JP2549627B2 - Digital color image reproduction processing method and apparatus - Google Patents

Description

Description: TECHNICAL FIELD The present invention relates to a color image reproducing process, and in particular, obtains image data (density data) of each color component by separating an original image.
A so-called halftone recording digital color image reproduction process in which this is processed into recording color component density data, and the recording color component density data specifies the halftone expression pattern for each recording color component and records the pattern. Regarding

2. Description of the Related Art In one type of conventional halftone image recording, when halftone recording is performed based on digital image data (density data) having a gradation range of 0 to MN, M × N of 1 to 1 are recorded. M ・
Each threshold value of threshold data indicating each of N is regularly or randomly dispersed in an M × N matrix, and each threshold value is compared with digital image data. The recording information bit is assigned to the position of the threshold value of, and the non-recording information bit is assigned if it is less than the threshold value, and 1 is assigned to the bit in the form of the recording and non-recording information bit matrix corresponding to the threshold value matrix.
Printing is performed by allocating the above dots.

In advance, the threshold matrix is compared with each of the image data indicating 1 to MN, and the recording and non-recording information bit matrix when the image data is 1 to MN, respectively,
M / N pieces are created and stored in a memory, and at the time of image reading and recording, one of the recording and non-recording information bit matrix is designated by the image data obtained by the image reading, and the bit matrix is specified. There is also a mode in which recording is performed corresponding to.

In the case of monochrome recording, a threshold matrix or M · N
The recorded and non-recorded information matrices may be one set. In the case of color recording, for example, in the case of color recording of three colors of Y (yellow), M (magenta) and C (cyan), color separation reading and read signal processing are used to allocate to Y, M and C recording. Y image record data, M image record data and C
Image recording data is obtained, and the above-mentioned halftone recording is performed for each of these image recording data.

However, when halftone recording of a plurality of colors is performed based on the same threshold matrix or the same set of recording and non-recording information bit matrices, moiré and the like appear in a reproduced color image and the image quality deteriorates. There is a problem that the vividness of colors is lost because all colors overlap on the same point.

Conventionally, in order to prevent moire, a threshold value matrix or a recording / non-recording information bit matrix is rotated by a predetermined angle, and a screen having a predetermined angle for each color and a different angle from other colors is used. The threshold value matrix or the recording / non-recording information bit matrix is used, which is modified to have a corner and has a unique screen angle for each. One example is disclosed in JP-A-58-182372.

However, in the related art, even in a low-density recording area, each color is overlapped and recorded at the same point, so that the improvement in color vividness is poor. Further, since the number of gradations is small, halftone representation of a reproduced image is poor, and in order to increase it, it is necessary to increase M × N. However, if M × N is increased, image data (original image That is, the recording area assigned to each of the data indicating the density of the entire predetermined small area of the reproduced image becomes large, and the reproduced image expands with respect to the original image. In order to prevent enlargement, it is necessary to set a large area to be read as one image data of the original image. This results in rough image reading and eventually impairs the fidelity of the reproduced image. As a result, even if the number of gradations is set to a wide range, the reproduction quality of an image is not substantially improved unless the area of one dot for recording is reduced. Therefore, conventionally, the halftone expression pattern (threshold matrix or recording / non-recording information bit matrix) cannot be made very large.

Further, conventionally, the same halftone expression pattern (orthogonal threshold matrix or orthogonal recording or non-recording information bit matrix) is rotated by a predetermined angle for each color, so that the degree of freedom in setting the screen angle is low and the vividness of the color is low. It was not possible to set a halftone expression pattern (threshold matrix or recording / non-recording information bit matrix) for improving the. Conventionally, since the halftone expression pattern is relatively small as described above, there is a problem that this further limits the degree of freedom.

SUMMARY OF THE INVENTION It is an object of the present invention to simultaneously realize sharpness and wide gradation range of recording colors for color image reproduction and smoothness of gradation expression.

Structure The present invention separates a color image into a plurality of colors, converts the image density into digital data for each color component, and processes the digital data into color component recording density data; Halftone expression pattern having a plurality of threshold data corresponding to one of the threshold data of a predetermined number of dots, or all of the threshold data is compared with each value of the recording density data in the planned range, and the predetermined number of dots is supported. The color component recording density data is converted into recording and non-recording bit information using a halftone expression pattern consisting of a number of bit distribution patterns corresponding to the range of recording density data, which is the recording and non-recording bit distribution. Recording / non-recording bit information is associated with one dot of the recording medium for each color component, and a predetermined color is recorded in the dot with which the recording information bit is associated; in digital color image reproduction processing: Medium The intertone expression pattern used for the first color recording and the second color recording is such that when a predetermined area is recorded by using it, the recording information bits correspond to the X and Y coordinates as the recording density increases in correspondence with the recording density. The recording information bit distribution spreads from a predetermined point, and the predetermined points are at different positions for each color, and the halftone expression pattern used for the third color recording has a recording information bit distribution spread from the predetermined point. One set for each color component different from the halftone expression pattern of the first color record and the second color record in which a plurality of predetermined points are dispersed between the predetermined points of the first color record and the second color record When the color component recording density data is converted into recorded and non-recorded bit information, the halftone expression pattern corresponding to each color is converted from the halftone expression pattern specified by the color component recording density data to Component recording concentration day Based on the coordinates of the pixels of the original image obtained, a region of a plurality of dots of a part of the halftone expression pattern is regularly cut out as a child matrix pattern corresponding to the pixels of the original image, and obtained from the child matrix pattern. The recorded and non-recorded bit information is output as image information corresponding to the color component recording density data.

According to this, as the designated density becomes higher on the halftone expression pattern, the recording area spreads from a different position for each color, that is, since the halftone dot center is different for each color, the low As the density recording is performed, there is no overlapping recording of different colors, and thus the color definition is remarkably improved. Moreover,
In the pattern of the third color, the halftone dots are dispersed in a plurality of dots, so that the color distribution is fine and the color expression is smooth. For example, assuming that an 8 × 8 matrix is used as a halftone expression pattern and recording is performed in yellow (Y), magenta (M), and cyan (C), the halftone dot center is temporarily divided into three divided regions of 8 × 8 matrix for each color patch. When set to the center of each, when each color is recorded with a density of 64/3 (decimal number) or less, there is no weight of the color at all. In this case the density is 64/3
The clarity of recorded colors up to is extremely high. The vividness of colors of M and C decreases due to color mixing with the other. On the other hand, when Y is mixed, the degree is low. Therefore, in the embodiment of the present invention, assuming that the M halftone expression pattern and the C halftone expression pattern are at positions where the halftone dots are separated, the Y halftone expression pattern is for M and C. It shall be dispersed between the dots of the pattern. That is, Y
The number of halftone dots 8 in the pattern for use is increased (thus, the distance between the halftone dots of Y and the halftone dots of M and C becomes shorter than the distance between the halftone dots of M and C). According to this, the deterioration of the vividness of M and C is reduced, and the color reproducibility of the color reproduction image is improved.

On the other hand, M × N matrix halftone expression pattern is M ×
Since N gradations can be expressed, the larger the halftone expression pattern, the wider the range of gradations that can be expressed, and the wider the range of gradations that can be expressed and the higher the sharpness. The larger the size, the larger the area unit for the halftone expression (recording area for pattern size: small area), and from this aspect, the smoothness of the halftone expression deteriorates. That is, the sharpness and wide gradation range of the recording color and the smoothness of gradation expression in size units of the halftone expression pattern conflict with each other.

