error diffusion method uses a fixed threshold value. When binarization is performed and the pixel density of interest is expressed in 8 bits, the difference from 255 if the binarization result is 1 (black) and 0 if it is 0 (white) is used as a new error and the density of the pixel of interest is expressed as a new error. This process diffuses the image into an unprocessed area, and has the characteristic of preserving the density of the original image in a global area.However, in reality, as shown in Figure 8, the gradation characteristics of printers are non-linear, and Since it is difficult to reproduce gradation in the density area, it is common to perform γ correction to lower the density of the entire original image before binarization.

0004

[Problem to be Solved by the Invention] However, in the error diffusion method that uses γ correction in combination, although the gradation in high density areas is reproduced, the stronger the correction is applied, the worse the gradation reproducibility in low density areas becomes. This method has the disadvantage of degrading the image quality, and the error diffusion method cannot take advantage of its ability to express continuous gradations in principle. γ correction is performed, for example, as shown in the graph shown in Figure 9 (when the image is expressed in 16 gradations), but all pixels in the range of original density 0 to 4 are converted to density 0. Therefore, there was a drawback that the gradation in the above range was completely lost.

The present invention has been made in view of the above-mentioned drawbacks of the conventional example, and its purpose is to create an image that can satisfactorily reproduce the gradation characteristics of the original image in the entire density range. The object of the present invention is to provide a processing device.

[0006]

[Means for solving the problem] Solving the above problems,
In order to achieve the object, an image processing apparatus according to the present invention binarizes input multi-valued density data and outputs the binarized data as binary data. a storage means for storing data; a calculation means for calculating an average density based on a plurality of binary data stored in the storage means; and a calculation means for calculating an average density based on a plurality of binary data stored in the storage means; a correction means for correcting based on gradation characteristics; and an error for calculating an error based on the average density corrected by the correction means, and distributing the calculated error to density data of non-binarized pixels after the pixel of interest. a binarization unit that binarizes data obtained by adding an error obtained from a pixel before the pixel of interest distributed by the error allocation unit and the density data of the pixel of interest, and the average density as a threshold; It is characterized by comprising:

[0007]

[Operation] According to this configuration, the storage means stores a plurality of binary data, and the calculation means stores a plurality of binary data stored in the storage means.
The correction means calculates the average density based on the value data, the correction means corrects the average density calculated by the calculation means based on a predetermined gradation characteristic corresponding to the density data of the pixel of interest, and the error distribution means corrects it with the correction means. The binarization means calculates an error based on the average density, allocates the calculated error to the density data of unbinarized pixels after the pixel of interest, and the binarization means calculates the error based on the density data of the unbinarized pixels after the pixel of interest, The data obtained by adding the error obtained from the pixel and the density data of the pixel of interest and the average density are used as a threshold value to binarize the data.

[0008]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings. <First Embodiment> FIG. 1 is a block diagram showing the configuration of a first embodiment of an image processing apparatus according to the present invention. The constituent elements will be explained in order below.

100 is a data line input to this device,
Digital data representing 8-bit (256 gradations) density is input. Reference numeral 200 indicates a data line that outputs binarized data. 1 is the binarization circuit of this embodiment, 2
3 indicates an average density calculation circuit of this embodiment, and 3 indicates an average density correction circuit of this embodiment. 400 again
, 500, and 600 indicate data lines, respectively.

As an operation of the above configuration, the binarization circuit 1
converts the digital data expressed by the aforementioned 8 bits from the data line 100 into binarized data according to the information from the average density calculation circuit 2 and the average density correction circuit 3, and converts the resulting 1(
(black) or 0 (white) is output to the data line 200. The average density calculation circuit 2 refers to a region consisting of binarized pixels surrounding the pixel of interest, calculates the average density weighted with a weight mask corresponding to each pixel in this region, and calculates the average density by weighting it with a weight mask corresponding to each pixel in this region.
Output to 00. The average density correction circuit 3 calculates a corrected average density based on digital data indicating gradation (data line 100) and binarized average density data (data line 500), and outputs it to a data line 400.

The details of this embodiment having the above configuration will be explained with reference to the drawings. Figure 2 shows the binarization circuit 1 shown in Figure 1.
3 is a diagram showing the relationship between the pixel of interest and the pixel to which errors are distributed, and FIG. 4 is a block diagram showing the configuration of the error distribution control circuit 16 shown in FIG. be. In FIG. 2, 11a to 11d are flip-flops (hereinafter referred to as "F/F") for latching data, 12a to 12d are adders, and 13 is a line memory for one line delay. Further, 14 indicates a comparator, and 16 indicates an error distribution control circuit.

