JPH11146201A - Correction data generating device and error processor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an error processor, with which the generation of a texture pattern is especially prevented and printing quality is improved in the error processor which uses an error diffusing method as a binarizing method for reproducing the gradation of gradation images. SOLUTION: A preprocessing part (correction data generating part) 13 composed of a detection filter 15, a random pattern switcher 17 and a random number generator 16 or the like detects pixels data of uniform density with a certain fixed width and makes them into pixel data so as to be random gradations. An error diffusing processing part 14 composed of a comparator 21, an error filter 22 and a filter selecting part 23 or the like performs error processing to those pixel data and at such a time, processing is performed while selecting the suitable filter according to the scanning direction of the error processing. Thus, random processing is performed in the preprocessing without changing the total data and further, the generation of a texture pattern is prevented by the suitable filter, based on the error filter.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to an error processing device using an error diffusion method as a binarization method for reproducing gradation of a grayscale image.

[0002]

2. Description of the Related Art Hitherto, a dither method using an independent decision method or an error diffusion method (conditional decision method) has been used as a binarization technique for reproducing the gradation of a grayscale image. However, in the independent determination method, since the threshold value is determined using random random numbers, granularity noise is present on the entire image, and the image quality is inferior. Therefore,
A dither method using an error diffusion method (conditional decision method) is widely used. In the error diffusion method, when a certain pixel is binarized, the density of the pixel is corrected by a weighted error from a neighboring pixel that has already been binarized, and the binarization is performed. The error of the pixel is used for calculation of a new pixel later.

FIG. 13 is a diagram for explaining a conventional error diffusion method. In the figure, D1, D2, D3, and D4 are reference pixels (peripheral pixels) for binarizing a target pixel indicated by *, and a predetermined weighting coefficient is set for the target pixel from the positional relationship. Have been. For example, the target pixel F
For (x, y), the weighting factor of the peripheral pixel D1 is “W1”, the weighting factor of D2 is “W2”, the weighting factor of D3 is “W3”, and the weighting factor of D4 is “W1”. W4 ". The error value of each of the peripheral pixels D1 to D4 is “e1” for the peripheral pixel D1, “e2” for D2, “e3” for D3, and “e4” for D4.

In the conventional error diffusion processing, assuming that the corrected image data of the target pixel * is f (x, y), an error (W1 × e1 +
W2 * e2 + W3 * e3 + W4 * e4) is calculated.
Then, the error and the data of the original target pixel (F
(X, y)) and f (x, y) = F (x, y) + W1 × e1 + W2 × e2
+ W3 × e3 + W4 × e4 is obtained.

[0005] Further, the above-mentioned added value is compared with an appropriate threshold value T by a comparator.
(X, y). That is, f (x, y) <T
Then h (x, y) = 0 and f (x, y) ≧ T
Then, h (x, y) = 1.

Further, the correction data f (x,
Assuming that the error of y) is e5, it can be expressed by e5 = f (x, y) -h (x, y). The error data e5 is used when performing the next target pixel correction process.

[0007]

However, the following problems occur in the printing process using the conventional error diffusion method. (A) First, in the case of an image of an equal density level where the density falls within a certain level, the level of peripheral pixels used when calculating the error of the target pixel is also constant, for example, as shown in FIG.
As shown in FIG. 6, a texture pattern is generated in an image of the same density level, and the image becomes difficult to see. (B) If the error signal created by the conventional processing has a periodicity, it may cause the generation of a texture pattern. For example, as shown in FIG. 15B, in an image in which the density gradually changes, a texture pattern is generated around 33% density.

An object of the present invention is to provide an error processing apparatus which prevents the occurrence of a texture pattern and has excellent print quality in view of the above-mentioned conventional circumstances.

[0009]

According to a first aspect of the present invention, there is provided a selection means for selecting a predetermined number of continuous image data from image data having a constant numerical value. For a predetermined number of image data selected by
Adding / subtracting means for adding / subtracting a numerical value whose sum of plus / minus is the same in the entire image data with respect to each image data, and corresponding to the average value of the predetermined number of image data with respect to the operation result of the adding / subtracting means This can be attained by providing a correction data creating apparatus having a superimposing unit that superimposes a numerical value and sets the correction data of the continuous predetermined number of image data.

