JPH06189121A - Color image forming method - Google Patents

Abstract

PURPOSE:To attain desired color reproduction by setting a second ink density signal on an input signal unreproducible by a printer as ink density, and also, changing an evaluation function which judges the optimization of color reproduction corresponding to the second signal. CONSTITUTION:First color correction is applied to a reproducible color out of the input signals R, G, and B to be recorded by an I/O means 24, and a first ink density signal can be obtained. It is judged whether or not the signals R, G, and B can be regenerated by the printer with the signal. Thus, a color correction arithmetic operation for finding the second ink density signal by which the optimum color reproduction can be performed out of the reproducible colors is performed on unreproducible signals R, G, and B. The first and second ink density signals are outputted from the means 204 as the ink density signals Y, M, and C used in recording. Accordingly, a control means 206 controls the heating value of a thermal head 207 in the plane sequence of yellow, magenta, and cyan, and forms a color image on reception image paper by recording gradation.

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 method such as a color printer or a color copying machine for printing out a color image.

[0002]

2. Description of the Related Art Color reproduction in a color image forming apparatus in the field of hard copy such as a color printer and a color copying machine is performed by using three primary colors (R,
G, B) are complementary colors of cyan (C), magenta (M),
The color reproduction is based on the subtractive color mixing principle in which the reflectance of color light is adjusted with the three color inks of yellow (Y).

FIG. 11 shows an example of spectral absorption characteristics of ink used in a sublimation thermal transfer recording type printer. As can be seen from the example of this ink, the spectral absorption characteristics of the actual ink have a center wavelength that is out of the ideal, and the absorption characteristics are broad, so there are unnecessary absorption components. In reproduction, the hue is different from the desired color, and a color with low saturation is reproduced. Therefore, it is necessary to perform color correction to reproduce a desired color.

Conventionally, in the field of hard copy, a method called masking has been used as this color correction.
Of the masking methods, the most commonly used one is to express the ink density signals (Y, M, C) as the complementary colors of the three primary color luminance signals (R, G, B) as shown in equation (1). This is called linear masking that is determined by a linear matrix operation for signals (D _R , D _G , D _B ).

[0005]

[Equation 1]

[0006] Linear masking is an additive law of density (Lambert-Beer) in color reproduction using real ink.
However, in the case of a color reproduction using an actual ink, for example, in the case of a sublimation-type thermal transfer recording system printer, the ink re-sublimation phenomenon, It is known that there are various nonlinear factors such as internal reflection of ink, and strictly speaking, the additive law and the proportional law do not hold.

Therefore, there has been proposed a non-linear high-order masking in which the ink density signal (Y, M, C) is determined by a high-order polynomial with respect to the three primary color density signals (D _R , D _G , D _B ). The simplest quadratic masking equation is shown in equation (2).

[0008]

[Equation 2]

This is to perform color correction by a quadratic equation including a non-linear factor existing in color reproduction using actual ink, and 27 correction coefficients a0 to a26 are determined by the least squares method regarding the density difference. This is used (for example, "Image processing for color reproduction", "Imaging Part 1", separate photo industry).

Further, there is a problem of color reproduction range in color reproduction in hard copy. The density range that can be recorded by the printer is equal to or lower than the maximum recording density peculiar to the apparatus and is equal to or higher than the surface density of the image receiving paper used for recording. The reproducible color reproduction range is limited by the spectral absorption characteristics of the real ink in which the recordable density range and the unnecessary absorption component exist, and the color reproduction range of the printer is generally narrower than that of the CRT.

FIG. 12 shows an example of the color reproduction range. FIG. 12 is a three-sided view showing the color reproduction range of the CRT and the color reproduction range of the printer in the L ^* u ^* v ^* system uniform color space recommended by the International Commission on Illumination CIE. Figure 12 (a) shows u ^*
In the v ^* plane, FIG. 12 (b) is in the L ^* u ^* plane, and in FIG.
(C) is a diagram in which each color reproduction range is projected on the L ^* v ^* plane. The color reproduction range of the printer uses the ink having the spectral absorption characteristics shown in FIG.
The color reproduction range of the RT is that of an NTSC CRT.

Thus, the color reproduction range of the printer is CR
Since it is narrower than the color reproduction range of T, a signal requesting a color exceeding the color reproduction range of the printer may be input as an input signal to be recorded. In that case, at least one color signal of the ink density signals (Y, M, C) obtained as a result of the above-described linear masking or non-linear higher-order masking is a density signal that cannot be recorded by the printer, that is, the paper surface density. There will be lower or higher than maximum concentrations.

In the prior art, when an ink density signal lower than the paper surface density is requested for this non-reproducible ink density signal, an ink density signal exceeding the maximum density is requested for the paper surface density. Was recording to the maximum density with a limiter.

[0014]

However, since the ink density signal and the color perceived by humans are in a non-linear relationship, the limiter for the ink density signal is not optimal in terms of color reproduction. .

FIG. 13 shows an example of color reproduction when a limiter is applied to the ink density signal of the masking calculation result. In FIG. 13, Pi (i = 1 to 3) is a target color represented by the input signal and is an input signal that cannot be reproduced by the printer. Qi (i = 1 to 3) represents a color reproduced when a non-recordable ink density signal among the ink density signals of the masking calculation result is subjected to a limiter for the paper surface density and the maximum density. It is a thing.

As in this example, in the prior art, when the target color corresponding to the input signal exceeds the color reproduction range of the printer, there are colors that humans feel more preferable among the colors that can be reproduced by the printer. Nevertheless, there is a problem in that the image quality may be deteriorated by reproducing largely different colors.

In view of the above, the present invention performs faithful color reproduction for an input signal reproducible by a printer, and human beings among colors reproducible for an input signal not reproducible by a printer. A color image forming method capable of performing the color reproduction that is most preferable, and suppressing a rapid change in color reproduction even when recording an image in which the input signal continuously changes with a color that cannot be reproduced by the printer It is an object of the present invention to provide a color image forming method capable of reducing discontinuous color reproduction.