The operation of this binarization circuit 1 will be explained. First, data input via the data line 100 (original image density data corresponding to the pixel position of interest (i, j)) is combined with the sum of errors allocated to the pixel position and the adder 12d.
The value is added to the comparator 14 via the signal line 300.
and is output to the error distribution control circuit 16. and comparator 1
4, the data on the signal line 300 and the average concentration calculation circuit 2
A comparison is made with the data (signal line 500) serving as a threshold value, and if the data on the data line 300 is larger than the threshold value, 1 (black) is output, and if it is smaller, 0 (white) is output to the signal line 200.

Next, the error distribution control circuit 16 outputs a corrected density signal 300 obtained by adding the distribution error to the original image density, and a corrected average density (signal from the average density correction circuit 3) based on the binary data around the pixel of interest. 400) is calculated as an error, and the error amount 160 to 190 is distributed to surrounding pixels.
control. As shown in FIG. 3, the error amount signals 160 to 190 are calculated from surrounding pixels (i-1, j+1), (i, j+1), (i
+1,j+1), (i+1,j) and the error amount already allocated to the adders 12a to 12d. Further, here, the number of pixels to which the error is distributed is set to four surrounding the pixel of interest, but the number is not limited to this, and can be easily increased or decreased using the same concept as above.

Details of the error distribution control circuit 16 are shown in FIG. 4 and will be described below. In the figure, 161 is a subtracter, 1
65a to 165d are multipliers that perform predetermined multiplication. The subtracter 161 calculates the difference (signal 300 - signal 400) is calculated and multiplier 165
It is output to a to 165d. Here, multipliers 165a to 1
65d, as shown in FIG. 3, surrounding pixels (i, j-
1, j+1) , (i, j+1) , (i+1, j+1
), (i+1, j), and by performing the multiplication shown below according to the weight, the signal lines 160, 170,
The results are output to 180 and 190. Multiplication is, for example,

[0015]

[Math 1]

Proceed as follows. FIG. 5 is a diagram showing the relationship between the pixel of interest and the pixel for calculating the average density, and FIG. 6 is a block diagram showing the configuration of the average density calculation circuit 2 in this embodiment. In FIG. 6, 22 is an adder, 23 is 1
Line memory for line delay; 25a to 25e are multipliers for multiplying input data by a certain constant; 26a to 26e;
indicate F/Fs that latch data.

In this circuit, first, binary data ("black"=1 or "white"=0) is inputted to the line memory 23 and the F/F 26d via the data line 200. As shown in Figure 5, when the pixel of interest is (i, j), F
/F26a to 26e have (i, j-1), (i-1
, j-1) , (i-2, j-1) , (i-1, j)
Binarized data (“black” = 1 or “white” = 0) corresponding to the position (i-2, j) is stored, and these data are weighted by multipliers 25a to 25e and sent to 22. Here, the weights for surrounding pixels are as shown in FIG.

[0018]

[Math 2]

is the output to the adder 22. The output from the adder 22 (signal line 500) is input to the average density correction circuit 3 and at the same time serves as a threshold value when binarizing in the binarizing circuit 1. This value is the binarized average density around the pixel of interest, and indicates the state of the density of the image around the pixel of interest. In this embodiment, in order to calculate the threshold value, the number of binarized pixels around the pixel of interest is set to five, but the number is not limited to this and can be easily increased or decreased.

FIG. 7 is a block diagram showing the configuration of the average density correction circuit 3 in this embodiment, and FIG. 10 is a diagram illustrating a part of the "black density" table in this embodiment. In the same figure, 61 is an LUT (lookup table)
, "black" density levels corresponding to each density value from 0 to 255 are stored. 62 represents a multiplier, and 63 represents a divider that performs a predetermined division.