However, according to the present invention, from the halftone expression pattern specified by the color component recording density data, based on the coordinates of the pixels of the original image from which the color component recording density data is obtained, a plurality of dots of a part of the halftone expression pattern Since the area is regularly cut out as a child matrix pattern corresponding to the pixels of the original image, the smoothness of halftone expression is not impaired, and the sharpness and wide gradation range of the recorded color and the smoothness of gradation expression are simultaneously achieved. To be realized.

More specifically, in the preferred embodiment of the present invention, further, in the patterns of M and C, the halftone dots are dispersed into two or more points to finely control the color dispersion to obtain a smooth color expression and the intermediate color. The key expression pattern MMP is divided into m × n child matrix patterns CMP _{11 to} CMP _mn , with m in the main scanning direction and n in the sub-scanning direction, and the beginning of the letter is the child matrix pattern in the MMP. The main scanning direction position of the, the latter half of the script indicates the position in the sub scanning direction. And m × n pieces of recording density data, which similarly consist of ICD ₁₁ to ICD _mn In assuming that obtain one image information of the halftone representation pattern content, records information of the child matrix pattern CMP _ij halftone expression pattern identified in the recording density data ICD _ij bit distribution for the recording density data ICD _ij Get as information.

Since a part of the halftone expression pattern is assigned to the recording density data in this way, the recording area corresponding to one piece of recording density data is smaller than the area corresponding to the halftone expression pattern, and therefore the halftone expression pattern is large. Even with a (wide gradation range), it is possible to perform recording in a mode in which the reproduced image does not expand with respect to the original image without making the recording 1 dot area extremely small. Nevertheless, since the halftone expression pattern is large, the gradation range of the recording density data is much larger than the gradation range obtained by the matrix pattern corresponding to the recording medium area (the conventional matrix pattern corresponds to this). Can be set to a large value. Even in such a setting, the small area of the original image that corresponds to one density data may be a small area that corresponds to the recording small area. Therefore, a wide halftone expression can be obtained without making the image quality particularly rough (without impairing smoothness). For example, the halftone expression pattern is set to 8 × 8, and 4 × 4 of one copy is recorded.
When using, the reading small area to which 1 density data is assigned corresponds to 4 × 4, the recording small area corresponds to 4 × 4, and the gradation range is 0 to 8 × 8. Becomes wider. However, the resolution (how many pixels are assigned to the small area: smoothness) does not deteriorate at all.

In addition to this, as the halftone expression pattern becomes larger, the degree of freedom in setting the predetermined point for each color, that is, the center of the halftone dot, for vividly recording the color in the halftone expression pattern becomes significantly higher. Moreover, the area allocated for recording is widened without overlapping with other colors. This is because the number of recording colors is a predetermined value, but the area (number of pixels) of the halftone expression pattern increases, and the area of the halftone expression pattern / the number of recording colors increases.

Therefore, according to the present invention, sharpness of a recording color and a wide gradation range and smoothness of gradation expression are realized at the same time.
In the embodiment in which halftone dots are dispersed into two or more dots even in the halftone expression pattern addressed to M and C, the sharpness of the recording color, the fineness of the color dispersion, the wide gradation range, and the smoothness of gradation expression are achieved. Realize at the same time.

The present invention will be described in more detail.
As shown in FIGS. a, 11b, and 11c, when the threshold data (1 to 64 shown in decimal in the figure) is distributed in the 8 × 8 matrix, the recording density data shows 16 (decimal). If it is a thing, the record of the distribution shown by the diagonal lines in the figure is
It is recorded in a small area corresponding to 8 matrices. Whichever pattern is used, the halftone expression pattern (threshold data matrix) for each color can be set so that the four colors do not overlap when the recording density data is 16 or less.

For example, a pattern obtained by shifting the halftone dot centers (threshold point centers 1 and 2) of the pattern shown in FIG. 11b by 2 in the X direction and 2 in the Y direction is used as an M (magenta) halftone expression pattern shown in FIG. 12a. Set the halftone dot center (threshold value data 1 and 2) of the pattern shown in FIG. 11b to 6 in the X direction, 2 in the Y direction,
The C (cyan) halftone expression pattern shown in FIG. 12b is set by shifting the squares of the above, and the halftone dot center (threshold data 1, 2, 3 and 4) of the pattern shown in FIG. 11c is set to 2 in the X direction. ,
When the Y (yellow) halftone expression pattern shown in FIG. 12d is set by shifting by 2 squares in the Y direction and each color is recorded with the recording density data 16 based on these,
The recording color distribution is as shown in FIG. 12c. As described above, when the recording density data of each color is 16, the respective colors do not overlap each other, and moreover, the entire small area corresponding to the 8 × 8 matrix is recorded with the same number of dots for each color, and the entire surface is recorded. When the recording density data for each color is 16 or less, therefore, there is no overlap between the colors, and vivid color recording is achieved. Note that in FIG.
When the halftone expression pattern (threshold matrix) of each color is set as shown in FIGS. 12b and 12d, the distance between the halftone dot centers (threshold data 1 and 2) between the patterns is the M pattern (FIG. 12a). ) And the C pattern (Fig. 12b).

Suppose that the pattern shown in FIG. 11a (or its halftone dot is X, Y shifted on the matrix) is assigned for M, and the pattern shown in FIG.
The pattern shown in the figure (or its halftone dot on the matrix X, Y
(Shifted pattern) is assigned to C, the halftone dot of the same pattern (pattern of FIG. 11b and its halftone pattern shifted on the matrix) is shifted as described above (FIG. 12a, The distance between the halftone dots of M and C becomes shorter than in the case of assigning (Fig. 12b) to M and C. Therefore, as described above, it is preferable that M and C are obtained by relatively maximally shifting the halftone dots of the same basic pattern (FIG. 11b).

With the same logic, the basic pattern shown in FIG. 11a may be transformed into two patterns in which halftone dots are relatively shifted to the maximum, and these patterns are assigned to M and C, respectively.
In this case, the pattern for Y is, for example, as shown in FIG. 12b (the one having a halftone dot between the halftone dots of the M and C patterns).
To use.

Similarly, let the pattern for M be as shown in FIG. 11c,
The pattern for C is entirely shifted by shifting the halftone dots of the pattern of FIG. 11c in the X direction by 2 meshes in the Y direction, and the pattern for Y is the pattern shown in FIG. 11d. May be shifted or modified so that the halftone dots exist between the halftone dots of the M and C patterns.

FIGS. 13a to 13c show an example of pattern allocation for implementing the present invention when a 10 × 10 matrix is used, as shown in FIG.
Figures 14 to 14c show another example. In FIGS. 13a to 13c, diagonal lines indicate areas to which a bit indicating recording is assigned when the recording density data is 40. Figure 13a ~
In the example shown in FIG. 13c, the screen angle differs by 45 ° for each color, and the halftone dot pitch is Different from each other. It can be applied to gradation image recording with a gradation number of 101 or less.

The diagonal lines in FIGS. 14a to 14c indicate areas where the recording density data is 20 or less and the recording dots are assigned. In this example, M
The patterns of C and C are the same patterns with halftone dots shifted so that the maximum phase is obtained, and these and the screen angle are 45
° Different, the dot pitch is The pattern obtained by phase shifting is set for Y so that the halftone dot is farthest from the halftone dots of M and C.
Even if recording is performed with the recording density data of 20 or less, as shown in FIG. 14d, not only the overlapping of the colors does not occur but there is still an empty space. No

In any of the above examples, the distance between the halftone dot centers of the M and C patterns is longer than the distance between them and the halftone dot center of the Y pattern. The halftone dot center of the Y pattern is located between the halftone dot center of the M pattern and the halftone dot center of the C pattern, and the number of halftone dot centers of the Y pattern is different from that of the M and C patterns. More than the number of halftone dots.

When the recording density data of each color has a small value, the colors do not overlap with each other, and M and C do not overlap with each other up to a relatively high gradation. Therefore, vivid color recording is achieved.

