JP2977583B2 - Halftone dot threshold generation method - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method of generating a halftone threshold value used when electrically creating a halftone dot or creating a contact screen in the field of image processing. is there.

[Prior Art] Conventionally, a method using halftone dots has been widely used in expressing the density of an image in the field of printing. As shown in FIG. 15, the halftone dots represent the print screen 1 in a grid 7 by virtual lines 5.
The shading is expressed by dividing the ink application area ratio in each minute area into a large number of minute areas 4 and changing the ink application area ratio in each minute area from 0 to 100%.

When printing is performed using the halftone dots 6, a contact screen is prepared in addition to the original film. contact·
The screen has the same transmittance so that the transmittance of light changes smoothly two-dimensionally periodically, that is, at the same corresponding position in each of the micro-regions arranged in a grid, and as a whole the transmitted light Is a film which has been processed in advance so as to draw a two-dimensionally periodic light and shade pattern.

When the original film and the contact screen are superposed and exposed on an unexposed film, and then developed, a negative film having a halftone dot pattern is produced, and a printing plate is produced using the negative film and is subjected to printing.

When halftones are created electrically by laser scanning using a computer instead of a contact screen, each minute area is further subdivided into laser scan unit areas, and a threshold is set for each scan unit area Then, the intensity of the transmitted light of the original film is compared with a threshold value. For example, when the threshold value is exceeded, the laser beam is transmitted (ON), and when the threshold value is not satisfied, the laser beam is not transmitted (OFF). Create

It is important that the appearance frequency of the threshold value in each minute area needs to be the same frequency for each value of 0 to 100 (%).

In many cases, the ink application area in the minute area of the halftone dot has an ink application area ratio of 0% as shown in FIG.
The threshold is distributed from the central part to the peripheral part as it increases to 100%. In this case, the threshold distribution is such that the threshold is larger in the central part of each minute area 4 and decreases in the peripheral part. I will be doing it. As a result, the unexposed film is less likely to be exposed at the center, so that light is more likely to be transmitted toward the center in a created negative film, and the ink is more easily applied to the center in a positive printing plate made using this, that is, If the ratio of the ink application area is small, the ink is applied from the center.

The threshold distribution is opposite for negative and positive, so depending on the application, the threshold may be smaller at the center by reversing the relationship between the threshold and the ON / OFF, but in any case a general network At the point, the magnitude of the threshold changes smoothly from the center to the periphery of each micro-region, and this is repeated in the direction of the lattice 7 of the micro-region. Therefore, the threshold periodically changes in a two-dimensional space. Will do.

Normally, in color printing, in order to prevent moire generated when each color of cyan, magenta, yellow, and black is overprinted, the direction of the halftone grid 7 is different from the direction of the screen for each color. Is rotated.

Even if the direction of the halftone grid 7 is rotated, the direction of laser scanning is usually the vertical or horizontal direction of the screen.
The order of appearance of the threshold values in the minute area at the same position on the screen differs between the case of a rotated halftone dot and the case of a halftone dot that does not rotate.

Conventionally, a method of obtaining the threshold value of the rotating halftone dot is such that the coordinates (i, j) correspond to the unit area of the laser scan _, and when the threshold value of the unit area before rotation is t _{i, j} , (i, j) j) is rotated to obtain rotated coordinates (i ', j') consisting of real values, which are converted to integers (I, J), and thresholds t _{i, j} are made to correspond to (I, J). . This correspondence is set in advance in a two-dimensional array on a storage device, and a method of retrieving the two-dimensional array using (I, J) to obtain a threshold value, or a method in which the rotation angle α is a rational number a / b
As long as the angle is such that (a, b; integer), a set of a ² + b ² unit areas is set in advance as one block, and an array of threshold values for one block after rotation is stored in a two-dimensional array on the storage device. In this case, the coordinates (i, j) are set using the fact that the same appearance order is repeated with this one block as a cycle as the threshold value after rotation.
For searching the two-dimensional array.

[Problems to be Solved by the Invention] However, in these methods, the balance of the appearance frequency of each threshold value is locally disrupted by rotation, and each threshold value is stored in a two-dimensional array on a storage device prepared in advance. However, when the method is applied to an image output apparatus having a relatively low resolution, the halftone dot shape is disturbed and a pattern is generated even in a monochrome image, which makes it difficult to see.

The present invention has been made in view of the above circumstances, and each threshold value can appear at even frequency even in a rotating halftone dot, and the coordinates of a unit area of a laser scan in which each threshold value should have the threshold value. It is an object of the present invention to provide a halftone dot threshold value generation method which can be finely determined according to the threshold value and makes it possible to obtain a beautiful halftone dot.