[0018]

In order to solve the above-mentioned problems, a color image forming method of the present invention performs an optimum color correction on a color that an input signal can be reproduced by a printer, and a first ink density signal ( Y1, M1, C1) first color correction calculation step, a judgment step of judging whether the input signal is a color reproducible by the printer or a color not reproducible by the printer, and the judgment step. The color reproduction of the printer when a recordable ink density signal is used is predicted for the input signal which is judged not to be reproducible by the printer, and the input signal and the color are calculated using different evaluation functions according to the input signal. By calculating the evaluation value from both the reproduction prediction and searching for the ink density signal with the best evaluation value, the optimum color among the colors that can be reproduced by the printer is reproduced for the input signal that cannot be reproduced by the printer. Second ink to do A second color correction calculation step for converting the input signal into a color signal (Y2, M2, C2), and if the input signal is a reproducible color according to the result of the determination step, the first color correction calculation step is performed. Ink density signals (Y1, M1, C1)
If the input signal has a color that cannot be reproduced, then the second
Ink density signals (Y2, M2, C2) are used to control the ink density to form a color image.

Further, in the color image forming method of the present invention, the evaluation function in the second color correction operation uses the information regarding the difference between the color represented by the input signal and the color reproduction prediction of the printer regarding the brightness, saturation and hue. In addition, information obtained by multiplying each weighting coefficient is used, and the weighting coefficient is changed according to an input signal.

Further, the color image forming method of the present invention is
A first color correction calculation step of performing optimum color correction on a color that can be reproduced by a printer and converting it into a first ink density signal (Y1, M1, C1), and reproducing the input signal by the printer. A determination step of determining whether the color is a color that can be reproduced or a color that cannot be reproduced, and an input signal that the determination step determines that the color cannot be reproduced by the printer are the optimum colors that can be reproduced by the printer. A second color correction calculation step for converting into a second ink density signal (Y2, M2, C2) for reproducing a color, and the input signal is a reproducible color according to the result of the judgment step. In this case, the first ink density signal (Y1, M1, C1) is used, and when the input signal is a color that cannot be reproduced, the second ink density signal (Y2, M1) is used.
2, C2) to determine the third ink density signal (Y3, M3, C3), and the third ink density signal (Y3, M3) corresponding to the input signal of interest.
3, C3) by subjecting the input signal located around the input signal of interest in the color space of the input signal to the neighborhood operation using the third ink density signal (Y3, M3, C3). Fourth ink density signal (Y
4, M4, C4), and the ink density is controlled by using the fourth ink density signal (Y4, M4, C4) to form a color image.

[0021]

With the above construction, the second color correction calculation for obtaining the optimum ink density signal with the color reproducible by the printer is performed for the input signal which cannot be reproduced by the printer. An evaluation function that determines the optimality in the color correction calculation is used in accordance with the input signal, and an evaluation function that determines the ink density signal for performing the color reproduction most perceived by humans according to the input signal is By using it, it becomes possible to reproduce the color that humans feel most preferable for all input signals.

Further, as the evaluation function, the information obtained by multiplying the information using the difference in lightness, saturation and hue between the color represented by the input signal and the reproduced color of the printer by each weighting coefficient is used. By changing this weighting coefficient in accordance with the input signal, the three attributes of color, lightness, saturation,
It is possible to continuously change which of the hues is to be emphasized and the degree of importance to be emphasized in accordance with the input signal.

Further, the third ink density signal obtained through the first color correction calculation step and the second color correction calculation step is an input located around the input signal of interest in the input color space consisting of the input signals. By performing the neighborhood calculation using the third ink density signal for the signal, it is possible to mitigate the abrupt change of the ink density signal due to the change of the input signal and maintain the smoothness of the gradation of the reproduced image. Becomes

[0024]

EXAMPLE A first example relating to the color image forming method of the present invention will be described below using a sublimation type thermal transfer type full color printer using three color inks of yellow, magenta and cyan, and three primary colors for driving a CRT. Luminance signal (R, G,
An example in which the ink density signals (Y, M, C) for B) are determined by software will be described.

FIG. 2 shows a block diagram of the experimental apparatus used in the first embodiment. In FIG. 2, 201 is a CPU that executes the color image forming method of this embodiment, and 202 is a CP.
U 201 is a RAM used as a work area when executing a program, 203 is a ROM that stores programs executed by the CPU 201, 204 is an input signal (R, G, B) to be recorded, and an ink density signal is also input. I / O means for outputting (C, M, Y), 205 is CP
U201, RAM202, ROM203, I / O means 2
A bus for connecting 04 to each other, and 206 for I / O means 204
The control means 207 controls the applied energy according to the ink density signals (Y, M, C) output from the control means 207, applies heat to the ink sheet (not shown) according to the applied energy controlled by the control means 206, It is a thermal head that forms a color image on an image receiving paper (not shown).

In the experimental apparatus thus constructed,
In the first embodiment, the color image forming method is executed by software. The overall flow of processing executed by the CPU 201 at this time is shown in FIG. 1, and more detailed flow of processing is shown in FIGS. 3 and 5.

First, the overall processing procedure will be described with reference to FIG. In step 101, input signals (R, G, B) to be recorded are input via the I / O means 204.
In step 102, a first color correction calculation for optimally correcting a color that can be reproduced by the printer is performed to obtain a first ink density signal (Y1, M1, C1).

In step 103, the first ink density signal (Y1, M1, C1) is used to input the input signal (R,
G, B) is a color that can be reproduced by the printer, but it is judged that the color cannot be reproduced. If it is judged that the color can be reproduced, the flg signal representing the judgment result is set to flg =
0, when it is determined that the color cannot be reproduced, flg =
Set to 1.

In step 104, if the input signal (R, G, B) is a reproducible color (flg = 0) according to the determination result of step 103, the process proceeds to step 106, and the input signal (R, G, B) is input. If B) is a non-reproducible color (flg = 1), the process branches to step 105.

In step 105, for the input signals (R, G, B) that cannot be reproduced by the printer, it is possible to perform the optimum color reproduction of the colors that can be reproduced by the printer.
The second color correction calculation for obtaining the ink density signals (Y2, M2, C2) of is performed.

In step 106, the first ink density signal (Y1, M1, C1) or the second ink density signal (Y2, M2, C2) is used for printing in accordance with the flg signal. It is output from the I / O means 204 as M, C).