Here, the selecting means is pixel data in which numerical values of a certain width, for example, the same numerical value or a numerical value closer thereto, are continuous, for example, 50, 50, 50, 50, 50, 51, 5
3, 52, 54, 53, and so on.
When a printing process is performed using image data in which such a numerical value having a constant width is continuous as it is, a so-called texture (false image) occurs. Therefore, a predetermined number (for example, three, four,
, 5 pieces, etc.) of pixel data, and, for example, add a specific numerical value within a range where the sum of the pixel data does not change.
/ Minus, and superimpose a numerical value corresponding to the average value of the predetermined number of pixel data on the calculation result of the addition / subtraction to obtain correction data.

With this configuration, pixel data of a fixed width in which texture (false image) occurs is corrected,
This is to prevent the generation of texture (fake image). According to a second aspect of the present invention, in the first aspect of the present invention, the numerical values to be added or subtracted by the addition / subtraction means are the same, and the image data to be added and the image data to be subtracted are random. It is a configuration that can be switched.

For example, the numerical value to be added or subtracted is set in advance, the numerical value is added to certain pixel data, the same numerical value is subtracted to certain pixel data, and the sum of a predetermined number of selected pixel data is calculated. Are the same, but pixel data having a difference exceeding the predetermined width between the respective pixel data is created. With this configuration,
This prevents the occurrence of texture.

According to a third aspect of the present invention, in the second aspect, the image data selected by the selection means is, for example, three, and the same image data is provided for the image data on both sides of three continuous image data. In this configuration, addition and subtraction of numerical values are performed.

In this embodiment, for example, a numerical value in the left pixel data of three consecutive pixel data is added, and the same numerical value is subtracted in the right pixel data. In the same manner, pixel data having a difference exceeding the predetermined width is created between the respective pixel data to prevent generation of texture.

In this case, the pixel data in the middle is the original value, and is not added or subtracted. According to the fourth aspect of the present invention, there are provided a plurality of filters, each of which is weighted by a position with respect to a target pixel; a selection unit that selects the filter according to a scanning direction of an error process of image data; An adding unit that calculates an error of the target pixel according to the selected filter and adds the value of the target pixel to the calculation result; and an output unit that compares the addition result of the adding unit with a threshold and outputs the result as image data. This can be achieved by providing an error processing device.

Here, the above-mentioned filter is a peripheral pixel specific to the target pixel, for example, a pixel immediately above the target pixel, pixels to the right and left of the target pixel, and pixels to the left of the target pixel. The selecting means selects the filter in accordance with the scanning direction of the error processing. For example, when scanning in the right direction, selecting the arrangement of the peripheral pixels with respect to the target pixel, and when scanning in the left direction, selecting other peripherals with respect to the target pixel. Select an array of pixels. If the weights are different, filters having the same arrangement of peripheral pixels can be selected.

With this configuration, the error of the target pixel is calculated in accordance with the selected filter, and the result of adding the value of the target pixel to the calculated result is, for example, print data when the result is larger than the threshold, and is smaller than the threshold. In this case, printing processing is performed as non-print data to prevent the occurrence of a texture pattern.

According to a fifth aspect of the present invention, there is provided a selection means for selecting a predetermined number of continuous pixel data from a plurality of pixel data having a constant width and a predetermined number of pixel data selected by the selection means. Adding / subtracting means for adding / subtracting a numerical value whose sum of plus / minus is the same in the entire pixel data with respect to the pixel data, and adding / subtracting the individual image data; A correction data generating unit having a superimposing unit that superimposes a value corresponding to an average value and obtains correction data of the continuous predetermined number of pixel data; a plurality of filters weighted according to a position with respect to the target pixel; Selecting means for selecting the filter according to the scanning direction of the data error processing; calculating an error of the target pixel according to the filter selected by the selecting means; The error processing unit includes an adding unit that adds a value created by the correction data creating unit, which is a target pixel, and an output unit that compares an addition result by the adding unit with a threshold and outputs the result as print image data. This can be achieved by providing an error processing device.

According to the present invention, error diffusion processing is performed using the pre-processing according to the first aspect, that is, the data created by the correction data creating means as target pixel data, and using the processing device according to the fourth aspect. In addition, even with such a configuration, it is possible to provide an error processing device that prevents the occurrence of a texture pattern, and therefore, it is possible to perform print processing with excellent print quality using image data that has been subjected to this error processing.