[Means for Solving the Problems] In response to this object, a halftone dot threshold generation method according to the present invention is a method for generating a two-dimensional spatially periodic threshold, and includes a method for generating a threshold value in the x direction and the y direction. Real values distributed almost uniformly
A function value generating process of outputting a function value z of a function z = f (x, y) taking a real value by inputting a two-dimensional coordinate (x, y) composed of x and y, and correcting z by inputting the z A function value correcting step of outputting a real value t obtained by using the real value t as a threshold value, and in the function correcting step, the real value t varies in a portion where the value is large with respect to a change width of z. The width is increased, and the width of the change is reduced relative to the width of the change of z in a small portion, so that the appearance frequency of the real value t is distributed almost uniformly along the distribution of (x, y). It is characterized by:

[Operation] When a halftone threshold value is generated by the halftone threshold value generation method configured as described above, first, the screen is divided into laser scan unit areas, and two-dimensional coordinates (x,
Make y) correspond. Here, the number of laser scan unit areas corresponding to the length of one side of the minute area for representing one halftone dot may be arbitrarily changed in some cases. If the coordinates (x, y) are determined, the coordinates (x, y) can be easily adjusted to the change in the number of unit areas, which is convenient.

Also, a function z = f (x, y) is selected. z is the threshold t
If this threshold value is represented by a number from 0 to 1, it is easy to cope with a change in the number of the unit areas of the laser scan. Therefore, z is correspondingly changed from 0 to 0. A real-valued function z = f that takes on values up to 1.
It is convenient if (x, y) is selected.

In the case of a non-rotating halftone dot, 0 ≦ X ≦
If a function z = f (X, Y) defined in the range of 1,0 ≦ Y ≦ 1 is selected, a function value f (x, y) for general coordinates (x, y)
Is obtained by repeating z = f (x, y) in the x direction and the y direction, so that f (x, y) = f (X, Y) where X = x− [x] and Y = y -[Y] can be conveniently defined as Where [x], [y]
Represents the largest integer not exceeding x and y, respectively.

Also, as described above, when it is desired that the threshold value be larger at the center of the minute area, X = 0.0 as f (X, Y).
5. A function that takes the maximum value when Y = 0.5 is selected.

Then, through the function value generation process, the values of z = f (x, y) are output as real values with the input of the coordinates (x, y) of the real values corresponding to each laser scan unit area.
(X, y) is uniformly distributed, but z
Is not always uniformly distributed due to the form of the function f (x, y), and (x, y) is {(x, y) | 0 ≦ X ≦ 1,0 ≦ y ≦ 1}
Does not necessarily change linearly as it changes from the central part to the peripheral part in the area of, the threshold value t cannot be used as it is.

Next, the real value z is corrected to the real value t by a function value correction process. Here, the cumulative appearance frequency of z or the cumulative density distribution of z or a function approximating to them is obtained in advance, and in a portion where the value is large, the width of the change of the threshold value t is increased with respect to the width of the change of z, and In the part, the width of the change of the threshold value t is made smaller than the width of the change of z so that t is uniformly distributed and (x, y) is ｛(x, y) | 0 ≦ x ≦ 1, 0 ≦ y ≦ 1｝
Are output in such a manner that t changes linearly as the central part changes from the central part to the peripheral part. This real value t is used as the threshold value of the coordinates (x, y).

In such a method of generating a halftone dot threshold, the coordinates (x, y) of the scan unit area are originally treated as real values, so that the number of scan unit areas corresponding to the length of one side of the halftone dot is determined. When the threshold value of the rotating halftone dot is to be obtained using real-valued coordinates (x ′, y ′) obtained by rotating the coordinates (x, y) having the threshold value t, the coordinates (x ′ , y '), if there is a scan unit area, the threshold of that unit area is set to t
_{x, y} , if not, two thresholds t _{x1, y1} and t in the vicinity
_{From x2, y2} , the threshold value of the scan unit area (X, Y) between these two coordinates is obtained by interpolation, and a threshold value having the same appearance frequency even locally corresponding to (X, Y) is obtained.

[Embodiment] Hereinafter, details of the present invention will be described with reference to the drawings showing one embodiment.

 FIG. 1 shows a method of generating a halftone dot threshold according to the present invention.

That is, first, the function value generating unit 2 receives the coordinates (x, y) of the scan unit area as an input, and a real-valued function z = f (x,
A function value generating process for outputting y) is performed.

Next, a function value correction unit performs a function value correction process of inputting the real value z and correcting the value to output a threshold value t of the real value. This modification causes z to make its output frequency uniform and vary linearly.

In the first embodiment, f (x, y) = h (x) + h (y) where z = f (x, y) where Is selected, and the graph of h (w) is shown in FIG.

