JP2001111829A - Dither generating device and its method and printer - Google Patents

Abstract

PROBLEM TO BE SOLVED: To efficiently generate an optimum blue noise mask in a short time where no texture is produced at a highlight part. SOLUTION: An initial dither matrix setting section 10 sets an initial dither matrix 16 where pixels are arranged in nearly a form of a regular rectangle corresponding to a prescribed gray level of the highlight side. An array generating section 12 generates an array 26 storing all pixels of a matrix picture in the arrangement order as the blue noise mask by seeking an air gap position placed remotest from a pixel placed at present as a position of a maximum value of a potential matrix and arranging the pixels in order based on the initial dither matrix 16.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a dither generating apparatus and method for inputting a matrix image having a large vertical and horizontal size and generating a blue noise mask serving as a base of a dither matrix used for binarization or multi-level conversion. In addition, the present invention relates to a printer, and more particularly, to a dither creating apparatus and a creating method capable of efficiently creating an optimal blue noise mask, and a printer.

[0002]

2. Description of the Related Art In recent years, with the spread of electronic computers and color printers, printing of color images and handling of color information such as color copies have become widespread. In the case of a printer that prints a color image, the original color information of, for example, 256 gradations of RGB must be converted into color information of a small number of gradations, for example, 5 gradations that can be printed by gradation change due to dot density and dot size. No.

In the case of a printer having a fixed dot size and capable of printing only two gradations of black and white, the original color information of 256 gradations of RGB must be binarized into two gradations of black and white.

[0004] Such gradation conversion for multi-level or binarization is called screening, and as a method for this, a dither method, a density pattern method, and a pixel distribution method are known. The dither method is widely used. The dither method is classified into a conditional decision method and an independent decision method. Among the conditional decision methods, the error diffusion method is one of the most widely used.

[0005] However, the conditional decision method has sharp edges and good gradation, but has a problem that it takes a long time to calculate and a unique texture appears. In addition, the independent determination method has a problem that the calculation time is short and the texture is difficult to appear, but the unique screen is conspicuous and the gradation property is not good.

As described above, in the screening in which the input gradation is multi-valued or binarized, a texture generated particularly on the highlight side causes a deterioration in image quality. Accordingly, a method has been proposed in which the pixel arrangement on the highlight side is a pattern in which the texture is not noticeable in advance.

[0007] This method is called a blue noise mask method, and is first proposed by Ulichney in "Digital Halftoning". The point of this blue noise mask method lies in the method of creating an initial blue noise mask. Ulichney uses the poid cluster method (Void
-Cluster method).

[0008]

However, there is a problem that it takes a very long time to generate a blue noise mask by such a conventional void cluster method. Several other methods of generating a blue noise mask have been proposed, but all have the problem that it takes a very long time to create the mask.

SUMMARY OF THE INVENTION It is an object of the present invention to provide a dither creation apparatus, a method and a printer capable of efficiently creating an optimum blue noise mask which does not cause texture in a highlight portion in a short time.

[0010]

FIG. 1 is a diagram illustrating the principle of the present invention. The present invention uses (128 × 128) pixels and (25
A dither matrix for inputting a multi-level gray scale signal of each pixel constituting a large vertical and horizontal size image such as a 6 × 256) pixel and converting it into a binary gray scale signal or another multi-level gray scale signal is a blue noise mask. It is intended for a dither generation device that generates a dither on the basis of.

With respect to such a dither creating apparatus, the present invention provides an initial dither matrix setting section 10 for setting an initial dither matrix 16 corresponding to a predetermined density on the highlight side.
And an array generation unit 12 that searches for a gap position furthest from the currently arranged pixels based on the initial dither matrix 16 and generates an array 26 in which all pixels are arranged in order as a blue noise mask. It is characterized by.

As described above, according to the present invention, since the initial dither matrix is generated and the blue noise mask is generated based on the initial dither matrix, the blue noise mask can be generated in a shorter time than in the void cluster method or the like. it can. Then, based on the blue noise max, a dither matrix for binarizing or multi-leveling an input image without a texture in a highlight portion can be created at high speed.

The initial dither matrix setting section 10 sets an initial dither matrix 16 in which pixels are arranged in a substantially equilateral triangle corresponding to a predetermined density on the highlight side. The initial dither matrix 16 defines a pixel arrangement serving as a basis for calculating the pixel arrangement of the blue noise mask (calculating the potential matrix).

When the inventors of the present invention examined a pattern that looks uniform to the human eye, it was found that if the pixels were arranged in a roughly triangular shape, it would look uniform. On the other hand, when the pixels are arranged in a shape close to a square, the directionality in the vertical direction and the horizontal direction is noticeable, and the pixels do not look uniform. Therefore, in the present invention, the initial dither matrix 16 is arranged in a pattern such that the pixels are arranged substantially in an equilateral triangle.

The initial dither matrix setting unit 10
When arranging the pixels in an equilateral triangle corresponding to the predetermined density on the highlight side, the arrangement position of the pixels is moved using a random number within a range of about one pixel. The movement of the pixels constituting the equilateral triangle by random numbers causes the shape to be broken, thereby preventing a plurality of positions where the potential becomes maximum from being generated in the equilateral triangle pixel arrangement in the potential calculation for searching for the void where the pixels are arranged. In.

The array generation unit 12 selects all the pixels of the initial dither matrix 16 in order from the farthest pixel based on a predetermined position and stores the pixels in the array 26. Subsequently, a potential matrix according to the potential curve is created based on all the pixels of the initial dither matrix 16, and the position of the gap where the potential farthest from the currently arranged pixel is maximized is located to arrange the pixels. The process of updating the potential matrix is repeated.

The array generator 12 uses a Gaussian distribution (Gaussian distribution) as a potential curve. The Gaussian distribution is a function that becomes 0.0 at the point of interest where pixels are arranged and becomes 1.0 at infinity, and may be another similar function.

Each time a new pixel is arranged, the array generation unit 12 updates the potential matrix by multiplying the previous potential matrix by the potential matrix of the arranged pixel.

The array generation unit 12 gives a priority to the selection of the gap in which the pixels are arranged. For example, the array generation unit 12 sets a priority according to a checkerboard matrix in selecting a gap in which pixels are arranged. For this reason, the arrangement of the pixels by the priority matrix is such that, for example, after all the pixels are arranged at the position of the first priority designated by the checkered pattern, the pixels are arranged at the remaining positions having the second priority designated by the checkered pattern. Deploy.