Then, in FIG. 2, the I / O means 204
The control unit 206 controls the amount of heat of the thermal head 207 in the order of yellow, magenta, and cyan in accordance with the ink density signals (Y, M, and C) output from the printer, so that gradation recording is performed on an image receiving paper (not shown). And a color image is formed.

Subsequently, the first executed in step 102
The color correction calculation will be described. In this embodiment,
The first operation that combines a linear matrix operation for a luminance signal and a non-linear masking operation for a density signal
It was used as a color correction calculation of.

The detailed processing will be described below.
First, with respect to the input signals (R, G, B), the deviation between the central wavelength of the spectral characteristics of the CRT phosphor, which is the target of color reproduction of the printer, and the central wavelength of the spectral absorption characteristics of the ink used in the printer, is calculated. For the purpose of correction, the luminance matrix calculation of the equation (3) is performed to obtain the second luminance signal (R ′, G ′,
B ').

[0035]

[Equation 3]

Next, (4) the second luminance signal by the complementary color conversion (R ', G', B ') 3 primary color density signals (D _R which respectively based on the subtractive color mixing principle of, D _G, D _B ).

[0037]

[Equation 4]

Then, a non-linear masking operation is performed for the purpose of correcting color turbidity due to unnecessary absorption components of the ink. First, the density signal is converted into a signal corresponding to the amount of ink color material. Specifically, letting a (a> 1) be a constant representing the non-linear degree of the relationship between the color material amount of ink and the density,
Each of the three primary color density signals (D _R , D _G , D _B ) obtained by the complementary color conversion by the first non-linear conversion shown in the equation (5),
Non-linear conversion into signals (C ′, M ′, Y ′) corresponding to the amount of color material of ink.

[0039]

[Equation 5]

Next, the output (C ', M', Y ') of the first non-linear conversion is subjected to the second non-linear signal (C ", M", Y "by the linear matrix operation of the equation (6). ).

[0041]

[Equation 6]

Further, each of (C ″, M ″, Y ″), which is an inverse function of the first non-linear conversion and obtained by the second non-linear conversion shown in the equation (7), is the first ink density. It is converted into a signal (Y1, M1, C1).

[0043]

[Equation 7]

In this embodiment, the CRT and the printer for a color chart of about 100 colors, which are approximately evenly located in the color reproduction range reproducible by the printer, are arranged so that the color correction for the input signal reproducible by the printer works optimally. In order to minimize the average value of the color reproduction error, {b _kl } in the luminance matrix calculation of the formula (3), {a _kl } (k = 1 to 3, l = in the linear matrix calculation of the formula (6)). 1 to 3), and the correction coefficient a in the equations (5) and (7) was determined by the convergence calculation. In the present embodiment, each correction coefficient used is shown in equation (8).

[0045]

[Equation 8]

In the first color correction calculation used in this embodiment, the average color difference Euv in the L ^* u ^* v ^* system uniform color space is Euv = 4.3 for the color reproducible by the printer. It was possible to make a highly accurate correction.

Next, a detailed description will be given of the processing of step 103 for determining whether the input signal (R, G, B) is a color that can be reproduced by the printer or a color that cannot be reproduced. Reproduction of the printer in a color space (hereinafter referred to as an ink density space) in which the density signals of the yellow (Y), magenta (M), and cyan (C) inks used in the printer are shown in FIG. 4 as axes of the orthogonal coordinate system. Indicates possible areas. Figure 4
4A is a perspective view of the ink density space, FIG. 4B is a view of the YC plane viewed from above the M axis, and FIG. 4C is M from above the Y axis.
A view of the C plane, and FIG. 4D is a view of the YM plane viewed from above the C axis.

As shown in FIG. 4, in the ink density space, the area that can be reproduced by the printer is represented by the density that can be recorded using each ink. That is, the paper surface densities of yellow, magenta, cyan, and each color are (Y0, M0,
C0), the maximum recordable density is (Ymax, Mmax,
Cmax), the reproducible area of the printer in the ink density space is Y = Y0, M = M0, C = C0, Y = Y
The area is represented by a rectangular parallelepiped surrounded by six surfaces of max, M = Mmax, and C = Cmax.

From this, when the first color correction calculation is a calculation for performing the optimum color correction for the color reproducible by the printer, the input signals (R, G, B) can be reproduced by the printer. In order to determine whether the color is a non-reproducible color or a non-reproducible color, the first ink density signal (Y1, M1,
Each of C1) is Y0 ≦ Y1 ≦ Ymax, M0 ≦ M
The determination can be made by checking that 1 ≦ Mmax and C0 ≦ C1 ≦ Cmax. In step 103, it is checked whether or not each of the first ink density signals (Y1, M1, C1) is higher than or equal to the paper surface density and lower than or equal to the maximum density.

FIG. 3 shows the detailed processing flow of step 103. In FIG. 3, in step 301, it is checked whether or not the yellow ink density signal Y1 of the first ink density signals (Y1, M1, C1) is in the range of the paper density or more and the maximum density or less. If it is not within the range, it is determined that the input signal is a color that cannot be reproduced, and flg = 1 is set in step 305.

In steps 302 and 303, the same comparison is performed with respect to the magenta ink density signal M1 and the cyan ink density signal C1 of the first ink density signal (Y1, M1, C1), respectively.

Then, the first ink density signal (Y1, M
1, C1) are all recordable density signals,
Assuming that the input signals (R, G, B) are reproducible colors, flg = 0 is set in step 304.

As described above, in this embodiment, the first ink density signal (Y1, M1,
C1) is used to determine whether the input signal (R, G, B) is a color that can be reproduced by the printer or a color that cannot be reproduced.

Next, the second color correction calculation in step 105 will be described. In the second color correction calculation in this embodiment, the color of the input signal (R, G, B), that is, the target color for color reproduction of the printer is obtained, the evaluation function is determined according to the target color, and the set ink is set. The color reproduction of the printer when the density signal is used is predicted, and the ink density signal having the optimum evaluation value calculated from the target color and the predicted value of the color reproduction is obtained.

