JPH04288771A - Color image forming device - Google Patents

Abstract

PURPOSE:To improve color reproducibility to a color frequently distributed in an original image and to obtain a satisfactory image. CONSTITUTION:Original data in respective B, G and R colors read by a CCD are converted to digital data by an A/D conversion circuit 11, transmitted to a LOG transformation circuit 12 and stored in a page memory 23 for the unit of a picture element after LOG transformation. Next, a CPU 20 reads the data on the page memory 23 and calculates the number of accumulated digital data levels concerning all the picture elements for each color. A masking coefficient is decided by using this calculated cumulative number and set to a color correction circuit 14 by the CPU 20. Afterwards, the data in the page memory 23 are transmitted to a UCR 13 and the color correction circuit 14 and after processing the data, the image is formed by a printer 15.

Description

[Detailed description of the invention]

0001

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a color image forming apparatus having a color correction function.

0002

2. Description of the Related Art Various color image forming apparatuses are known, among which a laser printer is often used, which exposes a photoreceptor to an image with laser light and develops it to form an image. . Such laser printers have advantages such as high image quality and high speed, and are widely used, for example, as output devices such as color copying machines and as ordinary printers.

A conventional color image forming apparatus will be described below with reference to FIGS. 9 and 10. FIG. 9 is a schematic configuration diagram of a color image reading device. In the figure, 1 is the original, 2
is the document table glass on which the document 1 is placed. 3 is an illumination device, and 4 is an imaging element array. 5 is an infrared cut filter, and 6 is a contact type CCD color sensor (hereinafter referred to as "CCD").
), and 7 is an optical system unit.

In the above configuration, first, the process of reading a color original will be explained. When a copy key (not shown) is pressed, the illumination device 3 illuminates the original 1, and the reflected light from the original 1 is transmitted to the imaging element array 4 and the infrared cut filter 5.
The image of the original is formed on the CCD 6. and,
The optical system unit 7 sequentially scans the original 1 in the direction of the arrow. As shown in FIG. 10, the CCD 6 has red (R), green (G), and blue (B) filters regularly attached to each pixel.

As the original 1 is scanned, electrical signals from the CCD 6 are processed by a signal processing circuit shown in FIG. In FIG. 11, 6B, 6G, and 6R represent signals from the B, G, and R elements on the CCD 6 shown in FIG. 10. Next, the R, G, and B signals are introduced into the A/D conversion circuit 11 and the LOG conversion circuit 12, and are converted into Y1, M1, and C1 signals in the form of digital signals.

The Y1, M1, and C1 signals are introduced into a circuit 13 that performs black extraction and UCR (undercolor removal) processing, and Y2, M2
, C2, and Bk2 signals are generated.

[0007]

[Math 1]

Here, (Y1, M1, C1) mi
n is the minimum signal among the signals of Y1, M1, and C1, and k1, k2, and k3 are coefficients. Next, these Y2, M2, C2 signals are introduced into the color correction circuit 14 in order to correct the color separation filter of the CCD 6 and the spectral distribution of the coloring material used by the printer 15, and are subjected to the arithmetic processing shown in the following equation. .

[0009]

[Math 2]

Here, a11 to a33 are masking coefficients for color correction, respectively. And the above coefficient k1 ~
k3, a11 to a33 are stored in the ROM (read only memory) 21, and after copying starts, the CPU
(Central processing circuit) 20 sets the circuits 13 and 14. The Y3, M3, C3, and Bk3 signals that are the calculation results are visualized by a color image forming means (printer 17) such as a color thermal transfer printer, a color inkjet printer, or a color laser beam printer. As a means of visualizing this image, laser printers use modulation that makes the laser emission time correspond to the size of the image signal.
Means for forming images by performing so-called pulse width modulation are often used.

Conventionally, the masking coefficient, which is the above-mentioned color correction means, has been determined, for example, by the following calculation. First, a standard color document whose chromaticity coordinates are known is read by the optical system unit 7 shown in FIG.
Y3, M3, C3 of the color correction circuit 14 shown in FIG.
, Bk3 signals have the above-mentioned chromaticity coordinates, the masking coefficients are determined using the following method of least squares procedure.

