JPH04342367A - Image forming device - Google Patents

Abstract

PURPOSE:To provide the image forming device making the improvement of clearness for achromatic color be compatible with the improvement of saturation for chromatic color when reproducing full color images. CONSTITUTION:It is stepwise judged according to level difference among three colors from digital data in red, green and blue obtained by scanning an original whether the color is achromatic or not and corresponding to the judged result, first and second optimum coefficients are selected out of the plural first and second coefficients stored in advance. The minimum value of the above- mentioned digital data is calculated, the product of the minimum value and the first coefficient is prepared as black data, and the product of the minimum value and the second coefficient is subtracted from the above-mentioned digital data. Based on the calculated result obtained in this way, the full color images are reproduced.

Description

[Detailed description of the invention]

0001

[Industrial Application Field] The present invention is an image reproducing device such as a copying machine or a printer that converts read data of red, green, and blue into reproduced color data of cyan, magenta, and yellow to form a multicolor image. Regarding.

0002

[Prior Art] In printers that reproduce images in full color, digital image data R, G, B of red, green, and blue read from a document is converted into three colors of color cyan, magenta, yellow C, M, and Y. Convert to data and reproduce the image. For this reason, the red, green, and
Performs data processing to convert three-color blue digital data into three-color data for image reproduction.

0003

[Problems to be Solved by the Invention] In the data processing of full-color images, consideration must be given to achieving both black vividness and chromaticity. In other words, in the reproduction of black in a full-color image, cyan C,
Even if black is reproduced by overlapping magenta M and yellow Y, it is difficult to reproduce clear black due to the influence of the spectral characteristics of each toner. Therefore, black reproducibility is improved by subtractive color mixing using reproduction color data Y, M, and C and black printing using black data K, and full color is realized. However, with this method, the sharpness of black improves as the degree of inking increases, but the saturation of chromatic colors decreases. Therefore, it is necessary to simultaneously improve the clarity of achromatic colors and the saturation of chromatic colors.

[0004] Therefore, it is conceivable to determine whether or not to perform black overprinting by determining whether the color is achromatic or chromatic. but,
If it is determined whether an achromatic color or a chromatic color is simply based on the read colors of red, green, and blue, there will be many erroneous determinations and the image will deteriorate significantly. The causes include halftone originals and moiré in the reading system, errors and noise in the reading system in areas where the original density is uniform, hue, and brightness gradually change.

[0005] An object of the present invention is to provide an image forming apparatus that can improve both the sharpness of achromatic colors and the saturation of chromatic colors in a full-color image.

[0006]

[Means for Solving the Problems] A first image forming apparatus according to the present invention includes an input means for inputting digital data of red, green, and blue reading colors obtained by scanning a document; A black generation means that calculates the minimum value of the digital data of the inputted reading color, and creates the product of the minimum value found by the black generation means and the first coefficient as black data, and then calculates the above from the digital data of this reading color. The under color removal black overprinting means subtracts the product of the minimum value of and the second coefficient, and the read color data processed by the undercolor removal black overprinting means is converted into cyan, magenta,
A data conversion means converts the digital data of the read colors of red, green, and blue inputted by the input means for each pixel into the reproduced color data of yellow, and determines whether or not it is an achromatic color based on the size of the level difference between the three colors. and an achromatic color determination means for determining whether the
The coefficient and the second coefficient are memorized, and the optimal first and second coefficients are selected according to the result of the determination by the achromatic color determination means, and the above-described method is performed using both of the selected coefficients. It is characterized by performing calculations. The second image forming apparatus according to the present invention further includes a smoothing unit that performs two-dimensional spatial digital filter processing for each color on the input digital data of the read color of the pixel of interest, and the data converting unit described above. is characterized in that the determination is made on digital data that has been subjected to spatial digital filter processing using a smoothing means. The third image forming apparatus according to the present invention further includes a monochrome mode, and the data conversion means sets both the first coefficient and the second coefficient for data processing to 0 in the monochrome mode. do.

[0007]

[Operation] The black generation means determines the minimum value of the digital data of the read color inputted by the input means. The achromatic color determining means performs stepwise determination as to whether the read data R, G, and B are achromatic or chromatic in a local area including the pixel of interest. Specifically, for example, one of the three read colors is used as a reference, and the level difference between the reference color data and the data of the other two colors is made to correspond to a plurality of levels. The undercolor removal ink reprinting means is
A first coefficient and a second coefficient optimally determined for each stage are stored. Then, the first coefficient and the second coefficient are selected according to the determination result (stage), and the product of the minimum value obtained by the black generation means and the first coefficient is created as black data (
The product of the minimum value and the second coefficient is subtracted from the digital data of the read color (undercolor removal), and the data conversion means converts the thus processed data into cyan, cyan,
Convert to magenta and yellow reproduction color data. This process is performed for each pixel. Furthermore, by performing smoothing (noise removal) on the R, G, and B read data by two-dimensional spatial filter processing (superimposed moving average processing in the example), achromatic color discrimination is performed, and each area is Improve judgment accuracy. Note that in monochrome mode, the first coefficient is
Both coefficients are set to 0, black data is not generated, and undercolor removal is not performed.

[0008]

DESCRIPTION OF THE PREFERRED EMBODIMENTS A digital color copying machine according to an embodiment of the present invention will be described in the following order with reference to the accompanying drawings. (a) Configuration of digital color copying machine (b) Image signal processing (b-1) Configuration of image signal processing section (b-2) Area determination results and outline of image signal processing (c)
Density conversion unit (d) Black generation unit (e) Undercolor removal/black printing automatic control in area discrimination unit (
Achromatic color chromatic color determination) (e-1) Purpose of undercolor removal/black printing automatic control (e-2)
Smoothing processing (e-3) Achromatic color chromatic color determination (f) Automatic masking control (hue determination) in area determination section (f-1) Purpose of automatic masking control (f-2) Hue determination (g) Color correction processing section (h) Edge enhancement/smoothing automatic control in area discrimination section (edge judgment) (h-1) Purpose of edge enhancement/smoothing automatic control (
n-2) Edge detection (i) MTF correction section

(a) Structure of digital color copying machine FIG. 1 is a longitudinal sectional view showing the overall structure of a digital color copying machine according to an embodiment of the present invention. A digital color copying machine is roughly divided into an image reader section 100 that reads a document image, and a main body section 200 that reproduces the image read by the image reader section.