The meaning of the term "black" density level will now be explained. Normally, when a gradation image having values in the range of 0 to 255 is binarized by the above-mentioned binarization circuit, the output is 0 (
Set the density level of output 1 (white) to 0, and set the density level of output 1 (black) to 2.
Generally, it is set to 55. However, as shown in Figure 8, actual printers have nonlinear gradation characteristics, and the degree of influence that a single black pixel has on the surrounding density differs depending on whether there are many black pixels or few surrounding black pixels. . That is, in the former case, whether a single pixel is black or white, there is no big difference in the density when looking at the area around the pixel of interest, but in the latter case, it has a large effect on the density. . Therefore, the density level of black pixels should not be uniformly set to 255, but it is better to increase the density level of black pixels when the density is low depending on the density of the image. In this way, by setting a higher threshold value and increasing the negative occurrence error compared to a method using a normal average density, it is possible to suppress the occurrence of black pixels, and it is possible to perform γ correction in accordance with the characteristics of the printer. The density level of a black pixel calculated based on this idea in accordance with the characteristics of the output printer is herein referred to as a "black" density level. FIG. 10 shows a part of a specific example of a "black" density table. For example, when binarizing a pixel with an original image density of 1, if the binarized average density of surrounding pixels is 35, in the conventional method, the threshold value is set to 35 and it is determined as 0 (white), and the generated error = 1-35 = -34 error is diffused into surrounding errors, but in this method, by referring to the table in Fig. 10, the average density is calculated as average density = 35 x 1416/2.
By setting 55=194, the generated error=1-194
= -193, which is about 6 times the negative error, so overall, the number of black pixels becomes about 1/6, and the density can be lowered. (In the conventional method, due to γ correction before binarization, pixels are treated as pixels with a density of 0, so they cannot be distinguished from pixels with an original density of 0, and no gradation is reproduced.) For all densities,
γ correction can be achieved by performing similar processing. Also,
By changing the values in the table, it is possible to perform optimal γ correction for any printer according to its characteristics.

Now, in FIG. 7, the LUT 61 calculates the product of the "black" density level according to the original image density (signal line 100), and outputs it to the divider 63 through the signal line 120. The divider 63 divides the input signal by 255 and sends the value to the signal line 4.
Output to 00. That is, the present average density correction circuit 2 uses (average density) x ("black" density level/255) as information for suppressing the occurrence of black pixels, that is, performing γ correction taking into account the characteristics of the printer. Output to the binarization circuit 1.

With the circuit of this configuration, it is possible to perform γ correction in accordance with the characteristics of the output printer without impairing the gradation in low density areas. Now, in this example, the image data is monochrome, but for example, Y (yellow), M
This embodiment can also be applied to color image data consisting of (magenta), C (cyan), and K (black), and the effects of the present invention can be applied to each of Y, M, C, and K data. It does not impair. <Second Embodiment> Now, in the first embodiment described above, the corrected density determined by the average density correction circuit 3 was obtained from the input data representing the density and the average density calculated by the average density calculation circuit 2. However, the present invention is not limited to this, and the corrected density may be calculated from the number of black strokes and the average density calculated by an average density calculation circuit.

FIG. 11 shows a second image processing apparatus according to the present invention.
FIG. 2 is a block diagram showing the configuration of an embodiment of the present invention. In the figure, 910 is a data line input to this device, which is an 8-bit (
Digital data representing density (256 gradations) is input. Reference numeral 920 indicates a data line that outputs binarized data. Reference numeral 901 indicates a binarization circuit of this embodiment, 902 an average density calculation circuit of this embodiment, and 903 an average density correction circuit of this embodiment. Also 94
0,950,960 indicate data lines, respectively.

As for the operation of the above configuration, the binarization circuit 9
01 binarizes the aforementioned 8-bit digital data from the data line 910 according to information from the average density calculation circuit 902 and the average density correction circuit 903, and converts the result to 1 (black) or 0 (white). ) is output to data line 920. The average density calculation circuit 902 refers to a region consisting of binarized pixels surrounding the pixel of interest, calculates the average density weighted with a weight mask corresponding to each pixel in this region, and outputs it to a data line 950. It also counts the number of black pixels within the reference area and outputs it to the data line 960. The average density correction circuit 903 collects binarized average density data (data line 950) and black pixel number data (data line 960).
) is used to calculate the corrected average density and output it to the data line 940.

The details of this embodiment having the above configuration will be explained with reference to the drawings. Note that in the following description, when referring to figures similar to those in the first embodiment, the description will be omitted. In particular, the binarization circuit 901 has the same configuration and function as the binarization circuit 1 of the first embodiment, so a description thereof will be omitted. FIG. 12 is a block diagram showing the configuration of the average density calculation circuit 902 in this embodiment. In FIG. 12, 922a and 922b are adders, 923 is a line memory for one line delay, 925a to 925e are multipliers that multiply input data by a certain constant, and 926a to 926e are F/Fs that latch data. It shows.