In the region {(x, y) | 0 ≦ x ≦ 1,0 ≦ y ≦ 1}, the x direction and y
Two-dimensional coordinates (x, y) consisting of real values x, y distributed almost uniformly in the directions, for example, (0,0), (0,0.01), (0,
0.02),…, (0,1.00), (0.01,0), (0.01,0.0
1), (0.01,0.02), ... (0.01,1.00), (0.02,0). (0.02,0.01), (0.02,
0.02), ... (0.02,1.00), ... (1.00,1.00), the values of z are (0,0) and (0,1), (1,
(0), (1,1), z = 0, (0.50,0.50), z = 1, and in other cases, a value between 0 and 1, which is 101 in increments of 0.01.
Values appear, and the appearance frequency of z is large near z = 0.5 at the center and small at the four ends. Further, based on this, a cumulative density distribution curve (a curve in which cumulative frequency / total frequency is made to correspond to z)
Is drawn as shown in FIG. This curve is , And it is analytically known that the number of coordinates (x, y) approaches g (z) as much as possible.

This g (z) has a large change width of t with respect to a unit change width of z near z = 0.5 where the appearance frequency of z is large,
In the vicinity of z = 0 or z = 1 where the appearance frequency of z is small, the width of change of t with respect to the unit change width of z is small, and if the value of g (z) is taken instead of z, the total z When the value is increased while appearing at an equal frequency, g (z) increases at equal intervals, so that g (z) can be used as the threshold t for (x, y).

From FIG. 3, a graph of t = g (z) can be used to determine the site of occurrence of an arbitrary threshold. By the way, when the threshold, that is, the range of z such that g (z) is in the range of 1.0 to 0.95 is obtained and the coordinates (x, y) giving the value of z in the range are obtained, the range of the point (x, y) is obtained. Is the central black part in FIG. 4. Each time the black part and the lattice pattern part change from the black part to the peripheral part, the range of the threshold value is 0.95 to 0.9, 0.9 to 0.85,.
From 0 to 0.

Thus, the threshold value changes smoothly from the center to the periphery.

In the second embodiment, f (x, y) = h (x) + h (y) hw) = (sin ((w−0.25) × 2π) +1) / 4 is selected as z = f (x, y). The graph of h (w) is shown in FIG.

0 ≦ x ≦ 1, 0 ≦ y ≦ 1 is divided into 300 equal parts each, and 9,000,000 coordinates (x, y) are created by combining them, and the value of z corresponding to (x, y) is obtained. 501 in almost ascending order
Values are extracted, and the values are expressed as z _i = i / 500 (i = 0 to 500)
A table (one-dimensional array on the storage device) is prepared as the value of the cumulative density distribution t corresponding to. The relationship between z and t is shown in FIG.

In this case, the function value correction process is performed using the table. That is, for z generated in the function generation process, z
= Z _i to become z _i the search to such a z _i, if there is t = i (0
≤ i ≤ 500), i may be used as a threshold, or 0 ≤ t
When ≤1, t = i / 500.

When there is no z _i that satisfies z = z _i , the threshold t is obtained by interpolation using z _p and z _{p + 1} before and after z. For example, z = 0.
023, z ₂₅ = 0.02, if the z ₂₆ = 0.03, the t = (25 + ((0.023 -.0.02) / (0.03-0.02)) / 500 = 0.0506.

As a result, a threshold having the distribution shown in FIG. 7 could be generated. The black part shown in the center of FIG. 7 is an area of a point (x, y) having a threshold value of 1.0 to 0.95, and the range of the threshold value is changed every time the black part and the lattice pattern part alternate from the black part toward the periphery.
The threshold value changes from 0.95 to 0.9, 0.9 to 0.85, ..., 0.005 to 0, and the threshold value changes smoothly from the center to the periphery.

The third embodiment is an example in which correction is performed using a table, as in the second embodiment, where z = f (x, y) and f
(X, y) = h (x) + h (y) Was selected. The graph of h (w) is shown in FIG.

Z that divides 0 ≦ z ≦ 1 into 500 equal parts in a one-dimensional array on the storage device
Cumulative appearance frequencies t _i corresponding to ₀ , z ₁ ,..., Z ₅₀₀ were set. z _i
The relationship between and _i is shown in FIG. In the correction process, similarly to the third embodiment, the given z is searched in the one-dimensional array, and t is obtained by interpolation.

As a result, a threshold having the distribution shown in FIG. 10 could be generated. FIG. 10 shows a point (x, y) region where the black part in the center has a threshold value of 1.0 to 0.95, and the threshold range is 0.95 to 0.9, 0 each time the lattice pattern part and the black part change from this black part toward the periphery.
The threshold value changes smoothly from 9 to 0.85,..., 0.05 to 0, and gradually changes from the center to the periphery.

The fourth embodiment is also an example of correction using a table,
f (x, y) = h (x) × h (y) h (w) = (sin ((w−0.25) × 2π) +1) / 2 where z = f (x, y) Yes, the graph of h (w) is shown in FIG.