In FIG. 1, a scanner 10 includes an exposure lamp 12 that irradiates an original, a rod lens array 13 that collects reflected light from the original, and a contact type CCD that converts the collected light into an electrical signal. It is equipped with a color image sensor 14. The scanner 10 uses a motor 1 when reading a document.
1 to move in the direction of the arrow (sub-scanning direction) and scan the original placed on the platen 15. The image of the document surface illuminated by the exposure lamp 12 is photoelectrically converted by the image sensor 14 . The three-color multi-value electrical signals of R, G, and B obtained by the image sensor 14 are processed by the read signal processing unit 20 to determine which of yellow (Y), magenta (M), cyan (C), and black (K). It is converted to gradation data. Next, the print head unit 31 performs correction (γ correction) on the input gradation data according to the gradation characteristics of the photoreceptor and dither processing as necessary, and then outputs the corrected image data. A laser diode drive signal is generated by D/A conversion, and the laser diode 221 (not shown) in the print head unit 31 is driven by this drive signal.
drive.

[0011] The laser diode 2 corresponds to the gradation data.
The laser beam generated from the laser beam 21 exposes the photosensitive drum 41, which is driven to rotate, through a reflecting mirror 37, as shown by the dashed line in FIG. As a result, an image of the document is formed on the photoreceptor of the photoreceptor drum 41. Photosensitive drum 41
The eraser lamp 42 is used before each copy is exposed to light.
, and is charged by a charger 43. When exposed to light in this uniformly charged state, an electrostatic latent image is formed on the photosensitive drum 41. Only one of the yellow, magenta, cyan, and black toner developers 45a to 45d is selected to develop the electrostatic latent image on the photoreceptor drum 41. The developed image is transferred by a transfer charger 46 to copy paper wound around a transfer drum 51.

The above printing process is repeated for yellow, magenta, cyan and black. At this time, the scanner 10 repeats the scanning operation in synchronization with the operations of the photosensitive drum 41 and the transfer drum 51. Thereafter, the copy paper is separated from the transfer drum 51 by operating the separation claw 47, passed through the fixing device 48, fixed, and discharged onto the paper discharge tray 49. The copy paper is fed from a paper cassette 50, and its leading edge is chucked by a chucking mechanism 52 on a transfer drum 51 to prevent positional deviation during transfer.

FIGS. 2 and 3 show general block diagrams of the control system of the digital color copying machine. Image reader section 100
is controlled by the image reader control unit 101. Image reader control unit 101 controls exposure lamp 12 via drive I/0103 based on a position signal from position detection switch 102 indicating the position of the document on platen 15, and also controls drum I/O 103 and parallel I/O. O1
04 to control the scan motor driver 105. The scan motor 11 is a scan motor driver 105
Driven by.

On the other hand, the image reader control section 101 is connected to the image control section 106 via a bus. Image control unit 1
06 is connected to each of the CCD color image sensor 14 and the image signal processing section 20 via a bus. The image signal from the CCD color image sensor 14 is
The signal is input to and processed by an image signal processing section 20, which will be described later.

The main body section 200 is provided with a printer control section 201 that controls copying operations in general, and a print head control section 202 that controls the print head section 31. Analog signals from various sensors 44, 60, 203 to 205 for automatic density control are input to the printer control unit 201. Further, by key input to the operation panel 206, the printer controller 2
Various data are input to 01. Printer control unit 201
is a control ROM 208 in which a control program is stored.
and a data ROM 209 storing various data are connected. The printer control unit 201 includes various sensors 44 and 6.
0, 203-205, operation panel 206 and data R
Data from the OM 209 controls the copy control unit 210 and the display panel 211 according to the contents of the control ROM 208, and furthermore, in order to perform automatic density compensation control, the grid of the charger is controlled via the parallel I/O 212 and the drive I/O 213. High voltage unit 214 for voltage generation
and controls the high voltage unit 215 for generating bias voltage for the developer.

The print head control section 202 has a control RO
It operates according to the control program stored in the M216, and is connected to the image signal processing section 20 of the image reader section 100 via an image data bus, and operates based on the image signal input via the image data bus. Then, the γ correction is performed by referring to the contents of the data ROM 217 in which the γ correction conversion table is stored, and furthermore, when the multilevel dither method is used as the gradation expression method, dither processing is performed, and the drive I/ The laser diode driver 220 is controlled via O218 and parallel I/O219. The laser diode 221 is a laser diode driver 22
0 controls its emission. Further, the print head control section 202 is connected to the printer control section 201, the image signal processing section 20, and the image reader control section 101 via a bus so that they are synchronized with each other.

(b) Image Signal Processing (b-1) Configuration of Image Signal Processing Section FIG. 4 shows a perspective view of the reading section. In the reading section, 3 wavelengths (
A light source (halogen lamp) 12 with a spectral distribution of R, G, B) irradiates the surface of the original 91, and the reflected light is transmitted to the light receiving surface of a contact type CCD color image sensor 14 by a rod lens array 13. Forms a line image at the same magnification. Rod lens array 13, light source 12 and CC
The optical system including the D color image sensor 14 performs line scanning in the direction of the arrow in FIG.

FIG. 5 is a block diagram of the image signal processing unit 20, and with reference to this figure, processing of image signals from the CCD color image sensor 14 to the print head control unit 202 via the image signal processing unit 20 is explained. Explain the flow. In the image signal processing unit 20, the image signal photoelectrically converted by the CCD color image sensor 14 is converted into an A/D
A converter 61 converts it into R, G, B multivalued digital image data. Note that the clock generator 70 generates a clock and transmits the clock to the CCD sensor 14 and the A/D converter 61. The converted image data is subjected to a predetermined shading correction in the shading correction unit 62, and since it is reflection data of the original 91, it is converted into density data of the actual image by log conversion in the density conversion unit 63. . Further, the black generation section 64 generates true black data K' from R, G, and B data.

On the other hand, for the shading-corrected image data, the region discrimination unit 65 extracts image features for each pixel in the local region to which the pixel belongs, and distinguishes between the flat density portions, edge portions, and intermediate portions of the image. A separation will be made. FIG. 6 shows a circuit configuration of the area determination section 65. After the edges of the image data R, G, and B output from the shading correction section 62 are detected by the edge detection section 84, the MTF correction control section 85 sends 2-bit filter selection signals FS1 and FS0 to the MTF correction section 67. send. On the other hand, the image data R, G, and B are subjected to filter processing in the smoothing processing unit 81 and then subjected to undercolor removal/blackening (U
CR/BP) control section 82 and color correction masking control section 83
2-bit (4 steps) achromatic and chromatic color determination signals US1, US0 and US2 are processed in the color correction processing section area 66.
Bit (4 types) masking coefficient selection signals MS1, M
Output S0.

