JPH10210313A - Image-processing method and image-processing unit - Google Patents

Abstract

PROBLEM TO BE SOLVED: To prevent a part of color of an input color image from being deformed and to apply optimum color reproduction to each page of the received color image, in the case that the input color image such as a color original is copied or printed out with a copying machine or a printer whose color region is narrower than the color region of the received image. SOLUTION: An image input section 100 reads an original twice for each page. In the case of a preliminary scanning, a coefficient decision means 250 decodes a color conversion coefficient, based on an input red, green blue (RGB) signal from the image input section 100 and an L*a*b* signal, obtained by converting the input RGB signal at a color conversion means 210. In this case, the color conversion coefficient is decided for each original page, so that a relative L*a*b* signal normalized from two points, the lightest part and the darkest part of the original is in matching with a relative L*a*b* signal normalized from two points, the lightest part and the darkest part of an object to be copied. In the case of main scanning, the color conversion means 210 converts the input RGB signal into the L*a*b* signal, based on the decided color conversion coefficient.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing method and apparatus for copying or printing an input color image such as a color original by a copying machine or a printer having a color gamut narrower than the color gamut.

[0002]

2. Description of the Related Art The basis of color reproduction is faithful reproduction. However, if the color gamut (range of colors that can be expressed) of a paper document or a so-called electronic document expressed as image information is wider than the color gamut of a copier or a printer, the wide part is not faithfully reproduced. It will be crushed.

Conventionally, this problem has been studied mainly as a problem of gradation compression in a density system from the viewpoint of how to make a tone curve, and it has been empirically said that a curve that is upwardly convex is good. Have been done.

[0004]

However, when considering color reproduction, it is difficult to understand at a glance the correspondence between the tone curve and the color. Therefore, conventionally, after adjusting the tone curve,
If the color is misaligned, the color is corrected or adjusted according to the misalignment, and the problem of gradation and color must be dealt with separately.

On the other hand, JP-A-2-126774 and JP-A-2-126775 disclose the above-mentioned collapse by uniformly compressing an input image by changing a tone curve and determining a color compression coefficient. It is shown to prevent the occurrence of.

However, this method has the drawback that it takes time to determine the coefficients and that the color gamut of the document is not constant for each document, so that the optimum color reproduction cannot be performed for each document.

In view of the above, the present invention does not cause the above-described collapse, and is capable of performing optimal color reproduction for each page of an input color image such as for each original.

[0008]

According to the present invention, there is provided an image processing method for obtaining an output color image by copying or printing an input color image. The relative brightness normalized from the brightest and darkest points of the image and the relative brightness normalized from the brightest and darkest points of the output color image match.

Further, according to the present invention, in an image processing method for obtaining an output color image by copying or printing an input color image, the image processing method comprises the steps of: The relative chromaticity normalized from the point and the brightest and darkest parts of the output color image.
The relative chromaticity normalized from the point is matched.

[0010]

According to the image processing method of the present invention according to the above method, an input color image signal, for example, an RGB signal which has been color-separated into three colors of red, green, and blue, is converted into, for example, L ^* a by a color conversion means. ^* B ^* is converted to a signal. At this time, for each page of the input color image, such as for each original, the relative brightness or relative chromaticity normalized from the brightest portion and the darkest portion of the input color image, and the brightest portion and the maximum brightness of the output color image. The color conversion coefficient of the color conversion means is determined so that the normalized relative lightness or relative chromaticity from the two points in the dark area matches.

Therefore, a part of the color gamut of the input color image is not destroyed, and optimal color reproduction is performed for each page of the input color image such as for each original.

[0012]

FIG. 1 shows an example of an image forming system for realizing an image processing method of the present invention, which is a digital color copying machine. The image forming system of this example,
That is, the digital color copying machine includes the image input unit 100,
It comprises an image processing unit 200 and an image output unit 300.

The image input section 100 optically reads the original 1 to obtain RGB signals as input color image signals. The input RGB signals are supplied to an image processing section 200. The document 1 is read twice for each page.

As will be described later in detail, the image processing unit 200 temporarily converts the input RGB signals from the image input unit 100 into L signals.
^{* A *} b ^* signal, and further converts the L ^* a ^* b ^* signal into a YMCK signal composed of yellow, magenta, cyan, and black color signals that are recording colors in the image output unit 300. , The YMCK signal is output to the image output unit 3
Sent to 00. The asterisk (*) in the figure
Is omitted.