According to a sixth aspect of the present invention, there is provided a detecting means for detecting a regularity of each pixel after error processing, and an addition / subtraction for adding or subtracting a random number when the regularity is detected by the detecting means. Means, an error calculating means for calculating an error of a result obtained by addition or subtraction by the addition / subtraction means, an addition means for adding the value of the target pixel to the calculation result, and an addition result by the addition means. Compare with threshold
This can be achieved by providing an error processing device having output means for outputting as value data.

That is, according to the present invention, when an error after comparison with a predetermined threshold value and output as binary data has a regularity, the regularity is corrected using a random number or the like. The error diffusion processing by the error filter is performed.

With such a configuration, the regularity of the data included in the error is eliminated, and the error processing is performed using the error data in which the regularity has been eliminated in advance, so that an image without a texture pattern can be obtained. Is to create.

According to a seventh aspect of the present invention, the numerical values to be added or subtracted by the adding / subtracting means are the same, and the pixel data to be added and the pixel data to be subtracted are the same. This is a configuration that can be switched at random.

According to an eighth aspect of the present invention, in the seventh aspect, the pixel data processed by the detecting means is 3 bits.
And the same numerical value is added to and subtracted from pixel data on both sides of the continuous pixel data.

The claims 7 and 8 correspond to the claims 2 and 3, and the claims 7 and 8 of this embodiment have the same configuration for the error. Corresponding. That is, for example, the numerical values of the random numbers to be added and subtracted are the same, the target pixel data is switched at random, and processing is performed on, for example, three pixels.

[0026]

Embodiments of the present invention will be described below in detail with reference to the drawings. <First Embodiment> FIG. 2 is a system configuration diagram of a printing apparatus for explaining an error processing apparatus according to a first embodiment of the present invention. That is, the error processing device used in the present example is included in a printer connected to a host device such as a personal computer.

In FIG. 1, a printer 1 has a CPU.
2, system ROM3, system RAM4, font R
OM 5, operation panel interface (hereinafter referred to as operation panel I / F) 6, video interface (hereinafter referred to as video I / F) 7, parallel interface (hereinafter referred to as parallel I / F)
8. Serial interface (hereinafter, serial I / F)
9), card interface (hereinafter, card I)
/ F) and a printer engine 11.

The system ROM 3 stores a program for controlling the system of the printer 1, and the CPU 2 controls the printing according to the program. The system ROM 3 also stores an error processing program 3a used in this embodiment, and the CPU 2 performs a binarization process according to this program during the above-described printing process.

The system RAM 4 is a memory used as a work area when the CPU 2 performs the above control processing.
a, an area of the frame memory 4b and the like are secured.
The font ROM 5 is a memory having character fonts such as a Mincho font, a Gothic font, and the like according to the character type and character size used for printing in the printer device 1 of the present embodiment. Character data is stored in a format.

The operation panel I / F 6 is connected to an operation panel (not shown), and outputs a key operation signal supplied from the operation panel to the CPU 2.
A display signal to the operation panel is output from the CPU 2 to the operation panel.

The video I / F 7 is connected to the printer engine 11 to transmit and receive control signals between the CPU 2 and the printer engine 11 and to control output of data developed in the frame memory 4b to the printer engine 11.
The printer engine 11 performs high-voltage control of a print head (not shown), a charger and a developing device in the image forming unit, and controls driving of a paper transport system.

On the other hand, the parallel I / F 8 controls input / output of parallel data such as a Centronics interface, exchanges data with, for example, a host device, and inputs print information to be printed and controlled by the printer device 1 from the host device. The serial I / F 9 is, for example, RS-232.
This interface is used to connect a C-compliant connector and transfer serial print data.

The card I / F 10 is a card in which programs and data are written in the printer 1 of the present embodiment,
In this configuration, a card on which font data is written, such as a font card, is inserted.

FIG. 1 is a diagram for explaining an error diffusion process performed by the CPU 2 using the system RAM 4 in the above-described printer system. The error diffusion processing of the present example includes a preprocessing unit (correction data creation unit) 13 and an error diffusion processing unit 14.
It consists of. The pre-processing unit (correction data creating unit) 13 includes a detection filter 15, a random number generator 16, a pattern random switching unit 17, a superposition calculator 18, and an adder 19.

Image data, which is multi-valued data, is supplied to the detection filter 15, for example, every three bits.
9 is output. The random number generator 16 generates a random number within a preset range. For example, in the description of this example,
It is assumed that a numerical value among “1” to “64” is randomly generated. The pattern random switch 17 randomly switches the pixels that supply the numerical values generated by the random number generator 16.