The color correction processing section 66 simultaneously performs black data generation processing and masking processing in response to the achromatic and chromatic color determination signals US1 and US0 and the masking coefficient selection signals MS1 and MS0 from the area discriminating section 65. That is, black data is generated and subtracted from the read data, and the read data is converted into three reproduced color data of C, M, and Y. Furthermore, in the MTF correction section 67, a digital filter is selected in response to the filter selection signals FS1 and FS0 from the area determination section 65, and smoothing processing or edge enhancement processing is performed. Next, variable magnification
The moving unit 68 changes the magnification. Further, the color balance unit 69 adjusts the color balance and outputs data to the print head control unit 202. In addition,
The print head control unit 202 sends a signal M/NC indicating whether the mode is monochrome mode or full color mode to the image signal processing unit 20 .

In FIG. 7, the register 87 connects data buses (D7 to D0) from the print head control section 202,
Various control parameters are set using address buses (MA2 to MA0), chip selection signal NCS1, and write signal NWR. And the control parameter REF0
, REF1, REF2 are sent to the undercolor removal/blackening control section 82, control parameters REF3, REF4 are sent to the MTF correction control section 85, and control parameter EDG is sent to the MTF correction section 67.

(b-2) Area Judgment Results and Outline of Image Signal Processing Each process of image signal processing will be explained in detail below, but before that, an outline of area discrimination and the corresponding processing will be explained. In this embodiment, the area determination unit 65 extracts the following three features from the R, G, and B read data for each pixel in a local area centered on that pixel. (a) Achromatic color or chromatic color (achromatic color chromatic color determination signal U
S) (see section (e-3)), (b) edge detection (filter selection signal FS) (see (h-2)), (c) hue judgment (masking coefficient selection signal MS) ((f-2) (see section)
. Then, in accordance with whether the color is an achromatic color or a chromatic color and the determination result of edge detection, the color correction processing unit 66 optimizes the amount of black (UCR ratio/BP ratio) and performs MTF correction. In section 67, smoothing or edge enhancement is performed by spatial digital filter processing. (1) When the color is achromatic and the density is flat: In addition to increasing the black amount according to the result of the achromatic and chromatic color determination,
Smooth the reproduced color M, C, and Y data. (2) When the edge portion is either achromatic or chromatic: The results of the achromatic and chromatic color determination are canceled, the amount of black is reduced, and edges are emphasized for C, M, and Y, which are the reproduced colors. (3) When the density is flat for chromatic colors: In addition to reducing the amount of black in accordance with the result of achromatic color chromatic color determination,
, C, Y data is smoothed. (4) In other cases, the amount of black is set to intermediate, and neither edge enhancement nor data smoothing is performed. In addition, linear masking coefficients corresponding to each of a plurality of color groups are prepared in the color correction processing unit 66, and according to the result of hue determination, masking processing is performed using the linear masking coefficient of the color group to which the hue belongs. conduct. Note that in the monochrome mode, the amount of black is set to 0, and a linear masking coefficient for the monochrome mode is used.

(c) Density Conversion Unit The density conversion unit 63 converts the output data of the CCD color image sensor 14 into an original density (OD) as seen from the human eye.
) to have linear characteristics. C.C.
The output of the D color image sensor 14 has photoelectric conversion characteristics that are linear with respect to the incident intensity (=original reflectance OR). On the other hand, the relationship between the original reflectance (OR) and the original density (OD) is -logOR=OD. Therefore, the nonlinear reading characteristics of the CCD color image sensor 14 are converted into linear characteristics using a reflectance/density conversion table. Specifically, reflectance/density conversion table 3 in FIG.
47, the R, G, B read data of the pixel of interest is converted into density data DR, DG, DB.

(d) Black generation part: Cyan, magenta, yellow, and other colors necessary for full color reproduction.
Each black color data C', M', Y', and K' is created for each scan using the field sequential method, and full color is reproduced by a total of four scans. The reason why black is also printed is that even if cyan, magenta, and yellow are superimposed to reproduce black, it is difficult to reproduce clear black due to the influence of the spectral characteristics of each toner. Therefore, in this full-color copying machine, black reproducibility is improved by subtractive color mixing using data Y', M', and C' and black overprinting using black data K'.
Achieve full color.

The black generation section 64 calculates the black amount K from the red, green, and blue components R, G, and B representing the brightness on the original as follows. Since DR, DG, and DB obtained from the density converter 63 are each density data of R, G, and B components, the components C of cyan, magenta, and yellow, which are complementary colors of R, G, and B, in the CCD reading section ', M', Y'. Therefore, as shown in FIG. 8, the minimum values of DR, DG, and DB may be taken as black data K' since C', M', and Y' on the document are components in which colors are superimposed. Therefore, the black generation section 6
In step 4, black data K'=MIN (DR, DG, DB) is detected.

As will be explained later, when the color correction processing section 66 creates the reproduced color data C, M, Y, the data K' from the black generation section 64 is used to create C', M'.
, Y' to create black data K, β·K' is output as the K amount. Here, α is the UCR (undercolor removal) ratio, which will be explained later, and β
is the BP ratio. Note that in order to improve black reproducibility, the parameters α and β are changed in four stages according to the characteristics of the local area including the pixel of interest after the smoothing process.

Next, the circuit of the black generating section 64 shown in FIG. 9 will be explained. Comparator 301 compares red data DR and green data DG, and two-input multiplexer 302 outputs the smaller value of DR and DG to comparator 303 based on the comparison result. This comparator 303 is
This output value is compared with the blue data DB, and the 2-input multiplexer 304 outputs DR, DG, and DB based on the comparison result.
The minimum value of is output to the two-input multiplexer 305. 2
Based on the signal M/NC, the input multiplexer 305
Output this minimum value (full color mode) or 0 (single color mode). That is, the signal M/NC determines the image output mode, and at "L" level, it is full color mode and outputs the above minimum value, but at "H" level,
This is a monochrome mode, and K' may be set to an appropriate value that optimizes the actual output result. In this embodiment, K' is fixed to 0, and data K' for blackening is cleared. Delay circuits 306, 307, and 308 are used to adjust timing. Note that the output data is read data R,
Although the display has been changed to C', M', and Y' in the sense that G and B are data of complementary colors cyan, magenta, and yellow, they essentially show the same data.

