JPS605669A - Multi-value output system - Google Patents

Abstract

PURPOSE:To improve the contrast of output density and to sharpen an intermediate tone picture by modulating a pulse width possible for setting output the density of each gradation at an equal interval in the system controlling the output density of a picture with a multi-value output. CONSTITUTION:A pulse width (b) is set to a spot diameter (a) so that the output density of each gradation of a picture obtained in the similar number of gradations as the output number of a multi-value output is an equal interval. That is, in case of a ternary output, a pulse width being nearly 80% of the binary pulse width is formed and in case of a 4-value output, two pulse widths of 64% and 90% of the binary pulse width are formed so as to reproduce a sharp intermediate tone picture with good contrast.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Application A.1] The present invention relates to a multi-value output method in which an optimum pulse width is set for multi-value output using a pulse width pressure 14.

[Prior art]

In general, as an image scanning device used for image recording, a scanning device using a rotating polygon mirror or a vibrating C is used for reasons such as a large scanning angle and less chromatic dispersion. It is widely used for image recording and display in Wit, various i-display devices, printing devices, etc. In particular, devices using rotating polygon mirrors are widely used as high-speed scanning devices. Conventionally, in order to record a half-tone ii!ji image using this German scanning device 1'1'', each pixel on the recording display surface was formed by nXn fine pixels arranged in a matrix (41♂). , the halftone f is determined by the black and white combination of the 02 micropixels constituting each pixel, that is, the distribution of white micropixels and black micropixels.
f: I try to express myself. For example, as shown in Fig. g1, each pixel IDi-2X2 micropixels 11 are arranged in a matrix. In order to express intermediate tones using micropixel groups arranged in a matrix like this, the dragon 0 shown in Fig.
~14), by increasing the number of black micropixels by one order, the brightness of the 5th floor iL',l is expressed, and in general, each pixel 10 is expressed as n x 11 pixels. If the fine pixels 11 form (1゛4), then n X n + 1
It is possible to record and display 1 ji 1 image of gradation. The advantage of a halftone image recording device in such a state is that each pixel is represented by two values, either pure white or full heat, so the signal processing system can be easily constructed. Become.

12) There is no problem even if the gamma characteristic of the recording/displaying medium is nonlinear, and any recording/displaying medium can be used.

There is an inn.

However, when performing halftone image recording using a combination of black and white in such a fine pixel matrix using, for example, a laser beam printer, there is a good correspondence between the number of black fine pixels in each pixel and one recording density. Linearity was not obtained.

In other words, the number of black micropixels in each pixel is 11Q
For l+, the recording degree of each pixel (f - the second for the vertical axis
As shown in the figure, when the correspondence between the two names is expressed, as the number of black micropixels increases, the recording type ``11'' generally becomes saturated. The shown percentage curve shows an example of the above-mentioned recording performance characteristic for a matrix of 4×4=16 micropixels, that is, a pixel matrix, as shown in FIG. In the figure, the number assigned to each fine pixel indicates the order of blackening.

As mentioned above with reference to FIGS. 2 and 3, there is no direct relationship between black micropixels in the pixel matrix and recording density, and the density sharply increases in areas with a small number of black micropixels. Changes are occurring. The reason why such density changes occur is that the recording dots are recorded extending beyond the fine pixels.

As a solution to this drawback, a method has been proposed in which the second half of +111, in which fine pixels are blackened, is simultaneously recorded with a plurality of blackened dots, as shown in Figure 4). As a result, as shown in the same figure (B1), if the order of blackening is plotted on the horizontal axis and the density is plotted on the vertical axis, the relationship can be expressed as a nearly straight line.Therefore, the order sound input data of blackening and Then, you can obtain a degree of turbulence that is proportional to the input data.

However, even with this method, the number of gradations that can be expressed is small because the output is only binary (black and white). Therefore, it is necessary to increase the number of gradations, and in addition to black and white, intermediate &'<I
It is effective to make all the boxes (t or dot). In other words, it is necessary to use a multi-value output method. As this multi-value output method, for example, when using a laser beam printer,
+Il laser brightness modulation multi-level output method and (2)
Two types of multi-level output methods using laser pulse width modulation are known. Among these, the former +1) multi-level output method using laser brightness modulation has some difficulties in control.
Moreover, it has drawbacks such as instability in response to environmental changes. On the other hand, the pulse width modulation of the laser on the rear right (2) causes multiple
(The 1-output method has the advantage of compensating for the disadvantages of +fl and further increasing the number of outputs by 6 pixels and 1 signal by modulating the pulse width to be shorter than the pulse width of 1 fine pixel. Figure 5 (IN to +13+ is a diagram showing a specific example of this pulse 111', '5 modulation. 1, that is, as shown in Figure 5 (at), it is a clock pulse output for each fine pixel. A certain fine pixel clock pulse is input to a one-shot multivibrator Ov, and from the one-shot multivibrator Ov, a clock pulse whose pulse width is modulated by changing the variable volume VL is input to an AND gate AND.
! In addition, the image of the beam 5 and the symbol number are input to one input terminal of the AND gate AN.
A signal for modulating the pulse width of the laser is output from the LJ. Therefore, by changing the variable volume V L /, the laser pulse width can be modulated as shown in FIGS. 5(b) and 5(c).