Other objects and features of the present invention will become apparent from the following description of embodiments with reference to the drawings.

First, referring to FIG. 1, a document 1 is placed on a platen (contact glass) 2 and a fluorescent lamp 3 ₁ , 3 ₂ for illuminating the document is provided.
First mirror that is illuminated by and is capable of moving its reflected light
4 ₁ , the second mirror 4 ₂ and the third mirror 4 ₃ pass through the imaging lens 5 and enter the dichroic prism 6 where light of three wavelengths, red (R), green (G) and blue. It is split into (B). The separated light enters CCDs 7r, 7g, and 7b, which are solid-state image pickup devices, respectively. That is, red light is CCD7r, green light is CCD7g,
The blue light is incident on the CCD 7b.

The fluorescent lamps 3 ₁ and 3 ₂ and the first mirror 4 ₁ are mounted on the first carriage 8, and the second mirror 4 ₂ and the third mirror 4 ₃ are mounted on the second carriage 9.
The second carriage 9 is mounted on the 1/3 of the first carriage 8
By moving at a speed of 2, the optical path length from the original 1 to the CCD is kept constant, and the first and second carriages are scanned from right to left when reading the original image. A carriage drive pulley 11 fixed to the shaft of the carriage drive motor 10
The first carriage 8 is coupled to the carriage driving wire 12 wound around the wire, and the wire 12 is wound around the moving pulley (not shown) on the second carriage 9. As a result, the first carriage 8 and the second carriage are moved forward (original image reading scan) and returned (return) by the forward and reverse rotations of the motor 10, and the second carriage 9 is half of the first carriage 8. Move at speed.

When the first carriage 8 is at the home position shown in FIG. 1, the first carriage 8 is detected by a home position sensor 39 which is a reflection type photo sensor. That is, when the first carriage 8 is driven to the right by exposure scanning and is out of the home position, the sensor 39 does not receive light (carriage is not detected), and when the first carriage 8 returns to the home position, the sensor is detected. Reference numeral 39 indicates light reception (carriage detection), and the carriage 8 is stopped when non-light reception is changed to light reception.

Referring now to FIG. 2, the outputs of the CCDs 7r, 7g, 7b are:
After being subjected to analog / digital conversion and subjected to necessary processing in the image processing unit 100, recording color information for recording energization of each of black (BK), yellow (Y), magenta (M) and cyan (C) Is converted into a binary signal. Each of the binarized signals is supplied to the laser driver 112bk, 11
The laser beams are input to 2y, 112m, and 112c, and the respective laser drivers energize the semiconductor lasers 113bk, 113y, 113m, and 113c, thereby emitting laser light modulated with a recording color signal (binary signal).

FIG. 1 is referred to again. The emitted laser light is reflected by the rotating polygon mirrors 13bk, 13y, 13m and 13c,
After passing through the f-θ lenses 14bk, 14y, 14m and 14c, the light is reflected by the fourth mirrors 15bk, 15y, 15m and 15c and the fifth mirror 16bk, 16y, 16m and 16c, and the polygonal mirror tilt correction cylindrical lenses 17bk, 17y, 17m. And 17c, then the photosensitive drum 18b
Image-irradiates k, 18y, 18m and 18c.

The rotary polygon mirrors 13bk, 13y, 13m and 13c are fixed to the rotary shafts of the polygon mirror drive motors 41bk, 41y, 41m and 41c,
Each motor rotates at a constant speed to drive the polygon mirror at a constant speed. By the rotation of the polygon mirror, the above-mentioned laser light is scanned in a direction perpendicular to the rotation direction (clockwise direction) of the photosensitive drum, that is, the direction along the drum axis.

The laser scanning system of the color recording apparatus is disclosed in detail in Japanese Patent Application No. 60-37213 filed by the present applicant.
The same applies to the laser scanning system shown in the figure.

The surface of the photoreceptor drum is uniformly charged by charge scorotrons 19bk, 19y, 19m and 19c connected to a negative voltage high voltage generator (not shown). When the laser beam modulated by the recording signal is applied to the uniformly charged surface of the photoconductor, the photoconductive phenomenon causes the charge on the surface of the photoconductor to flow to the equipment ground of the drum body and disappear. Here, the laser is not turned on in the portion where the original density is high, and the laser is turned on in the portion where the original density is low. As a result, the surface of the photoconductor drums 18bk, 18y, 18m, and 18c has a potential of −800V at the portion corresponding to the dark portion of the original, and −100V at the portion corresponding to the light portion of the original. An electrostatic latent image is formed corresponding to the light and shade. This electrostatic latent image is developed by a black developing unit 20bk, a yellow developing unit 20y, a magenta developing unit 20m and a cyan developing unit 20c, respectively, and black, yellow and magenta are respectively formed on the surfaces of the photoconductor drums 18bk, 18y, 18m and 18c. And forming a cyan toner image.

The toner in the developing unit is positively charged by agitation, and the developing unit is biased to about -200 V by a developing bias generator (not shown). A corresponding toner image is formed.

On the other hand, the recording paper 267 stored in the transfer paper cassette 22 is fed out by the feeding operation of the feed roller 259, and sent to the transfer belt 25 at a predetermined timing by the registration roller 24. The recording paper placed on the transfer belt 25 is the transfer belt.
By moving 25, photoconductor drums 18bk, 18y, 18m and 18c
Of the black, yellow, magenta and cyan toners on the recording paper by the action of the transfer corotron at the lower part of the transfer belt while passing through the lower part of the transfer belt sequentially and passing through the photosensitive drums 18bk, 18y, 18m and 18c. Are sequentially transferred.
The transferred recording paper is then sent to the thermal fixing unit 36, where the toner is fixed to the recording paper, and the recording paper is discharged to the tray 37.

On the other hand, the residual toner on the surface of the photoconductor after the transfer is removed by the cleaner units 21bk, 21y, 21m and 21c.

The cleaner unit 21bk for collecting the black toner and the black developing unit 20bk are connected by the toner collecting pipe 42, and the black toner collected by the cleaner unit 21bk is collected in the developing unit 20bk. The black toner is reversely transferred from the recording paper to the photoconductor drum 18y at the time of transfer, and the yellow, magenta, and cyan toners collected by the cleaner units 21y, 21m, and 21c are different from the different color developing device in the preceding stage of those units. Since the toner is mixed, it will not be collected for reuse.

The transfer belt 25 that feeds the recording paper in the direction from the photosensitive drums 18bk to 18c is stretched around the idle roller 26, the drive roller 27, the idle roller 28, and the idle roller 30, and is driven to rotate counterclockwise by the drive roller 27. To be done. Drive roller
27 is pivotally attached to the left end of a lever 31 pivotally attached to a shaft 32. At the right end of the lever 31, a black mode setting solenoid plunger 35 (not shown) is pivotally mounted. A compression coil spring 34 is arranged between the plunger 35 and the shaft 32,
This spring 34 gives the lever 31 a rotational force in the clockwise direction.

When the black mode setting solenoid is not energized (color mode), as shown in FIG.
25 is in contact with the photosensitive drums 44bk, 44y, 44m and 44c. In this state, when the recording paper is placed on the transfer belt 25 and toner images are formed on all the drums, the toner image of each image is transferred onto the recording paper as the recording paper moves (color mode). When the black mode setting solenoid is energized (black mode), the lever 31 rotates counterclockwise against the repulsive force of the compression coil spring 34, the drive roller moves down 5 mm, and the transfer belt 25
Is separated from the photosensitive drums 44y, 44m and 44c and remains in contact with the photosensitive drum 44bk. In this state, the recording paper on the transfer belt 25 is only in contact with the photoconductor drum 44bk, so only the black toner image is transferred to the recording paper (black mode). Since the recording paper does not contact the photosensitive drums 44y, 44m and 44c,
No attached toner (residual toner) of 4m and 44c is attached, and no stain such as yellow, magenta, and cyan appears.
That is, in copying in the black mode, a copy similar to that of a normal monochrome black copying machine can be obtained.