FIG. 5 shows a detailed processing flow of the second color correction calculation. First, in step 501, input signals (R, G,
B) is converted into a color signal of a target color for color reproduction of the printer.
In this embodiment, three primary color luminance signals (R, G, B) for driving a CRT are used as an input signal, and NTSC system C is used.
The tristimulus values (Xo, Yo, Zo) reproduced when the three primary color luminance signals were input to the RT were obtained by the output equation (9) of the NTSC CRT.

[0056]

[Equation 9]

Furthermore, considering the human visual characteristics, 3
The extreme value (Xo, Yo, Zo) is L by the formula (10).^* u^* 
v^* System uniform color space coordinates (Lo^* , Uo^* , Vo^* ) To
Converted Here, Equation (10) is the International Lighting Commission CIE
L recommended by^* u^* v ^* Conversion formula to system uniform color space
Is.

[0058]

[Equation 10]

In step 502, an evaluation function for evaluating the optimality of the color reproduced by the printer is determined according to the input signal. Next, in step 503, the value of the ink density signal is set in order to search for the optimum ink density signal.

In step 504, the color reproduction (Li ^* , ui ^* , vi ^* ) of the printer when the ink density signal set in step 503 is used is predicted. In step 505, the target color of the color reproduction obtained in step 501 (Lo
^* , Uo ^* , vo ^* ) and the color reproduction prediction value (Li ^* , ui ^* , vi ^* ) obtained in step 504, based on the evaluation function determined in step 502. Obtain an evaluation value for a signal.

In step 506, it is determined whether the evaluation value obtained in step 505 is the minimum, and if it is determined that it is the minimum, the ink density signal set in step 503 is used as the second ink density signal (Y2, It is stored as M2, C2).

Then, in step 508, it is judged whether or not the search for the ink density signal is completed. If the search is not completed, the process branches to step 503 to set a new ink density signal, and step 503. ~ Step 50
8 is repeated.

Next, the prediction of color reproduction for the set ink density signal in step 504 will be described.
The first color correction calculation used in this embodiment includes nonlinear conversion such as equations (4), (5), and (7), and is a non-linear operation as a whole. Since it is represented by the existing function, it is possible to perform the conversion from the ink density signal to the three primary color luminance signals by the inverse operation of the first color correction operation. Therefore, in the present embodiment, the inverse operation of the first color correction operation is used to convert the set ink density signal into the three primary color luminance signals which are the color space of the input signal, and L ^* is obtained as in the case of obtaining the target color ^. Converted to color signals in the u ^* v ^* uniform color space.

First, the set ink density signal (Y2, M
2, C2) is transformed into (Y2 ″, M2 ″, C2 ″) by subjecting it to the nonlinear transformation of the equation (11).

[0065]

[Equation 11]

Then, (Y2 ″, M2 ″, C2 ″) is changed to (Y) by the equation (12) which is the inverse matrix operation of the equation (6).
2 ', M2', C2 ').

[0067]

[Equation 12]

Then, by the non-linear conversion of the equation (13),
The three primary color density signals (D2 _R , D2 _G , D2 _B ) corresponding to the set second ink density signals (Y2, M2, C2) are converted.

[0069]

[Equation 13]

Further, by the inverse complementary color conversion of the equation (14), the luminance signals (R2 ', B2', C2 ') of additive color mixture are converted.

[0071]

[Equation 14]

Then, it is converted into three primary color luminance signals (R2, G2, B2) by the inverse luminance matrix calculation of the equation (15).

[0073]

[Equation 15]

Furthermore, the three primary color luminance signals (R2, G2, B2) obtained from the ink density signals are input signals (R, G,
As in the case of B), conversion to tristimulus values (Xi, Yi, Zi) displayed on the CRT according to equation (9), coordinates (Li ^* , ui ^* , vi) of the uniform color space according to equation (10). By performing conversion to ^* ), color reproduction is predicted when the set ink density signal is used.

As described above, the first color correction calculation used in this embodiment is L ^* for the color reproducible by the printer ^.
The average color difference Euv in the u ^* v ^* system uniform color space is Euv =
A highly accurate correction of 4.3 is possible, and the color reproduction prediction in the present embodiment using the inverse function of the first color correction operation is also highly accurate in the average color difference Euv = 4.3. It was possible.

Next, the evaluation function for evaluating the optimality of the reproduced color of the printer in this embodiment will be described. The evaluation function in this embodiment has the target color (Lo ^* , uo ^* , vo ^* ) and the reproduced color (L
i ^* , ui ^* , vi ^* ) lightness (Lo ^* , Li ^* ), hue (θo, θi), and saturation (So, Si) squared are multiplied by a weighting coefficient. Then, the weighting coefficients a, b, and c are changed according to the input signal to set the evaluation function E according to the input signal. Note that the saturation and hue here are the saturation and hue in the L ^* u ^* v ^* system uniform color space. Regarding the hue, the difference between the hues of the two colors is set to 2 in order to align the unit with the brightness and the saturation. A value obtained by multiplying the average value of color saturation was used.

[0077]

[Equation 16]

Then, an evaluation function suitable for the target color represented by the input signal was determined by a subjective evaluation experiment. Red (R), green (G), as colors that cannot be reproduced by the printer,
In the input color space consisting of each blue (B) luminance signal, the CRT reproduction color when the CRT is driven by 26 input signals located on the wall surface of the CRT reproduction range is set as the target color, and several evaluation functions are used. A color chart is created by the printer using the ink density signal determined by using each color chart, each color chart is compared with the target color, and the color chart that is felt to be optimal for each target color is selected. An evaluation function suitable for the color was determined. As the evaluation function, three types of evaluation functions were used: an evaluation function that emphasized lightness reproduction, an evaluation function that emphasized hue reproduction, and an evaluation function that equally emphasized lightness, hue, and saturation.
Specifically, in the evaluation function E of the equation (16), the lightness weighting evaluation function has a lightness weighting coefficient a, and the hue weighting evaluation function has a hue weighting coefficient b that is larger than the other weighting values. In the evaluation function that equally attaches importance to hue and saturation, each weighting coefficient has the same value.

