JPS6335078A - Color masking parameter determining device - Google Patents

Abstract

PURPOSE:To optimize color reproduction by providing a color distribution calculating means that calculates the statistic of distribution in a color space and a weight calculating means that calculates the weight for each color patch from the statistic and the color coordinates of color patches and making convergence calculation that minimizes the evaluation value. CONSTITUTION:A color distribution calculating means 4 calculates parameter muand S, and a weight calculating means 5 calculates the weight Wn for each color patch based on these parameters, and the results are stored in a weight storing means 6. A black quantity calculating means 8 finds Kmax from the density of three primary colors and [alJ] and calculates the black quantity Kn' according to a given thershold value T. An evaluation value calculating means 11 finds mesh point effective area rate cn', mn', yn', kn' from four main density signal Cn', Mn', Yn', Kn', and simulates three stimulus values (Xn', Yn', Zn') and reproduced colors (L*n', u*n', v*n') by Neugebauer equation. The evaluation value calculating means 11 evaluates the value E of (L*n', u*n', v*n') stored in a color patch storing means 1 and three primary colors and outputs it.

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial Application Field] The present invention is applicable to color plate making or color printers, in which the luminance signals of red (R), green (G), and blue (B) of an image are converted into cyan (C), The present invention relates to a device for determining parameters necessary for color conversion into main density signals of magenta (M), yellow (Y), and black (K) inks.

[Conventional technology]

In color plate making or color printers, for example, the three primary color luminance signals (
R, G, B) to enclose (C, M
, Y, K) When obtaining the main density signals of the inks, first the three primary color luminance signals (R, G.

B) to the three primary color density signals (D, , Dg, Db) using the formula (1
), determine the amount of black within the range of formula (2), and satisfy K≦KIT16.
)<=min (Dr+Dg, Db)
(2) Furthermore, conversion into main density signals (C', M', Y') is performed using a matrix (acj) as shown in equation (3). This (acJ) is called a color masking parameter.

By performing this conversion as accurately as possible, the colors printed with the amount of ink of the main density signals (C', M', Y') will be very similar to the original (R, G, B) colors. In order to make it visible, the color masking parameter (aIj) has conventionally been calculated by performing the least squares method on the main density.

Specifically, black ink is not used, and N (N is a natural number) known Cn, Mn, yn (n = 1°...,
N), and the three primary color densities (D rn , D gn , D b
For example, for cyan ink, e2=Σ(Cn an Drn ad2Dgn
at3 Dbn) aII that minimizes 2, a12
+ Find at3. This is obtained by solving simultaneous equations (5).

Similarly, for magenta ink and yellow ink, nine color masking parameters of (apl) can be obtained.

However, in the above method, the color masking parameters are affected by the set of colors of the color patches used, and are not optimal for the image actually converted. In addition, for evaluation where the final color difference is minimal when seen by the human eye, L II U Kyo V system or L * a *
Although a uniform color space such as the b* system is used, the minimization of e2 in equation (4) is a minimization of the main densities of the three primary color inks, so it cannot be used as a method for determining the optimal color masking parameter. was insufficient.

In order to solve this problem, a color masking parameter determination device has been proposed that takes into account the color distribution in the target image and minimizes the color difference seen by the human eye for each image through simulation. 60-122683
), this device will be briefly described in the third drawing. - Color values of virtual color patches uniformly selected in uniform color space (L'n+ u”n+ v”n) (n =
1 + '..., N) is the color patch storage means 1
is stored in The parameter storage means 2 also stores initial values of masking parameters (aIj). The color conversion means 3 converts the three primary colors (R,
, Glt, B, ) values into (L"l* "at "n) values, and the color distribution calculation means 4 converts each (L"z+ u"Q
, calculate the color distribution of the target image from the V'lli values.

The weight calculation means 5 calculates the weight Wn for each color patch from the obtained color distribution, and the result is stored in the weight storage means 6.

The brightness value calculation means 15 calculates the virtual color patch L"n+
"n+"n) value is converted into the three primary color values (Rn, an, Bn), and the masking calculation means 16 calculates each ink main density (Cn', M.',
Find Yn'). The evaluation value calculation means 17 (crl'
.

M, ', Yn') to ink effective area ratio (cn', m
fi'.

yn') and apply the Neugebauer equation to this to predict the color (L*□ + u”n' * v*□') that will be reproduced by printing with this ink combination.Evaluation values are for each E2 (Formula (6)
).

