JPS6367076A - Picture processor - Google Patents

Abstract

PURPOSE:To reduce the feeling of roughness of an image, by setting a density pattern, and a pulse width having mutivalues, so that the distribution of energy in a laser beam matrix becomes a function having a curved surface of intermediate density area. CONSTITUTION:A picture signal read by a scanner unit B is added on a shading correction circuit 44 which corrects optical illuminant unevenness. Next, the correction of each chrominance signal is performed at a gamma correction circuit 45, and at a masking processing circuit 46, the optimum quantities of a Y (yellow), an M (magenta), and a C (cyan) at time of printing, are calculated. At a UCR processing circuit 47 at the next stage, the quantity of a black(BK) appropriate for the generation of a black color is calculated from the Y, the M, and the C. Furthermore, at a density pattern processing circuit 48, pattern distribution corresponding to each multilevel out of 64 multilevels, is outputted, and at a mutilevel processing circuit 49, the modulation of the pulse widths corresponding to values 1-4 in a pattern are performed. At this time, the value of the pulse width is decided by a Gaussian function, and in this way, no high harmonic wave which becomes a dotted image, is generated, thereby, the image in which a feeling of roghness is reduced, can be reproduced.

Description

DETAILED DESCRIPTION OF THE INVENTION (Technical Field) The present invention relates to an image processing apparatus, and more particularly to an image processing apparatus that can be applied to a laser printer, a digital color copying apparatus, and the like.

(Prior Art) Recently, in image processing devices applied to digital color copying machines, etc., small 71-lix There is a way to represent multiple gradations. For example, in JP-A-58-85671 and JP-A-59-204060, one dot is further expanded by three
By changing the pulse width by 1/3, for example, in a 4x4 dot matrix, 4X4X3+1
=49 value (+1 is 0 level).

By the way, the radiation contraction line of the latent image pattern on the photoreceptor produced by irradiation with the beam energy obtained by the above method becomes a rectangular wave or a triangular wave. The grainy or noisy feel that occurs in digital images is caused by the harmonic components of this rectangular wave or triangular wave. The spatial frequency component (MTF) of a rectangular wave or triangular wave is processed by means such as Fourier transform, and when the above waveform is subjected to Fourier transform, a high 11 wave is generated at a frequency higher than the fundamental frequency of the dot.

On the other hand, in the case of an analog copying machine, the image generated by the lens optical system becomes a blurred point image due to the aberration of the optical system and the oblique light image shift caused by the curved photoreceptor, and the image is blurred close to a Gaussian distribution. It is thought to be formed from a collection of images. The MTF obtained by Fourier transform of a Gaussian distribution is still a Gaussian distribution, and high frequencies are not generated. In other words, it is determined that there is a difference between digital images and analog images here.

The transfer function of an analog image is a Gaussian distribution, but the transfer function of a digital image is a pulse distribution including harmonics.

To explain the conventional means based on the drawings, FIG. 9 shows an example of area gradation using the density pattern method without pulse modulation.
Case 2 will be shown. In the figure, each pixel l is made up of 2 x 2 micropixels 2 arranged in a matrix, and the intermediate tones are determined by sequentially increasing the number of black micropixels like (bl, (cl, fdl, tel) in the figure. , expresses five gradations of brightness.

FIG. 10 shows a graph of area gradation by the conventional density pattern method [FIG. 10(a)] and reproduced gamma characteristics [FIG. 10(bl)] in a 4×4 matrix. As shown in the figure, in the case of only area gradation, the linearity of the gamma characteristic is not good due to the reproduced dots.

This is especially true when reproducing highlights, where the rise in density becomes steep and the 8-dimensionality deteriorates.

Figure 11(a), mountain 1. (cl indicates the moving state of the laser beam, the width of the pulse signal, and the light intensity distribution of the obtained micropixel. From this figure, when the pulse signal is shorter than the moving state of the laser beam, the light intensity distribution of the obtained micropixel is It is understood that the peak value of is lower than the peak value in a state where the pulse signal is longer than the moving state of the laser beam.