FIG. 6 shows an evaluation function determined to be optimum for each input signal determined by subjective evaluation. FIG. 6 shows 26 input signals selected as the target colors of the subjective evaluation experiment in the input color space in which the luminance signals of red (R), green (G), and blue (B) are represented as axes of the orthogonal coordinate system. On the other hand, FIG. 6A shows an evaluation function determined to be optimum, FIG. 6A is a diagram of the CRT reproduction range viewed from the white (W) side, and FIG. 6B is a CRT reproduction range. It is the figure seen from the black (Bk) side which is the origin of the input color space. For example, when the target color is blue (B) of CRT, the color chart created by using the ink density signal determined by the evaluation function in which each weighting coefficient has the same value is most preferable, and the target color is CRT. In the case of green (G), the color chart created using the ink density signal determined by the evaluation function that emphasizes the lightness feels most preferable, and when the target color is red (R) of CRT, the hue is emphasized. The color chart created using the ink density signal determined by the evaluation function felt most preferable. Further, in FIG. 6, CR indicated by a black circle
For white (W) and black (Bk) of T, the same ink density signal (YM for W is used) regardless of which evaluation function is used.
(C is the density on the paper, and Bk is the maximum density on YMC)
was gotten. In this embodiment, as shown in FIG. 6, the ink density signal for each input signal is determined using an evaluation function in which weighting factors for lightness, hue, and saturation are continuously changed according to the input signal.

Next, an example of the color reproduction experiment result using the ink density signal determined by the color image forming method in the first embodiment will be shown. FIG. 7 shows C as an input signal in which the color reproduction of the printer is impossible in the L ^* u ^* v ^* system uniform color space.
The results of color reproduction of the printer for red, green, and blue of RT and each input signal are shown. In FIG. 7, P4 and R4 represent red target color and printer reproduced color, P5 and R5 represent green target color and printer reproduced color, and P6 and R6 represent blue target color and printer reproduced color, respectively. Is. As described above, for the red input signal, the ink density signal is determined by using the hue-focused evaluation function, so that the target color and the hue of the colors that can be reproduced by the printer are almost the same. Was done. Similarly, for a green input signal, color reproduction with almost the same lightness as the target color is performed, and for a blue input signal, among the colors that can be reproduced by the printer, the target color and the uniform color space are reproduced. Color reproduction was performed to reduce the distance.

As described above, the first color image forming method of the present invention
With regard to the embodiment, the configuration and operation of the experimental apparatus, the process of determining the ink density signal with respect to the input signal, the evaluation function suitable for the input signal, and the experimental result have been described. As described above, in the present embodiment, the ink density signal is determined by using the evaluation function determined by the subjective evaluation experiment according to the input signal, so that the color reproduction that humans feel is most preferable compared with the target color is performed. , It has become possible to greatly improve the image quality.

Although the evaluation function shown in FIG. 6 is used in this embodiment, the color reproduction range of the printer varies depending on the recording method of the printer and the spectral characteristics of the ink used, and therefore the target The evaluation function suitable for the color may have different results depending on the printer, and the present invention is not limited to this evaluation function.

Next, a second embodiment of the color image forming method of the present invention will be described. The second embodiment of the present invention is the first
The experiment was performed using the same experimental apparatus as in the above example. In the second embodiment, the upper 5 bi for each color of the input signal with 8-bit precision are used.
The ink density signals (C, M, Y) corresponding to the 32 × 32 × 32 discrete representative points given by t are previously L
RAM of FIG. 2 as UT (Look Up Table)
The ink density signal for the input signal, which is stored in 202 and recorded in the middle of the representative point, is determined by an 8-point interpolation method which is a three-dimensional linear interpolation method.
Further, in the color image forming method of this embodiment, the ink density signal for the input signal of 32 × 32 × 32 discrete representative points including the input signal which cannot be reproduced by the printer is determined, and the input color space is further determined. In (1), the ink density signal used for recording is determined by subjecting the neighborhood calculation.

The entire table creating process in this embodiment will be described with reference to FIG. In order to determine ink density signals for all input signals of 32 × 32 × 32 discrete representative points, first, in step 801, input signals (R, G,
Set B).

In step 802, a first color correction calculation is carried out for optimum color correction for colors that can be reproduced by the printer for the set input signals (R, G, B), and the first ink The density signals (Y1, M1, C1) are obtained. The first color correction calculation in this embodiment is similar to the first color correction calculation in the first embodiment in that it performs a linear matrix calculation on a luminance signal, a complementary color calculation, and a non-linear masking calculation on a density signal. is there.

In step 803, each of the first ink density signals (Y1, M1, C1) of the first color correction calculation result is Y0≤Y1≤Ymax, M0≤M1≤Mm.
By checking that ax and C0 ≦ C1 ≦ Cmax, it is judged whether the input signal (R, G, B) is a color reproducible by the printer or an unreproducible color, and reproducible. If it is determined that the color is a color, the flg signal representing the determination result is set to flg = 0, and if it is determined that the color is an unreproducible color, flg = 1 is set.

In step 804, if the input signal (R, G, B) is a reproducible color (flg = 0), in step 806, the input signal (R, G, If B) is a non-reproducible color (flg = 1), the process branches to step 805.

In step 805, for the input signals (R, G, B) that cannot be reproduced by the printer, the optimum color reproduction can be performed among the colors that can be reproduced by the printer.
The second color correction calculation for obtaining the ink density signals (Y2, M2, C2) of is performed. Similarly to the first embodiment, the second color correction calculation in this embodiment is to obtain an ink density signal having an optimum evaluation value calculated from a target color represented by an input signal and a predicted value of color reproduction. In the present embodiment, unlike the first embodiment, the distance in the L ^* u ^* v ^* system uniform color space is used as the evaluation function for all input signals at the representative points.

In step 806, either the first ink density signal (Y1, M1, C1) or the second ink density signal (Y2, M2, C2) is selected according to the flg signal, and the third ink density signal (Y1, M1, C2) is selected. Ink density signal (Y3, M3, C
3) is stored in the RAM 202.