Here, the signals for a standard color document of one color are Y', M', C' (Y2, M2 in FIG.
, C2) and Y, M, C are the chromaticity coordinates as target values, the deviation amount of each color from the target ΔY , ΔM ,
ΔC is expressed by the following formula.

[0013]

[Math 3]

[0014] This is carried out using, for example, a standard color original of 20 colors as shown in FIG. 12, and the method of least squares is applied so that the amount of deviation is minimized. Let the evaluation functions of yellow, magenta, and cyan be φY, φM, and φC, respectively. Then, for yellow, φY is ΔY squared by 20
The addition of color is as follows.

[0015]

[Math 4]

Here, Yi ′, Mi ′, Ci ′ are signals for the i-th standard color original, and Yi
represents the target value of yellow among the color components of the i-th standard color document. Since the coefficients are determined so that the deviation with respect to the above evaluation function is zero,

[0017]

[Math 5]

Similarly, when magenta and cyan are processed, the following equation is obtained.

[0019]

[Math 6]

[0020] Here, if the above equation is expressed as [C]・[A]=[D], [A]=[C-1]・[D], so a
The coefficients 11 to a33 are found. Note that [C-1] is [C
] is the inverse matrix of The masking coefficients determined in this manner are stored in advance in the ROM 21, and color correction is performed using the masking coefficients to reproduce the colors of the original.

[0021]

However, the above conventional example has the following drawbacks in color reproduction. When determining the masking coefficient using the least squares method described above, for example, for the yellow evaluation function,

[
0022

[Math 7]

However, the first term of the above equation is
Similarly, the final term is the chromaticity error when reproducing the 20th color in FIG. 12. Similarly, the final term is the chromaticity error when reproducing the 20th color in FIG. Here, since the squared error is added with the same weight for the chromaticity errors from the 1st to the 20th, the masking coefficients finally created are set so that the average of the reproducibility errors for each color is minimized. It has a masking coefficient.

Therefore, when color reproduction is performed using this masking coefficient, each color of the original always has a certain error from its chromaticity, and is reproduced as shown in FIG. 13. Therefore, for example, even when reproducing an original with many greenish colors, the image is reproduced with the above-mentioned errors, and depending on the image, it is not always possible to obtain an image with good color reproducibility.

The present invention has been made to solve the above-mentioned problems, and it is possible to improve color reproducibility for colors that are widely distributed in a document image, and to obtain a color image that is good in quality. The purpose of the present invention is to provide an image forming apparatus.

[0026]

Means for Solving the Problems and Operations In order to achieve the above object, a color image forming apparatus of the present invention has the following configuration. That is, in a color image forming apparatus that inputs an original image, performs color correction, and then forms a color image, there is a generating means for generating color information of the original image, and a calculation is performed based on the color information in the generating means. It has a calculating means and a determining means for determining the color characteristics of the original image according to the result of the calculating means, and determines the processing coefficient of the color correction based on the result of the determining means, and the color image is It is characterized by forming.

[0027] Furthermore, the arithmetic means is characterized in that it performs an arithmetic operation of accumulating the output level of the generation means for each color. Furthermore, the calculation means is characterized in that it performs calculation to create a frequency distribution of the output level of the generation means for each color.

[0028]

DESCRIPTION OF THE PREFERRED EMBODIMENTS A preferred embodiment of the present invention will be described in detail below with reference to the accompanying drawings. FIG. 1 is a block diagram of an image forming apparatus, and explanations of the same numbers as in FIG. 11 will be omitted. In the figure, 20 is a CPU, which performs arithmetic processing according to a processing procedure (program) stored in a ROM 21, which will be described later. 21 is ROM, CP
Stores the processing procedure of U20. 22 is a RAM, which stores the calculation results of the CPU 20 and stores the calculation results of the CPU 20 as necessary.
20 is the one from which data can be extracted. 23
is a page memory that stores digital data after reading the original, and data transfer is controlled by the CPU 20.