The image output unit 300 forms an output color image on paper using yellow, magenta, cyan, and black color materials in response to the YMCK signal from the image processing unit 200.

FIG. 6 shows an example of the image output unit 300,
This is a case of a laser scanning type electrophotographic method. In the image output unit 300, the image data from the image processing unit 200 is converted by the screen generator 390 into a binary signal whose pulse width is modulated according to the pixel value, and the binary signal is used to convert the binary signal into the laser light scanner 380. Is driven, and a laser beam L is obtained from a laser beam scanner 380, and the laser beam L is
Irradiated on 10.

The photosensitive drum 310 is charged by the charger 320, and the laser light L from the laser light scanner 380 is charged.
Is applied, an electrostatic latent image is formed on the photosensitive drum 310. The developing devices 331, 332, 333, and 334 of the respective colors of the rotary developing device 330 abut on the photosensitive drum 310 on which the electrostatic latent image is formed, and thereby the respective colors of the colors formed on the photosensitive drum 310 are contacted. The electrostatic latent image is developed into a toner image.

Then, the paper on the paper tray 301 is sent onto the transfer drum 340 by the paper supply unit 302 and corona discharge is applied from the back of the paper by the transfer charger 341, so that the paper on the photosensitive drum 310 is The developed toner image is transferred onto a sheet. When the output image is a multi-color image, the paper is repeatedly brought into contact with the photosensitive drum 310 two to four times, so that the images of two to four colors are multiplex-transferred.

The transferred sheet is sent to a fixing device 370, and the toner image is fixed on the sheet. The photosensitive drum 310 is
After the toner image is transferred onto the sheet, the toner image is cleaned by the cleaner 350 and is prepared for reuse by the pre-exposure device 360.

In the example shown in FIG. 1, the image processing unit 200 includes a color conversion unit 210, an inverse color conversion unit 220, a dot area ratio conversion unit 230, a black plate generation unit 240, and a coefficient determination unit 25.
0.

The input RG from the image input unit 100
B signals, the color converter ^210, L ^* a * ^{b *} are converted signals, the ^L * ^a * ^{b *} signal, the inverse color transform 2
20 is converted into a YMC signal composed of yellow, magenta and cyan color signals, and the YMC signal is converted by a dot area ratio conversion means 230 into a YMC dot area ratio signal Yo.
Converted to MoCo. Further, the black signal generation means 240 generates a black signal Ko from the YMC halftone dot area ratio signal YoMoCo.
The K signal YoMoCoKo is obtained.

For example, as shown in FIG. 2, the color conversion means 210 converts the input RGB signal into a transmission neutral density signal ReGeB.
e, a transmission neutral density conversion means 211 composed of an LUT (look-up table), and a matrix conversion means 212 for performing a matrix operation on the transmission neutral density signal ReGeBe to convert it into an L ^* a ^* b ^* signal. , Can be configured.

As shown in FIG. 3, the color conversion means 210 includes a DLUT (direct look-up table).
Can also be configured.

The color conversion coefficient of the color conversion means 210 is determined and changed by the coefficient determination means 250 for each page of the document 1. That is, document 1 is read twice per page, and as shown below,
The coefficient determination means 250 determines the color conversion coefficient of the color conversion means 210, and at the time of the main scan, the color conversion means 210 converts the input RGB signal into an L ^* a ^* b ^* signal according to the determined color conversion coefficient. Then, as described above, the image data is converted into an output YMCK signal by the inverse color conversion unit 220, the dot area ratio conversion unit 230, and the black plate generation unit 240.

FIG. 4 shows an example of the coefficient determining means 250. At the time of pre-scan, the color conversion coefficient storage unit 251 stores a color conversion coefficient of a default value obtained from the standard color chart, and the input RGB signal from the image input unit 100 is
In addition to being stored in the RGB storage unit 252, the color conversion unit 210 uses the default color conversion coefficient
It is converted to an L ^* a ^* b ^* signal, and the converted L ^* a ^* b
^{* The} signal is stored in the Lab storage means 253.

Further, the L ^* a ^* b ^* signal stored in the Lab storage unit 253 is converted into an XYZ signal by the XYZ conversion unit 254, and the white point / black point detection unit 255 calculates the maximum value of each component of the XYZ signal. The signal XwYwZw is the brightest part of the document, and the signal XkYkZ of the minimum value of each component
k is detected as the darkest part of the document. Hereinafter, the brightest part is referred to as a white point, and the darkest part is referred to as a black point.