In this circuit, first, binarized data ("black"=1 or "white"=0) is inputted to the line memory 923 and the F/F 926d via the data line 920. As shown in FIG. 5, when the pixel of interest is (i, j), F/Fs 926a to 926e have (i, j-1),
(i-1, j-1) , (i-2, j-1) , (i-
Binarized data (“black” = 1 or “white” = 0) corresponding to the positions of (1, j) and (i-2, j) are stored, and these data are processed by multipliers 925a to 925e. It is weighted and becomes an input to adder 922a. Here, the weights for surrounding pixels are as shown in FIG.

[0028]

[Math 3]

##EQU1## becomes the output to the adder 22a. The output from this adder 22a (signal line 500) becomes an input to the threshold value setting circuit 3. This value is the binarized average density around the pixel of interest, and indicates the state of the density of the image around the pixel of interest. Further, the data of F/Fs 926a to 926e are also input to an adder 922b. Adder 922b
Since the input to signal line 600 is not weighted,
shows the number of black pixels in the reference area around the pixel of interest. This is input to an average density correction circuit 903, which will be described later, and is used for the purpose of correcting the average density.

In this embodiment, in order to calculate the threshold value, the number of binarized pixels around the pixel of interest is set to five, but the number is not limited to this and can be easily increased or decreased. Figure 13
FIG. 3 is a diagram illustrating a part of a "black" density table according to the second embodiment. Average density correction circuit 9 according to second embodiment
The configuration of 03 is the same as that of the first embodiment shown in FIG. 7, so its explanation will be omitted, but since the data used is different, this point will be explained below. Further, although the numbers of each unit are not shown in the drawings, the numbers used in FIG. 7 will be explained with dashes added. In this embodiment, the input data to the average density correction circuit 903 are the average density and the number of black pixels from data lines 950 and 960.

For example, when a pixel with an original image density of 1 is to be binarized, if the binarized average density of surrounding pixels is 35, in the conventional method, the threshold value is set to 35 and it is determined to be 0 (white).
The generated error = 1-35 = -34 error is diffused to the surrounding errors, but in this method, by referring to the table in Fig. 13, the average density is calculated as average density = 35 x 1272/255 =
By setting it to 174, the generated error = 1-174 = -1
73, which is about a five-fold negative error, so overall the number of black pixels is reduced to about 1/5, and the density can be lowered. In the conventional method, the pixels are treated as pixels with a density of 0 due to γ correction before binarization, so they cannot be distinguished from pixels with an original density of 0, and no gradation is reproduced. γ correction can be achieved by performing similar processing on all densities. Furthermore, by changing the values in the table, it is possible to perform optimal γ correction for any printer according to its characteristics.

When compared with FIG. 7 of the first embodiment,
In the second embodiment, the LUT 61' outputs a "black" density level corresponding to the number of black pixels (signal line 960) to the signal line 110'. Multiplier 62' calculates the product of the average density and the "black" density level and outputs the product to divider 6 through signal line 120'.
Output to 3'. The divider 63' divides the input signal by 255 and outputs the result to the signal line 400. That is, the present average density correction circuit 940 calculates (average density) x (“black” density level/
255) as a threshold for binarization, the binarization circuit 901
Output to.

With the circuit of this configuration, it is possible to perform γ correction in accordance with the characteristics of the output printer without impairing the gradation in low density areas. (Modification of Second Embodiment) Now, in the second embodiment, the average density and the number of black pixels are calculated separately, but the present invention is not limited to this, and the circuit configuration can be simplified. Therefore, the average density may be used instead of the number of black pixels without calculating the number of black pixels.

FIG. 14 is a block diagram showing the configuration of an average density correction circuit according to a modification of the second embodiment. The average density correction circuit 903 can be configured by using one LUT, as shown in FIG. The corrected average density corresponding to the average density of surrounding pixels is stored in the LUT. FIG. 15 shows a part of a specific example of this density correction table. According to this method, for example, if the density of the pixel of interest is 1 and the binarized average density is 5, in the conventional method, the generated error = 1
By first referring to the correction table, the error of -5=-4 is diffused to the surrounding pixels, and the average density is
As a result, a negative error of approximately 6 times is generated (error = 1-27 = -26), so overall, the number of black pixels is reduced to approximately 1/6, and the density can be lowered.

In the second embodiment, the image data is monochrome; however, for example, color image data consisting of Y (yellow), M (magenta), C (cyan), and K (black) may also be used. This does not impair the effect of the present invention, which allows the present embodiment to be applied to each of Y, M, C, and K data. As explained above, according to the second embodiment, it is possible to correct γ without deteriorating the gradation in the low-density area in any printer with any gradation, so it is possible to obtain a high-quality reproduced image. Can be done. <Third Example> In the first example, the average density was treated as a threshold value during binarization and as data related to error allocation, but in the third example, the average density is It is treated as data for calculating a threshold value, and at this time, the tone characteristics are determined by the density of the pixel of interest.