In step 807, it is determined whether the processing for determining the third ink density signal has been performed for all the input signals at the representative points. If the processing for all the input signals of the representative points is not completed, the process branches to step 801, and the value represented by the upper 5 bits of each color of the input signal is increased to set the input signal to be processed next, The processing from 802 to step 806 is performed. Then, when the processing for all the input signals of the representative points is completed, the process branches to step 808.

After step 808, the third ink density signal (Y3, M3, C3) for the input signal of interest is output.
And the third ink density signal (Y3, M) for the input signal located around the input signal of interest in the input color space.
3, C3) is a process for performing a neighborhood calculation. In this embodiment, the neighborhood calculation is applied to the third ink density signals (Y3, M3, C3) corresponding to the input signal of the representative point.

In step 808, the input signals are set in order to perform the neighborhood calculation on the third ink density signals (Y3, M3, C3) corresponding to all the input signals of the representative points. At step 809, a third ink density signal (Y3, M3, C) corresponding to the input signal set at step 808 is obtained.
Neighbor calculation is performed in 3).

In step 810, the ink density signal subjected to the neighborhood calculation is converted into the fourth ink density signal (Y4, M4,
It is stored in the LUT on the RAM 202 as C4). In step 811, it is determined whether the third ink density signals (Y3, M3, C3) corresponding to all the input signals of the representative points have been subjected to the proximity calculation processing. If the processing is not completed, the process branches to step 808, and the upper 5b of each color of the input signal
By increasing the value represented by it, the input signal to be processed next is set, and the same processing is performed. Then, when the processing is completed for the third ink density signals (Y3, M3, C3) for all the input signals of the representative points, the table creation processing is completed.

At the time of recording, an input signal (R, G, B) with 8-bit precision to be recorded is input from the I / O means 204 of FIG. 2, and the CPU 201 outputs the higher order of the input signal (R, G, B). By referring to the LUT on the RAM 202 according to 5 bits, the input signals (R, G,
B) Obtaining a fourth ink density signal (Y4, M4, C4) for the input signal of the representative point represented by the upper 5 bits of B), and performing interpolation calculation using the lower 3 bits of the input signal (R, G, B). The ink density signal corresponding to the input signal (R, G, B) is determined by the above, and is output from the I / O means 204. The interpolation calculation in the present embodiment is performed by the fourth ink density signal (Y4) with respect to the input signals of the eight representative points located at the vertices of the unit cube including the input signals (R, G, B) in the input color space. , M4, C4)
And a conventional three-dimensional interpolation calculation using a weighting factor corresponding to the volume ratio selected by the lower 3 bits of the input signal.

Next, the neighborhood calculation in step 809 will be described. The third ink density signal corresponding to the input signal of each representative point is the first color correction calculation for the input signal recordable by the printer and the first color correction calculation for the input signal not reproducible by the printer as described above. Although any one of the results of the two color correction calculations is selected, the shape of the color reproduction range of the printer is determined because the optimum ink density signal is obtained for each input signal in the second color correction calculation. In some cases, in the case of an image in which the input signal changes continuously, the reproduced color changes abruptly and the smoothness of the gradation of the reproduced image is impaired. Therefore, the neighborhood calculation of this embodiment is a three-dimensional spatial filter calculation in the input color space in which each of the red (R), green (G), and blue (B) luminance signals is an axis of the orthogonal coordinate system. is there.

FIG. 9 shows a schematic diagram of a three-dimensional spatial filter in the input color space. FIG. 9A shows three-dimensionally the input signal of interest among the input signals of the representative points existing in the input color space, which should be subjected to the neighborhood calculation, and the input signal of the representative point used for the neighborhood calculation. The input signals of the representative points used for the neighborhood calculation are the input signals of the six representative points that are adjacent to each other in the R, G, and B axial directions of the input signal of the representative point of interest. The signal represented by the upper 5 bits of the input signal of the representative point of interest is (R0, G0, B0), and the input signals of the six representative points used for the neighborhood operation are (R0-1, G0,
B0), (R0, G0-1, B0), (R0, G0, B
0-1), (R0 + 1, G0, B0), (R0, G0 +
1, B0), (R0, G0, B0 + 1), for example, the upper 5 bits of each element of (R0, G0, B0) are 2
In the case of a base number (11011, 10111, 00110), the upper 5 bits of each element of the input signals of the six representative points used for the neighborhood calculation are shown in (Table 1).

[0097]

[Table 1]

FIGS. 9B and 9C show the third ink density signal for the input signal of interest and the third ink luminance signal for the input signal of the representative point used for the neighborhood calculation in the spatial filter calculation. Is shown in two drawings separately. Y of the third ink density signal,
Of the M and C elements, the Y signal corresponding to the input signal of interest is Y3 (R0, G0, B0), and the Y signal corresponding to the input signal at the representative point used for the six neighborhood operations is Y3 (R0-1,
G0, B0), Y3 (R0, G0-1, B0), Y3
(R0, G0, B0-1), Y3 (R0 + 1, G0, B
0), Y3 (R0, G0 + 1, B0), Y3 (R0, G
0, B0 + 1), in this embodiment, as shown in FIG.
As shown in (c), the fourth Y signal Y4 (R0, G0, B0) for the input signal of interest is determined by the calculation of equation (17).

[0099]

[Equation 17]

Similar calculations are performed for M and C of the Y, M and C elements of the third ink density signal to obtain the M and C elements of the fourth ink density signal. Next, the result of the color reproduction experiment when the color image forming method of this embodiment is applied will be described with reference to FIG. FIG. 10A shows the color reproduction range of the printer on the u ^* v ^* plane of the L ^* u ^* v ^* system uniform color space, and FIG. 10B shows the portion surrounded by the circle in FIG. 10A. It is the figure which expanded. FIG. 10B shows the target color of the color reproduction experiment and the reproduced color of the printer.
White circles are target colors and are colors that cannot be reproduced by the printer. The triangles represent the color reproduction of the printer when the third ink density signal before the neighborhood calculation of this embodiment is used,
The squares represent the color reproduction of the printer when the fourth ink density signal obtained by subjecting the third ink density signal to the neighborhood calculation is used. Further, the white circle of the target color and the triangle for color reproduction using the third ink density signal are connected by a wavy line, and the square for color reproduction using the fourth ink density signal is connected by a solid line. As shown in this figure, for a continuously changing target color, if the third ink density signal that does not perform neighborhood calculation is used, the color reproduction may change discontinuously. , The discontinuous color reproduction was alleviated when the fourth ink density signal subjected to the neighborhood calculation was used.