(e) Undercolor removal/black printing automatic control in area discrimination unit (achromatic and chromatic color determination) The area discrimination unit 65 performs undercolor removal/black printing automatic control, automatic masking control, edge emphasis/smoothing. Area discrimination processing is performed for automatic control, and correction parameters are output. Undercolor removal/blackening automatic control will be described here.

(e-1) As explained in the destination of undercolor removal/black printing automatic control (achromatic and chromatic color determination), in the black generation section 64, the black data K'
Then, K'=MIN(DR, DG, DB) is detected. Then, in the color correction processing section 66, C', M', Y'
When creating data K by subtracting α・K' from β
・Output K' as K amount. Here, α is the UCR ratio and determines the black amount. β is the BP ratio and lowers the color data. The UCR ratio/BP ratio has an influence on the saturation of color reproduction and the sharpness of achromatic colors.

[0030] The reproducibility of achromatic colors is determined by the UCR ratio/BP ratio (-
If α/β) are respectively increased, pure black K' is reproduced, improving the color saturation, but on the other hand, the saturation of chromatic colors decreases because the output ratio of K' increases. Therefore, it is considered that by controlling the UCR ratio/BP ratio by determining whether or not an achromatic color is present, it is possible to simultaneously improve the sharpness of an achromatic color and the saturation of a chromatic color.

However, simply determining whether the read data R, G, B (primary colors) is achromatic or not is likely to result in erroneous determination, which may conversely lead to image deterioration. The causes of this misjudgment include misjudgment due to moiré of halftone dot documents in the reading system, and misjudgment due to errors and noise in the reading system in documents with uniform density or portions where the hue or brightness changes gradually.

Therefore, first, the read data R of the pixel of interest,
For G and B, smoothing processing (weighted moving average processing in the embodiment) is performed on a local area including the pixels. Next, we decided to compare the levels of R, G, and B to determine whether the color is achromatic. Furthermore, to improve black reproducibility,
The parameters α and β were changed in four stages (see FIG. 11) according to the characteristics of the region including the pixel of interest after the smoothing process, and the closer to the achromatic color the higher the UCR ratio/BP ratio.

(e-2) Smoothing process For under color removal/black printing automatic control, first, 5× centering on the pixel of interest
Smoothing processing is performed on a region of 5=25 pixels. In the smoothing processing unit 81 shown in FIG. 10, the 8-bit (level 0 to 255) R, G
, B data is used to perform a moving average of weighted addition for the pixel of interest for each color using a 5×5 filter 344. That is, first, four line memories 340, 341, 34
Four lines of data are sequentially stored in 2 and 343. Then, these 4 lines of data R2 (G2, B2), R3 (
G3, B3), R4 (G4, B4), R5 (G5, B5
) and the data R1 (G1, B1) of the line currently being output, the center pixel of the center line is smoothed by the filter 344 for smoothing processing, and then the smoothed data RS (GS, BS) is obtained. Undercolor removal/
The data is sent to the ink printing control section 82 and the color correction masking control section 83. In the filter 344, weighting is performed gently around the pixel of interest.

Smoothing processing using this filter 344 makes it possible to prevent false resolution of high frequencies of pixels and to extract low frequency components. This results in
For areas to be determined, (a) noise removal from images with uniform density, (b) moiré removal caused by the halftone image reading system, and (c) gradual changes in hue, brightness, and saturation. This has the effect of smoothing the image and improves discrimination accuracy. The data RS (GS, BS) smoothed in this way is
It is used for hue determination (automatic masking control) in the undercolor removal/blackening control unit 82, and is used in the color correction masking control unit 8.
Achromatic and chromatic color determination in 3 (UCR/BP automatic control)
used for.

(e-3) Achromatic color chromatic color determination The under color removal/black printing control unit 82 determines achromatic color (black) by utilizing the fact that the three-color read data satisfies R≒G≒B. I'm doing it. FIG. 11 shows a distribution area diagram for this determination. This determination is performed for each pixel. In FIG. 11, the smoothed data RS, GS, and BS of each pixel are in a range KLVL0 (GS-REF0<RS,
BS<GS-REF0), range KLVL1 (GS-REF1<RS, BS<GS-
REF1), the range KLVL2 (
GS-REF2<RS, BS<GS-REF2), it is classified into four distribution areas from achromatic to chromatic as shown in the selection table in Table 1, and achromatic and chromatic color determination is performed accordingly. Signals US1 and US0 are generated. Here, if it is within the area indicated by KLVL0, it is determined to be an achromatic color, and if it is outside the area indicated by KLVL2, it is determined to be a chromatic color, but two areas are provided in between. . Then, 2-bit achromatic and chromatic color determination signals US1, U
Corresponding to the values 0, 1, 2, 3 represented by S0, the UCR ratio/
The value of BP ratio (-α/β) is changed stepwise. This ratio is set to a higher value as the color becomes more achromatic.

[0036]

[Table 1]

Undercolor removal/black printing control section 82 shown in FIG.
In this circuit, it is determined to which distribution region in FIG. 11 the data RS, GS, and BS of each pixel belong, and an achromatic color is determined. In this determination, a certain constant value is added or subtracted from the reference color data (GS) to determine the level difference between it and the other two colors. That is, for the GS data, the subtracter 36
1 and REF0 are subtracted and added in an adder 362 to obtain GS-REF0 and GS+REF0. and,
Four of the 12 comparators 367 compare these with RS, GS, and BS. Then, the result is passed through the NAND gate 369 to G
N if S-REF0<RS, BS<GS+REF0
KLVL0="L" is output to the table (ROM) 372. Similarly, subtracter 363 and adder 364 perform subtraction and addition with REF1 for GS data, and
Obtain S-REF1 and GS+REF1. Then, four comparators among the comparators 367 compare these RS and BS. And the result is N
Through AND gate 370, GS-REF1<RS,
When BS<GS+REF1, NKLVL1="L" is output to the table 372. Furthermore, subtractor 365 and adder 366 perform subtraction and addition with REF2 on the GS data to obtain GS-REF2 and GS+REF2. Then, four comparators among the comparators 367 compare these with RS and BS. Then, the result is passed through the NAND gate 371 to the G
N if S-REF2<RS, BS<GS+REF2
KLVL2="L" is output to the table 372. In table 372, according to the selection table of Table 1,
Outputs 2-bit achromatic and chromatic color determination signals US1 and NS0.