The console board 300 has a copy start switch, a color mode / black mode designation switch 302 (the switch key is turned off immediately after the power is turned on, and the color mode is set;
When the switch is closed twice, the switch key lights up and the black mode is set, and the black mode setting solenoid is energized; when the switch is closed twice, the switch key is turned off and the color mode is set and the black mode setting solenoid is de-energized.) In addition, it is equipped with other input key switches, charactor displays and indicators.

Next, the operation timing of the main part of the copying mechanism will be described. The modulation energization of the laser 43bk based on the recording signal is started at substantially the same timing as the start of the exposure scanning of the first carriage 8, and the lasers 43y, 43m and 43c are respectively the photosensitive drums.
The modulation bias is started with a delay of the movement time Ty, Tm, and Tc of the transfer belt 25 by a distance of 44y, 44m, and 44c from 44bk. The transfer corotrons 29bk, 29y, 29m, and 29c respectively take a predetermined time (the time required for the laser irradiation position on the photosensitive drum to reach the transfer corotron) from the start of the modulation bias of the lasers 43bk, 43y, 43m, and 43c. Fired after a delay.

Please refer to FIG. The image processing unit 100 includes CCDs 7r and 7
The three color image signals read by g and 7b are converted into black (BK), yellow (Y), magenta (M) and cyan (C) recording signals required for recording. The BK recording signal is given to the laser driver 112bk as it is, but the Y, M and C recording signals are stored in the buffer memories 108y, 108m and 108c, respectively, which are the original recording color gradation data, and then the delay time Ty, Tm. After a time delay of reading out after Tc and Tc and converting into a recording signal, the laser driver
Give to 112y, 112m and 112c. The image processing unit 100 includes the CCDs 7r, 7g and 7g in the copier mode as described above.
Although three color signals are given from b, in the graphics mode, the three color signals are supplied from the outside of the copying machine to the external interface.
Given through 117.

Shading correction circuit 101 of image processing unit 100
Is the color gradation data obtained by A / D converting the output signals of the CCD7r, 7g and 7b into 8 bits, and correcting the uneven light intensity and the sensitivity variation of the internal unit elements of the CCD7r, 7g and 7b. Create read color gradation data.

The multiplexer 102 is a multiplexer that selectively outputs one of the output grayscale data of the correction circuit 101 and the output grayscale data of the interface circuit 117.

The γ-correction circuit 103 that receives the output (color gradation data) of the multiplexer 102 changes the gradation (input gradation data) according to the characteristics of the photoconductor, and also the gradation by the operation buttons of the console 300. And the input 8-bit data is changed to output 6-bit data. Since the output is 6 bits, data indicating one of 64 gradations will be output. The 6-bit 3-color gradation data indicating the respective gradations of red (R), green (G), and blue (B) output from the γ correction circuit 103 is applied to the correction generation / black separation circuit 104.

The correction generation and the complementary color generation in the black separation circuit 104 are the reading of the names of the respective recording color signals of the color reading signals, and the red (R) gradation data is the cyan (C) gradation data and the green (G) floor. The tone data is converted (read) into magenta (M) tone data, and the blue tone data (B) is converted into yellow tone data (Y). C, M and Y
The gradation data is given to the averaged data compression circuit 105 as it is. If all of these gradation data indicate high density, black recording may be performed. Therefore, the digital comparator in the circuit 104 converts the C, M, and Y gradation data to respective threshold setting switches. Compare with the set reference value data. Each of the digital comparators compares 8-bit data with each other.
Data with the upper 2 bits of the level added (input data)
Is set to L level with the least significant digit 1 bit and the uppermost digit 3 bits, and is compared with 8-bit data (reference value data) in which the second to fourth bits from the lowermost are reference value data set by the switch for threshold setting. , L is given to the NAND gate if the input data is less than or equal to the reference value data, and H is given if it is over the reference value data.
The NAND gate is L (black) when all the comparators are giving an L signal, and is H when one of them is giving an H signal.
(White) is output and given to the data selector 110. To explain this in more detail, the gradation data input of the comparator is in the range of 0 to 3FH in hexadecimal 6-bit data. When the value is 0, black is output, as the value increases, white is output, and the output black At the time of insertion, L is black and H is white. Therefore 8
The MSB side 2 bits (Q6, 7) of the bit input data are input to L, and the lower 6 bits (Q0 to 5) are input C, M, and Y gradation data, respectively. The comparison data side uses a rotary type dip switch so that the comparison level can be set to 7 steps.
Furthermore, since the black level is set, if black is included including too much white color, halftone (gray) can be recorded as black with high resolution, but black is often generated in terms of color balance, which is not preferable. So for the time being 7 to intermediate level
Since the 5th and 6th bits are also set to L so that it can be set in stages, and there is no need to set it very finely, 1 bit on the LSB side is set to L and the set value from the dip switch is input to the middle 3 bits (P1 to 3). There is. The dip switch setting is now 010
, The reference value is 0000010, and when the data of each of C, M, and Y are all less than this value, that is, the decimal number 0 to 3
During this period, the output of the comparator is L and the black (BK) output is L
(Black) Here, the DIP switch for setting is C,
It is commonly used for comparison judgment of M and Y, but by using three sets, it is possible to set each color, or by setting the setting range width of each color using the minimum and maximum setting switches, the specific color It is also possible to output with a black pattern with high resolution.

The averaged data compression circuit 105 of the image processing unit 100
4 × 4 with 6-bit gradation data for one image
The image data is averaged and output as 6-bit gradation data. In the case of this embodiment, the processing mode in which the sizes of the input image and the output image are the same is standard, and the input data (read value from CCD) is A / D converted into 8-bit data and converted into 6-bit data by γ correction. However, the output data to the laser driver is laser on / off (1
Bit) data. It is possible to separate 64 levels of density by inputting 6-bit data. Therefore, the input data 8 ×
The density of 8 pixels is averaged to obtain density data. Further, the data amount and the processing speed are compressed to 1/64 by this averaging, and the data capacity when storing and the cost of the hardware part are reduced.

Next, the masking processing circuit 106 and the UCR processing circuit 107 will be described. The operation formula of the masking process is uniform, Yi, Mi, Ci: data before masking, Y ₀ , M ₀ , C ₀ : data after masking.

In addition, UCR processing as a general formula, Can be represented by

Therefore, in this example, using these equations and using the product of both coefficients, Is calculated to find a new coefficient. Masking process and
Coefficients (a ₁₁ ″, etc.) in the above equation for performing both UCR processing at the same time are calculated in advance and substituted into the above equation,
Scheduled inputs Yi, Mi and Ci of masking processing circuit 106
The calculated value (Y ₀ ′ etc .: which becomes the output of the UCR processing circuit 107) associated with each (6 bits) is stored in the ROM in advance.
Therefore, in this embodiment, the masking processing circuit 106
The UCR processing circuit 107 is composed of a set of ROMs, and the data of the addresses specified by the inputs Y, M and C to the masking processing circuit 106 are output from the UCR processing circuit 107 as buffer memories 108y, 108m, 108c and It is given to the gradation processing circuit 109. Note that generally speaking, the masking processing circuit 106
Is adjusted to the characteristics of the spectral reflection wavelength of the recording image forming toner.
The Y, M, and C signals are corrected, and the UCR processing circuit 107 performs correction for color balance in superimposing the color toners.