FIGS. 12(5) and 12(b) respectively show a conceptual diagram of a pattern and gamma characteristics when a 4×4 matrix is subjected to ternary equal interval pulse width modulation. According to experiments using this method, even when multivalued pulse width modulation is performed, the density rises quickly at levels corresponding to values 1 to 3 of the rise of the highlight, as shown in Figure 12(b). It was confirmed that the gamma characteristic is not a straight line after all.Also, as mentioned earlier, the energy distribution is rectangular, and as shown in the example fat pattern in Figure 12, this pattern does not include harmonics. It has a rough texture.

(Objective) The present invention has been made to eliminate the above-mentioned drawbacks of the conventional art, and its purpose is to provide an image processing device that can reduce the roughness of images, which is a drawback of digital images. It is in.

(Structure) In order to achieve the above-mentioned object, the present invention provides image processing comprising an image processing means for expressing halftone density by area gradation and a modulation means for expressing multi-value by modulating the pulse width of a pulse signal of a laser beam. The apparatus is characterized in that the concentration pattern and the multivalued pulse width are set so that the energy distribution in the matrix of the laser beam is not rectangular in the intermediate concentration region but is a function having a certain curved surface. .

Hereinafter, a detailed explanation will be given based on one embodiment of the present invention.

FIG. 1 is a schematic diagram showing the photosensitive drum and its surroundings of a digital color copying apparatus to which the present invention is applied. In the figure, 3 is a photosensitive drum, 4 is a charger, 5 is a yellow developer, and 6 is a magenta developer. , 7 is a cyan developer, 8 is a black developer, 9 is a pulse sensor, 10 is a pre-transfer lamp, 11 is a pulse counter, 12 is a transfer drum, 13 is a transfer charger, 14 is a pre-cleaning static eliminator, 15 is a separation charger, 16 is a cleaning device. Further, 17 is a yellow replenishment signal line, 18 is a magenta replenishment signal line, 1
9 is a cyan replenishment signal line, 20 is a black replenishment signal line, 21 is a yellow development bias signal, 22 is a magenta development bias signal, 23 is a cyan development bias signal,
24 is a black developing bias signal, 25 is a drum rotation pulse signal line, and 2G is a pulse counter output line.

In this way, this digital color copying apparatus has four developing units 5. 6. 7. 8, a photosensitive drum 3, a transfer drum 12, and a writing laser optical system which is omitted from this figure as it will be explained in detail later.

The process of toner replenishment control in the above configuration will be explained. First, apart from the image, a standard detection pattern is exposed onto the photoreceptor drum 3 at the rear end of the image signal by a laser beam directed toward the photoreceptor drum 3 from the direction of the arrow, thereby forming a detection pattern latent image. This latent image is developed with the same developing bias value as the image signal portion. The detection pattern image of the photoreceptor drum 3-H is rotated and sent to the pulse sensor 9 i1.
When one period has elapsed, the amount of reflected light of the pattern is measured by the pulse sensor 9 at the timing measured by the pulse counter 11. This process is repeated four times for each color.

FIG. 2 is a schematic cross-sectional view showing an image scanner that photoelectrically converts and reads a document as an example of a writing laser optical system. In the figure, 27 is an image scanner, 28 is a video processing circuit, 29 is a contact glass on which the original is placed,
30 is a light source, 31 is a first scanning mirror, 32 is a second scanning mirror, 33 is an imaging lens, 34 is a photoelectric conversion unit that electrically reads the reflected light of the document, 35 is a lighting circuit that drives the light source 3o, 36 is a DC that performs mechanical scanning (sub-scanning)
It's a motor.

This image scanner 27 uses a fixed original scanning method to accommodate the diversity of originals, so it uses a light source 3o and a first scanning mirror so that the optical path length of the original reflected light 37 is always constant during sub-scanning. 31 and the first to mount
The carriage Cal and the second carriage Ca2 on which the second scanning mirror 32 is mounted are driven in the sub-scan by the DC motor 36 at a speed ratio of 2:1.

In addition, the photoelectric conversion unit 34 uses two sets of CCDs (solid-state imaging device arrays) to reduce the burden on the microlenses of the reduction imaging optical system.
Documents can be read at a high resolution of (dots/open 2). Therefore, the main scanning is performed using these two sets of C
Divided imaging scanning is performed using the OD and two sets of imaging lenses 33.