Z that divides 0 ≦ z ≦ 1 into 500 equal parts in a one-dimensional array on the storage device
Cumulative appearance frequencies t _i corresponding to ₀ , z ₁ ,..., Z ₅₀₀ were set. z _i
The relationship between and _i is shown in FIG. Using this table, z was corrected by interpolation, and as a result, a threshold having the distribution shown in FIG. 13 could be generated. The way of displaying the distribution is the same as in the second and third embodiments.

In the fifth embodiment, assuming that z = f (x, y), f (x, y) = h (x) × h (y) where h (w) = (sin ((w−0.25) × 2π) +1 ) / 2, and g (z) = 0.5 − ((0.5−z) × 2) ^a / 2 (when 0 ≦ z ≦ 0.5) 0.5 − ((z −0.5) × 2) ^a / 2 (when 0.5 <z ≦ 1.0). The result depends on the value of a.
Error when 665 <a <0.753 (difference between desired value and actual value)
Can be within 2%. Correction was performed with a = 0.717, and as a result, the distribution threshold shown in FIG. 14 could be generated.

[Effects of the Invention] As is apparent from the above description, according to the present invention, each threshold value can appear at even frequency even in a rotating halftone dot, and the unit area of a laser scan in which each threshold value should have the threshold value. Can be finely determined according to the coordinates of the halftone dot, and a halftone dot threshold value generation method capable of obtaining a beautiful halftone dot can be obtained.

[Brief description of the drawings]

FIG. 1 is a diagram showing a method of generating a halftone dot threshold value according to the present invention.
The figure shows a function h (w) for constructing the function f (x, y) of the first embodiment. FIG. 3 shows the function h (w) of the first embodiment.
FIG. 4 is a diagram showing a distribution of halftone threshold values generated in the first embodiment, and FIG.
The figure shows a function h (w) for constructing the function f (x, y) of the second embodiment. FIG. 6 shows the function h (w) of the second embodiment.
Set in the primary array on the storage device to correct
FIG. 7 is a diagram showing the relationship between z _i and its cumulative frequency t, FIG. 7 is a diagram showing the distribution of threshold values of halftone dots generated in the second embodiment, and FIG. 8 is a function f (x, function h for constructing y)
FIG. 9 is a diagram showing (w), FIG. 9 is a diagram showing the relationship between the value z _i set in the primary array on the storage device for correcting z in the third embodiment and its cumulative frequency t, FIG. FIG. 11 is a diagram showing a distribution of threshold values of halftone dots generated in the third embodiment, and FIG. 11 is a diagram showing a function h (w) for constructing a function f (x, y) of the fourth embodiment. FIG. 12 is a diagram showing the relationship between a value z _i set one-dimensionally on the storage device to correct z of the fourth embodiment and its cumulative frequency i, and FIG. 13 is a diagram showing the fourth embodiment. FIG. 14 is a diagram showing a distribution of halftone thresholds generated, FIG. 14 is a diagram showing a distribution of halftone thresholds generated in the fifth embodiment, and FIG. 15 is a print screen using halftone dots. 1 ... print screen, 2 ... function value generation unit, 3 ... function value correction unit, 4 ... minute area, 5 ... virtual line, 6 ... halftone dot, 7 ... grid

──────────────────────────────────────────────────続 き Continued on the front page (58) Field surveyed (Int.Cl. ⁶ , DB name) H04N 1/40-1/409 H04N 1/46 H04N 1/60

Claims (3)

(57) [Claims] 1. A method for generating a two-dimensional spatially periodic threshold value, comprising two-dimensional coordinates (x, y) consisting of real values x, y distributed almost uniformly in the x and y directions, respectively. ) As input and outputting a function value z of a function z = f (x, y) taking a real value, and a function value correcting process of outputting a real value t obtained by correcting z with the z as an input. And the real value t is set as a threshold value. In the function correction process, the real value t increases a change width with respect to a change width of z in a portion where the value is large, and a change width of z in a small portion. A method of generating a halftone dot threshold value, wherein the width of change is made smaller than the width, and the appearance frequency of the real value t is corrected so as to be distributed substantially uniformly with the distribution of (x, y). 2. In the function correcting step, the function z =
A patent wherein a cumulative density distribution function or a cumulative frequency distribution function of f (x, y) or a function approximated to them is calculated as a real value t = g (z), and g (z) is calculated as a threshold value t. A method for generating a halftone dot threshold according to claim 1. 3. In the function correcting step, the function z =
g (z) is calculated as a real value t = g (z) using a cumulative density distribution function or a cumulative frequency distribution function of f (x, y) or a function approximating them, and a finite number of A patent, wherein a sequence of function values of t = g (z) for a sequence of values of z is set in advance in a one-dimensional array on a storage device, and t corresponding to the input z is obtained interpolatively. A method for generating a halftone dot threshold according to claim 1.