Next, the buffer memories 108y and 108m of the image processing unit 100
And 108c. These simply generate a time delay corresponding to the distance between the photosensitive drums.
The write timing of each memory is the same, but the read timing is the memory 108y according to the modulation activation timing of the laser 43y, the memory 108m according to the modulation activation timing of the laser 43m, and the memory 108c is the modulation timing of the laser 43c. It is performed according to the timing of energization, and is different for each.
The capacity of each memory is based on the maximum size of A3.
At 108y, 24% of the maximum required amount of A3 original, memory at 108m
It may be 48%, and about 72% in the memory 108c. For example, if the CCD reading pixel density is 400 dpi (dots per inch: 15.75 dots / mm), the memory 108y is about 87 kbytes, the memory 108m is about 174 kbytes, and the memory 10
8c should have a capacity of about 261 bytes. In this embodiment, since 64 gradations and 6-bit data are handled, the memory
108y, 108m and 108c have capacity of 87K, 174K and 2 respectively
It is 61K bytes. When the memory address is calculated in 6-bit units from the byte unit (8 bits), the memory becomes 108y: 116K × 6 bits, memory 108m: 232K × 6 bits, and memory 108c: 348K × 6 bits.

Next, the density pattern processing circuit 109 of the image processing unit 100
Will be explained. This circuit 109 is a circuit for generating a pattern corresponding to the density from the recording density data of each of Y, M, C and BK, and is a BK gradation processing circuit 109bk, y gradation processing circuit 109y, M gradation. It is composed of a processing circuit 109m and a C gradation processing circuit 109c.

The 6-bit grayscale data can represent density information of 64 grayscales (65 grayscales including 0 to which no pattern is assigned).
Ideally, if the dot diameter of one dot can be varied to 64 steps, the resolution will not be reduced, but dot diameter modulation is only stable at most about 4 steps in the laser beam electrophotographic method. There are many combinations of density patterning and beam modulation. Here, with an 8x8 matrix, 64
A gradation expression processing method is used.

FIG. 3 shows the configuration of the Y gradation processing circuit 109y. The pattern memory 1012 is a ROM, and as shown in FIG.
From a halftone expression pattern (threshold distribution pattern) in which threshold data is distributed in 8 matrices, recording information bits are written only at one position of the threshold and non-recording information bits are written at other threshold positions. Density 1 pattern specified by density 1... Recording information bits are written at positions below the threshold value i and non-recording information bits are written at other threshold positions. And a total of 64 halftones of the recorded information bit representation of the pattern,... And the density 64 patterns specified by the density 64 in which the recorded information bits are written at positions (that is, the entirety) of those below the threshold value of 64 An expression pattern has been written. The Y allocation pattern has a halftone dot center in the shaded area shown in FIG. 12c. The hatched squares in FIG. 12c indicate the positions to which recording information bits are assigned when the density data indicates 16.

The other gradation processing circuits 109m, 109c and 109bk have the same hardware configuration as the circuit 109y. However, since the halftone pattern data stored in the pattern memory is different, 64 patterns created based on the original pattern shown in FIG. 12a are stored in the pattern memory (not shown) of the circuit 109m in the circuit 109c. In the pattern memory (not shown), 64 patterns created based on the original pattern shown in FIG.
The k pattern memory (not shown) stores 64 patterns created based on the original pattern shown in FIG. 11a.

Next, the halftone processing of one color Y will be described using the gradation processing circuit 109y as an example. The processing operations of the other gradation processing circuits are the same. In this embodiment, Y, M and C are recorded on the same recording paper in the full color recording mode, but the gradation recording of BK is only one color in the black (single color) gradation recording mode.

As described above, one halftone expression pattern (mother matrix pattern) in one group is specified by the recording density data, and information of a child matrix pattern forming a part of the mother matrix pattern is extracted. Image information is obtained in a form assigned to gradation data. According to this, since the gradation pattern is updated in the unit of the child matrix pattern, the resolution is increased, and thereby, the reproducibility of the edge portion of the face outline, line drawing, etc. of the photographic image is improved. For example, in the outline of the image, the corresponding child matrix pattern becomes a part of the high-density mother matrix pattern, so that the outline clearly appears, and in the low-density portion outside the outline, the corresponding child matrix pattern. Since the pattern becomes part of the low-density mother matrix pattern, a low-density image appears and the outline becomes clear. Further, since the large (mother) matrix pattern is specified by the recording density data, the smoothness is improved in the portion where the poorly changing gradation gradually changes. In other words, the number of child matrix patterns, each of which constitutes a mother matrix pattern corresponding to the recording density data, is arranged in an area corresponding to one mother matrix pattern of the reproduced image.

However, since the mother matrix pattern has similar patterns when the expression density is close, the number of child matrix patterns constituting one mother matrix pattern is small in the image portion where the density is gradually changing. Is similar to one specific mother matrix pattern, and the number of expressed gradations is about the same as the number of expressed gradations represented by the mother matrix pattern. Moreover, as shown in FIGS. 12a to 15c, a unique arbitrary halftone dot center can be set for each color. As described later, the mother matrix pattern MMP is
It is divided into m × n child matrix patterns CMP _{11 to} CMP _mn with m in the main scanning direction and n in the sub-scanning direction, and the beginning of the letter is in the main scanning direction of the child matrix pattern in the mother matrix pattern. , The latter half of the character indicates the position in the sub-scanning direction. And m × n pieces of grayscale data, which are also ICD ₁₁ to ICD _mn If image information for one mother matrix pattern is to be obtained with, the information of the child matrix pattern CMP _ij of the mother matrix pattern specified by the gradation data ICD _ij is used as the image information of the bit distribution for the gradation data ICD _ij . obtain. That is, the position of the child matrix pattern for obtaining the image information corresponding to each of the arrangement of one mother matrix pattern and the number m × n of gradation data corresponds to the position of the gradation data in the mother matrix pattern. Position. According to this, in the area corresponding to one mother matrix pattern of the reproduced image, the information is that of each mother matrix pattern corresponding to each gradation data, but the position is a predetermined number that constitutes one mother matrix pattern as a whole. It has a shape in which m × n child matrix patterns at positions are arranged. According to this, since the mother matrix pattern is similar when the expression density is close, in the image portion where the density is slowly changing, the reproduced image by the m × n child matrix patterns is specified. The similarity with one mother matrix pattern is further increased, the number of expressed gradations becomes equal to the number of expressed gradations represented by the mother matrix pattern, and the density is expressed by the conventional fixed density pattern method using the mother matrix pattern. Be equivalent. Further, for example, in the contour line of the image, the corresponding child matrix pattern becomes a part of the high-density mother matrix pattern, so that the contour line clearly appears, and in the low-density portion outside the contour line, it corresponds thereto. Since the child matrix pattern becomes a part of the low-density mother matrix pattern, a low-density image appears, and the outline becomes clearer.

The mother matrix pattern of one group has density No. 1 to 64
Each of the mother matrix patterns has a structure of 8 × 8 bits (pixels) that expresses 64 gradations (65 gradations with a density of 0, though it does not have a pattern of density 0) As described above, is created based on the original mother pattern (FIG. 15a) having 64 pieces of threshold data.

When the mother matrix pattern is divided into two, A and B shown in FIG. 5A or 5B are child matrix patterns. In the child matrix pattern division shown in FIG. 5a, the odd-numbered one of the recording density data for one row is
Mother matrix pattern for density (1 out of 64 types)
Is specified and the left half A is extracted as recording data, the mother matrix pattern corresponding to the density of the even number is specified, and the right half B thereof is extracted as recording data. In the child matrix pattern division shown in FIG. 5b, the mother matrix pattern (one of 64 types) corresponding to the density is specified in each of the recording density data of the odd numbered rows, and the upper half A is extracted as the recording data. Then, the mother matrix pattern corresponding to the density is specified in each of the recording density data of the odd-numbered rows, and the lower half B thereof is extracted as the recording data.