E2=ΣWn ((L”n' Hゝn) 2nbal + (u'n' -u”n) 2 + (v”□'-v
*n) 2) The control means 18 updates the masking parameter (acJ) stored in the parameter storage means 2 so as to minimize E2. By repeating the above steps, the optimal color masking parameters are determined by convergence calculation.

[Problem that the invention seeks to solve]

However, the above method makes the image cyan.

When reproducing with the three primary color inks of magenta and yellow, it is possible to determine the color masking parameters to optimize the color reproduction of the image, but as with normal printing, some of the masking parameters are replaced with black ink and the surrounding area is This effect is not taken into consideration when packaging printing is performed. Therefore, when enclosing printing is performed, the obtained color masking parameters are not necessarily optimal.

SUMMARY OF THE INVENTION An object of the present invention is to provide a color masking parameter determination device that solves the above problem by taking into consideration the process of generating the amount of black ink.

[Means for solving problems]

The color masking parameter determining device of the first invention includes:
Color patch storage means for storing color coordinate values of a plurality of virtual color patches uniformly selected in a uniform color space; Parameter storage means for storing nine color masking parameters; All or part of a target image. color conversion means for converting the three primary color values of the pixels into color coordinate values in a uniform color space; color distribution calculation means for calculating statistics of the distribution of the color coordinate values in the color space; weight calculation means for calculating weights for each color patch from color coordinate values; weight storage means for storing the calculated weights; density value calculation means for calculating three primary color density values from the color coordinate values of the color patches; a black amount calculation means for calculating a black main density from the three primary color density values and a given black generation threshold; a UCR calculation means for calculating the three primary color density values after undercolor removal from the black main density and the three primary color density values; a masking calculation means for performing masking calculation on the three primary color density values after removal of the undercolor using the color masking parameters to calculate the three primary color main densities; Reproduced color coordinate values in the color space are obtained, color differences with the color coordinate values stored in the color patch storage means are obtained, and evaluation values are determined using weights for each color patch stored in the weight storage means. It is characterized by comprising an evaluation value calculation means for calculating, and a control means for updating the color masking parameter so as to minimize the evaluation value, and determining the optimal color masking parameter for the target image by convergence calculation. shall be.

The color masking parameter determining device of the second invention includes:
A color patch storage means for storing color coordinate values of a plurality of virtual color patches uniformly selected in a uniform color space, a parameter storage means for storing nine color masking parameters, and a means for generating black ink. a black threshold storage means for storing a black threshold, which is a threshold; and a color conversion means for converting the three primary color values of all or some pixels of the target image into color coordinate values in a uniform color space; color distribution calculation means for calculating distribution statistics; weight calculation means for calculating weights for each color patch from the statistics and color coordinate values of the color patches; weight storage means for storing the calculated weights; density value calculation means for calculating three primary color density values from the color coordinate values of the color patch; black amount calculation means for calculating a black main density from the three primary color density values and the black threshold value; a JCR calculation means for calculating three primary color density values after removing the undercolor from the values; and a masking subtraction means for performing masking calculation on the three primary color density values after removing the undercolor using the color masking parameters to calculate the three primary color main densities of the ink. and, from the three primary color main densities of the inks and the black main density, calculate the reproduced color coordinate values in the uniform color space by this combination of inks, and obtain the color difference with the color coordinate values stored in the α means, which are written in the color patch. , and further the weight description 4
．． an evaluation value calculation means for calculating an evaluation value using weights for each color patch stored in the two means; and a control means for updating the color masking parameter and the black threshold so as to minimize the evaluation value. It is characterized by determining the optimal color masking parameters and black threshold for the target image through convergence calculation.

[Effect]

When color printing is performed by dot printing of cyan, magenta, yellow, and black surrounding inks, the effective area ratios of each ink are c' and m', respectively.

y', ni', the main density of each ink is C', .
M'.

It is assumed that Y' and N' are connected by the relationship shown in equation (7).

Reproduced colors by halftone printing are predicted by the Neugehauer equation of equation (8) using the CIE-1931XYZ system.

Here, α+ = (1-c') (1-m')
(1-y') (1-to') α2 = C'
(1'-m') (1-y') (1-ni')
α3 = (1-c") m' (t-y
' ) (1-to' ) α4=C"m' (1
-y'>(1-to') αs = (1-c'
) ('1-m') y'(1-ni'
)αs=c'(1m')y'(1'k')
αv = (1-c') m'Y' (1
-to' ) αB = c'm'y' (1
-to' ) αs = (1-c' ) (1
-m') (1-y') k"α+o=c
'(1-m') (1-y')k'αo=
(1-c') m'(1-y'
) k'α12=c'm' (1-y
')k'α13=(1-c' ) (1m'
)V'k'α14=c'(1-m')
y'k'α+s=(1-c' ) m' (
1-)")k'α16"C'm'y'
k' where (X+, Yt, Z+) are the measured (X, Y, Z) values for all combinations of printing and non-printing for each ink, and Y= for white. It is normalized to 1. Table 1 shows the correspondence between i and the printed ink.