FIG. 16 shows the third image processing apparatus according to the present invention.
FIG. 2 is a block diagram showing the configuration of an embodiment of the present invention. In the figure, reference numeral 1100 is a data line input to the apparatus, into which digital data representing 8-bit (256 gradations) density is input. Reference numeral 1200 indicates a data line that outputs binarized data. 1001 is the binarization circuit of this embodiment, 1
002 is the average density calculation circuit of this embodiment, and 100
3 indicates a threshold value setting circuit of this embodiment. Further, 1400 and 1500 indicate data lines, respectively.

As for the operation of the above configuration, the binarization circuit 1
001 binarizes the aforementioned 8-bit digital data from the data line 1100 according to information from the threshold value setting circuit 1003, and converts the resulting 1 (black) or 0 (
white) is output to the data line 1200. The average density calculation circuit 1002 refers to a region consisting of binarized pixels surrounding the pixel of interest, calculates the average density weighted with a weight mask corresponding to each pixel in this region, and calculates the average density by weighting the data line 1500.
Output to. The threshold setting circuit 1003 calculates a binarization threshold based on the input data (data line 1100) and the binarized average density data (data line 1500), and outputs it to the data line 1400.

The details of this embodiment having the above configuration will be explained with reference to the drawings. Note that in the following description, when referring to figures similar to those in the first embodiment, the description will be omitted. Figure 1
7 is a block diagram showing an example of a specific configuration of the binarization circuit 1001 shown in FIG. 16. The configuration of the binarization circuit 1001 is almost the same as that in FIG.
The data to be compared in this embodiment is not the average density as in the first embodiment, but the threshold value data from the threshold value setting circuit 1003 is used. The overall operation of the comparator 14' is the same as that in FIG. 2, so the overall explanation will be omitted.

Error distribution control circuit 1 of binarization circuit 1001
6, the difference between the corrected density signal 300', which is the original image density plus the distribution error, and the signal from the threshold setting circuit 1003 (data line 1400), which is the average density of binary data around the pixel of interest, is calculated. The error amount 160' to 190' calculated as an error and distributed to surrounding pixels is controlled. As shown in FIG. 3, the error amount signals 160' to 190' are calculated based on the surrounding pixels (i-
1, j+1), (i, j+1), (i+1, j+1),
The amount of error already allocated to (i+1,j) and the adder 12a
'~12d' is added. Further, here, the number of pixels to which the error is distributed is set to four pixels around the pixel of interest, but the number is not limited to this, and can be increased or decreased in the same manner as described above. The details of the error distribution control circuit 16' are the same as those in FIG. 1, so the explanation will be omitted.

The configuration and functions of the average density calculation circuit 1002 and the threshold value setting circuit 1003 are as shown in FIG. 1 of the first embodiment.
Since this circuit is similar to the average density calculation circuit 2 and the average density correction circuit 3 described in Section 3, the description thereof will be omitted. As explained above, according to the third embodiment, it is possible to obtain a high-quality reproduced image because it is possible to perform γ correction without deteriorating the gradation characteristics in the low density area even in a printer having any gradation characteristics. Can be done. <Fourth Example> In the third example, the gradation characteristics were determined based on the density of the pixel of interest, but in the fourth example,
It is based on the number of black pixels of the surrounding pixels that have been binarized.

FIG. 18 shows a fourth image processing apparatus according to the present invention.
FIG. 2 is a block diagram showing the configuration of an embodiment of the present invention. In the figure, reference numeral 2100 is a data line input to the apparatus, into which digital data representing 8-bit (256 gradation) density is input. 2200 indicates a data line that outputs binarized data. 2001 is the binarization circuit of this embodiment, 2
002 is the average density calculation circuit of this embodiment, and 200
3 indicates a threshold value setting circuit of this embodiment. Further, 2400, 2500, and 2600 indicate data lines, respectively.