FIG. 5c shows an example in which the mother matrix pattern is divided into four child matrices A to D. In this example, the mother matrix pattern (one of 64 types) is specified by the odd-numbered recording density data of the odd-numbered row, and the data of the upper left 1/4 minute (corresponding to part A in FIG. 5c) is extracted. , The mother matrix pattern (one of 64 types) is specified by the even numbered recording density data of the odd numbered row, and the data of the upper right 1/4 minute (corresponding part B in FIG. 5c) is extracted and The mother matrix pattern (one of 64 types) is specified by the odd-numbered recording density data of the row, and the data of the lower left 1/4 minute (corresponding part C in FIG. 5c) is extracted and the even-numbered row is even. The mother matrix pattern (one of 64 types) is specified by the recording density data, and the data in the lower right quarter (D corresponding portion in FIG. 5c) is extracted.

When the mother matrix pattern is divided into 16 child matrix patterns A to P as shown in FIG. 5d, and when the mother matrix pattern is divided into 64 child matrix patterns A, B, C, as shown in FIG. 5e. Similarly, when dividing into ..., similarly, the mother matrix pattern is first specified by the recording density data, and then the image information of the child matrix pattern at the position corresponding to the allocation position of the density data for the mother matrix pattern is extracted. .

Now, the recording density data shown in FIG. 7a has arrived, and the mother matrix pattern (recording information bit distribution) is
Assuming that there are 64 types corresponding to densities 1 to 64 shown in the figure, and when four divisions are specified, the gradation data is Therefore, the reproduced image data has the distribution shown in FIG. 8a (the numerical values in FIG. 8a correspond to the density numerical values in FIG. 10, and Alphabet indicates the divided portions in FIG. 5c). That is, the child matrix patterns are arranged as follows in accordance with the distribution of the incoming recording density data (FIG. 7a).

The number at the beginning refers to the mother matrix pattern of the mother matrix pattern 1 assigned to the density indicated by the number.

In the above, the rectangular area surrounded by the line is the size of one mother matrix pattern. In FIG. 8a, a rectangular area surrounded by a bold line is the size of one mother matrix pattern. The reproduced image has the form shown in FIG. 9a.

When the gradation data is arranged as shown in FIG. 7b,
When the image information is reproduced in 16 divisions (as shown in FIG. 5d),
The arrangement of the child matrix pattern shown in FIG.

In FIG. 8b, the rectangular area surrounded by the thick line is the size of one mother matrix pattern. In FIG. 8a, the numeral indicates the mother matrix pattern of the mother matrix pattern 4 assigned to the density indicated by the numeral. The reproduced image has the form shown in FIG. 9b.

The child matrix pattern A shown in FIG.
The extraction of the child matrix patterns A and C shown in FIG.
This is performed by taking the logical product of the 1-byte mask pattern FOH shown in FIG. 4a and the 1-line data of the mother matrix pattern to be extracted in the main scanning direction. When the logical product is obtained, the logical product is stored in the page memory or the buffer memory. This is performed for eight lines.

In the mask pattern shown in FIG. 4a, "1" (shown by oblique lines) is stored in a portion to be extracted, and "0" is stored in a portion not to be extracted. That is, the mask pattern FOH shown in FIG. 4a is data indicating FOH.

The child matrix pattern B shown in FIG.
The extraction of the child matrix patterns B and D shown in FIG.
This is performed by taking the logical product of the 1-byte mask pattern OFH shown in FIG. 4a and the 1-line data of the mother matrix pattern to be extracted in the main scanning direction. When the logical product is taken, "0" of the non-extracted portion of the previous logical product memory is stored in the page memory or the buffer memory, so the data in the memory target area of the page memory or the buffer memory is read and this is now obtained. The logical sum is calculated with the logical product data, and this logical sum is updated and stored in the page memory or the buffer memory. This is performed for eight lines. Also in the mask pattern 0FH shown in FIG. 4a, "1" (indicated by diagonal lines in the figure) is stored in the part to be extracted, and "0" is stored in the part that is not extracted. The mask pattern 0FH shown in FIG. 4a is data indicating 0FH.

Similarly, in the extraction of the pattern information in the child matrix pattern division shown in FIG. 5d, the extraction of the child matrix patterns A, E, I, and M uses the mask pattern C0H shown in FIG. For the extraction of F, J and N, a mask pattern of 30H is used, for the extraction of C, G, K and O, the mask pattern of 0CH is used, and for the extraction of D, H, L and P, 03H is used. A certain mask pattern is used.

Then, the data that extracted A, E, I and M (logical product)
Is written to the page memory or buffer memory as it is, but the data (logical product) obtained by extracting A, E, I and M, C, G, K and O, and D, H, L and P is written to the page memory or buffer memory. It is logically ORed with the already written data and then updated in the page memory or buffer memory.

In the child pattern division shown in FIG. 5e, information extraction of the child matrix pattern is similarly performed.

In the above description, the mother matrix pattern is 1 byte in the main scanning direction and is in byte units, and the child matrix patterns are all the same size. Note that the number of bits in the sub-scanning direction of the mother matrix pattern and the child matrix pattern is arbitrary because it does not matter whether it is a byte unit in terms of information processing.

However, in the main scanning direction, it is preferable to set both the mother matrix pattern and the child matrix pattern in byte units first, because information can be processed in byte units at high speed. Therefore, in the above-described embodiment, the child matrix pattern is processed in byte units by the logical processing using the mask pattern. Therefore, according to this logical processing, all the child matrix patterns constituting one mother matrix pattern do not have to be the same size. When the mother matrix pattern is in units of bytes, the child matrix pattern can be easily processed in units of bytes as described above.

However, when both the number of bits in the main scanning direction of the mother matrix pattern and the number of bits in the main scanning direction are fractions of bytes, the processing becomes complicated.

Therefore, in such a case, paying attention to the number of bits c in the main scanning direction of the child matrix pattern, and assuming that c × d = e bytes and d and e are minimum integers, the arrangement of the child matrix patterns in the main scanning direction is arranged. The data of one column is continuously written to the e byte c times, and the logical product data is written to the page memory or the buffer memory by taking the logical product with the mask pattern which leaves only the data of the required one column in the e byte. When the child matrix pattern is the leftmost one, the logical product data is written as it is to the page memory or the buffer memory, but in the case of the child matrix pattern at other positions, the data of the page memory or the buffer memory is further written. The logical sum is obtained and then written to the page memory or the buffer memory.

As described above, since a large mother matrix pattern is used, the number of gradations can be increased, and since the mother matrix pattern can be easily configured in bytes, information processing can be easily performed in bytes. Since the child matrix pattern is assigned to the gradation data, there is an advantage that the resolution is increased.

Further, there is an advantage that the magnification of the reproduced image can be changed. For example, one child matrix pattern is assigned to one gradation data, but in the child matrix pattern division shown in FIGS. 5a to 5e, the size (number of bits, that is, the number of dots) of each child matrix pattern is different. Therefore, the magnification of the reproduced image differs between the child matrix pattern divisions shown in FIGS. 5a to 5e.

That is, assuming that one piece of gradation data now indicates the density of the entire area of 4 dots (corresponding to the child matrix shown in FIG. 5d) of the original image, the child matrix pattern division shown in FIG. 5d. Then, the reproduced image is 1: 1 with respect to the original image.
However, in the child matrix pattern division of FIG. 5a, the reproduced image is enlarged by 2 times in the main scanning direction and 4 times in the sub scanning direction, and by the child matrix pattern division shown in FIG. In the sub-matrix pattern division shown in FIG. 5c, the reproduced image is enlarged by 4 times in the vertical direction and by 2 times in the sub scanning direction. In the child matrix pattern division shown in (1), a reproduced image is reduced to 1/2 in both the main scanning direction and the sub scanning direction.