Further, the superimposition value calculator 18 calculates a superimposition value corresponding to the average value of the three pixel data input via the detection filter 15. FIG. 3 is a diagram for explaining this processing. The average value (di) of three pieces of input data (pixel data) is shown.
n) is “128”, the superimposition rate is 100%, and even if the average value (din) of the input data (pixel data) is large,
The calculation is performed based on a curve in which the superimposition rate decreases as shown in FIG. For example, input data (pixel data)
When the average value (din) is “128”, the superimposition rate is 100%, the data output from the pattern random switch 17 is multiplied by the superimposition value “1”, and the result is output to the adder 19. Further, for example, when the average value (din) of the input data (pixel data) is “100”, the superimposition rate is set to 50%, the data output from the pattern random switch 17 is multiplied by the superimposition value “0.5”, and added. Output to the container 19.

The adder 19 adds the data output from the superimposition calculator 18 to the above-mentioned three pixel data supplied via the detection filter 15 and outputs the result to the error diffusion processor 14.

The error diffusion processor 14 comprises an adder 20, a comparator 21, an error filter 22, a filter selector 23, and an adder / subtractor 24. The error diffusion processor 14 is supplied from the above-described preprocessor (correction data generator) 13. The correction data is input to the adder 20.

The error filter 22 calculates an error value of each pixel according to the filter selected by the filter selection unit 23.
Here, the filter selection unit 23 sequentially selects a filter from, for example, six filters and outputs the selected filter to the error filter 22. FIG. 4 shows an example of the error filter. In the figure,
D1, D2, D3, and D4 are reference pixels (peripheral pixels) when binarizing the target pixel indicated by *, and a predetermined weighting coefficient is set for the target pixel from the positional relationship. For example, in the example of FIG. 4, the target pixel F (x, y)
In contrast, the weighting factor of the peripheral pixel D1 is “1”, the weighting factor of D2 is “5”, the weighting factor of D3 is “3”, and the weighting factor of D4 is “7”.

In the example of FIG. 5, the target pixel F (x,
y), D1 ′, D2 ′, D3 ′, and D4 ′ are reference peripheral pixels when binarizing the target pixel, the weighting factor of D1 ′ is “2”, and the weighting factor of D2 ′ Is "3", the weighting coefficient of D3 'is "2",
The weighting coefficient of D4 'is "5". Although not shown, in this example, there are four other filters as described above.

The filter selecting section 23 selects one of the above six filters and calculates an error value by the error filter 22. This calculation result is supplied to the adder 20, added to the data after the preprocessing described above, and supplied to the comparator 21.

The comparator 21 compares the added data with the threshold value, converts the value into a binary value of "1"("255") print data and "0" non-print data, and outputs the binary value. The adder / subtractor 24 performs an addition process or a subtraction process on the output supplied from the comparator 21 and outputs the result to the error filter 22. In the error processing device having the above configuration,
The processing operation will be described below.

First, print data supplied from an external host device is supplied to the system RAM 4 and the system RA 4
After being stored in the reception buffer in the M4, it is analyzed under the control of the CPU 2, and for example, multi-valued image data is converted into binary image data.

Next, the above-described image data is output to the detection filter 15 shown in FIG. 1, for example, three consecutive pixel data are input to the detection filter 15. FIG. 6A shows three data areas of the detection filter 15, and areas A,
This indicates that pixel data is input serially in the order of B and C. Therefore, three consecutive pixel data are stored in the area A,
B and C are stored. The superimposition value calculator 18 calculates the average value of the input pixel data, and obtains the superimposition rate corresponding to the average value from FIG. For example, when three consecutive pixel data are “100”, “100”, and “100”, the average value is “100”, and thus the superimposition rate of “0,5” is set.

On the other hand, the random number generator 16 outputs "0" to "6".
A certain numerical value from among “4” to “4” is supplied to the pattern random switch 17 as a random number. For example, at this time, when the value of the numerical value “64” is supplied to the pattern random switch 17 as a random number, the pattern random switch 1
The data shown in FIG. 6B is stored in the area 7. That is, a random number of −r (−64) is input to the left area, a random number of + r (+64) is input to the right area, and “0” data remains in the middle area.