FIG. 3 is a block diagram illustrating the general operation of the image scanner 27. In the figure, 30 is a light source, 34 is a photoelectric conversion unit, 34a, 34b are CCDs, 34c, 34
d is an amplifier, 35 is a lighting circuit, 36 is a DC motor, 38
is a timing circuit, 38a is a timing generator,
38b is an oscillator, 39 is a scanner control circuit, 4
0 is a position sensor, 41 is a sub-scanning speed control circuit, 42 is a speed signal generation circuit, 43 is an encoder, and a is a start signal.

In FIGS. 2 and 3, a lighting circuit 35 drives a light source 30. In FIG. The light reflected from the original on the reading line (sub-scanning line) irradiated by the light source 30 is reflected by the first scanning mirror 31 and the second scanning mirror 32, and is guided to the photoelectric conversion unit 34 by the 2M imaging lens 33.

Here, the divided images are formed on the light receiving surfaces of the two CCDs 34a and 34b, respectively, and main scanning is performed. These CCD34
The main scanning by a and 34b is performed by the timing generator 38.
CCD34a by the shift pulse from a.

34b is sequentially shifted and performed.

FIG. 4 is a block diagram showing a schematic configuration for performing color image processing. In the figure, a scanner unit B and a printer unit C are shown centered around an image processing unit A. In FIG. 4, 44 is a shading correction circuit, 45 is a gamma (γ) correction circuit, 46 is a masking processing circuit, 47 is a UCR processing circuit, 48 is a density pattern processing circuit, 49 is a multi-value processing circuit, and 50 is a laser driver. , 51 is a laser unit, 52 is a main body control section, 5
3 is a synchronous control circuit.

In FIG. 4, the image signal read by the COD light receiving unit (not shown) of scanner unit B is processed by a shading correction circuit 44 for correcting optical illuminance unevenness (yellow color).
Correction is performed for each of Y), magenta (M), and cyan (C). Next, the gamma correction circuit 45 corrects the gradation of each color signal.

The masking processing circuit 46 calculates the optimum appropriate amounts of Y, M, and C during printing. Next UCR processing circuit 4
In 7, black (Bk) suitable for creating black color:! i
is calculated from Y, M, and C.

Further, a density pattern processing 48 outputs a pattern distribution according to each of the 64 gradations, and a multi-value processing circuit 4
At step 9, pulse width modulation is performed in accordance with values 1 to 4 in the pattern.

The pulse of this pulse width modulation will be explained with reference to FIG. FIG. 5 shows a portion according to the present invention, and in this figure, four values of pulse width are provided. The values during this pulse are determined by a Gaussian function. Figure 5 (al is represented by a Gaussian distribution and each level. Here, M represents the pulse duration allowed for one dot (unit: time n 5ee), which is 125 nsec in this case. Figure 5 fal,
The value of each level in (bl, (cl) is determined as follows. First, level 1 is the smallest pulse that can be reproduced by an actual printer, and depends on the beam diameter, but in an experimental machine it is about 10 nsac. Therefore, level 1 is determined to be 1Qnsec. Next, M is 125ns
Since it is determined to be ec, 125,y
Assign 10 to , and 3 to X (the level 1 distance is 3 distance from the center of Gauss) to determine the shape parameter a.

For the other levels 2 and 3, 2.3 is substituted for each of X to obtain the value of y. In this way, four values of pulse width are obtained.

Figure 5 (il+ indicates the pulse width of each level within one dot.
The values during each of the 11 pulses are shown. In addition, Fig. 5 (dl shows an example of the gamma characteristic in this case. This gamma characteristic is equal to the gamma characteristic of halftone dots in printing. The reason is that the shape of the dots approximates a Gaussian distribution. , it is thought that the same tendency as in the case of halftone dots is shown.

In addition, FIG. 6 fal, (bl, (cl) shows the case of 7-level pulse width modulation in the same method as above,
Figure 6 (a) shows the Gaussian distribution and the setting status of each level with dotted lines, Figure 6 (b) shows 1 dot 7 values, Figure 6 (C1 is the same as Figure 5 (c+) The value in each pulse of the seven values calculated according to the formula % is therefore determined as in the table for each level of pulse.