As for the operation of the above configuration, the binarization circuit 2
001 transfers digital data expressed in 8 bits from the data line 2100 to a threshold value setting circuit 200 (described later).
The information from 3 is binarized and the resulting 1 (black) or 0 (white) is output to the data line 2200. The average density calculation circuit 902 refers to a region consisting of binarized pixels surrounding the pixel of interest, calculates the average density weighted with a weight mask corresponding to each pixel in this region,
00, and also counts the number of black pixels in the reference area and outputs it to the data line 2600. Threshold setting circuit 200
3 is binarized average density data (data line 2500)
A binarization threshold is calculated based on the data on the number of black pixels (data line 2600), and is output to the data line 2400.

The details of this embodiment having the above configuration will be explained with reference to the drawings. Note that in the following description, when referring to figures similar to those in the second embodiment, the description will be omitted. In particular, the average density calculation circuit 2002 and the threshold value setting circuit 2003 are respectively similar to the average density calculation circuit 9 of the second embodiment.
02, has the same configuration and function as the average density correction circuit 903, and the binarization circuit 2001 has the same configuration and function as the binarization circuit 1001 of the third embodiment, Detailed explanation will be omitted.

In particular, the threshold value setting circuit 2003 has the same configuration as that shown in FIG. Therefore, according to the fourth embodiment, it is possible to correct γ without deteriorating the gradation properties in the low density area in any printer having any gradation characteristics, so it is possible to obtain high-quality reproduced images. can.

Now, a modification of the fourth embodiment is the second
Since this is the same as the modification of the embodiment, the explanation will be omitted. <Fifth Embodiment> The fifth embodiment is based on a method of controlling the gradation of an output device by correcting distribution errors.

FIG. 19 shows the fifth image processing apparatus according to the present invention.
FIG. 2 is a block diagram showing the configuration of an embodiment of the present invention. In the figure, 3100 is a data line input to the device of this embodiment,
Digital data representing 8-bit (256 gradations) density is input. 3200 indicates a data line that outputs binarized data. 3001 indicates a binarization circuit,
3002 indicates an error calculation circuit.

The binarization circuit 3001 connects the data line 3100
The data expressed in 8 bits is binarized based on information from the error calculation circuit 3002, which will be described later, and the result is 1 (black).
Or outputs 0 (white) to the data line 200. The error calculation circuit 3002 receives input data (data line 31
00), binarized output data (data line 3200)
Then, the binarization error that occurs when the pixel of interest is binarized based on the corrected input data (data line 3300) is calculated and output to the data line 3400.

The details of this embodiment having the above configuration will be explained below. FIG. 20 shows the binarization circuit 3001 shown in FIG.
FIG. 2 is a block diagram showing a specific example of the configuration. In the binarization circuit 3001 of this embodiment, the same number as the first embodiment described above is given the same number with a double dot. The binarization circuit 3001 first processes data input via the data line 3100 (target pixel position (i,
The original image density data for j) is added to the sum of errors allocated to the pixel value in an adder 12d'', and the value is output to an error calculation circuit 3002 and a comparator 14'' via a signal line 3300. be done. Then, the comparator 14'' compares the data on the signal line 3300 with the binarization threshold T (T is a predetermined value) from the signal line 3500.
If the data above 0 is larger than the threshold T, "1" is output to the signal line 320, and if it is smaller, "0" is output to the signal line 320.

Next, the error distribution control circuit 16'' calculates the error amount (error amount signal 160'') to be distributed to surrounding pixels based on the generated error (signal line 400) from the error calculation circuit 3002.
~190'').Error amount signal 160''~190
As shown in FIG. 3 (same as the first embodiment), when the pixel position of interest is (i, j), the surrounding pixel (i-
j, j+1), (i, j+1), (i+1, j+1),
(i+1, j) are given weights W1, W2, respectively, as shown in the figure.
It is weighted by W3 and W4 and added to the already distributed error amount by adders 12a'' to 12d''. Further, here, the number of pixels to which the error is distributed is set to four pixels around the pixel of interest, but the number is not limited to this and can be increased or decreased in the same manner as above.

The detailed configuration of the error distribution control circuit 16'' is shown in FIG. 21 and will be explained below.
3165a to 3165d, as shown in FIG.
, j+1), (i+1, j), the following multiplication is performed according to the weight, and the signal lines 160'', 170'', 1
Output the results to 80" and 190". Multiplication is, for example,

[0051]

[Math 4]

Proceed as follows. FIG. 22 is a block diagram showing the configuration of an error calculation circuit 3002 according to the fifth embodiment. In the figure, 3051 is an LUT in which "black" density levels corresponding to each density value from 0 to 255 are stored. 3052 indicates a subtracter, and 3053 indicates a selector. In this embodiment, as in the first embodiment, a negative error of about 6 times occurs, and overall, the number of black pixels is reduced to about 1/6, and the density can be lowered.