This data is supplied to the above-described superimposed value calculator 18, where the data of (a) and (b) shown in FIG. 6 are added and output to the error diffusion processing unit 14. For example,
As shown in FIG. 6, the data of the areas A, B, and C is "10".
In the case of “0”, “100”, and “100”, the superimposition rate is “0,5” as described above, and the outputs of the superimposition value calculator 18 are A-, B-, and C-“-32”, “ 0 ”and“ 32 ”, and the output results from the adder 19 are“ 68 ”and“ 10 ”.
0 "and" 132 ".

As shown in FIG. 7, when three consecutive pixel data are “130”, “128”, and “126”, the superimposition ratio is “1”. Are "98", "128", and "158". The same applies to other numerical values.

The correction data created by the preprocessing unit (correction data creation unit) 13 in this manner is added to the adder 20.
Is supplied to the comparator 21 via Here, the error filter 22 calculates an error with respect to the target pixel using the filter selected by the filter selection unit 23.

FIG. 8 is a diagram showing the relationship between the scanning direction of the filter and the filter to be selected. For example, the filter 1 is used in the + direction of the nth line, and the filter 2 is used in the + direction of the (n + 1) th line. In the negative direction of the (n + 2) th line, the filter 3 is used.
In the case of the direction, the filter 4 is used. Further, the filter 5 is used in the + direction of the (n + 4) th line, the filter 6 is used in the + direction of the (n + 5) th line, and the process returns to the previous use of the filter 1 in the (n + 6) th line. The above-described filter selection processing is repeated, and the error of the target pixel is calculated using the selected error filter 22.

For example, if the filter 1 is configured as shown in FIG. 4, when scanning in the + direction, the weight of the peripheral pixel to the target pixel * is as shown in FIG. 9A, and the filter 3 is configured as shown in FIG. Then, when scanning in the negative direction, the weighting of the peripheral pixel to the target pixel * is as shown in FIG. 9B.

The error filter 22 calculates an error value of the target pixel based on the selected filter. This calculation formula is as shown below. ΔDm, n = (1 / Σαi, j) ΣΔi, j ・ en-1, n-1 Formula (1) where ΔDm, n is an error value of the target pixel, and this error The value is output to the adder 20, and the data (D
m, n). The calculation formula is as shown below. D'm, n = Dm, n + .SIGMA.Dm, n ... Equation (2) Further, the added data is compared with a threshold value by the comparator 21 to perform printing or non-printing. This calculation formula is as shown below. D′ m, n ≧ Tga, n = 1 (print / black) D′ m, n <Tga, n = 0 (non-print / white) Expression (3) That is, the addition value is a threshold. If the value is larger than a value (eg, “128”), data “1” is output. If the added value is smaller than a threshold value (eg, “128”), data “0” is output. Output.

The error data generated at the target pixel is calculated by the following formula. em, n = △ Dm, n−gm, n (4) This error data is used in the next error diffusion process.

In the binary print data output as described above, uniform pixel data is detected by the pre-processing unit (correction data creation unit) 13, and an equal density portion excluding the edge portion is extracted. Since the processing has been performed, the data is data with the texture pattern removed.

The error diffusion processing unit 14 performs the error diffusion processing using six types of filters having different weights, and uses an appropriate filter depending on the scanning direction of the error processing. This is to realize a smooth image while reducing the texture pattern generated in the density level area.

In the above description of the embodiment, "64" is used as a random number. However, the random number is not limited to "64", and the range of the random number is also limited to the range of "0" to "64". Absent.

The number of pixels stored in the detection filter 15 is not limited to three, but may be four, five, six,.... In this case, the corresponding number of registers (areas)
Is required for the detection filter 15 and the pattern random switch 17. <Second Embodiment> Next, a second embodiment of the present invention will be described.

This embodiment differs from the above embodiment in that data supplied to the error filter 22 is different. That is, in the present example, as shown in FIG.
In addition, it comprises an error detection unit 30, a random number generation unit 31, a pattern random switching unit 32, and an error calculation unit 33. Here, the comparator 27 and the error filter 28 have the same configuration as in the above-described embodiment, but an error detection unit 30, a random number generation unit 31, a pattern random switching unit 32, and an error calculation unit 33 are newly added. It is.