By giving priority to the positions where pixels are arranged in accordance with the potential matrix as described above, the graininess at the highlight portion is reduced, and the roughness at the transition to the medium density or high density is reduced. .

When updating the potential matrix, the array generating section 12 reconstructs the entire potential matrix according to the state of the potential. For example, array generator 1
2, when the maximum value of the potential matrix becomes smaller than the specified value, the entire potential matrix is reconstructed so as to exceed the specified value.

In this case, if the multiplication is repeated for updating the potential matrix, the potential level decreases as a whole, and the calculation accuracy falls below. Therefore, when the maximum value of the potential becomes equal to or less than the specified value, the whole is stretched so that the maximum value of the potential becomes 1, thereby improving the calculation accuracy.

The array generator 12 uses a template for updating the potential matrix. The present invention is based on calculating the potential matrix each time for each pixel arrangement. However, a template in which a potential curve is calculated in advance for each number of arranged pixels is created, and the speed is increased by using this template. it can.

The dither generator of the present invention further includes a dither generator 14 for generating a dither matrix corresponding to the input gradation value based on the generated blue noise mask array.
Is provided.

When the output device has two gray levels, the dither generator 14 calculates the number of pixels in the array from the ratio of the specified input gray level to the full gray level, and starts the pixel allocation from the top of the array. A binary dither matrix in which up to a number of pixels are black pixels and the remaining pixels are white pixels is generated.

When the output device has a multi-valued gradation such as 5 gradations, the dither generating section 14 generates a predetermined multi-valued level based on the ratio of the designated input gradation value to the full gradation value. Calculate the output gradation value of the image, find the upper and lower high-level gradation values, and calculate the output gradation value according to the ratio between the two output pixel values obtained by subtracting the lower-level gradation value from the output gradation value. To obtain a multi-valued dither matrix in which pixels from the top of the array to the number of pixels arranged are pixels of a high-level output gradation value, and the rest are pixels of the low-level output floor value. I do.

For example, based on the ratio (108/256) of the designated input tone value 108 to the full tone value 256 (108/256), an output tone value 2.12 having five levels is calculated, and the upper and lower high-level Tone value 3 and low level tone value 3 are obtained, and pixel arrangement in the array according to the ratio 0.12 between two output pixel values obtained by subtracting lower level tone value 2 from output tone value 2.12. Then, a multi-valued dither matrix is generated in which the pixels from the top of the array to the number of arranged pixels are pixels having three-level output gradation values, and the remaining pixels are pixels having two-level output floor values.

Further, according to the present invention, a dither matrix for inputting a multi-level gradation signal of each pixel constituting an image and converting it into a binarization gradation signal or another multi-level gradation signal is based on a blue noise mask. An initial dither matrix setting process for setting an initial dither matrix 16 corresponding to a predetermined density on the highlight side, and a currently arranged pixel based on the initial dither matrix 16. A pixel arrangement generating step of generating, as a blue noise mask, an array 26 in which all the pixels are sequentially arranged by searching for a gap position furthest from the pixel.

The details of the dither generation method are basically the same as those of the device configuration.

Further, the present invention provides a printer, and the printer according to the present invention uses the initial dither matrix corresponding to the predetermined density on the highlight side to determine the position of the gap farthest from the currently arranged pixel. And a storage unit storing a dither matrix created by using a blue noise mask in which all pixels are sequentially arranged, and converting the multi-valued gradation signal of the pixel into a binary value or another multi-valued gradation signal by the dither matrix. And a conversion unit for converting. The details of the printer are basically the same as those of the device configuration.

[0031]

FIG. 2 is a block diagram showing a functional configuration of a dither generating apparatus according to the present invention.

In FIG. 2, the dither generating apparatus of the present invention comprises an initial dither matrix setting unit 10, an array generating unit 12 for generating an array as a blue noise mask, and a dither generating unit 14 for generating a dither matrix. .

The initial dither matrix setting section 10 sets an initial dither matrix 16 corresponding to a predetermined density on the highlight side of the target image. FIG. 3 is an example of the initial dither matrix 16 set by the initial dither matrix setting unit 10 in FIG. This initial dither matrix 1
6 is composed of a matrix of (vertical × horizontal) = (256 × 256) pixels, for example.

If the density on the highlight side is, for example, 256 gradations at full gradation, an initial dither matrix 16 corresponding to the density corresponding to the highlight gradation value (1/256) is created. I have. The pixel arrangement of the initial dither matrix 16 is based on repeating the pixel pattern in which three pixels 16-1, 16-2, and 16-3 are arranged in an equilateral triangle as shown in FIG.

The initial dither matrix 16 defines the pixel arrangement on the highlight side, which is the basis of the calculation of the potential matrix for the pixel arrangement in the array generation unit 12 in FIG. According to the study by the inventors of the present application, when a pattern that looks uniform to human eyes is examined, it is found that the pattern looks uniform when pixels are arranged in a substantially triangular shape.

On the other hand, when the pixels are arranged in a shape close to a square, the directionality in the vertical and horizontal directions is conspicuous, so that the appearance is not uniform as compared with the case where the pixels are arranged in an equilateral triangle. Based on such considerations, in the present invention, a pattern in which three pixels 16-1, 16-2, and 16-3 are arranged in an equilateral triangle in FIG. The initial dither matrix 16 is set by arranging it corresponding to the predetermined density on the light side.

Here, the pixels 16-1 to 16-1 of an equilateral triangle in FIG.
The length a of one side of the equilateral triangle in the arrangement of 6-3 is calculated as follows. First, the initial dither matrix 16
The target reflectance corresponding to the predetermined density on the highlight side, which is the target image for which is created, is determined.

In the embodiment of the present invention, for example, (1/25)
6) is set. This gradation value (1/256) is 0.9961 in terms of reflectance. Further, between the reflectance and the density, since the reflectance is given by 10 to the power of minus density, the reflectance of the target image is given by 1.7 × 10 −3.

Thus, the target reflectance is 0.99.
If 61 is designated, the pixels 16-1 to 16- in FIG.
The area S of the equilateral triangle formed by 3 is calculated by the following equation.

[0040]

(Equation 1)

As is apparent from the equation (1), when the target image becomes bright and the reflectance increases (approaching 1), the area S increases. Conversely, when the target image becomes dark and the reflectance decreases, the area S increases. The area S has a relationship of becoming smaller.

If the area S of the equilateral triangle in which the pixels 16-1 to 16-3 are arranged can be calculated in this way, the length a of one side may be calculated by the following equation.