Therefore, for example, as shown in FIGS. 5c to 5e, a plurality of child matrix pattern divisions are set, and one division mode is specified in accordance with magnification instruction data M (M indicates the number of divisions). Then, the magnification of the reproduced image can be selected. In order to increase the selectable magnification, it is preferable to increase the size of the mother matrix pattern.

The configuration and operation of the Y gradation processing circuit 109y shown in FIG. 3 will be described here. The pattern memory 1012 is a ROM storing 64 mother matrix patterns created based on the original pattern shown in FIG. 15a. , One of the patterns is specified by the recording density data output from the memory 108y. Data of a specific part (all rows: 8 bits) of the specified pattern is specified by the microprocessor 1010,
Read from the memory 1012 is supplied to the AND gate LG ₁ in. Andgate LG ₁ , microprocessor 1010
Gives the aforementioned mask pattern (1 byte). In logical processing by the AND gate LG _1, so that the data in the child matrix pattern is removed. The extracted data is given to the data selector G ₃ . The selector G ₃ and the OR gate LG ₂ are for processing the extracted data into a bit distribution corresponding to the recording surface, and by the read / write control of these and the microprocessor, a memory capacity of at least one row (8 dot width) or more is provided. The extracted data is surface-expanded in the buffer memory included therein. The data expanded in the buffer memory is transferred to the laser driver 112y on a row-by-row basis.

6a and 6b show the data processing operation of the microprocessor 1010. To explain this, the computer
In 1010, when proceeding to gradation data processing for converting the recording density data received from the memory 108y into recording data (data indicating the recording dot distribution), first, magnification instruction data M (M is the number of divisions of the mother matrix pattern = child matrix pattern) Read the number), the number of divisions in the main scanning direction and the sub scanning direction,
Ie Is set in the register L (step 1: hereinafter, the word step is omitted in parentheses). In this example, M is
Only one of 4 (Fig. 5c), 16 (Fig. 5d) and 64 (Fig. 5e) is used. It is to be noted that M = 16 (extraction of the child matrix in FIG. 5d) is the standard, and the reproduced image is the same size as the original image.

Next, the microprocessor 1010 sets 1 (j = 1) to a counter V for grasping the position (j) of the sub matrix pattern to be processed in the sub-scanning direction (step 2),
The counter H for grasping the position (j) in the main scanning direction is set to 1 (i = 1) (step 3), and the data from the memory 108y is read (step 4). When the input data is the recording density data, the content of the line counter LC is set to a value obtained by multiplying the content of the register L by the value obtained by subtracting 1 from the content of the counter V (8).

Next, the microprocessor 1010 specifies a mask pattern based on the magnification instruction data M and the contents of the counters V and H (9). That is, the child matrix pattern CMP _ij from which the image data is to be extracted is specified from the number of divisions M and the contents of the counters V and H (i is the content of the counter H, j is the content of the counter V, and M is FIG. The number of divisions indicating which division mode in FIG. 5e), and the matrix pattern to be assigned to this child matrix pattern (for example, FIGS. 4a and 4b).
Figure) is specified.

Next, the microprocessor 1010 uses the pattern memory 1012.
1 byte data of the line (aligned in the main scanning direction) indicated by the content of the line counter LC is stored in a buffer memory BUF (internal register) (10), and the data of the buffer memory BUF and the data of the mask pattern are ANDed. ANDs given to the gate LG _1, to update the memory of the logical data in the buffer memory BUF (11), the counter H
(12).

When the content of the counter H is 1, this indicates that the child matrix pattern for extracting information is the leftmost one in the mother matrix pattern.
Write the BUF data as is to the buffer memory 1020 (1
Five).

If the content of the counter H is not 1, the data of the leftmost child matrix pattern has already been written in the memory 1020, and by this writing, the mask pattern data “0” is written in the other child matrix pattern writing section. Since it is stored in memory, the LC line (LC is the content of the counter LC) (1 byte) of the previously written pattern data is read from the memory 1020 and stored in the buffer memory MBUF (internal register). The data in the buffer memory MBUF and the data in the buffer memory BUF are given to the OR gate LG ₂ to perform a logical sum, the logical sum data is updated and stored in the buffer memory BUF (14), and the data in the buffer memory BUF is updated and stored in the page memory 1020 (15). .

Next, the microprocessor 1010 increments the line counter LC by one (16), and the contents of the line counter LC and the number of lines of the child matrix pattern (17), and the content of the line counter LC is equal to the number of lines. If not, the process proceeds to the next line of image extraction, but if it does, the counter H is incremented by one (18), and the contents of the counter H are compared with the contents of the register L (19). If the former is greater than the latter, image extraction has been completed for the child matrix pattern located at the leftmost position in the main scanning direction in the mother matrix pattern,
To advance the next process to the leftmost child matrix pattern, the content of the counter H is set to 1 (20), and the process proceeds to the next data reading (4).

When the data read in the data read (4) indicates the end of halftone processing, the microprocessor
1010 returns to the main routine. When the data is line feed "LF", the counter V is incremented by one (21), and the contents of the counter V are compared with the contents of the register L (22). If the former is larger than the latter, it means that the image processing for one mother matrix pattern has been completed, so the counter V is set to 1 (23) and the process returns to the data reading (4). When the data is the carriage return “CR”, it means that the image processing for the width of one mother matrix pattern in the main scanning direction has been completed.
Is set (3), and the process proceeds to data reading (4).

In the above description, a mother matrix pattern (halftone expression pattern) of 64 types as one group is formed using the original pattern (FIG. 12c) having threshold data, and this is stored in the memory 1012 in advance. With reference to the mode, the memory 1012 stores the original pattern as a halftone expression pattern, compares the recording density data given from the memory 108y with each threshold of the original pattern, and A mother matrix pattern may be created, or recording density data may be compared with a partial threshold value of the original pattern to obtain recording / non-recording bit information, and recording may be performed based on this. Alternatively, prior to gradation processing, the original pattern and each of the recording density gradation data (representing 1 to 64) are compared to create one group (64) of mother matrix patterns, which is stored in the RAM. It may be stored in a memory such as. In this way, the data in the memory 1012 can be reduced.