The second embodiment of the color image forming method of the present invention is as described above.
With regard to the embodiment of (1), the process of determining the ink density signal with respect to the input signal and the experimental results have been described. According to the color image forming method of the present embodiment, an optimum ink density signal is obtained for each input signal with respect to an input signal that cannot be reproduced by the printer,
Furthermore, by performing a three-dimensional neighborhood calculation in the input color space, it is possible to suppress a rapid change in color reproduction with respect to a continuously changing target color, and to mitigate discontinuous color reproduction.

In this embodiment, the calculation by the equation (17) is performed using the third ink density signals of the input signals of the six representative points as the neighborhood calculation. The number of the ink density signals of No. 3 and the coefficient of the spatial filter calculation are not limited to those of this embodiment.

In this embodiment, the upper 5 bi of the input signal
A neighborhood operation was performed on the third ink density signal corresponding to the input signal of the representative point represented by t, but the representative point b
The it accuracy is not limited. For example, by selecting the coefficient of the spatial filter calculation, the same effect can be obtained by subjecting the third ink density signal to all the input signals represented with 8-bit precision to the proximity calculation.

Next, the third embodiment of the color image forming method of the present invention.
The embodiment will be described. The third embodiment of the present invention is carried out by using the same experimental apparatus as the first embodiment, and as in the second embodiment, the upper 5 colors of each color of the input signal of 8-bit precision are obtained.
The ink density signals (C, M, Y) corresponding to 32.times.32.times.32 discrete representative points given by the bit are set in advance as a LUT (look-up table) in FIG.
When the data is stored in the AM 202 and recorded, the output for the input signal located in the middle of the representative point is determined by the 8-point interpolation method which is a three-dimensional linear interpolation method.

In the color image forming method of this embodiment, the ink density signal corresponding to the input signal of the representative point is determined, and the ink density signal corresponding to the input signal of interest is located in the input color space around the input signal of interest. The ink density signal used for recording is determined by subjecting the signal to a proximity calculation using the ink density signal.

In the third embodiment, the flow of the entire processing for creating the table of the fourth ink density signal for the input signal of the representative point is the same as the flow chart of FIG. 8 explained in the second embodiment. . That is, the result of either the first color correction calculation or the second color correction calculation is applied to all the input signals of the representative points, and the input signal is a color that can be reproduced by the printer, or a color that cannot be reproduced. It is selected according to the result of determination as to whether it is present, stored in the memory as a third ink density signal, further subjected to proximity calculation, and stored as a LUT in the RAM 202 as a fourth ink density signal.

However, in the second color correction calculation in the second embodiment, the evaluation function for evaluating the optimality is set to the distance of the L ^* u ^* v ^* system uniform color space with respect to the input signals of all the representative points. However, in the second color correction calculation in this embodiment, the evaluation function of the equation (16) is used as the evaluation function as in the first embodiment, and each weighting coefficient is changed according to the input signal. The concentration signal was determined.

When an image is formed by the printer using the color image forming method of this embodiment, it cannot be reproduced by the printer because the evaluation function is changed according to the input signal in the second color correction calculation. With respect to the input signal, it has become possible to perform color reproduction that humans feel is most preferable compared to the target color. In addition, the ink density signal corresponding to the input signal of interest is subjected to a proximity calculation using the ink density signals corresponding to the input signals located around the input signal of interest, so that the color of the input signal cannot be reproduced by the printer. It was possible to realize smooth gradation reproduction without causing a sharp gradation change even in the case of an image that continuously changes at.

The embodiments relating to the color image forming method of the present invention have been described above. Here, in these embodiments, the input signal is a luminance signal for driving the CRT, and the first color correction operation is an operation in which a linear matrix operation in the luminance signal and a non-linear masking operation in the density signal are combined. The first color correction calculation of the color image forming method of the invention is not limited to the above calculation. For example, in the case of the color correction calculation expressed by a function having an inverse function like the conventional linear masking, the color reproduction prediction for the ink density signal set in the second color correction calculation is performed by the first color correction calculation. It is possible to use the inverse operation.

Alternatively, if the inverse calculation of the first color correction calculation is not used for the color reproduction prediction, for example, the conventional nonlinear high-order masking is used as the first color correction calculation, and the input signal to the ink density signal is changed. It is also possible to express the conversion by conventional linear masking and perform color reproduction prediction for the ink density signal set by using the inverse function of this linear masking. In this case, non-linear high-order masking can be performed with high accuracy in color correction for input signals that can be reproduced by the printer, and linear masking accuracy can be adjusted for input signals that cannot be reproduced by the printer. It is possible to perform optimum color reproduction in the color reproduction prediction.

Although the sublimation type thermal transfer recording type color printer is used in this embodiment, it is obvious that the color image forming method and the color image forming apparatus of the present invention are not limited to the recording type of the printer. is there.

[0112]

As described above, according to the present invention, for the input signal reproducible by the printer, the first ink density signal obtained by the first color correction calculation is changed to the input signal not reproducible by the printer. In contrast, the second ink density signal obtained by the second color correction calculation is used as the ink density used for recording, and the evaluation function for determining the optimum color reproduction in the second color correction calculation is changed according to the input signal. By doing so, it is possible to perform color reproduction using a color that humans feel most preferable among colors that can be reproduced by the printer for all input signals, and it is possible to greatly improve deterioration of image quality. .

Further, as an evaluation function for an input signal which cannot be reproduced by the printer, each weighting coefficient is added to the information using the difference in brightness, saturation and hue between the color represented by the input signal and the color reproduction prediction of the printer. By using the multiplied information and judging the optimality by continuously changing each weighting coefficient according to the input signal, the lightness which is the three attributes of the color,
It is possible to reflect which of the hue and the saturation is emphasized in the color reproduction, and it is possible to smoothly change the evaluation function with respect to the change of the input signal.