Now, in FIG. 22, LUT 3051 outputs a "black" density level corresponding to the original image density (signal line 3100) to signal line 3110. The subtracter 3052 is
The difference between the corrected image density (signal 3300) and the signal 3110 (signal 3300-signal 3110) is calculated and output to the selector 3053 through the signal line 3120. If the binary signal 3200 is "1", that is, the result of the binarization is black, the selector 3053 transfers the error calculated as described above (input from the signal line 3120) to the signal line 3400.
Also, if it is "0", that is, the result of binarization is white, the original image density (=input from the signal line 3300) is regarded as a generated error and is switched to the signal line 3400 and output.

With the above configuration, in the fifth embodiment, it is possible to perform γ correction in accordance with the characteristics of the output printer without impairing gradation in low density areas. Now, in the error calculation circuit 3002 in the fifth embodiment described above, the "black" density level is made variable according to the density of the pixel of interest, but other methods may be used. For example, the binarized output pixels located around the pixel of interest may be stored, and the "black" density level may be controlled according to the number of black pixels among the output pixels near the pixel of interest. In this case, a binary line buffer may be added as hardware.

[0055]

[Effects of the Invention] As explained above, according to the present invention,
Since γ can be corrected without deteriorating the gradation properties in low-density areas in any printer having any gradation characteristics, a high-quality reproduced image can be obtained.

[Brief explanation of the drawing]

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

2 is a block diagram showing an example of a specific configuration of the binarization circuit 1 shown in FIG. 1. FIG.

FIG. 3 is a diagram showing the relationship between a pixel of interest and pixels to which errors are allocated.

4 is a block diagram showing the configuration of the error distribution control circuit 16 shown in FIG. 2. FIG.

FIG. 5 is a diagram showing the relationship between a pixel of interest and pixels for calculating average density.

FIG. 6 is a block diagram showing the configuration of the average density calculation circuit 2 in this embodiment.

FIG. 7 is a block diagram showing the configuration of the average density correction circuit 3 in this embodiment.

FIG. 8 is a diagram illustrating a case where the density level of a black pixel is not constant due to nonlinear gradation characteristics of the printer.

FIG. 9 is a diagram illustrating a case where gradation is lost in a low density area due to normal γ correction.

FIG. 10 is a diagram illustrating part of a "black density" table in this embodiment.

FIG. 11 is a block diagram showing the configuration of a second embodiment of the image processing apparatus according to the present invention.

FIG. 12 is a block diagram showing the configuration of an average density calculation circuit 902 in this embodiment.

FIG. 13 is a diagram illustrating a part of a “black” density table according to the second example.

FIG. 14 is a block diagram showing the configuration of an average density correction circuit according to a modification of the second embodiment.

FIG. 15 is a diagram showing a part of a specific example of a density correction table according to a modification of the second embodiment.

FIG. 16 is a block diagram showing the configuration of a third embodiment of the image processing apparatus according to the present invention.

17 is a block diagram showing an example of a specific configuration of the binarization circuit 1001 shown in FIG. 16. FIG.

FIG. 18 is a block diagram showing the configuration of a fourth embodiment of the image processing apparatus according to the present invention.

FIG. 19 is a block diagram showing the configuration of a fifth embodiment of an image processing apparatus according to the present invention.

20 is a block diagram showing a specific example of the configuration of the binarization circuit 3001 shown in FIG. 19. FIG.

FIG. 21: Error distribution control circuit 16'' according to the fifth embodiment.
FIG. 2 is a block diagram showing the detailed configuration of the device.

FIG. 22 is a block diagram showing the configuration of an error calculation circuit 3002 according to a fifth embodiment.

[Explanation of symbols]

1,901,1001,2001,3001 Binarization circuit 2,902,1002,2002 Average density calculation circuit 3,903 Average density correction circuit 11a-11d, 11a'-11d', 11a"-11
d” F/F 12a~12d, 12a'~12d', 12a''~12
d" Adder 13, 13', 23, 923 Line memory 14, 1
4', 14" comparator 16, 16', 16" error distribution control circuit 22, 92
2a, 922b Adders 25a to 25e, 62, 165a to 165d, 925a
~925e Multipliers 26a~26e, 926a~926e F/F61,
71,3051 LUT 63 Divider 100,200,300,400,500,600
Data lines 161, 3052 Subtracters 1003, 2003 Threshold setting circuit 3002 Error calculation circuit 3053 Selectors 3165a to 3165d Multiplier