The error detecting section 30 detects whether or not the print data output from the comparator 27 has a regularity. If the print data has a regularity, the random number generator 31 outputs a random number. Here, the random number to be used is “0” as described above.
The random number is output to the pattern random switch 32. The error calculator 33 calculates data output from the pattern random switch 32 and data output from the error detector 30 and outputs the calculated data to the error filter 28. In addition, the pattern random switch 3
2 is the same as that of the pattern random switch 17 shown in FIG. The processing operation of the error processing device having the above configuration will be described below.

Also in this example, print data supplied from an external host device is supplied to the system
The multivalued image data is converted into binary image data according to the control of PU2. That is, the adder 20 shown in FIG.
For example, multi-valued data of "0" to "255" is input and compared with a threshold value in the comparator 27. For example, multi-valued data larger than the threshold value of "128" is regarded as print data "1" and a threshold value. Multi-value data smaller than the value “128” is output as non-print data “0”.

On the other hand, the error when the multi-value data is converted into the binary data as described above is added or subtracted by the adder / subtractor 30 and output to the error detector 30. The error detector 30 stores a plurality of data supplied from the adder / subtractor 34, and when the data has regularity, converts the data and the random number data output from the pattern random switch 32 into an error calculator. 33. Then, both data are added in the error calculator 33.

For example, as shown in FIG. 11, data having regularity in the n-th to (n-7) -th pixels before the target pixel, for example, "80", "0", "80", "0",
In the case of data of “80”, “0”, “80”, and “0”, these data are detected by the error detection unit 30, and the random number data supplied from the pattern random switch 32 in the error calculator 33 Is added. In this case, the pattern random switch 32 is driven in the same manner as the pattern random switch 17 used in the above-described first embodiment, and outputs a random number r output from the random number generator 31, for example, “64” to three pixels. Are added and subtracted on the left and right sides, and output to the error calculator 33.

For example, as shown in FIG. 11, n-7 to n-5
"-64" supplied from the pattern random switch 32 to the second data "80", "0", "80",
"0", "+64" are added, and "16", "0", "+
144 ". Also, for the next n−4 to n−2 data “0”, “80”, “0”, “−64” supplied from the pattern random switch 32,
“0” and “+64” are added, and the error data is changed to “−6”.
4 "," 80 ", and" +64 ". Hereinafter, the arithmetic processing is performed in the same manner to correct the error data having regularity.

The error data whose regularity has been corrected in this way is output to the error filter 28, and error processing is performed on the target pixel in accordance with the error data having no regularity. This processing is the same as that of the above-described embodiment, and the error filter 2
2 calculates △ Dm, n from the above error data according to the above-described calculation formula (1), and then calculates the Dm, n ′ by the adder 20 using the calculation formula (2). This data is the data of the target pixel after the error correction according to the present embodiment. By comparing this data with the threshold value by the comparator 27, it is possible to output image data in which a texture pattern does not occur.

That is, according to the present embodiment, the error detector 33 corrects the regular error data by the random number data supplied from the pattern random switch 32 in the error calculator 33, and generates a texture pattern. Can be supplied to the printing unit.

In this example, "64" is used as a random number used for error diffusion. However, the number is not limited to "64", and the range of the random number is in the range of "0" to "64". It is not limited.

The error processing is not limited to every three pixels,
Every pixel, every 5 pixels, every 6 pixels,...
Further, in the above two embodiments, 4
Although the reference pixels are used, a configuration may be used in which 12 reference pixels are referred to a target pixel indicated by * as shown in FIG. It should be noted that the numerical values shown in the figure are weighting values of the reference pixels at the respective positions.

[0067]

According to the present invention, generation of a texture pattern can be prevented, and an image having excellent print quality can be created.

Further, when uniform data continues, preprocessing is performed, so that the generation of a texture pattern can be further prevented, and an image with more excellent print quality can be obtained.

[Brief description of the drawings]

FIG. 1 is a diagram illustrating an error diffusion process performed by a CPU using a system RAM.

FIG. 2 is a system configuration diagram of a printing apparatus illustrating an error processing apparatus according to the first embodiment.

FIG. 3 is a diagram illustrating a process of calculating a superimposition value.

FIG. 4 is a diagram illustrating an example of an error filter.

FIG. 5 is a diagram illustrating another example of the error filter.

FIGS. 6A and 6B are diagrams for explaining preprocessing.

FIGS. 7A and 7B are diagrams illustrating pre-processing.