A = √ (2S / sin60 °) (2) A case where the basic pattern of an equilateral triangle in which the three pixels 16-1 to 16-3 of FIG. 4 are arranged as described above is repeatedly arranged as shown in FIG. The position at which the pixel is arranged is moved by about one pixel based on the pixel 36 at the upper left corner by random numbers. By moving the arrangement position of about one pixel in accordance with such a random number, the arrangement pattern of the triangle on the initial dither matrix 16 in FIG. 3 is different from the equilateral triangle shown in FIG. Are repeatedly arranged as a substantially equilateral triangle pattern shifted in a direction according to

As described above, the reason why a pixel is shifted by about one random number when a pattern of an equilateral triangle is actually arranged is that, in the calculation of the potential matrix performed by the array generator 12 in FIG. In order to avoid this, a plurality of positions are arranged at positions shifted by about one pixel from the equilateral triangle, so that the position of the maximum value of the potential always occurs at one point on the potential matrix. It has a triangular arrangement.

FIG. 5 shows another embodiment of the initial dither matrix 16 used in the present invention. In this embodiment, the three pixels 16-1 to 16-3 of FIG. 4 are arranged in an equilateral triangle. These basic patterns are repeatedly arranged in the lower right diagonal direction. Also in this case, similar to the embodiment of FIG. 3, when arranging the pixels of the equilateral triangle, the pixels are moved by about one pixel based on the pixel 36 at the upper left corner to the position according to the random number.

Referring back to FIG. 2, the array generation unit 12
Generates an array 26 of (256 × 256) pixels corresponding to the matrix size, that is, a blue noise mask, based on the initial dither matrix 16 set by the initial dither matrix setting unit 10. The generation 26 of the array as the blue noise mask in the array generation unit 12 is basically performed based on the initial dither matrix 16 by arranging the pixel at the position farthest from the currently laid pixel.

In the calculation for this, in the present invention, a potential curve having a Gaussian distribution (Gaussian distribution) is used, a position where the potential is greatest is searched for, and pixels are arranged there.

For this reason, the array generation unit 12 includes a potential update unit 18, a potential matrix storage unit 20, a maximum value detection unit 22, and an array 2 as a blue noise mask.
6 is stored in a dither array storage unit 24 for storing the number 6 in the memory.

The potential updating unit 18 first stores the pixels in the array 26 for the pixels of the initial dither matrix 16. In the initial dither matrix 16 shown in FIG. 3, since the gradation value is 1/256, the number of pixels arranged in a substantially regular triangle in the initial dither matrix 16 is, for example, 256. .

Therefore, in the potential updating unit 18, all the pixels of the initial dither matrix 16 are selected from the farthest pixel with reference to the pixel 36 at the upper left corner of FIG. Arrange the selected pixels.

The pixel stored in the array 26 is information indicating the coordinate position of the pixel. The horizontal pixel address in FIG. 3 is i = 0 to 255, and the vertical pixel address is j = 0.
If it is assumed to be 255, the array 26 stores the position (i, j) of the selected pixel.

Therefore, when binarizing the input image, if the input image has a uniform gradation value (1/256), the dither matrix generated from the array 256 as a blue noise mask is used. By the conversion using the used tissue dither, a repetition pattern of substantially the same triangle as the initial dither matrix 16 of FIG. 3 is obtained as a conversion pattern.

After the arrangement of the pixels in the array 26 based on the initial dither matrix 16 is completed, the potential updating section 18 has a Gaussian distribution for each pixel arrangement in order from the top for all the pixels arranged in the initial dither matrix 16. A potential matrix is generated by calculation using a potential curve.

The calculation of the potential curve is shown in FIG.
First, the matrix space 38 (two-dimensional) of (256 × 256) pixels is converted into a cylindrical matrix space 38 as shown in FIG. ) Is handled as the shape of the torus-like matrix space 38.

The reason why the potential curve is calculated as the torus-shaped matrix space 38 of FIG. 6C in which the upper and lower sides and the left and right sides of the planar matrix space are closed is as follows.
Since the dither matrix created by the present invention is repeated at the cycle of the initial dither matrix 16 shown in FIG. 3, the potential curve is calculated in a space closed to this repeating unit.

Therefore, in the potential updating unit 18 shown in FIG. 2, the torus-like matrix space 3 shown in FIG.
8 to the dither array storage 2 from the initial dither matrix 16
The potential matrix is calculated while sequentially taking out the pixels stored at the head position of the array 26 of No. 4 and arranging them one by one.

FIG. 7 shows a three-dimensional Gaussian distribution when one pixel is arranged in the matrix space. The three-dimensional space in FIG. 7 is defined as a potential matrix 40. When a pixel is arranged at a pixel arrangement point 42-1 and a three-dimensional shape of a Gaussian potential curve is obtained, a potential is obtained at the pixel arrangement point 42-1. The potential becomes the minimum value, and the potential becomes the shape of a Gaussian distribution of the potential that reaches the initial position as going outward.

This is because the potential curve using the Gaussian distribution is a function that is 0.0 at the point of interest at the pixel arrangement point 42-1 and 1.0 at the point at infinity.

FIG. 8 shows a pixel arrangement point 42-2 following FIG.
3 shows a three-dimensional shape of the potential matrix 40 when a second pixel is arranged. The pixel arrangement point 42- in FIG.
The calculation of the potential matrix 40 when the pixels are arranged at each of the pixels 1 and 42-2 is performed by calculating the potential matrix 40 of FIG. 7 when the pixels are arranged at the pixel arrangement point 42-1.
Is multiplied by a potential matrix obtained when a pixel is placed at the pixel arrangement point 42-2 in FIG.

By updating the potential matrix for each pixel arrangement in such a matrix space, a potential matrix in a state where all the pixels of the initial dither matrix of FIG. 3 are arranged is generated in the potential matrix storage unit 20.

Subsequently, the potential updating unit 18 uses the potential matrix to arrange the remaining pixels not arranged in the initial dither matrix 16 with respect to the array 26.

First, the current potential matrix storage unit 2
For the potential matrix stored in 0,
The maximum value detection unit 22 detects the position of the gap where the potential becomes the maximum value. The position where the potential becomes maximum on this potential matrix gives the position of the gap furthest from the currently arranged pixel.

Therefore, the arrangement address of the pixel at the position of the maximum value of the potential detected by the maximum value detecting section 22 is
It is stored in the array 26 of the dither array storage 24. When a new pixel is arranged in this way, the potential updating section 18 updates the potential matrix storage section 20 by multiplying the current potential matrix by a potential matrix according to the potential curve of the newly arranged one pixel. I do.