Figure 7 (al) shows the density pattern of a 4x4 matrix. This shows the priority order for filling in the patterns within the 4x4 matrix. value -6
The pulse level distribution within the matrix determined in the case of 37 values out of 4 gradations is shown. FIG. 7 [C] shows the processing flow of the level allocation determination program. This processing flow sets the gamma curve by inputting the gamma parameter (xt
+y2=a) and 7-level level pulse width input P(1) to P
Energy calculation for 0 to 63 gradations using (7) E (x)
and loop =0~63. If K>63,
Arrange each P (x) according to the order of the density pattern, and
Outputs 4 matrix O to 63 gradations. On the other hand, in the loop = 0 to 63. If K>63, the number of 1-value P (1) is calculated from the 7-value P (7).

Figure 8 (a) shows the horizontal and diagonal level values of the determined 4x4 matrix, Figure 8 (8th appearance) shows its exposure energy distribution, and Figure 8 (C1 shows the latent image electric field distribution). This shows the radiation line.The latent image electric field distribution in Fig. 8(C) is approximately Gaussian distribution.Fig. It is something.

As described above, according to this embodiment, the toner adhering to the photosensitive drum 3 shown in FIG. 1 has a shape approximating a Gaussian distribution, resulting in an analog dot image and no harmonics.

Therefore, an image with less roughness is reproduced.

(Effects) As described above, according to the present invention, the concentration pattern and the multivalued pulse width are set so that the energy distribution in the matrix of the laser beam is not rectangular in the intermediate concentration region, but is a function with a certain curved surface. As a result, the adhering toner on the photoreceptor drum can be made to have a shape that approximates a Gaussian distribution, resulting in an analog dot image that does not generate harmonics and therefore reproduces an image with less graininess. It is possible to provide an image processing device that performs the following functions.

[Brief explanation of the drawing]

FIG. 1 is a schematic diagram showing the photosensitive drum and surroundings of a digital color copying apparatus to which the present invention is applied, FIG. 2 is a schematic sectional view showing an image scanner, and FIG. 3 is a block diagram illustrating the general operation of the image scanner. FIG. 4 is a block diagram showing a schematic configuration for color image processing, FIG. 5 (al is a graph showing a Gaussian distribution, FIG. 5 is an explanatory diagram showing the pulse width of each level within one dot, Fig. 5 (c+ is a diagram showing the relationship between each level and pulse [11], Fig. 5 fdl is a graph showing gamma characteristics, Fig. 6 (al is a graph showing Gaussian distribution in the case of 7-value pulse width modulation, Figure 6 (1+)
Figure 6 (C) is a diagram showing the relationship between each level and the pulse, Figure 7 (fa) is a diagram showing the density pattern of a 4 x 4 matrix, and Figure 6 (C) is a diagram showing the 4 x 4 matrix density pattern. A diagram showing pulse level distribution in a 4×4 matrix, FIG. 7 fcl is a processing flowchart of a level distribution determination program, and FIG.
Figures (Al is an explanatory diagram showing the horizontal and diagonal level values of the determined 4x4 matrix, Figure 8 (b) is a graph showing the exposure energy distribution, Figure 8 (C1 is a graph showing the latent image electric field distribution) Graph, FIG. 8 (dl is a conceptual three-dimensional diagram showing the toner image distribution of a 4×4 matrix, FIG. 9 (al,
bl, (cl, (di), (e) is a diagram showing the brightness of five gradations in the density pattern method without pulse modulation, the first
Figure 0 (al is a diagram showing area gradation by the density pattern method of 4 x 4 matrix, Figure 10 (bl is a graph showing reproduced gamma characteristics, Figure 11 (al, (bl, (c)
are the moving state of the laser beam, the width of the pulse signal, and
FIG. 12 (
81, (bl is a pattern conceptual diagram and gamma characteristic diagram when ternary equal interval pulse width modulation is applied to a 4×4 matrix. A...Image processing unit, B...Image scanner unit 1 -1C...printer unit, 48...
Density pattern processing, 49...multi-value processing circuit. Amended string procedure amendment (voluntary) February 7, 1986

Claims (2)

[Claims] (1) In an image processing apparatus comprising an image processing means that expresses halftone density by area gradation and a modulation means that expresses multi-value by modulating the pulse width of a pulse signal of a laser beam, the energy in the matrix of the laser beam is An image processing apparatus characterized in that a density pattern and a multivalued pulse width are set so that the distribution is not rectangular in an intermediate density region but is a function having a certain curved surface. (2) The concentration pattern of the matrix is a concentrated type or a spiral type, and the reference distribution is a Gaussian function that approximates the pulse width distribution of each level of the multivalue. The image processing device described.