Here, in the table 372, MT
Filter selection signal FS input from F correction control section 85
When 0 is "L" (edge portion), the above-mentioned determination is canceled. That is, achromatic and chromatic color determination signals US1 and US0="3" are output that reduce the amount of black. The filter selection signal FS0 is set so that the edge detection amount in either the main scanning direction or the sub-scanning direction of R, G, B data is at a constant level (REF3
) or more. In this case, since it is an edge part of the image, it is thought that achromatic/chromatic color determination is likely to be incorrect, so for pixels determined to be at the edge part of the image, achromatic/chromatic color determination is performed in advance. Set it so that no processing is performed to reduce the rate of false positives. In other words, since the undercolor removal/blackening control is optimized only in areas where the image density is relatively uniform, image deterioration due to misjudgment is eliminated.

As explained above, the undercolor removal/blackening control section 82 and the color correction masking control section 83 shown in FIG.
Now, various selection signals (US1
, US0, MS1, MS0, FS1, FS0) and outputs them to the color correction processing section 66 and the MTF correction section 67.

(f) Automatic masking control (hue determination) in area determination section (f-1) Purpose of automatic masking control In order to reproduce full-color input data into an image, masking calculation is performed in the MTF correction section 37. The linear masking coefficient is set so that the average color difference is minimized over almost the entire color gamut, but in some parts of the gamut, the color difference is not necessarily minimized and color reproduction errors and gradation errors occur. It gets bigger. Therefore, DR, DG,
DB term has its square term and DR・DG, DG・DB, D
Although it is said that secondary masking processing in which B/DR terms are added is better, the circuit configuration becomes complicated and large-scale. Therefore, in this embodiment, linear masking processing is used, but the color correction processing unit 66 calculates a plurality of linear masking coefficients corresponding to four groups (primary color group, complementary color group, and two intermediate groups) to which each hue belongs. The color correction masking control unit 83 determines the hue of the image data, selects a masking coefficient that minimizes the color difference within that hue according to the hue, and achieves color reproduction equivalent to secondary masking processing. Get sex. In addition, in the masking process as well, in the case of achromatic and chromatic color determination, the data R that has been smoothed by the smoothing processing unit 81 in FIG. 10 is used.
S, GS, and BS are used to make false judgments less likely to occur.

(f-2) Hue Judgment The hue judgment in the color correction masking control section 83 is as shown in FIG.
As shown in FIG. 2, this is done using a color correction table (ROM) 368. Here, three high-order bits each of RS, GS, and BS data are input to addresses A8 to A0 of the color correction table 368, and correspondingly, 2-bit masking coefficient selection signals MS1 and MS0 are output. .

[0042] Hue is R, G, B, W system (primary color system),
They are classified into two groups: C, M, Y, and BK (complementary colors), and an intermediate group. Here, consider a regular cube whose coordinate axes are each RS, GS, and BS data as shown in FIG. Each vertex of the cube is cyan (C), green (G), blue (B)
, represents the pure component of white (W). Therefore, R, G, B,
The W-based group is located within four small regular cubes containing vertices (R), (G), (B), and (W). C,M,
The same applies to Y and BK groups. One of the intermediate groups is located in the outline of a large cube surrounding the regular cube of the R, G, B, W system group, and the other intermediate group is located in the C
, M, Y, BK system group is located in the outer shape of a large cube surrounding the regular cube. As shown in the figure, (R), (M)
, (BK), (B) plane, the area (I) is C
, M, Y, BK group, and region (II) is C
, M, Y, BK system, and is an intermediate group close to the area (I
II) is an intermediate group close to the R, G, B, W group, and region (IV) is an R, G, B, W group. In this way, R, G, B masking and C, M
, Y-based masking because the hues of both sets (R, G, B) and (C, M, Y) are appropriately sparsely distributed. In other words, if appropriately sparsely distributed color samples are used, the masking coefficients described below will not take significantly different values. Therefore, the (R, G, B) system and (C, M,
Y) When determining the system, there is no major problem even if there is a misjudgment. Furthermore, the reason why the middle part is divided into two is to make it difficult to cause problems even if an erroneous determination is made.

In the color correction table 368, the input R
It is determined which group of the regular cube it belongs to according to the S, GS, and BS data, and when the determination result is (I), the MS
1,0="3" is output, and when (II), MS1,
0="2" is output, and when (III), MS1,0
="1" is output, and when (IV), MS1,0="
0” is output.

(g) Color correction processing unit The color correction processing unit 66 includes the CCD color image sensor 14.
The transmission characteristics of the filters R, G, and B in the printer section and the reflection characteristics of the toners C, M, and Y in the printer section are corrected to match the color reproducibility to characteristics close to ideal. Taking the G filter and M toner as an example, as shown in the transmission characteristics in Figure 14 and the reflection characteristics in Figure 15, the characteristics of the G filter and M toner are different from ideal characteristics as shown in the shaded area. There is a non-ideal wavelength range. In order to perform this correction, the color correction processing section 66 performs linear correction using the following masking equation in addition to the blackening process described above. (Note that since printing is performed in plane sequential fashion, this masking equation is executed line by line.

[0045]

[Math 1]

[0046] K'=MIN(DR, DG, DB)

00
47] In the circuit of the color correction processing unit 66 shown in FIG. 16, the parameter UCR ratio/
Black printing control and color correction masking processing are performed according to the BP ratio (-α/β). These various coefficients are set in the color correction register 826 shown in detail in FIG. 17 according to the nature of the area.

As shown in the circuit diagram of FIG. 17, the color correction register 826 stores masking coefficients (Ac
i, Bci, Cci, Ami, Bmi, Cmi, Ayi
, Byi, Cyi) (which is the element (M)ji (j=c,m,y;i=1,2,3) of the 3×3 matrix Mk) and the UCR ratio/BP ratio (−α/ β), outputs d. Here, the first masking coefficient M0 (k=0) selected when MS1,0="0" is the primary color R, G
, B is a coefficient that reduces the color difference, and MS1,0
="3", the fourth masking coefficient M3 (
k=3) reduces the color difference with respect to C, M, and Y, which are complementary colors. Furthermore, the second masking coefficient (k=1) selected when MS1,0="1" is (2/3) M0 + (1/3) which is weighted to the masking coefficient for primary colors.
Choose a masking coefficient that is M3 and also MS1,0
The third masking coefficient (k
= 2) is weighted to the complementary color masking coefficient (1
/3) Select a masking coefficient that is M0+(2/3)M3. The reason why the masking coefficient for the intermediate group is set by mixing the primary color masking coefficient and the complementary color masking coefficient is to make hue changes due to misjudgment less noticeable during color reproduction.