The signals XiYiZi at the white point and black point
(I = w, k) is supplied to the relative Lab conversion means 256, and the relative Lab conversion means 256 converts the XYZ signal from the XYZ conversion means 254 into a relative L
It is converted into a signal L ^* 'a ^* ' b ^* 'which is a ^* a ^* b ^* signal, and the signal L ^* ' a ^* 'b ^* ' is stored in the relative Lab storage means 257.

Further, in the color conversion coefficient determining means 258, the relative L ^* a stored in the relative Lab storing means 257 is stored.
^From the signal L ^* 'a ^* ' b ^* ', which is a ^* b ^* signal, and the input RGB signal stored in the RGB storage means 252, RG
A color conversion coefficient for converting the B signal into an L ^* a ^* b ^* signal is calculated, and the calculated color conversion coefficient is stored in the color conversion coefficient storage unit 251 instead of the color conversion coefficient having the default value. You.

[0029] Here, the relative normalized from two points of white spots and black spots of the original L ^* a ^* b ^* signals, and the white point of the copies and normalized from the two points of black spots are relative L ^* a ^* b ^*
The signals are given by equation (1a) or (1b), equation (2) and equation (3), respectively.

L ^* wk = 116 {Y / (Yw−Yk)} − 16 (1a) [However, when Y / (Yw−Yk)> 0.008856] L ^* wk = 903.29Y / (Yw −Yk) (1b) [However, when Y / (Yw−Yk) ≦ 0.008856] a ^* wk = 500 [{X / (Xw−Xk)} ^1/3 −ΔY / (Yw−Yk) )｝ ^1/3 ] (2) b ^* wk = 200 [{Y / (Yw−Yk)} ^1/3 − {Z / (Zw−Zk)} ^1/3 ] (3) 2), Equation (3) is expressed as X / (Xw-Xk),
When Y / (Yw-Yk) and Z / (Zw-Zk) are both greater than 0.008856, X / (Xw-
Xk), Y / (Yw-Yk), Z / (Zw-Zk)
When 0.008856 or less, ΔX / (Xw−X
k) { ^1/3 is 7.787 {X / (Xw-Xk)}
^1/3 +16/116, {Y / (Yw-Yk)}
^1/3 is 7.787 {Y / (Yw-Yk)} ^1/3 +
On 16/116, {Z / (Zw−Zk)} ^1/3
7.787 {Z / (Zw-Zk)} ^1/3 +16/11
6 is replaced.

As described above, Xw, Yw, and Zw are tristimulus values of white points, and Xk, Yk, and Zk are tristimulus values of black points.

Then, the obtained relative L ^* a ^* b ^* signal of the original is converted into L ^* wkI, a ^* wkI, b ^*
wkI, the relative L ^* a ^* b ^* signal for the copy is
Assuming that ^* wkO, a ^* wkO, b ^* wkO, the color conversion coefficient determining means 258 calculates L ^* wkI = L ^* wkO (4) a ^* wkI = a ^* wkO (5) b ^* wkI = b ^* WkO... (6) The color conversion coefficients for converting the RGB signals into L ^* a ^* b ^* signals are determined, and the determined color conversion coefficients are stored in the color conversion coefficient storage unit 251. .

FIG. 5 shows the above-described coefficient determination procedure. First, in step 101, the image input unit 100 pre-scans the original 1, and converts the obtained input RGB signals into RGB signals.
B storage means 252 and the color conversion means 21
0 to convert it to an L ^* a ^* b ^* signal, and the L ^* a ^* b ^*
The signal is stored in the Lab storage means 253.

Next, in step 102, the XYZ conversion means 254 stores the L stored in the Lab storage means 253.
^{The *} a ^* b ^* signal is converted to an XYZ signal, then step 1
In step 03, the white point and black point are detected by the white point / black point detecting means 255.
The b conversion means 256 converts the XYZ signal into a relative L ^* a ^* b ^* signal based on the detected white point and black point, and stores the converted signal in the relative Lab storage means 257.

Next, at step 105, the color conversion coefficient determining means 258 sends the RGB signal and the relative L ^* a from the RGB storage means 252 and the relative Lab storage means 257.
^{The *} b ^* signal is read out, the color conversion coefficient is determined by the least squares method, and the determined color conversion coefficient is stored in the color conversion coefficient storage unit 251.

At the time of the main scan, the image input unit 1
The input RGB signals from 00 are converted into L ^* a ^* b ^* signals by the color conversion coefficients provided from the color conversion coefficient storage means 251 in the color conversion means 210 as described above.