In this configuration, first, the CCD sensor 6
The B, G, and R color original data read by
The A/D conversion circuit 11 converts the data into digital data. Next, the signal is sent to the LOG conversion circuit 12, subjected to LOG conversion, and then stored in the page memory 23 in units of pixels. Here, when the color-separated digital image data for each pixel is as shown in the schematic diagram shown in FIG. 2, the digital image data (8 bits) is stored in Y, M,
The data are written to predetermined addresses in the order of C and in the order of pixels.

Next, the CPU 20 reads out the data on the page memory 23 and calculates the cumulative number of digital data levels for all pixels for each color. Note that the cumulative numbers corresponding to each color of yellow, magenta, and cyan are NY and NM.
, NC. Here, a method for determining masking coefficients using the above calculation results will be described below with reference to FIGS. 3 to 5.

[0031] Here, the above-mentioned NY, NM, NC
The case where NY > NM > NC will be explained. First, according to the flowchart shown in FIG.
The difference between and NM and the degree of the difference between NM and NC are determined. Specifically, it is checked whether NY is 20% or more larger than NM, which has an intermediate cumulative number (step S1), and the degree of difference between yellow and magenta is determined. In other words, YES
If so, it is determined to be step S2, and if NO, it is determined to be step S3. Also, it is checked whether NC is 20% or more smaller than NM (step S4), and the degree of difference between magenta and cyan is determined. That is, if YES, it is determined to be step S5, and if NO, it is determined to be step S6.

Then, based on the above-mentioned judgment results, the characteristics of the document as shown in FIG. 4 are determined. For example, S2 and S
If it is 5, that is, there is no big difference in the cumulative numbers of yellow, magenta, and cyan, so it is determined that ``Either each color is distributed to the same degree in the manuscript, or there is a large distribution of gray.'' do. Also, if S3 and S5, '
There is a large distribution of yellow in the manuscript. ” is decided. Similarly, if it is S2 and S6, the cumulative number of yellow and magenta is much larger than that of cyan, so 'red has a large distribution in the document. ” is decided. And S3 and S6
If so, ``There is a large distribution of yellow and red in the manuscript. ” is decided.

In this manner, after determining the color distribution in the document, appropriate masking coefficients are determined based on the determination results, thereby making it possible to perform appropriate color reproduction. In this embodiment, the masking coefficient is determined by selecting or calculating from seven types of masking coefficients stored in advance in the ROM 21. Here, 7
The types of masking coefficients are a total of seven masking coefficients, including the masking coefficients used in the conventional example and six types of masking coefficients that have particularly excellent reproducibility of each of the six colors: yellow, magenta, cyan, blue, green, and red. , these are assumed to be masking coefficients of types 0 to 6, respectively.

Masking coefficients corresponding to each of FIG. 4 are determined from these seven types of masking coefficients as shown in FIG. 5. For example, if S2 and S5, masking coefficient type 0, which is similar to the conventional one, is selected, and if S3 and S5, masking coefficient type 1, which has particularly excellent yellow reproducibility, is selected. In addition, if it is S2 and S6, select masking coefficient type 6, which has particularly excellent red reproducibility,
If it is S3 and S6, the masking coefficient is calculated as the average of masking coefficient type 1, which has particularly excellent reproducibility for yellow, and masking coefficient type 6, which has particularly excellent reproducibility for red.

The CPU 20 sets the masking coefficient determined by the above processing in the color correction circuit 14. Thereafter, the data in the page memory 23 is sent to the UCR 13 and the color correction circuit 14, and after being processed, an image is formed by a printer. Next, a method for calculating seven types of masking coefficients stored in advance in the ROM 22 will be described.

In this embodiment, seven types of masking coefficients are calculated by weighting each term of the evaluation function of the least squares method described in the conventional example and changing the weighting method. For example, in order to obtain a masking coefficient with particularly good reproducibility for yellow, it is necessary to use the conventional yellow evaluation function.