The arrangement of the pixels and the updating of the potential matrix are repeated until the last pixel of (256 × 256) pixels. The array 26 of the dither array storage unit 24 obtained in this manner is generated as a blue noise mask that is used as a basis for creating a dither matrix.

Next, the dither generator 14 of FIG. 2 generates a dither matrix based on the array 26 storing the pixel addresses of (256 × 256) pixels created by the array generator 12. The creation of the dither matrix in the dither creation unit 14 can be divided into creation of a binarized dither matrix and creation of a multi-valued dither matrix.

The binarized dither matrix includes a binarized dither matrix generator 28 and a binarized dither matrix storage 3
0 is generated. The multi-valued dither matrix is generated by the multi-valued dither matrix generation unit 32 and the multi-valued dither matrix storage unit 34.

The binarized dither matrix is, for example, RGB
A dither matrix for converting the input image of 256 gradations into a black and white binary signal which is a gradation expression of an output device such as a printer is created.

The method of generating a binarized dither matrix by the binarized dither matrix generating section 28 is as follows. When the gradation value A of the input image is, for example, A = 0 to 255, the input gradation values 0 to 2
A dither matrix is generated for each 55 and stored. Details of the generation of the binarized dither matrix and the multi-valued dither matrix will be described later.

The binarized dither matrix generated by the dither generator 14 is stored in a printer or the like capable of expressing black and white two gradations.
The input image of gradation is converted into a binary image of black and white two gradations and printed.

The multi-valued dither matrix generated by the dither generator 14 is stored in a color printer capable of expressing multiple gradations, and for example, each of the 256-level input image of YCM can be expressed by a color printer. Each of the YMCs is converted into an image of 5 gradations (gradation level 5) by the method of tissue dither, and a color image is printed out.

FIG. 9 is a flowchart of the dither creation processing by the apparatus of FIG. In FIG. 9, first, the initial dither matrix 16 set by the initial dither matrix setting unit 10 in step S1 is input. Since this initial dither matrix 16 repeats the basic regular triangular pattern shown in FIG. 4, it may be generated automatically or manually by an operator.

Next, in step S2, an array 26 is created as an empty arrangement. Next, in step S3, the position addresses of the pixels are arranged in the array prepared in step S2 from the top in the order of pixels farthest from the pixel 36 at the upper left corner of the initial dither matrix 16 in order.

Next, in step S4, a potential matrix is created by calculating a potential curve while arranging pixels of the initial dither matrix one by one in a torus-like matrix space as shown in FIG. 6C. Next, in step S5, a position where the potential becomes maximum is searched for in the potential matrix obtained by arranging all the pixels of the initial dither matrix. In step S6, the searched pixel position where the potential becomes maximum is added to the array. .

Subsequently, in step S7, it is checked whether or not all the pixels have been arranged. If all the pixels have not been arranged, the process returns to step S8, and the potential matrix of the newly arranged pixels is generated by the previous pixel arrangement. The updated potential matrix is multiplied and updated, and the process returns to step S5 again to repeat the process of searching the updated potential matrix for the location of the maximum potential.

When the arrangement based on the potential matrix of all the pixels is completed in step S7, in step S9, the arrangement of the generated pixels (blue noise max)
A dither matrix corresponding to the input tone value is created based on.

FIG. 10 shows the dither matrix creation processing in step S9 in FIG.
9 is a flowchart of binarized dither generation processing by a binarized dither generation unit 28 provided in FIG.

This binarized dither creation processing is performed in step S
1 designates an initial value A = 0 of the input gradation A = 0 to 255. Next, in step S2, the number of pixel arrays NB in the array 26 as the blue noise mask corresponding to the input gradation A is calculated. That is, the pixel array number NB is obtained by NB = total array number × (A / 255).

Subsequently, in step S3, bit 1 representing a black pixel is placed at the B-th pixel address calculated from the head of the array as a blue noise mask, and bit 0 representing a white pixel is stored in the remaining pixel positions. Create a dither matrix.

Subsequently, in step S4, it is checked whether or not the input tone A has reached the maximum value 256. If not, the input tone A is incremented by one in step S5, and the process returns to step S2. A binarized dither matrix for the input gradation A is created, and this processing is repeated until the input gradation value A = 256.

FIG. 11 is a flowchart of the multi-valued dither creation process in the dither matrix creation in step S9 in FIG. 9, and is a process performed by the multi-valued dither matrix creation unit 32 provided in the dither creation unit 14 in FIG. It is.

In FIG. 11, in the multi-valued dither generation processing, first, in step S1, the input gradation is set to 255 from 255 to 255.
If the gradation is to be set, an initial value A = 1 is specified. Next, in step S2, an output gradation B corresponding to the input gradation A is calculated.
The output gradation B is calculated by B = the number of output gradations × (A / 255). That is, the number of output gradations is multiplied by the ratio of the input gradation value to the full gradation. This step S
The output gradation B calculated by 2 includes a numerical value smaller than a decimal number that cannot be expressed by a single output pixel.

Next, at step S3, the calculated output gradation B
The ratio K between two high and low gradation values BL and BH located above and below is calculated. Two high and low gradation values BL and BH located above and below the output gradation B including the calculated decimal number.
Are integer values that can be represented by a single output pixel, and 2
The ratio K of the calculated output gradation B between the two integer values is calculated as K = B-BL. As a result, the ratio K becomes a decimal value less than 1.

Next, in step S4, based on the ratio K, the arrangement number N from the top of the pixels of the high-level gradation BH in the array
Calculate H. That is, the pixel arrangement number NH of the high-level gradation BH in the array is calculated as NH = total pixel number NT × K.

Subsequently, in step S5, the multi-level dithering storing the high-level gradation BH at the pixel addresses from the first to the calculated NH-th in the array and storing the low-level gradation BL at the remaining pixel positions. Create a matrix and store it in memory. The processing of steps S2 to S5 is repeated until the input gradation A reaches 256 in step S6.
Is repeated while incrementing the input tone A.

FIG. 12 illustrates a specific example of the multi-valued dither creation processing of FIG. FIG. 12A shows gradation values 0 to 255.
Is the scale of the input tone 255 of step S1.
It is assumed that A = 108 is designated as the input gradation A. On the other hand, it is assumed that the output gray scale is a gray scale 5 of gray scale values 0 to 5.