The following matrix shows the masking coefficients M0 for R, G, and B.

##EQU00002## Also, the following matrix shows the masking coefficient M3 for C, M, and Y.

##EQU00003## For comparison, conventional masking coefficients are shown below.

[Math 4]

That is, the color correction register 826 in FIG.
In the circuit diagram, a level 861 is a data bus (MD7 to MD0), an address bus (MA4 to MA0), and a chip selection signal N
Data is set by the CS0 and NWR signals. That is, data of the four types of matrix coefficients for each color described above are output to multiplexers 862, 863, and 864, respectively. Multiplexers 862, 863, and 864 select one type of matrix coefficient in response to selection signals MS1 and MS0 from color correction masking control section 83, and output the selected matrix coefficients to two-input multiplexers 865, 866, and 867. On the other hand, coefficients Dc, Dm, and Dy for monochrome mode are also input to the two-input multiplexers 865, 866, and 867. Multiplexers 865, 866, and 867 receive mode selection signals M/
One is selected by NC and output. On the other hand, four types of UC
The R ratio/BP ratio is determined by multiplexer 868 using the signal US
Selected by 1,0. Note that the constant d is used to increase the saturation at low density, but its explanation will be omitted here. In the above processing, d=0 is used to simplify the explanation.

To explain the circuit of the color correction processing unit 66 in FIG. 16, first, in the black reprinting unit, during the under color removal control for printing C, M, and Y, the multiplier 824 performs control from the black generation unit 64. The black data K' of is multiplied by the UCR ratio (-α) from the color correction register 826. and,
This multiplication value (-α・K') is added to adders 821, 822, 8
At step 23, it is added to the complementary color data C', M', Y' and output as under color removed values C1, M1, Y1. On the other hand, during black printing control for printing K, the multiplier 824 multiplies the BP ratio β from the color correction register 926, and this multiplied value (β·K') is sent to the adder via the limiter circuit 834. 835
send to

Furthermore, the color correction masking section uses a linear masking process with a simple circuit configuration, and when controlling undercolor removal, multipliers 831, 832, and 833 apply color correction registers to data C1, M1, and Y1. Each masking coefficient (Ac~Cc, Am~Cm, Ay
~Cy). Then, this multiplication value C2, M2,
Y2 is added by an adder 835 and output as VIDEO1 data. At this time, the output from the limiter circuit 834 is cleared to "00", and the adder 835 outputs
Outputs the addition result of C2, M2, and Y2.

On the other hand, during black overprint control, the color correction register 826 sets multipliers 831, 832, and 833 to "00", so C2, M2, and Y2 are cleared, and K1(
=K2) passes through the limiter 834 and is output as VIDEO1 data.

In order to show the masking correction effect, first, FIG.
8 shows the original color (represented by a white circle) and the reproduced color (represented by a black circle) when using the above-mentioned normal masking coefficient.
It is expressed using the L*a*b* color system of the 1976 uniform color space. (The a*-b* plane shown in the figure shows hue and saturation, and the L* direction perpendicular to the paper plane shows lightness.)
The deviation between the original color and the reproduced color corresponds to the color difference. Here, R,
The average pigment for only G and B is 10.5335, and for C and M
, Y only has an average color difference of 4.0029. Figure 19 shows
FIG. 3 is a diagram showing original colors (white circles) and reproduced colors (black circles) on the a*-b* plane when the above-described masking coefficient M0 for R, G, and B is used. Here, the average color difference of only R, G, and B is 3.8576, and the hue shift is smaller than that in the case of normal masking coefficients. Note that the average color difference of only C, M, and Y is 12.1797. Moreover, FIG. 20 shows
FIG. 4 is a diagram showing original colors (white circles) and reproduced colors (black circles) on an a*-b* plane when the above-described masking coefficient M3 for C, M, and Y is used. Here, the average color difference of only C, M, and Y is 2.43782, and the hue shift is smaller than that in the case of normal masking coefficients. Note that the average color difference of only C, M, and Y is 12.7757. As clearly shown above, the selected masking coefficient reduces the color difference between the corresponding hues.

Note that when the single color mode is set by the M/NC signal, color reproduction is performed in a single color. Monochromatic reproduction means that the gradation information based on the human sense of brightness sensitivity (relative luminous sensitivity) is expressed as C, M, Y, K, R (M+Y),
The purpose is to reproduce with either G (B+Y) or B (C+M) toner. Therefore, similarly to the masking process, the relative luminous efficiency information (MC) is used as the masking coefficient DR, DG, DB.
It can be created by linear processing. MC=Dc・C'+Dm・M'+Dy・Y', that is,
The color correction register 826 has Dc,
Dm and Dy are set, thereby outputting MC data as VIDEO1 data. This masking coefficient D
c, Dm, and Dy are determined depending on the type of toner and correspond to the relative luminous efficiency. Note that, at this time, as already explained, the blacking process is not performed. That is, in the black generation unit 64 shown in FIG. 9, in the single color mode, K'=“00” is always set.
is output.

(h) Edge enhancement/smoothing automatic control in area discrimination section (h-1) Purpose of edge enhancement/smoothing automatic control Generally speaking, for monochrome images, characters/photos are adjusted according to the density change or density distribution of the image. Automatic identification is performed, edge enhancement is performed for character images, and smoothing processing is performed for photographic images, thereby achieving both image sharpening and smoothing, and optimizing MTF correction.

However, in the case of a color image, such discrimination does not necessarily work well with simple edge enhancement because the image density also changes with changes in hue and saturation. For example, when changing from white to red, the edges may be emphasized, but when changing from red to cyan, the hue changes at the edges strangely, so it is better not to emphasize the edges. Skin color is particularly affected. Therefore, it is necessary to extract and control only the change in image brightness.