Table 1 It can be considered that normally in the area where black ink is present, the influence of other inks does not appear on the color, and equation (8) can also be replaced by equation (9).

sound しZ′　−壽Z,　β.

Here, βI −α1. β2=α2. ..., β8=α
8. β, = .

At this time, βS−Σα holds true.

1=9 Thus, the main ink density C' of the surrounding of each color patch
, M', Y', if ' is given, the color is (
X', Y', Z').

Therefore, the color of the original color patch is virtually (R, G
, B), and the color masking parameter (
aU) is given, the main ink densities (C', M', Y') are determined by equations (1) to (3).

To determine whether the halftone-printed reproduced color is actually close to the original (R, G, B) color, convert both to coordinate values in a uniform color space. It can be evaluated by To obtain the optimal masking parameter for a certain target image, the color distribution in the target image is expressed as a weight for a color patch uniformly taken over the golden space, and the sum of squares of the color reproduction error including the weight E2 (formula (6)) is determined to minimize.

Since this method of least squares is not linear, it cannot be solved in a partial manner as in equation (5), but it can be solved sequentially using a numerical calculation algorithm called the hill-climbing method.

Therefore, regarding the color distribution of the target image, the average amount, μ2. Calculate μ3 and covariance matrix S.

Here, Σ indicates addition for all pixels, and M is the total number of Q=I pixels. Then, the color distribution of the target image is expressed by normal distribution approximation as shown in equation (11).

■ ・exp (--(X-μ)'S-1(X-μ)) where X= (L", u"l V")' 1 μ=(μm
.

μ2. μ3)t. p (L", u", v') is
Since (L*, u”, V”) indicates the color distribution density in the uniform color space, the weight Wn for each virtual color patch is p (L”n+U”□+
It is appropriate to use ``n''.

Note that the color represented by (X, Y, Z) is a uniform color space (L*
, uII, vII) is the formula (
12), the formula for converting (R, G, B) to (X, Y, Z'') is given by formula (13), for example, if (R, G, B) is for the NTSC color television system. .

Here, X Y X+15Y+32 u o' =0.201, V o' =0.4
61 Therefore, the virtual color patch is (L'n+
u'rl+v"n), n=1.2. . . . and then perform the inverse operation of equations (12) and (13) to obtain (Rn, On.

Bn), and use equations (1), (2), (3),
By (7), (Cn'+ rnfi', )' n
'+kn'), and from these calculate formula (8) or (9
), and the reproduced color ("n' * u") by equation (10)
Color difference evaluation is performed by determining Q' + V%'), and the masking parameters can be optimized based on this result. Appropriately, the virtual color patch set is based on the previously proposed technique (Japanese Patent Application No. 122683/1983).

Black bonds can be arbitrarily determined within the range of formula (2), but
For example, it can be determined using the threshold value T as follows.

Note that KmaX can be easily calculated using equation (2), but it should be taken into consideration that the values of C', M', and Y' obtained by masking after UCR (formula (3)) cannot be negative. Therefore, it is more accurate to determine by equation (15).

By determining a2+ + 322 + 323 T, the amount of black changes, and the amount of ink for each color patch changes accordingly as shown in equation (3). Therefore, the optimal masking parameter (acj) is determined for each given T.

Moreover, conversely, T can be used as a parameter to be optimized together with (au). In this case, the optimal masking parameter (acj) as well as the optimal T can be determined.

〔Example〕

FIG. 1 is a block diagram of an embodiment of the first invention constructed based on the principle of the color masking parameter determination method described above.

The L*, u*, v* values (L*n,
u''n, V''n) (n=1. . . +, N) are stored in the color patch storage means 1. The parameter storage means 2 stores appropriate color masking parameters (acj) (i=1.2,3; j=1) as initial values.
゜2.3) is stored.

The target image is scanned and the R, G, B values (Rρ, G
λ, Bx) (β = 1.
v*I2). Conversion formula (12), (1
3) is executed. The color distribution calculation means 4 executes equation (lO) and calculates parameters μ and S regarding the color distribution.

The weight calculation means 5 calculates equation (11) based on these parameters.
) is used to calculate the weight wn for each color patch, and the result is stored in the weight storage means 6.