With the same hardware configuration and control operation as the gradation processing circuit 109y described above, the gradation processing circuits 109m, 109c and 109bk
Generates recording image data of magenta M, cyan C and black BK, respectively. In these, the halftone dot centers of the mother matrix (halftone expression pattern) stored in the pattern memory are shown in FIGS. 12a to 12c and FIG. 11a (BK), respectively.
The difference is that the positions are different based on the original pattern shown in FIG.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of a mechanical structure according to an embodiment of the present invention. FIG. 2 is a block diagram showing the configuration of the electric system of the embodiment. FIG. 3 is a block diagram showing a configuration of the gradation processing circuit 109y shown in FIG. FIGS. 4a and 4b are plan views showing mask patterns used for extracting a child matrix pattern. 5a, 5b, 5c, 5d and 5e show the number of modes in which the halftone expression pattern is divided into child patterns (partial patterns) A, B, C,. It is a top view which shows a seed. 6a and 6b are flow charts showing the recorded image data processing operation of the microprocessor 1010 shown in FIG. 7a and 7b are plan views showing the distribution of the recording density data corresponding to the recording surface, and the numbers in the figures indicate the densities (decimal numbers) indicated by the recording density data. 8a and 8b respectively correspond to the recording density data distributions shown in FIGS. 7a and 7b, respectively, and extract the recording data from the halftone expression pattern by dividing FIGS. 5c and 5d, respectively. FIG. 4 is a plan view showing a print data distribution when assigned to a print surface. 9a and 9b are plan views showing recording information distributions (shaded areas) when recording data is extracted from the halftone expression pattern shown in FIG. 10 in the mode shown in FIGS. 8a and 8b, respectively. It is. FIG. 10 is a plan view showing a group of halftone expression patterns created based on original data different from the original data shown in FIGS. 14a to 14d. 11a, 11b, 11c, and 11d are plan views showing basic pattern examples of halftone expression patterns used in the present invention, and the hatched lines in the drawings indicate that the recording density data is 16. Indicates the position to which the recording information bit is assigned. Figures 12a, 12b and 12c show Figures 11b and 11
FIG. 12 is a plan view showing a halftone expression pattern assigned to each recording color based on the basic pattern shown in FIG.
FIG. d is a plan view showing the color distribution of the surface recorded based on these patterns when the recording density data for each color is shown by 16. 13a, 13b, and 13c are plan views showing other examples of the halftone expression pattern assigned to each recording color. FIGS. 14a, 14b and 14c are plan views showing still another example of the halftone expression pattern assigned to each recording color, and FIG. 14d shows that each color recording density data shows 20. FIG. 6 is a plan view showing a color distribution of a surface recorded based on these patterns when 1: Original, 2: Platen 3 ₁ , 3 ₂ : Fluorescent lamp, 4 ₁ to 4 ₃ : Mirror 5: Variable lens unit 6: Dichroic prism 7r, 7g, 7b: CCD, 8: 1st carriage 9: 2nd Carriage 10: Carriage drive motor 11: Pulley, 12: Wire (1 to 12: Color image reading means) 13bk, 13y, 13m, 13c: Polyhedral mirror 14bk, 14y, 14m, 14c: f-θ lens 15bk, 15y, 15m , 15c, 16bk, 16y, 16m, 16c: Mirror 17bk, 17y, 17m, 17c: Cylindrical lens 18bk, 18y, 18m, 18c: Photosensitive drum 19bk, 19y, 19m, 19c: Charge scorotron 20bk, 20y, 20m, 20c: Developing device 21bk, 21y, 21m, 21c: Cleaner 22: Paper feeding cassette, 23: Paper feeding roller 24: Registration roller, 25: Transfer belt 26, 28, 30: Idle roller 27: Driving roller 29bk, 29y, 29m , 29c: Transfer corotron 31: Lever, 32: Axis 33: Pin, 34: Compression coil spring 35: Black copy mode setting solenoid plunger 36: Fixer, 37: Tray (13 to 37, 41 to 46, 112: Recording means ) 39: Home Position Sensor 40: carriage guide bar 41bk, 41y, 41m, 41c: polygon mirror drive motor 42: toner collection pipe 43bk, 43y, 43m, 43c: laser 44bk, 44y, 44m, 44c: beam sensor 45: photoconductor drum drive motor 46: Motor driver 100: Image processing unit 104y, 104m, 104c: Digital comparator 104sh: Rotary dip switch (101 to 107: Component data processing means) 109: Gradation processing circuit, 109y: Y gradation processing circuit 109m: M gradation processing circuit, 109c: C gradation processing circuit 109bk: BK gradation processing circuit 1012: Pattern memory 1010: Microprocessor (pattern information reading means) 200: Microprocessor system 300: Console 301: Copy start key switch 302: Full color / single color black mode selection key switch

Claims (5)

(57) [Claims] 1. A color image is color-separated into a plurality of colors, image density is converted into digital data for each color component, and the digital data is processed into color component recording density data; Halftone expression pattern having a plurality of threshold data corresponding to one of the threshold data of a predetermined number of dots, or all of the threshold data is compared with each value of the recording density data in the planned range, and the predetermined number of dots is supported. The color component recording density data is converted into recording and non-recording bit information using a halftone expression pattern consisting of a number of bit distribution patterns corresponding to the range of recording density data, which is the recording and non-recording bit distribution. Recording / non-recording bit information is associated with one dot of the recording medium for each color component, and a predetermined color is recorded in the dot with which the recording information bit is associated; in digital color image reproduction processing: The expression pattern used for the first color recording and the second color recording is such that when recording a predetermined area using it, the recording information bits correspond to the predetermined points of the X and Y coordinates as the recording density increases in correspondence with the recording density. The recording information bit distribution spreads from the predetermined point, and the predetermined points are different from each other for each color, and the halftone expression pattern used for the third color recording has the recording information bit distribution spreading from the predetermined point. 1 set corresponds to each color component, which is different from the halftone expression pattern of the first color record and the second color record in which the predetermined point of is dispersed between the predetermined points of the first color record and the second color record. When the converted color component recording density data is converted into recorded and non-recorded bit information, the attached halftone expression pattern for each color is recorded from the halftone expression pattern specified by the color component recording density data. Concentration data Based on the coordinates of the pixels of the obtained original image, a region corresponding to a plurality of dots of a part of the halftone expression pattern is regularly cut out as a child matrix pattern corresponding to the pixels of the original image, and a recording obtained from the child matrix pattern, A digital color image reproduction processing method, wherein non-recording bit information is output as image information corresponding to the color component recording density data. 2. The halftone expression pattern has a size M.N.
The digital color image reproduction processing method according to claim (1), which represents a gradation of a maximum dot number M · N determined by 3. The halftone expression pattern used for the first color recording and the halftone expression pattern used for the second color recording are substantially the same pattern when many of them are surface-developed. The threshold data or the recording or non-recording data is distributed such that the predetermined point of the second color record and the predetermined point of the second color record are distributed at positions separated by the maximum distance. The digital color image reproduction processing method according to the item 1). 4. A halftone expression pattern MMP is divided into m × n child matrix patterns CMP _{11 to} CMP _mn with m in the main scanning direction and n in the sub-scanning direction, and the beginning of the letter is MMP
, The position of the child matrix pattern in the main scanning direction, and the latter half of the subscript indicate the position in the sub-scanning direction. And m × n pieces of recording density data, which similarly consist of ICD ₁₁ to ICD _mn In assuming that obtain one image information of the halftone representation pattern content, records information of the child matrix pattern CMP _ij halftone expression pattern identified in the recording density data ICD _ij bit distribution for the recording density data ICD _ij The digital color image reproduction processing method according to claim (1), (2) or (3), which is obtained as information. 5. A color image reading means for separating a color image into a plurality of colors and converting the image density into digital data for each color component; a color component data processing means for processing the digital data into color component recording density data. When a predetermined area is recorded using it, the recording information bits have a recording information bit distribution that spreads from a predetermined point of the X and Y coordinates as the recording density increases, and the predetermined points are different for each color. The halftone expression patterns of the first color group and the second color recording group at different positions, and the recording information bit distribution extending from a predetermined point, and a plurality of predetermined points are the first color record and the second color record. A memory hand storing a halftone expression pattern of the third color recording group, which is different from the halftone expression pattern of the first color recording and the second color recording, which is dispersed between predetermined points of the color recording. A group is specified from the memory means corresponding to a color component, a set of halftone expression pattern information in the group is specified based on the color component recording density data, and then the color component recording density data is specified. Based on the coordinates of the pixels of the obtained original image, a region corresponding to a plurality of dots of a part of the halftone expression pattern is regularly cut out as a child matrix pattern corresponding to the pixels of the original image, and a recording obtained from the child matrix pattern, Pattern information reading means for outputting non-recording bit information; and recording means for recording a predetermined color on a recording medium at a position to which the recording information bit should be allocated, in response to the output recording bit information. Digital color image reproduction device.