In step S2 in FIG. 11, the output gradation B corresponding to the input gradation A = 108 is calculated. In this case, the output gradation B = 2.12 is calculated. In the case of this output gradation B = 2.12, for example, assuming a dither matrix of (5 pixels × 5 pixels) = 25 pixels as shown in FIG. 12B, the output gradation B = 2.12 The output gray scale B = 2.12 is realized by the ratio of the low level gray scale BL = 2 and the high level gray scale BH = 3 with respect to.

In this case, the number of level 3 pixels may be determined so that the gradation value is increased by 0.12 with respect to the level 2 pixels. Therefore, the dither arrangement number NH of the pixel of level 3 which becomes the high level gradation BH is calculated as NH = the total number of pixels NT × K = 25 × 0.12 = 3.

The arrangement of the three pixels of level 3 as the high-level gradation BH is to be arranged at the first to third pixel positions of the array based on the initial dither matrix according to the arrangement approximated by an equilateral triangle. From, for example, FIG.
As shown in (B), a multi-valued dither matrix in which pixels of level 3 are arranged at positions substantially in an equilateral triangle is created.

Of course, the multi-valued dither matrix actually generated by the present invention is a dither matrix in a matrix space of (256 × 256) pixels.

The multi-valued dither creation processing shown in FIG.
For example, when the output device is a printer, a multi-valued dither matrix corresponding to each of the 255 gradations of each color component of YMC is created.

The binarized dither matrix created in FIG. 10 and the multi-valued dither matrix created in FIG.
The printer is set in a printer such as a jet ink storage device, which will be described later, and converts an input image into a binarized image or a multi-valued image by a method of tissue dither and prints the image.

In the tissue dithering method, (256.times.256) pixels of the same size as the dither matrix are cut out from the input image, the corresponding dither matrix created by the present invention is read out according to the gradation value of each cut out pixel, and the input dither matrix is read out. 2 stored at the position of the dither matrix corresponding to the key value
Outputs a quantified value or a multi-valued value as a converted value.

FIG. 13 is a flowchart of another creation process in the dither creation device of FIG. The embodiment of FIG. 13 is characterized in that, when arranging pixels by searching for the location of the maximum potential, the location where the pixel is searched is given priority.

In FIG. 13, the creation of a potential matrix based on the initial dither matrix in steps S1 to S4 is the same as in the flowchart of FIG. In the next step S5, the position of the maximum potential is searched in order from the position with the highest priority, and pixels are arranged. That is, in step S5, since a pixel arrangement position with the highest priority is assigned, a place where the potential is highest among the highest priorities is searched, and in step S6,
Add the searched location to the array.

Subsequently, in step S7, it is checked whether or not all the pixel arrangement positions of the current priority have been added to the array, and if all have been added, the lower priority is updated in step S8, and the update is performed in step S5. The lower priority set as the current priority is used to search for the location of the maximum potential in the pixel portion of the priority and to add it to the array. Steps S9 to S8 other than the arrangement according to the priorities of steps S6 to S8
S11 is the same as the process in FIG.

FIG. 14 shows a part of the priority matrix used for the pixel arrangement having the priority shown in FIG. The priority matrix 44 arranges priority images represented by white pixels and low-priority pixels represented by black pixels with low priority in a checkered pattern, and sets two levels of priority.

For this reason, in the processing of steps S5 to S8 in FIG. 13, first, the position of the maximum potential for the pixel position of the priority pixel indicated by the white pixel is searched and added to the array. Is completed, the position of the maximum potential is searched for the position of the lower pixel indicated by the black pixel and added to the array.

FIG. 15 shows another embodiment of the priority matrix used for the pixel arrangement having the priorities shown in FIG.

Priority matrix 46 in FIG.
FIG. 1 shows three levels of priority of a first priority pixel represented by a white pixel, a second priority pixel represented by a gray pixel, and a low-priority pixel represented by a black pixel having the lowest priority.
5 is set in the same checkered pattern. In this case, the search for the location of the maximum potential is performed by the first priority pixel and the second priority pixel.
This is performed sequentially in three stages of priority pixels and lower-order pixels.

By giving priority to the positions of the pixels to be arranged based on the potential matrix in the arrangement as described above, it is possible to reduce the granularity in the highlight portion and the roughness when shifting to the central portion or the high density portion. Can be reduced.

Particularly, in a printer, when the density is higher than the medium density, the pixels overlap, so that the advantage of the blue noise mask method is diminished. Therefore, according to the embodiment having the priority shown in FIG. 13, the dither pattern similar to the tissue dither can be obtained for the region having the medium density or higher, and the image quality at the time of transition from the highlight portion to the medium density and further to the high density is obtained. To eliminate the deterioration of

FIG. 16 is a flowchart of another embodiment of the dither creation processing in the apparatus shown in FIG. This figure 1
The sixth embodiment is characterized in that a template is used for updating the potential matrix.

That is, in the dither creation processing of FIGS. 9 and 13, the potential matrix is calculated and updated each time according to the potential curve each time a pixel is arranged. In the embodiment of FIG. In step S2, a template of the potential matrix is created in advance.

This potential template is a potential matrix according to a potential curve calculated in advance for each pixel arrangement number. By using this potential template for updating the potential matrix in step S9, the potential curve The calculation becomes unnecessary, and the processing can be speeded up accordingly. The other points are the same as those in the flowchart of FIG.

FIG. 17 is a flowchart of another embodiment of the dither creation processing by the apparatus shown in FIG. 2, which shows that a decrease in calculation accuracy due to repetition of potential curve multiplication when updating the potential matrix is suppressed. Features.

When the multiplication for updating the potential matrix for each pixel arrangement is repeated according to the potential curve, the potential curve takes a value of 1 or less, so that the maximum potential value sequentially decreases, and the calculation accuracy falls below the value. Therefore, in the embodiment of FIG. 17, when the potential maximum value becomes equal to or less than a specified value corresponding to the calculation accuracy, a process of extending the entire potential matrix so that the potential maximum value is set to 1 is added.

That is, in the flowchart of FIG. 17, steps S1 to S9 are the same as those in the flowchart of FIG. 16, but in the next step S10, it is checked whether or not the maximum potential value is below a predetermined value. If so, the potential matrix is reconfigured in step S11 so that the maximum potential value becomes 1, and the process returns to step S6 to search for the location of the maximum potential value.

FIG. 18 is an explanatory view of a printer in which schooling dithering is performed by a systematic dither using a binarized dither matrix or a multi-valued dither matrix created by the dither creating apparatus of the present invention. For example.