FIG. 8 is a diagram illustrating a relationship between a filter scanning direction and a filter to be selected.

9A is an example of a filter for a target pixel * when scanning in the + direction, and FIG. 9B is an example of a filter for the target pixel * when scanning in the − direction.

FIG. 10 is a diagram illustrating an error processing device according to a second embodiment.

FIG. 11 is a diagram illustrating an example of data having regularity in nth to n-7th pixels that are pixels before a target pixel.

FIG. 12 is a diagram illustrating an example of another filter.

FIG. 13 is a diagram illustrating a conventional error diffusion method.

FIG. 14 is a system diagram of error processing by an error filter.

FIG. 15A is an example in which a texture pattern is generated in an image at an equal density level, and FIG. 15B is an example in which a texture pattern is generated in the vicinity of a density of 33%, for example, in an image whose density gradually changes. .

[Explanation of symbols]

 Reference Signs List 1 printer device 2 CPU 3 system ROM 4 system RAM 5 font ROM 6 operation panel I / F 7 video I / F 8 parallel I / F 9 serial I / F 10 card I / F 11 printer engine 13 preprocessing unit (correction data) Creation unit) 14 Error diffusion processing unit 15 Detection filter 16 Random number generator 17 Pattern random switch 18 Superposition calculator 19, 20 Adder 21 Comparator 22 Error filter 23 Filter selection unit 24 Addition / subtraction unit 27 Comparator 28 Error filter 30 Error Detector 31 Random number generator 32 Pattern random switch 33 Error calculator

Claims (8)

[Claims] 1. A selecting means for selecting a predetermined number of continuous pixel data for pixel data having a constant width of numerical values, and a sum of plus / minus for a predetermined number of pixel data selected by the selecting means. Is an addition / subtraction unit that adds / subtracts a numerical value that is the same in the entire pixel data, and superimposes a numerical value corresponding to an average value of the predetermined number of pixel data on an operation result of the addition / subtraction unit, And a superimposing means for making correction data of the continuous predetermined number of pixel data. 2. A numerical value to be added or subtracted by said adding / subtracting means is the same, and pixel data to be added and pixel data to be subtracted are randomly switched. The described correction data creation device. 3. The pixel data selected by the selection means is 3
3. The correction data generating apparatus according to claim 2, wherein the same numerical value is added to and subtracted from pixel data on both sides of the continuous pixel data. 4. A plurality of filters whose weights are set according to their positions with respect to the target pixel; a selecting means for selecting the filter according to the scanning direction of the error processing of the pixel data; and a filter for selecting the target pixel according to the filter selected by the selecting means. Adding means for calculating an error and adding the value of the target pixel to the calculation result; and output means for comparing the addition result by the adding means with a threshold value and outputting the result as error-processed pixel data. Characteristic error processing device. 5. A selecting means for selecting a predetermined number of continuous pixel data for pixel data having a constant numerical value, and a selecting means for selecting a predetermined number of pixel data selected by the selecting means.
Adding / subtracting means for adding / subtracting a numerical value whose sum of plus / minus is the same in the entire pixel data with respect to each image data, and corresponding to the average value of the predetermined number of pixel data with respect to the operation result of the adding / subtracting means A correction data generating unit having a superimposing unit that superimposes a value to generate correction data of the continuous predetermined number of pixel data; a plurality of filters weighted according to a position with respect to the target pixel; Selecting means for selecting the filter according to the scanning direction, calculating an error of the target pixel according to the filter selected by the selecting means,
An adding unit that adds a value created by the correction data creating unit, which is the target pixel, to the calculation result; and an output unit that compares the addition result by the adding unit with a threshold and outputs the result as print image data. An error processing device, comprising: an error processing unit. 6. A detecting means for detecting the regularity of each pixel after the error processing, an adding / subtracting means for adding or subtracting a random number when the regularity is detected by the detecting means, Error calculating means for calculating an error of a result obtained by subtraction; adding means for adding the value of the target pixel to the calculation result; comparing the addition result by the adding means with a threshold value; An error processing device, comprising: output means for outputting as an output. 7. A numerical value to be added or subtracted by said adding / subtracting means is the same, and pixel data to be added and pixel data to be subtracted are randomly switched. An error processing device as described. 8. The pixel data processed by the detection means is 3
8. The error processing apparatus according to claim 6, wherein the same numerical value is added to and subtracted from pixel data on both sides of the continuous pixel data.