[0037]

[Math. 8]

Instead of using an evaluation function that gives the same weight to all terms, as in Just use it. That is, according to the Munsell color standard shown in Figure 12, 10Y8/10,5Y6/4,5G
Evaluation function weighted for the color Y6/8, 1YR6/4

[0039]

[Math. 9]

Use [0040]. Similarly, φM, φC
Also, by creating a masking coefficient using an evaluation function that weights the third, eighth, ninth, and fifteenth terms, it is possible to obtain a masking coefficient with particularly excellent yellow reproducibility. . The color reproducibility when performing color correction using this masking coefficient is shown in Figure 1 above.
3, as shown in Figure 6, the color reproducibility is inferior to blue colors, which are complementary colors of yellow, but the color reproducibility is particularly excellent for yellow colors. I understand that.

By changing the above-mentioned weighting method, other colors (magenta, cyan, blue, green,
It is possible to obtain masking coefficients with particularly excellent reproducibility for red). In this embodiment, a page memory and a CPU are used to accumulate digital data, but the same effect can be obtained by using an adder.

As explained above, according to this embodiment,
The color distribution in the original image is judged from the color information of the original image, and the color correction means is determined based on the judgment result, thereby improving color reproducibility for colors that are widely distributed in the original image. You can obtain good images. It also has the effect of preventing deviations in color reproduction as much as possible when creating a grandchild copy such as a copy of a copy.

[0043]

[Other Embodiments] Next, other embodiments according to the present invention will be described below. In this embodiment, the characteristic of the document is determined from the color information of the document in the previously described embodiment: ``Either each color is distributed to the same degree in the document, or there is a large distribution of gray. ”, then it is further determined which of the above-mentioned characteristics the document has.

The configuration of the device is the same as that shown in FIG. 1, and the data flow is the same as in the previous embodiment until the color-separated digital data is stored in the page memory 23, so the explanation will be omitted. . In this embodiment, page memory 2
The CPU 20 reads out the data of No. 3 and creates a histogram of image data for each color as shown in FIG.

FIG. 8 is a flowchart showing the processing procedure performed by the CPU 20, in which processing is performed for each color of Y, M, and C. First, data for one pixel is read from the page memory 23 (step S11), the frequency of the nth level of digital data from 0 to 255 is incremented (step S12), and when processing is completed for all image data ( Step S13), SY (
n), SM(n), and SC(n) are created. In addition,
The memory of the RAM 22 is used as necessary, and the final frequency distributions SY (n), SM (n), SC (n) (
n=0 to 255) are also stored in the RAM 22. and,
Using the frequency distribution Si (n) (i=Y, M, C), the CPU 20 calculates, for example, the following amounts for each color. (1) Digital data level indicating the maximum frequency of the histogram. (2) Cumulative number of digital data levels for all pixels.

Here, the amounts (1) and (2) will be explained using, for example, a yellow signal. (1) is yellow (Y) in Figure 7
Ymax is calculated from the histogram. here,
At low levels of the CCD output (levels 0 to 32), the maximum frequency is not detected. This is CC
The low level side of the D output corresponds to the white area of the original, so
This is to prevent this from being detected.

(2) is based on the yellow (Y) histogram in FIG.

[0048]

[Math. 10]

It is obtained as follows. Similarly, Nmax
, Cmax, NM, and NC are generated. Next, a method of determining masking coefficients using the above calculation results will be described below. The degree of difference between NY, NM, and NC is determined using the same method as in the above-mentioned embodiment. be. ” If it is determined that
Determine the values of Ymax, Mmax, and Cmax.

This determination is performed using a process similar to the process shown in FIG. That is, Ymax, Mmax, C
It determines whether the other two values are relatively close to the intermediate value of max, and as a result,
If the values of Ymax, Mmax, and Cmax are equal or approximate, the above judgment result is further narrowed down to ``There is a large distribution of gray in the manuscript.'' ”, and also Ymax
, Mmax Cmax is relatively large, and there is a large difference in size.
It is distributed. ”.