(h-2) Edge Detection The edge detection section 84 shown in FIG.
Edge detection is performed in both the main scanning and sub-scanning directions for each color in the area around the pixel of interest. In other words, first 4
4 line memories 340, 341, 342, 343
Line data is stored sequentially. Then, these four lines of data R2 (G2, B2), R3 (G3, B3),
From R4 (G4, B4), R5 (G5, B5) and data R1 (G1, B1) of the line currently being output, the data of the center pixel of the center line is sent to a filter 345 for edge detection in the main scanning direction. After an edge is detected by the edge detection filter 346 in the sub-scanning direction, output data RE1 (GE1, BE1) in both directions is sent to the MTF correction control unit 85.
, RE2 (GE2, BE2), respectively.

In edge detection using the filters 345 and 346 described here, the amount and direction of inclination of the pixel of interest in two directions, the main scanning direction and the sub-scanning direction, are extracted. Here, the amount of inclination is the output data RE1 (GE1
, BE1), and RE2 (GE2, BE2), and the inclination direction is its sign (positive, negative). At the edge portions of the image (portions where the brightness changes rapidly), a hue change typified by a color ghost phenomenon occurs. Therefore, this output data is used by the MTF correction control unit 85 to extract areas where erroneous achromatic and chromatic color judgments are likely to occur and to selectively separate areas for MTF correction.

Note that the data of the pixel of interest on the center line is sent to the density conversion section 63, as described above, and is converted into density data DR using the reflectance/density conversion table 347.
(DG, DB).

FIG. 21 shows a circuit of the MTF correction control section 85 for performing automatic MTF control.
Based on the signal from the edge detection section 84, the image is divided into a flat density portion, an edge portion, and an intermediate portion thereof. This density flat area refers to R, G, and G in both the main scanning and sub-scanning directions.
, B data have a certain threshold value (R
EF3) or below. That is, the density flat area is an area where the change in brightness is small for any color data. At this time, the filter selection signal FS0="L" is output. On the other hand, an edge portion is one in which the edge detection amounts of R, G, and B data are all greater than or equal to a certain threshold value (REF4) in either the main scanning or sub-scanning direction, and the R, G,
This is an area where the inclination directions of the B data match. At this time, the filter selection signal FS1="L" is output. Here, in order to prevent misjudgments due to changes in hue, changes in brightness between achromatic colors (white ← → background level such as black ← → black characters/thin lines) or changes between achromatic colors and chromatic colors (white ← →
Background levels such as color patches (red/blue characters, thin lines) are extracted as edges. If both filter selection signals are not "L", it is an intermediate portion.

In the MTF correction control circuit in FIG. 21, the edge detection amounts RE1, GE detected in the main scanning direction
1 and BE1 are absolute value detection circuits 381 and 382, respectively.
, 383 into absolute values RE1', GE1', BE1'. These absolute values RE1', GE1', BE1'
is compared with the threshold value REF3 in the comparator 390, and if the total absolute value is smaller than the threshold value REF3, it passes through the negative logic AND gate 391 and is output to the negative logic AND gate 3.
A signal is sent to 95. Absolute value RE1', GE1', B
E1' is similarly compared to a threshold value REF4 in comparator 390 and is signaled through negative logic AND gate 392 to negative logic AND gate 396 if the total absolute value is greater than threshold value REF4. On the other hand, for the edge detection amounts RE1, GE1, and BE1, the slope (sign) at the edge is detected by the slope discrimination circuit 384, and the sign is output to the negative logic AND gate 396. Therefore, the negative logic AND gate 396 operates when the edge detection amount is larger than the threshold value REF4 in the main scanning direction and the inclination directions of the R, G, and B data match (when it is determined that it is an edge portion). A filter selection signal FS1 (="L") is outputted through a negative logic OR gate 397.

Similarly, the edge detection amounts RE2, GE2, BE2 detected in the sub-main scanning direction are converted to absolute values RE2', GE by absolute value detection circuits 385, 386, 387, respectively.
2', BE2'. This absolute value RE2',G
E2' and BE2' are compared with the threshold value REF3 in a comparator 390, and the total absolute value is compared with the threshold value REF3.
If it is smaller, a signal is sent through negative logic AND gate 393 to negative logic AND gate 395. Therefore, the negative logic AND gate 395 outputs the filter selection signal FS0 when the edge detection amount is smaller than the threshold value REF3 in both the main scanning direction and the sub-scanning direction (density flat portion). Similarly, the absolute values RE2', GE2', BE2' are
It is compared with the threshold value REF4 in the comparator 390, and if the total absolute value is larger than the threshold value REF4, it is passed through the negative logic AND gate 394 to the negative logic AND gate 39.
A signal is sent to 7. On the other hand, edge detection amount RE2, GE2
, BE2, the slope (sign) at the edge is detected by the slope discrimination circuit 388, and the sign is output to the negative logic AND gate 396. Therefore, negative logic AND gate 39
6, the edge detection amount in the sub-scanning direction is the threshold RE
If it is larger than F4 and the slope directions of R, G, and B data match (edge part), the negative logic OR gate 39
7, the filter selection signal FS1 (=“L”) is output. FIG. 22 schematically shows the edge detection amount (GE1) using the filter 345 when the G image data changes in the main scanning direction as shown below. This edge detection amount (
GE1) is the threshold value REF3,RE as shown on the upper side.
F4 is compared, and filter selection signals FS0 and FS1 are outputted in accordance with the result. Note that FS1 is output when either the slope signal NSINA or NSINB is "L".

Threshold R in MTF correction control section 85
EF3 and REF4 can be adjusted externally by sharpness settings (see FIG. 7). For example, if you want to increase sharpness, you can set REF3 and REF4 low.

Note that in this embodiment, REF3<REF4
However, depending on the purpose of processing, REF3>RE
It may also be set to F4.

(i) MTF correction unit Filter selection signal F set by MTF correction control unit 85
S1 and FS0 are spatial filter selection signals of the MTF correction section 66. FS with no edges in both main scanning and sub-scanning directions for any color of read data R, G, B
When 0=“L” (flat density portion), smoothing processing is performed on the data converted to C, M, Y, and K data, and achromatic and chromatic color determination is permitted. Even if achromatic and chromatic color determination is allowed, K
If the amount changes rapidly, it looks like random image noise. Therefore, the MTF correction section 67 performs smoothing processing to make the change less noticeable. (This is also why we classify achromatic and chromatic color judgment into four stages.
Since this was still insufficient, measures were taken to further reduce noise for pixels determined to be achromatic. ) Image noise and moiré caused by the reading system are also reduced, making it possible to smooth areas with gradual changes in brightness, saturation, and hue, such as in photographs.