The above-described determination of the color conversion coefficients at the time of pre-scanning and the color conversion at the time of main scanning based on the determined color conversion coefficients are performed for each page of the document 1.

Therefore, the color gamut part of the document is not destroyed, and optimal color reproduction is performed for each document.

[0039]

As described above, according to the present invention, even when an input color image such as a color original is copied or printed by a copying machine or a printer having a color gamut narrower than the color gamut, the input color image can be copied. The color gamut portion does not collapse, and optimal color reproduction can be performed for each page of the input color image such as for each document.

[Brief description of the drawings]

FIG. 1 is a diagram illustrating an example of an image forming system that realizes an image processing method according to the present invention.

FIG. 2 is a diagram illustrating an example of a color conversion unit.

FIG. 3 is a diagram showing another example of the color conversion means.

FIG. 4 is a diagram illustrating an example of a coefficient determining unit.

FIG. 5 is a diagram illustrating an example of a coefficient determination procedure.

FIG. 6 is a diagram illustrating an example of an image output unit.

[Explanation of symbols]

 Reference Signs List 100 image input unit 200 image processing unit 210 color conversion unit 250 coefficient determination unit 300 image output unit

Claims (4)

[Claims] An image processing method for obtaining an output color image by copying or printing an input color image, comprising: a brightest portion and a darkest portion of an input color image in a uniform color space for each page of the input color image; An image processing method characterized in that the relative brightness normalized from the two points and the relative brightness normalized from the brightest part and the darkest part of the output color image coincide with each other. 2. An image processing method when an output color image is obtained by copying or printing an input color image, wherein each page of the input color image is normalized from two points of a brightest part and a darkest part of the input color image. An image processing method characterized by making the calculated relative chromaticity coincide with the relative chromaticity normalized from two points of the brightest part and the darkest part of the output color image. 3. An image processing method in which an output color image is obtained by copying or printing an input color image, wherein each page of the input color image is normalized from two points, a brightest part and a darkest part of the input color image. The relative lightness and relative chromaticity thus normalized, and the relative lightness and relative chromaticity normalized from two points of the brightest part and the darkest part of the output color image, and tristimulus values of the brightest part are defined as Xw, Yw, and Zw. Let L ^* wk = 116 {Y / (Yw−Yk)} − 16 (1a) [where Y / (Yw−Yk)> 0.008856, where Xk, Yk, and Zk are the tristimulus values of the darkest part. Time] L ^* wk = 903.29Y / (Yw-Yk) (1b) [However, when Y / (Yw-Yk) ≦ 0.008856] a ^* wk = 500 [｛X / (Xw-Xk) ^{} 1/3 - {Y / (Yw} -Yk)} ^1/3 ] ... (2) b ^* wk = 200 [{Y / (Yw-Yk)} ¹ /3-{Z / (Zw-Zk)} ^1/3 ] ... (3) ) (3) is X / (Xw−Xk), Y
/ (Yw-Yk) and Z / (Zw-Zk) are all
X / (Xw-X
k), Y / (Yw-Yk), Z / (Zw-Zk)
When 0.008856 or less, ΔX / (Xw−X
k) { ^1/3 is calculated as 7.787 {X / (Xw-Xk)}.
^1/3 +16/116, {Y / (Yw-Yk)}
^1/3 is calculated as 7.787 {Y / (Yw-Yk)} ^1/3 +
On 16/116, {Z / (Zw-Zk)} ^1/3
7.787 {Z / (Zw-Zk)} ^1/3 +16/11
6), and the obtained relative lightness and relative chromaticity of the input color image are represented by L ^*.
wkI, a ^* wkI, b ^* wkI, and the relative lightness and relative chromaticity of the output color image are represented by L ^* wkO, a ^* w
When kO and b ^* wkO, L ^* wkI = L ^* wkO (4) a ^* wkI = a ^* wkO (5) b ^* wkI = b ^* wkO (6) Image processing method. 4. A method according to claim 1, wherein each of the pages of the input color image includes red, green,
An input color image signal separated into three colors of blue is normalized relative L ^* a ^* b ^* from two points of the brightest part and the darkest part ^.
A color conversion means for converting the input color image signal and the relative L ^* a ^* b ^* signal into a signal; a coefficient determination means for determining a color conversion coefficient of the color conversion means from the input color image signal; and the relative L ^* a ^* b ^* Converting means for converting the signal into an output color image signal composed of yellow, magenta, cyan, and black color signals.