Claims (14)

[Claims] Claim 1: Binarize input multi-value density data,
In an image processing device that outputs the binarized data as binary data, a storage means for storing a plurality of binary data, and a calculation for calculating an average density based on the plurality of binary data stored in the storage means. means, a correction means for correcting the average density calculated by the calculation means based on a predetermined gradation characteristic corresponding to density data of the pixel of interest, and calculating an error based on the average density corrected by the correction means, error allocation means for allocating the calculated error to density data of non-binarized pixels after the pixel of interest, and errors obtained from pixels before the pixel of interest distributed by the error allocation means and the density of the pixel of interest; An image processing apparatus comprising: binarization means for binarizing the data obtained by adding the data, using the average density as a threshold value. 2. The image processing apparatus according to claim 1, wherein the predetermined gradation characteristic corresponds to a gradation characteristic of an output device. 3. The image processing apparatus according to claim 2, wherein the correction means has a table storing the tone characteristics. Claim 4: Binarize the input multivalue density data,
An image processing device that outputs the binarized data as binary data includes a storage unit for storing a plurality of binary data, and a storage unit for calculating an average density based on the plurality of binary data stored in the storage unit. a second calculation means for calculating the number of black pixels based on a plurality of binary data stored in the storage means; and a second calculation means for calculating the average density calculated by the first calculation means. a correction means for correcting based on the number of black pixels calculated by the means and a predetermined gradation characteristic; an error is calculated based on the average density corrected by the correction means; An error allocation means that allocates the density data of the valued pixel, and data obtained by adding the error obtained from the pixels before the pixel of interest distributed by the error allocation means and the density data of the pixel of interest, is added to the average density. An image processing device comprising: binarization means for binarizing as a threshold value. 5. The image processing apparatus according to claim 4, wherein the predetermined gradation characteristics correspond to gradation characteristics of an output device. 6. The image processing apparatus according to claim 5, wherein said correction means has a table storing said gradation characteristics. Claim 7: Binarize the input multivalue density data,
In an image processing device that outputs the binarized data as binary data, a storage means for storing a plurality of binary data, and a calculation for calculating an average density based on the plurality of binary data stored in the storage means. means, a correction means for correcting the average density calculated by the calculation means based on a predetermined gradation characteristic corresponding to density data of the pixel of interest, and calculating an error based on the average density corrected by the correction means, error allocation means for allocating the calculated error to density data of non-binarized pixels after the pixel of interest, and errors obtained from pixels before the pixel of interest distributed by the error allocation means and the density of the pixel of interest; an image processing apparatus comprising: binarization means for binarizing the data obtained by adding the data, using the average density corrected by the correction means as a threshold value. 8. The image processing apparatus according to claim 7, wherein the predetermined gradation characteristic corresponds to a gradation characteristic of an output device. 9. The image processing apparatus according to claim 8, wherein the correction means has a table storing the tone characteristics. 10. An image processing device that binarizes input multivalued density data and outputs the binarized data as binary data, comprising: a storage means for storing a plurality of binary data; a first calculation means for calculating an average density based on a plurality of binary data stored in the storage means; a second calculation means for calculating the number of black pixels based on a plurality of binary data stored in the storage means; , a correction means for correcting the average density calculated by the first calculation means based on the number of black pixels calculated by the second calculation means and a predetermined gradation characteristic corresponding to the density data of the pixel of interest; error distribution means for calculating an error based on the average density corrected by the correction means, and distributing the calculated error to density data of non-binarized pixels subsequent to the pixel of interest; and The present invention is characterized by comprising a binarization means for binarizing data obtained by adding an error obtained from a pixel before the pixel of interest and the density data of the pixel of interest, using the average density corrected by the correction means as a threshold value. Image processing device. 11. The image processing apparatus according to claim 10, wherein the predetermined gradation characteristic corresponds to a gradation characteristic of an output device. 12. The image processing apparatus according to claim 11, wherein said correction means has a table storing said gradation characteristics. 13. A binarization device for inputting multivalued image data and outputting binary image data, comprising: error allocation means for allocating a binarization error of a pixel of interest to unprocessed pixels; Binary determination means for performing binary determination of a pixel of interest based on the error distributed by the means and the original multilevel image density, and storage means for storing gradation characteristics of an output device; 2. A binarization device, characterized in that the gradation of an output device is controlled by correcting the distribution error in the error distribution means based on the error distribution means. 14. The image processing apparatus according to claim 13, further comprising a table storing the tone characteristics.