When FS1="L" (edge portion), the output of the Laplacian filter 324 is permitted, addition is made to the pixel of interest, and edge emphasis processing of the image is performed. Therefore, in the achromatic edge portion, the sharpness of the image is improved without increasing the UCR ratio/BP ratio, so that the sharpness of the image is improved.

In the MTF correction section 66 shown in FIG.
By using an R two-dimensional digital filter, edge enhancement and smoothing processing is performed. First, four lines of data are sequentially stored in four line memories 320, 321, 322, and 323. Then, the data of each line and the data of the line currently being output are processed by a 5×5 digital filter 324 for second-order differentiation (edge emphasis) and a 5×5 digital filter 325 for smoothing. and are output to a multiplier 326 and a two-input multiplexer 328, respectively. The multiplier 326 outputs the product of the output value of the digital filter 324 and the value EDG to the 2-input multiplexer 327.
7 is a filter selection signal FS1="L" (edge part);
In response to “H”, the output value or “00” is output to the adder 329. On the other hand, the multiplexer 328 outputs the smoothed output value of the digital filter 325 and the output value of the pixel of interest that is not smoothed from the line memory 321 to the filter selection signal FS0="L".
(density flat part), selected corresponding to "H", adder 32
Output to 9. Adder 329 adds the two input values and outputs the result as signal VIDEO2.

Here, the second-order differential filter 324 used for edge enhancement detects the edge enhancement amount of the image, and the edge enhancement is performed by linearly converting the edge enhancement amount obtained by this filter and the center pixel. (original image + second-order differential). In other words, FS1="L"
(edge portion), the adder 329 adds the edge-enhanced data to the data of the pixel of interest.

On the other hand, the filter 325 used for smoothing processing reduces image noise by using a moving average based on weighted addition of surrounding pixels to create smooth image data. (The weighted addition prevents false resolution such as moiré caused by filter processing.) In other words, FS0="L
” (density flat part), only the smoothed data is output from the adder 239.

The EDG value that affects edge enhancement in the MTF correction section 66 can be adjusted externally by sharpness settings (see FIG. 7). For example, if you want to increase sharpness, you can increase the EDG value.

The MTF processing described above sharpens and smoothes the image data of the read data R, G, and B. This is because if the same processing is performed on the reproduced color data C, M, and Y, edges will be emphasized even in hue-changing portions, which will conversely deteriorate color reproducibility. Therefore, MTF correction was selectively performed by extracting the brightness change of the read data.

[0073]

[Effects of the Invention] Since the determination as to whether the color is achromatic or chromatic is divided into multiple stages according to the saturation, the number of false determinations is reduced. Therefore,
Judgment accuracy has been improved for images with uniform density and hue, such as color patches. In addition, since the blackening process is performed in stages on the uniform density portions of the halftone black original, image noise is reduced and the changes in the image become more natural. Furthermore, by making the determination on the read color data that has been subjected to smoothing processing, the number of false determinations is further reduced and reproduction can be performed more accurately.

[Brief explanation of drawings]

FIG. 1 is a sectional view of a full-color copying machine.

FIG. 2 is a block diagram of a part of the control system.

FIG. 3 is a block diagram of other parts of the control system.

FIG. 4 is a perspective view of the reading section.

FIG. 5 is a block diagram of an image signal processing section.

FIG. 6 is a circuit diagram of an area determination unit.

FIG. 7 is a circuit diagram of registers for setting various parameters.

FIG. 8 is a graph showing the relationship between density data and black amount K'.

FIG. 9 is a circuit diagram of a black generation section.

FIG. 10 is a circuit diagram of a smoothing processing section and an edge detection section.

FIG. 11 is a graph showing four regions from achromatic colors to chromatic colors.

FIG. 12 is a circuit diagram of an undercolor removal/blackening control section and a color correction control section.

FIG. 13 is a diagram of a regular cube of hue distribution.

FIG. 14 is a graph of the characteristics of the G filter.

FIG. 15 is a graph of the characteristics of M toner.

FIG. 16 is a circuit diagram of a color correction processing section.

FIG. 17 is a circuit diagram of a color correction register.

FIG. 18 is a diagram illustrating color differences when normal masking coefficients are used.

FIG. 19 is a diagram showing color differences when using R, G, and B masking coefficients (M0).

FIG. 20 is a diagram showing color differences when using masking coefficients for C, M, and Y (M3).

FIG. 21 is a circuit diagram of the MTF correction control section.

FIG. 22 is a diagram of an example of edge detection.

FIG. 23 is a circuit diagram of the MTF correction section.

[Explanation of symbols]

65... Area discrimination unit, 66... Color correction processing unit, 67
...MTF correction section, 81...Smoothing processing section, 82
...Undercolor removal/black printing control section, 83...Color correction masking control section, 84...Edge detection section, 85...MTF correction control section, US1, US0...Achromatic color chromatic color determination signal, MS
1, MS0...masking coefficient selection signal, FS1, FS0
...filter selection signal.

Claims (3)

[Claims] 1. Input means for inputting digital data of red, green, and blue reading colors obtained by scanning a document, and black generating means for calculating the minimum value of the digital data of reading colors inputted by the input means. and an under color removal black that creates the product of the minimum value obtained by the black generation means and the first coefficient as black data, and subtracts the product of the minimum value and the second coefficient from the digital data of this read color. an overprinting means; a data conversion means for converting the read color data processed by the undercolor removal black overprinting means into reproduced color data of cyan, magenta, and yellow; , an achromatic color determining means for determining whether digital data of the read colors of green and blue are achromatic in a stepwise manner based on the level difference between the three colors; , stores a plurality of first coefficients and a plurality of second coefficients, and stores a plurality of first coefficients and second coefficients, and determines the first coefficient and second coefficient that are optimal for the result, corresponding to the determination result of the achromatic color determining means.
An image forming apparatus that selects a coefficient and executes the above calculation using both of the selected coefficients. 2. The image forming apparatus according to claim 1, further comprising a smoothing unit that performs two-dimensional spatial digital filter processing for each color on the input digital data of the read color of the pixel of interest, An image forming apparatus characterized in that the data conversion means makes a determination on digital data that has been subjected to spatial digital filter processing by the smoothing means. 3. The image forming apparatus according to claim 1 or 2, further comprising a monochrome mode, wherein the data conversion means sets both the first coefficient and the second coefficient to 0 in the monochrome mode. Features of the image forming device.