In FIG. 18, the printer 48 is provided with a paper insertion guide 50 and a paper discharge guide 64. The paper insertion guide 50 is a guide for inserting unprinted paper into the printer 48. The paper discharge guide 64 stores the discharged paper.

FIG. 19 shows the internal structure of the printer 48 shown in FIG. The sheet stored in the sheet insertion guide 50 is picked up by a pickup roller 52 and guided by a sheet guide 54. The sheet sent by the sheet guide 54 is fed to the head 7 by a sheet pressing roller 56.
It is held down before 0. The sheet printed by the head 70 is sent backward by the feed roller 60, and at this time, the sheet pressing roller 62 sandwiches the sheet between the sheet and the feed roller 60.

The ink jet head 70 has a nozzle face 7
2 facing downward. In addition, the inkjet head 70 moves along a shaft 74 extending in the depth direction in the drawing. The cleaning mechanism 58 cleans the nozzle surface 72 of the inkjet head 70. The cleaning mechanism 58 is provided outside the inkjet head 70 and below the nozzle surface 72 of the inkjet head 70.

FIG. 20 is a block diagram of the functional configuration of the printer 48 of FIG. In FIG. 20, the host 76 transfers the print data to the printer 48 for printing.
A CPU 78 provided in the printer 48 controls an ink jet recording operation. The interface 80 receives a print command and print data from the host 76.

The pulse control section 82 generates a drive pulse to be supplied to the recording head 27. The head drive circuit 84
Upon receiving a drive pulse from the pulse control unit 82, the ink ejection drive unit in the recording head 86 is driven to control the medium transport unit 90 and the head transport unit 92.

The medium transport section 90 transports the printing paper to a predetermined position for printing by the recording head 86. The head transport unit 92 is provided with the recording head 8 along the shaft 22 in FIG.
By the movement of 6, the recording head 86 is transported to a predetermined value for printing.

In the ink jet recording operation performed by the printer 48, when a print command and print data are received from the host 76 by the interface unit 80, the print data is stored in the storage unit 94, and the printing is performed in accordance with the firmware in the storage unit 94. To start.

At this time, a dither matrix created based on the blue noise mask of the present invention is stored for each gradation value of the input image in the storage unit 94, and the input image is subjected to ink jet recording in accordance with the method of tissue dither. The image is converted into a binarized image or a multi-valued image that can be represented by an operation. If necessary, the blue noise mask itself may be stored in the storage unit 94.

The CPU 78 controls the mechanism control unit 88 in accordance with a prescribed format of the print data to move the positions of the recording paper and the recording head 86. When the recording paper and the recording head 86 reach predetermined positions, a pulse The controller 82 is controlled to supply a drive pulse to the head drive circuit 84.

The head drive circuit 84 supplies a drive pulse to the recording head 86 to eject ink, sprays the ejected ink droplets on recording paper, and prints characters and pictures.

Note that the present invention is not limited to the above embodiment,
Includes appropriate modifications that do not impair its purpose and advantages. Further, the present invention is not limited by the numerical values shown in the above embodiments.

[0120]

As described above, according to the present invention, an initial dither matrix for providing a dither matrix on the highlight side is created, and based on this initial dither matrix, the initial dither matrix is located at a position farthest from the currently placed pixel. Since an array as a blue noise mask is created so that pixels are arranged, it is possible to create a blue noise mask in a shorter time than with the void cluster method used for creating a conventional blue noise mask, and By creating a binarized or multi-valued dither matrix using a blue noise mask,
An appropriate dither matrix that does not generate texture on the highlight side can be created at high speed.

The following is a list of further features of the present invention.

A. 3. The dither creating apparatus according to claim 2, wherein the initial dither matrix setting unit uses a random number to arrange the pixels when arranging the pixels in an equilateral triangle corresponding to the predetermined density on the highlight side. Are moved and arranged within a range of about one pixel.

B. In the dither creation apparatus of the item a, the array generator reconstructs the entire potential matrix so as to exceed the specified value when the maximum value of the potential matrix becomes smaller than the specified value.

C. In the dither generating apparatus according to claim 3,
The array generation unit uses a template for updating the potential matrix.

D. The dither generation device according to claim 1, further comprising a dither generation unit that generates a dither matrix corresponding to the input gradation value based on the array.

E. In the dither creation device of d, the dither generation unit obtains the number of pixels arranged in the array from the ratio of the designated input tone value to the full tone value, and calculates the number of pixels arranged from the top of the array to the number of pixels arranged. Is characterized in that a binarized dither matrix is generated in which pixels are black pixels and the remaining pixels are white pixels.

F. In the dither generating apparatus of d, the dither generating section calculates an output tone value having a predetermined multi-value level based on a ratio of the designated input tone value to the full tone value, and calculates the upper and lower heights. The level gradation value and the low level gradation value are obtained, and the lower level gradation value is subtracted from the output gradation value.
The number of pixels arranged in the array is determined according to the ratio between the two output pixel values, and pixels from the beginning of the array to the number of arranged pixels are defined as high-level output gradation value pixels, and the rest as low-level output floor value pixels. The multi-valued dither matrix is generated.

[Brief description of the drawings]

FIG. 1 is a diagram illustrating the principle of the present invention.

FIG. 2 is a block diagram of a functional configuration of the present invention.

FIG. 3 is an explanatory diagram of an initial dither matrix set in the present invention.

4 is an explanatory diagram of the length of a side of an equilateral triangle in which pixels are arranged in FIG.

FIG. 5 is an explanatory diagram of another initial dither matrix set in the present invention.

FIG. 6 is an explanatory diagram of a matrix space in a torus state to which a potential curve is applied in the present invention.

FIG. 7 is an explanatory diagram of a potential matrix when one pixel is arranged in the present invention.

FIG. 8 is an explanatory diagram of a potential matrix when two pixels are arranged in the present invention.

FIG. 9 is a flowchart of a dither creation process of the present invention according to the embodiment of FIG. 2;

FIG. 10 is a flowchart of a binarized dither matrix creation process in FIG. 9;

FIG. 11 is a flowchart of a multi-valued dither matrix creation process in FIG. 9;

FIG. 12 is an explanatory diagram of a specific example of creating a multi-valued dither matrix in FIG. 11;

FIG. 13 is a flowchart of a dither creation process according to the present invention in which pixel arrangements have priorities;

FIG. 14 is an explanatory diagram of a checkered priority matrix having two levels of priority used in the processing of FIG. 13;

FIG. 15 is an explanatory diagram of a checkered priority matrix having three levels of priority used in the processing of FIG. 13;

FIG. 16 is a flowchart of a dither generation process of the present invention using a potential template.