Masking coefficients are then selected based on the above results. That is, if it is the former, a masking coefficient with particularly excellent gray color reproducibility is selected, and if it is the latter, the same masking coefficient as described in the conventional example is selected. Here, in order to obtain a masking coefficient with particularly excellent gray reproducibility, as in the previous embodiment, a large weight is given to the gray color in the evaluation function term of the least squares method, and the masking coefficient is All you have to do is decide. That is, the gray in FIG. 12 is indicated by brightness, Y=8.
5%, Y=50%, Y=30%, Y=15%, Y=5%
The masking coefficients can be determined using an evaluation function that weights these colors.

[0052] By using the above masking coefficients, the image created for a color document with a large distribution of gray in the document image, for example, a document with a color photograph pasted on a portion of a black text document, can be improved. It is possible to obtain an image with good gray reproduction without any tint in the text. <Modified example> Next, set the cumulative numbers for each color as NY, NM, N
C, and the case where the color distribution of the document is determined by the following calculation will be explained.

First, the cumulative numbers NY, NM, N for each color
Let the sum of C be Nall, Nall =NB +N
Calculate G +NR. Next, if the degrees of B, G, and R color components in the original are aB, aG, and aR, then aB
=NB/Nall,aG=NG/Nall,
aR /Nall, and aB +aG +aR =
It is 1. This indicates that the ratio of color components in the original is blue:green:red=aB:aG:aR.

On the other hand, in the modified example, there are three types of masking coefficients stored in advance, which are masking coefficients with particularly excellent reproducibility for blue, green, and red, and these are designated as types 1, 2, and 3. Ratio aB of the above color components,
The masking coefficient (M) calculated from aG, aR and the masking coefficient is calculated as (M)=aB・(Type 1)+aG・(Type 2)+aR・(Type 3).

By determining the masking coefficient using the method described above, complicated judgment is not required, and since the basic masking coefficient is less than half, the memory capacity of the ROM can be saved.

[Effects of the Invention] As explained above, according to the present invention,
Color reproducibility can be improved for colors that are widely distributed in the original image, and a good image can be obtained.

[Brief explanation of the drawing]

FIG. 1 is an image processing block diagram in this embodiment.

FIG. 2 is a schematic diagram showing a color-separated document in units of pixels.

FIG. 3 is a flowchart for comparing cumulative numbers for each color.

FIG. 4 is a diagram for determining document characteristics.

FIG. 5 is a diagram for determining masking coefficients according to document characteristics.

FIG. 6 is a diagram showing color reproduction using masking coefficients with particularly excellent yellow reproducibility.

FIG. 7 is a diagram of frequency distribution for each color.

FIG. 8 is a flowchart for creating a frequency distribution.

FIG. 9 is a diagram showing the configuration of a document reading system.

FIG. 10 is an enlarged schematic diagram of a CCD sensor.

FIG. 11 is a conventional image processing block diagram.

[Figure 12-1],

[Figure 12-2],

FIG. 12-3 is a diagram showing the correspondence between conventional color patches and Munsell color standards.

FIG. 13 is a diagram showing color reproduction using conventional masking coefficients.

[Explanation of symbols]

6 CCD 11 A/D conversion circuit 12 LOG conversion circuit 13 Black extraction and UCR circuit 14 Color correction circuit 15 Color printer 20 CPU 21 ROM 22 RAM 23 Page memory

Claims (3)

[Claims] Claims: 1. A color image forming apparatus that inputs a document image and forms a color image after performing color correction, comprising: a generation means for generating color information of the document image; comprising a calculating means for performing calculations and a determining means for determining color characteristics of the original image according to the results of the calculating means, and determining processing coefficients for the color correction based on the results of the determining means; A color image forming apparatus characterized by forming a color image. 2. A color image forming apparatus according to claim 1, wherein said calculation means performs a calculation of accumulating the output level of said generation means for each color. 3. A color image forming apparatus according to claim 1, wherein said calculation means performs calculation for creating a frequency distribution of output levels of said generation means for each color.