However, with the multi-value output method based on (2), the intermediate 6t·l! Iiii! When reproducing the Il image, each lif effect 1. There was a drawback that the contrast of the output density of 4 was often poor and unclear.

[Purpose of the invention]

This invention was made by focusing on the above-mentioned drawbacks, and has good output density contrast and clear intertone images.
The purpose is to obtain a multi-level output method that can be used to express

〔Example〕

An embodiment of the present invention will be described below with reference to FIGS. 6 to 10.

The light beam emitted during recording generally has a Gaussian-shaped cross section, but in order to proceed with the following explanation as a qualitative particle analysis, the spot shape of the light beam is changed to a spot width a (i71t) as shown in FIG. Think of it as a rectangular shape.

The light I intensity distribution when such a spot-shaped light beam (pulse width as shown in Figure 11-6) is already recorded is i
f, rl = 8 +, K P +, ) <However, 8
(, zl-rect(5) is a function representing the original spot, P(tech)-rect(king) is a function representing the pulse width, and ■ is a convolution integral). The results of this calculation are shown in FIG.

In Fig. 7, (AJ shows the case where the pulse width is larger than the spot diameter a, and the value of the light intensity divided by the convolution IILt is the spot diameter a, that is, the integral value of the light beam. At the same time, the saturated intensity continues until the pulse width is reached, and then attenuates.

In the same figure, (13) is the pulse width and the spot diameter a
If they are equal, the M4rIJ intensity becomes a point.

In the same figure, (C1 indicates the case where the pulse width is smaller than the spot diameter a, and the peak intensity is proportional to the pulse width. This saturation is (spot diameter a - Hals 1iv
It continues for a distance equivalent to b) and eventually attenuates.

In the figure, (1)) shows the case where the pulse width becomes extremely small, and the saturation intensity corresponding to the pulse width is (
It continues for a distance equivalent to the spot diameter a minus the pulse width b). Note that since the pulse width is extremely small, (spot diameter a-
The pulse width b) is very close to the pulse width b).

Such a light intensity distribution is converted into a potential distribution and developed in the case of a photon photograph.

Now, if the development level is an open chain type (that is, high gamma), the formed potential distribution is cut at a gate value of 6,
A black and white binary output is obtained. The light quantity versus 'superordinate characteristic is generally a nonlinear function. However, a certain potential value corresponds to one amount of light. Now, this amount of light is C in Figure 7.
Assume that the level is The length l of the cross section of the light distribution cut by such an INJ value is 1, and is expressed as A=a+b-2c (b≧c, a≧C≧0) (a). Here, in the case of b(c), the gate value becomes stronger than the peak, and becomes l-0. This is shown in Figure 8 (
1)).

From this relationship, the length l of the cross section of the existing light component I
is surrounded by a straight line expressed as d=ba−a and 13=b+a (the shaded part shown in FIG. 8).

For example, to simplify the calculation, let us assume C-7. That is, it is assumed that t7 is 1 of the peak intensity of the amount of light.

At this time, the above formula (a) becomes z = b, b ≧ f - old... (b), and since the length e of the blackened part is proportional to the pulse width, the pulse width is It can be seen that a spot diameter a of 1 or less does not contribute to blackening.

In other words, when the gate value type development method f: is set, the pulse width is 44] light [1 (a ↓ violet 1! (j iM), the development blackening width l is the pulse width This is proportional to
This also applies when the pulse width is smaller than the spot diameter a, and holds true up to -L < 2 of the spot diameter a. If the value is less than i, the value becomes less than 1jlJ value C, and no blackening occurs.

In reality, the situation is somewhat different when considering the simple model described above. 1. At t, the midpoint of the change in the potential distribution is somewhat shifted from the midpoint (c-2) of the light extraction distribution α.

However, it has been experimentally shown that the above-mentioned model can be matched with reality with only minor modifications.

Therefore, based on the above conclusion, the pulse width can be set so that the densities determined from the area ratio of the halftone dots formed are equally spaced (that is, linearly).