FIG. 17 is a flowchart of dither generation processing of the present invention for expanding a potential matrix.

FIG. 18 is an external view of an ink jet recording apparatus using a dither matrix created by the present invention.

FIG. 19 is an explanatory diagram of the internal structure of FIG. 18;

FIG. 20 is a functional block diagram of FIG. 18;

[Explanation of symbols]

 10: Initial dither matrix setting unit 12: Array generation unit 14: Dither generation unit 16: Initial dither matrix (initial pattern) 18: Potential update unit 20: Potential matrix storage unit 22: Maximum value detection unit 24: Dither array storage unit 26 : Array (blue noise mask) 28: Binarized dither matrix generator 30: Binarized dither matrix storage 32: Multi-level dither matrix generator 34: Multi-level dither matrix storage 36: Reference position 40: Point Char matrixes 42-1 and 42-2: Pixel arrangement positions (points of interest) 44 and 46: Priority matrix 48: Ink jet recording apparatus

────────────────────────────────────────────────── ───

[Procedure amendment]

[Submission date] May 22, 2000 (2000.5.2)
2)

[Procedure amendment 1]

[Document name to be amended] Statement

[Correction target item name] 0007

[Correction method] Change

[Correction contents]

[0007] This method is called a blue noise mask method, and is first proposed by Ulichney in "Digital Halftoning". The point of this blue noise mask method lies in the method of creating an initial blue noise mask. Ulichney is, Bo Ido-cluster method in this creation (Void
-Cluster method).

[Procedure amendment 2]

[Document name to be amended] Statement

[Correction target item name] 0020

[Correction method] Change

[Correction contents]

By giving priority to the positions where pixels are arranged in accordance with the potential matrix as described above, the feeling of graininess in the highlight portion and the feeling of roughness in the transition to the medium density or high density are reduced. .

[Procedure amendment 3]

[Document name to be amended] Statement

[Correction target item name] 0027

[Correction method] Change

[Correction contents]

For example, based on the ratio (108/256) of the designated input tone value 108 to the full tone value 256 (108/256), an output tone value 2.12 having five levels is calculated, and the upper and lower high-level Tone value 3 and low level tone value 3 are determined, and pixel arrangement in the array is determined according to the ratio 0.12 between two output pixel values obtained by subtracting lower level tone value 3 from output tone value 2.12. Then, a multi-valued dither matrix is generated in which the pixels from the top of the array to the number of arranged pixels are pixels having three-level output gradation values, and the remaining pixels are pixels having two-level output floor values.

[Procedure amendment 4]

[Document name to be amended] Statement

[Correction target item name] 0052

[Correction method] Change

[Correction contents]

Therefore, when the input image is binarized, if the input image has a uniform gradation value (1/256), the dither matrix generated from the array 26 as a blue noise mask is used. By the conversion using the used tissue dither, a repetition pattern of substantially the same triangle as the initial dither matrix 16 of FIG. 3 is obtained as a conversion pattern.

[Procedure amendment 5]

[Document name to be amended] Statement

[Correction target item name] 0057

[Correction method] Change

[Correction contents]

FIG. 7 shows a three-dimensional Gaussian distribution when one pixel is arranged in the matrix space. The three-dimensional space in FIG. 7 is defined as a potential matrix 40. When a pixel is arranged at a pixel arrangement point 42-1 and a three-dimensional shape of a Gaussian distribution potential curve is obtained, a potential is obtained at the pixel arrangement point 42-1. becomes the minimum value, the potential toward the outside becomes the shape of the Gaussian distribution of the potential to reach the initial value.

[Procedure amendment 6]

[Document name to be amended] Statement

[Correction target item name] 0070

[Correction method] Change

[Correction contents]

The multi-valued dither matrix generated by the dither generation unit 14 is stored in a color printer capable of expressing multiple gradations. For example, each input image of 256 gradations of YMC can be expressed by a color printer. Each of the YMCs is converted into an image of 5 gradations (gradation level 5) by the method of tissue dither, and a color image is printed out.

[Procedure amendment 7]

[Document name to be amended] Statement

[Correction target item name] 0075

[Correction method] Change

[Correction contents]

When the arrangement based on the potential matrix of all the pixels is completed in step S7, in step S9, the arrangement of the generated pixels (blue noise mask )
A dither matrix corresponding to the input tone value is created based on.

[Procedure amendment 8]

[Document name to be amended] Statement

[Correction target item name] 0078

[Correction method] Change

[Correction contents]

Subsequently, in step S3, bit 1 representing a black pixel is placed at the NB- th pixel address calculated from the head of the array as a blue noise mask, and bit 0 representing a white pixel is stored in the remaining pixel positions. A dither matrix is created.

[Procedure amendment 9]

[Document name to be amended] Statement

[Correction target item name]

[Correction method] Change

[Correction contents]

Next, at step S3, the calculated output gradation B
The ratio K between two high and low gradation values BL and BH located above and below is calculated. Two high and low gradation values BL and BH located above and below the output gradation B including the calculated decimal number.
Are integer values that can be represented by a single output pixel, and 2
The ratio K of the calculated output gradation B between the two integer values is calculated as K = B-BL. As a result, the ratio K is <br/> a fractional value less than 1.

[Procedure amendment 10]

[Document name to be amended] Statement

[Correction target item name] 0091

[Correction method] Change

[Correction contents]

The binarized dither matrix created in FIG. 10 and the multi-valued dither matrix created in FIG.
Inkjet Symbol you clear in the later explanation is set in the printer of憶device, so that printing by converting the input image into a binary image or multivalued image by a technique ordered dither.

[Procedure amendment 11]

[Document name to be amended] Statement

[Correction target item name] 0096

[Correction method] Change

[Correction contents]

FIG. 14 shows a pixel having the priority shown in FIG.
Extracting part of the priority matrix used for placement
You. The priority matrix 44 is a priority image indicated by white pixels.Elementary
And low-priority pixels with low priority indicated by black pixels in a checkerboard pattern
In addition, two levels of priority are set.

[Procedure amendment 12]

[Document name to be amended] Statement

[Correction target item name] 0101

[Correction method] Change

[Correction contents]

Particularly, in a printer, when the density is higher than the medium density, the pixels overlap, so that the advantage of the blue noise mask method is diminished. Therefore the embodiments which gave priority 13, the medium concentration or more regions to allow a similar dither patterns and ordered dither, medium density from the highlight portion, the image quality when even a transition to a high concentration To eliminate the deterioration of

[Procedure amendment 13]

[Document name to be amended] Statement

[Correction target item name] 0108

[Correction method] Change

[Correction contents]

[0108] Figure 18 is an explanatory view of a printer scan Cleaning by ordered dither using binary dither matrix or multilevel dither matrix created by the dither generating apparatus of the present invention is performed, the ink jet type Take a printer as an example.