Further, the first ink density signal by the first color correction calculation is applied to the input signal reproducible by the printer, and the second ink color correction operation is applied to the input signal not reproducible by the printer. The fourth ink obtained by selecting each of the second ink density signals according to (3) as the third ink density signal and subjecting the third ink density signal to a three-dimensional neighborhood operation in the input color space. By using the density signal for recording, even when recording an image in which the input signal changes continuously with colors that cannot be reproduced by the printer, abrupt changes in color reproduction are suppressed and discontinuous color reproduction is achieved. It becomes possible to relax.

In the second color correction calculation, the evaluation function is changed according to the input signal, and the fourth ink density signal obtained by performing the proximity calculation on the third ink density signal is used for recording. It is possible to achieve both color reproduction that humans feel most preferable for all input signals and smooth gradation reproduction for images in which the input signals continuously change with colors that cannot be reproduced by the printer. Therefore, it is possible to significantly improve the image quality of the reproduced image.

[Brief description of drawings]

FIG. 1 is a flowchart showing the overall flow of processing of a color image forming method according to a first embodiment of the present invention.

FIG. 2 is a block configuration diagram of an experimental device used in an example of the present invention.

FIG. 3 is a detailed flowchart for determining whether the input signal is a reproducible color or an unreproducible color in the embodiment.

FIG. 4 is a diagram showing a region that can be reproduced by a printer in an ink density space.

FIG. 5 is a detailed flowchart of a second color correction calculation for an unreproducible color in the embodiment.

FIG. 6 is a diagram showing an evaluation function according to an input signal in a second color correction calculation in the same embodiment.

FIG. 7 is a diagram showing a color reproduction experiment result in a uniform color space in the example.

FIG. 8 is a flowchart showing a process of creating a table of ink density signals for input signals in the color image forming method according to the second embodiment of the present invention.

FIG. 9 is a schematic diagram illustrating neighborhood calculation in the same embodiment.

FIG. 10 is a diagram showing a color reproduction experiment result in a uniform color space in the example.

FIG. 11 is a diagram showing an example of spectral absorption characteristics of ink used in a sublimation thermal transfer recording type printer.

FIG. 12 is a diagram showing an example of a color reproduction range of a CRT and a printer.

FIG. 13 is a diagram showing color reproduction by a conventional color image forming method when an input signal is a color that cannot be reproduced by a printer.

[Explanation of symbols]

 201 CPU 202 RAM 203 ROM 204 I / O Means 205 Bus 206 Control Means 207 Thermal Head

[Procedure amendment]

[Submission date] January 12, 1994

[Procedure Amendment 1]

[Document name to be corrected] Drawing

[Name of item to be corrected] Fig. 4

[Correction method] Change

[Correction content]

[Figure 4]

[Procedure Amendment 2]

[Document name to be corrected] Drawing

[Name of item to be corrected] Figure 7

[Correction method] Change

[Correction content]

[Figure 7]

[Procedure 3]

[Document name to be corrected] Drawing

[Correction target item name] Figure 9

[Correction method] Change

[Correction content]

[Figure 9]

[Procedure amendment 4]

[Document name to be corrected] Drawing

[Name of item to be corrected] Fig. 10

[Correction method] Change

[Correction content]

[Figure 10]

[Procedure Amendment 5]

[Document name to be corrected] Drawing

[Name of item to be corrected] Fig. 12

[Correction method] Change

[Correction content]

[Fig. 12]

[Procedure correction 6]

[Document name to be corrected] Drawing

[Name of item to be corrected] Fig. 13

[Correction method] Change

[Correction content]

[Fig. 13]

Claims (4)

[Claims] 1. A first ink density signal (Y1,
M1, C1) first color correction calculation step, a judgment step of judging whether the input signal is a color reproducible by the printer or a color not reproducible by the printer, and a printer in the judgment step. The color reproduction of the printer when a recordable ink density signal is used is predicted for the input signal determined to be unreproducible by, and the input signal and the color reproduction are performed by using different evaluation functions according to the input signal. The second ink density signal (Y2, M2, which reproduces the optimum color among the colors reproducible by the printer is calculated by calculating the evaluation value from both predictions and searching for the ink density signal having the best evaluation value. C2) and a second color correction calculation step, and if the input signal is a reproducible color according to the result of the determination step, the first ink density signal (Y1, M1, C1) cannot reproduce the input signal Color image forming method, wherein the second ink density signals (Y2, M2, C2), to control the ink concentration using respectively, to form a color image when it is. 2. The evaluation function in the second color correction calculation is
The brightness of the color represented by the input signal and the color reproduction prediction of the printer,
Claims characterized in that information obtained by multiplying information using differences in saturation and hue by respective weighting factors are used, and the weighting factors are changed according to an input signal. The method for forming a color image according to item 1. 3. A first ink density signal (Y1,
M1 and C1), a first color correction calculation step, a judgment step for judging whether the input signal is a color reproducible by the printer or an unreproducible color, and the judgment step is a printer. A second color correction calculation step of converting an input signal determined to be unreproducible by the method into a second ink density signal (Y2, M2, C2) that reproduces an optimum color of colors reproducible by the printer. According to the result of the determination step, if the input signal has a reproducible color, the first ink density signal (Y1, M1, C1) cannot be reproduced by the input signal. If the color is a different color, the second ink density signal (Y2, M2, C2) is selected to determine the third ink density signal (Y3, M3, C3), and the input signal of interest is selected. The third ink density signal (Y3, M3 To C3), wherein for an input signal located around the input signal the interest in color space of the input signal a third ink density signal (Y3, M3,
The fourth ink density signal (Y4, M4, C4) is obtained by performing a neighborhood calculation using C3), and the ink density is controlled using the fourth ink density signal (Y4, M4, C4). A method for forming a color image, which comprises forming a color image. 4. The second color correction calculation predicts the color reproduction of the printer when a recordable ink density signal is used, and the input signal and the color reproduction are calculated by using different evaluation functions depending on the input signal. The second ink density signal (Y2, M2, which reproduces the optimum color among the colors reproducible by the printer is calculated by calculating the evaluation value from both predictions and searching for the ink density signal having the best evaluation value. The color image forming method according to claim 3, wherein the color image is converted into C2).