That is, in Fig. 1O, if Bl is the one in which all the fine pixels are blackened, and Bl is the one in which only the width of X is blackened, then the reflectance 1 is 13u c hd a According to the old formula, at is also W(1-x)-1-R'r-x --------
,,,(c) However, l (w indicates the reflectance of the white paper, and RT indicates the reflectance of the blackened portion (toner, etc.).

is given by

Now, 11. When determining the concentration ot with w=1, 1)--l
og (L, r+RTex)−'Og (L
x(L-1(rr)) - , , , , , . Here, in the case of FIG. 10 A), that is, since the solidity degree Dmax is at X-1, lamax = -1o
g LL'l・Therefore, 1t・1・= 10−, rh+
ax, and the formula is D'=-log(t-, x(t-1o Month...Town...
).

Therefore, tB) for that shown in 1lOIA(5)
The concentration ratio P of those shown in is expressed as, and the following is obtained.

From the above conclusion, this X is the pulse Ilv. Since it is proportional to 1, the final pulse width ratio α multiplied by:

Therefore, for example, in the case of three-value output, in order to make the output rotational pressure of each gradation equal intervals, that is, P = 0.5, Dmax
If == 1.2, then the pulse width should be about 80 inches of the binary pulse width.

In addition, in the case of four-value output, the output shapes of each one tone are equally spaced, that is, p == 7. , to make it 1, Dma
If x4 = 1.2, then for P-100, for P-g, it becomes: Approximately 64% of the binary pulse width and 90 x 02 pulse widths need to be created.

As described above, the pulse width for each gradation can be easily set.

〔Effect of the invention〕

As explained above, the multi-value output method of the present invention is a multi-value output method in which the image output state 1jJ is full IQIed by multi-value output. By adjusting the pulse width so that output symptoms 1-IE of each gradation of the image, which are balanced by the number d, can be installed at equal intervals, the contrast of the output temperature of each gradation is good, l1ttl: clear intermediate arA Ii
! This has the effect of being able to reproduce the J image.

[Brief explanation of the drawing]

Figure 1 shows lf'j
Figure 2 shows the blackening state of the pixel that has been created.
A diagram showing the relationship between t* and the number of blackened dots. Figure 4 tA) is a diagram showing the order 1k of blacking of micropixels in the i'i element, and the latter part'f of the order of blackening of the micropixels entirely; , tlJl in the same figure is a diagram showing the relationship between the order of blackening and the degree of rivetness, and Figure 5 is a diagram showing a specific example of laser pulse 1 (i modulation). Block diagram, +=Figure +131 is the same figure (5)
Output wave, shape diagram in the same figure (C1 is the same diagram (C1 is the output waveform diagram showing the case where the laser pulse width is narrowed by further changing the variable volume 'CK' from the bottom 71Q of 131, Fig. 6 1
FIG. 7 is a diagram showing the spot shape of the original beam in an embodiment of the present invention, FIG. 7 is a diagram showing the pulse width of the light beam, and FIG. 8 is a diagram showing the light intensity distribution of the light beam. b) "A diagram showing the case of a. Figure (B) is a diagram showing the case of b = a. Figure tc+ is a diagram showing the case of b (a). Figure (L)) is a diagram showing the case of b = a. - Figure 9 shows the length of the cross section of the existing light distribution of the one shown in Figure 7. Figure 10 (5) shows an example of Jin's invention. Figure (13) shows a case in which all of the micropixels are blackened, and FIG. ii!+i 11・・・・・・Fine pixel D・・・・・・Density Dmax---solid (M:! a・・・・・・Pulse II!iab・・・・・・
...Spot diameter Figure 7 Figure 2 (2) (1197f no P) Procedural amendment q (method) % formula % 2 invention 0 name Multi f16 output method 3, relationship with person Jj+ who does sleeves 2 Application Person f3 Nisho (residence) Ki (name) (100) Representative of Gyanon Co., Ltd. (6606) Name 41 Hiroyuki Niwa 5 Lottery jl: Order of IJ Ij Sho I'll 5f3 year
9B27th (Shipping date) 6 Sleeve jl: Due to jl.
Number of inventions to be added 7, #li it two λ゛j illustration, 1
111? ! f. Ki-Shizuku review left (inset of "Detailed explanation of the invention") 8. Contents of "sleeve" r As attached 8. Contents of the amendment 1 ■ First line from the bottom of page 1 of the book Diary 1J2 f-(
? Insert ``30 shots 1! Detailed explanation of IJ'' into 1151 of ``Multi-value output method described in item 1'' in the second line from the bottom of the first section.

Claims (1)

[Claims] +ll In a multi-value output method in which the output of the image (9,) is output by pulse width modulation, the open side 1 is made by the multi-value output,
The main feature of the multi-level output is that the pulse ll'1r1kg is adjusted so that the output 0 degrees of each floor R1'J of the art world, which is obtained by the same number of gradations as the number of outputs of the multi-value output, can be set at equal intervals. Value output method. (2) The pulse width to be modulated is obtained by setting the ratio α to the pulse width before the change ≦175 to the density ratio of the solid color P of the ink. Multi-value output method of ili&.