[Procedure amendment 14]

[Document name to be amended] Statement

[Correction target item name]

[Correction method] Change

[Correction contents]

The pulse control section 82 generates a drive pulse to be supplied to the recording head 86 . The head drive circuit 84
Upon receiving a drive pulse from the pulse control unit 82, the ink ejection drive unit in the recording head 86 is driven to control the medium transport unit 90 and the head transport unit 92.

[Procedure amendment 15]

[Document name to be amended] Statement

[Correction target item name]

[Correction method] Change

[Correction contents]

The medium transport section 90 transports the printing paper to a predetermined position for printing by the recording head 86. The head transport section 92 is provided with the recording head 8 along the shaft 74 in FIG.
6 move to <br/> convey the recording head 86 at a predetermined position for printing.

[Procedure amendment 16]

[Document name to be amended] Statement

[Correction target item name]

[Correction method] Change

[Correction contents]

[0118] The head drive circuit 84 causes ejecting ink by supplying drive pulses to the recording head 86, spraying the ink droplets ejected on the recording paper, prints characters, etc. picture.

[Procedure amendment 17]

[Document name to be amended] Statement

[Correction target item name] Brief description of drawings

[Correction method] Change

[Correction contents]

[Brief description of the drawings]

FIG. 1 is a diagram illustrating the principle of the present invention.

FIG. 2 is a block diagram of a functional configuration of the present invention.

FIG. 3 is an explanatory diagram of an initial dither matrix set in the present invention.

4 is an explanatory diagram of the length of a side of an equilateral triangle in which pixels are arranged in FIG.

FIG. 5 is an explanatory diagram of another initial dither matrix set in the present invention.

FIG. 6 is an explanatory diagram of a matrix space in a torus state to which a potential curve is applied in the present invention.

FIG. 7 is an explanatory diagram of a potential matrix when one pixel is arranged in the present invention.

FIG. 8 is an explanatory diagram of a potential matrix when two pixels are arranged in the present invention.

FIG. 9 is a flowchart of a dither creation process of the present invention according to the embodiment of FIG. 2;

FIG. 10 is a flowchart of a binarized dither matrix creation process in FIG. 9;

FIG. 11 is a flowchart of a multi-valued dither matrix creation process in FIG. 9;

FIG. 12 is an explanatory diagram of a specific example of creating a multi-valued dither matrix in FIG. 11;

FIG. 13 is a flowchart of a dither creation process according to the present invention in which pixel arrangements have priorities;

[14] a checkered pattern priority illustration of Conclusions Rikusu having 2 priority levels used in the process of FIG. 13

[15] a checkered pattern priority illustration of Conclusions Rikusu having 3 priority levels used in the process of FIG. 13

FIG. 16 is a flowchart of a dither generation process of the present invention using a potential template.

FIG. 17 is a flowchart of dither generation processing of the present invention for expanding a potential matrix.

FIG. 18 is an external view of an ink jet recording apparatus using a dither matrix created by the present invention.

FIG. 19 is an explanatory diagram of the internal structure of FIG. 18;

FIG. 20 is a functional block diagram of FIG. 18;

Claims (10)

[Claims] 1. A dither matrix for inputting a multi-level gradation signal of each pixel constituting an image and converting it into a binary gradation signal or another multi-level gradation signal is created based on a blue noise mask. In the dither creation device, an initial dither matrix setting unit that sets an initial dither matrix corresponding to a predetermined density on the highlight side, and a gap position furthest from the currently arranged pixel based on the initial dither matrix. A dither generation device, comprising: an array generation unit that generates an array in which all pixels are searched and arranged in order as a blue noise max. 2. A dither generating apparatus according to claim 1, wherein
The dither creating apparatus according to claim 1, wherein the initial dither matrix setting unit sets an initial dither matrix in which each pixel is arranged so as to be substantially arranged in a regular triangle corresponding to the predetermined density on the highlight side. 3. A dither generating apparatus according to claim 1, wherein
The array generation unit selects all the pixels of the initial dither matrix in order from a pixel farthest from a reference of a predetermined position, stores the pixels in the array, and subsequently sets a potential based on all the pixels of the initial dither matrix. Generating a potential matrix according to the curve, searching for the position of the gap where the potential farthest from the currently arranged pixel is the largest, arranging the pixel, and repeating the process of updating the potential matrix after arranging the pixel. A dither creation apparatus characterized by the above-mentioned. 4. A dither generating apparatus according to claim 3, wherein
The dither creation device, wherein the array generation unit uses a Gaussian distribution as the potential curve. 5. A dither generating apparatus according to claim 3, wherein:
The dither creation apparatus according to claim 1, wherein the arrangement generating unit updates the potential matrix by multiplying the previous potential matrix by the potential matrix of the arrangement pixel every time a new pixel is arranged. 6. A dither generating apparatus according to claim 3, wherein
The dither creation device, wherein the array generation unit sets a priority in selecting a gap in which pixels are arranged. 7. A dither generating apparatus according to claim 3, wherein
The dither creation device according to claim 1, wherein the array generation unit sets a priority in accordance with a checkered matrix in selection of a gap in which pixels are arranged. 8. A dither generating apparatus according to claim 3, wherein:
The dither generation device, wherein the array generation unit rebuilds the entire potential matrix according to the state of the potential when updating the potential matrix. 9. A dither matrix for inputting a multi-level gray scale signal of each pixel constituting an image and converting it into a binary gray scale signal or another multi-level gray scale signal is created based on a blue noise mask. In the dither creation method, an initial dither matrix setting step of setting an initial dither matrix corresponding to a predetermined density on a highlight side, and a gap position furthest from a pixel currently arranged based on the initial dither matrix. And generating an array in which all pixels are arranged in order as a blue noise mask. 10. A blue noise mask in which all the pixels are arranged in order by searching for a gap position furthest from a currently arranged pixel based on an initial dither matrix corresponding to a predetermined density on a highlight side. A printer comprising: a storage unit that stores a created dither matrix; and a conversion unit that converts a multi-level gray scale signal of a pixel into a binary value or another multi-level gray scale signal using the dither matrix. .