At the stage where the above preparations are complete, the density value calculation means 7 calculates each of the virtual color patches (L*, , u*., V
*n), calculate the amount of equations (12) and (13),
Furthermore, the three primary color densities (Drn, Dg
n, Dbn). The black amount calculating means 8 calculates equation (2) or equation (1) from these three primary color densities and (au).
5) to obtain Km&X, and calculate the black amount Kn' according to the given T.

The UCR calculation means 9 (D rn , D gn , D
bn) and K. ’ after removing the undercolor from the three primary color densities (Drn
', Dgn'.

Dbn') is calculated according to equation (16).

The masking calculation means 10 was obtained here (Drn'
+ Dgn' l Dbn' ), the masking calculation is performed according to the current masking parameter (au) stored in the parameter storage means 2.

The calculation corresponds to equation (3), and the main concentration signal (Cn'.

Mn', Yn') are obtained.

The evaluation value calculation means 11 calculates four main concentration signals Cn', M
By inversely solving equation (7) from n', Y n', K,', the effective halftone area ratio cn'+ mn'+
Find yn' + kn' and use the Neugebauer equation (
The tristimulus value (Xn
', Yn', Z. '), and further according to equation (12) (L gn I , ulln /.

■Simulate *,, /). Evaluation value calculation means 11
is also stored in the color patch storage means 1 (
The evaluation value E2 of the three primary colors with L ``n+u''n+v''n) is evaluated and output according to equation (6).

The control means 12 repeatedly changes the masking parameter (a+jl) in the parameter storage means 2 from the obtained E2, executes a hill climbing method by numerical calculation, and obtains E2.
(au) is determined so that it takes the minimum value. Control means 1
2, the execution ends when it is determined that the improvement of the masking parameter (aU) has reached the limit according to a certain criterion, and the optimum masking parameter for the target image is obtained in the parameter storage means 2. Such a control means can be realized by a well-known electronic computer or the like.

Next, the second invention will be described in detail with reference to FIG. In the second invention, a black threshold storage means 14 is added to the configuration of the first invention. Therefore, the control means 13
The function is different from that of the control means 12 shown in FIG. The other parts are exactly the same as the first invention. That is, the operation is similar until the evaluation value E2 is obtained from the evaluation value calculation means 11.

In the second invention, the black threshold T (Equation (14)) is also subject to optimization, similar to the masking parameter (apl).

The black threshold storage means 14 stores an appropriate T at the start of operation.
Stores the initial value of. The black amount calculating means 8 calculates black ilK using this T. ′ is determined.

The control means 13 can change the black threshold T at the same time as the masking parameter (aul), and the second invention uses a numerical sequential calculation method such as the hill-climbing method to change the masking parameter (a+j) that becomes the minimum of the evaluation value E2. The black threshold T is determined at the same time.

In addition, in each of the above embodiments, Lg u
Although the *■* system was used, the present invention can be constructed in exactly the same manner using the L*a*b' system.

The relationship between the L*a*b8 system and the XYZ system is given by equation (17) under the same conditions as equation (12).

In addition, the relationship between ′ and ′ in equation (7) is stated assuming that the solid density of black ink is infinite, but if this approximation does not hold, then the solid density is ,
Formula (18) is used.

K'=-ffog(1-k(L-10-'k)l
(18) Furthermore, in each of the above embodiments, the color distribution calculation means 4 calculates the statistics μ and S for the colors of all pixels of the target image, but in reality, similar colors are distributed in neighboring pixels. Therefore, instead of scanning all pixels, l/m (m
is a natural number), it is sufficient for the purpose of the present invention to calculate the statistics for the pixels obtained by thinning out the pixels. , using virtual color patches and image data, it is possible to determine the optimal color masking parameter (a+j) for color reproduction of the target image during envelope printing, and in the second invention, the optimal color masking parameter ( aU) and the black threshold T can be determined.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing an embodiment of the first invention, FIG. 2 is a block diagram showing an embodiment of the second invention, and FIG. 3 is a block diagram showing a conventional example. ■...Color patch storage means 2...Parameter storage means 3...Color conversion means 4...Color distribution calculation means 5...Weight calculation means 6... ... Weight storage means 7 ... Density value calculation means 8 ... Black amount calculation means 9 ... OCR calculation means 10, 16 ... Masking calculation means 11, 17... Evaluation value calculation means 12, 13.18, control means 14, black threshold storage means 15, brightness value calculation means

Claims (2)

[Claims] (1) Color patch storage means for storing color coordinate values of a plurality of virtual color patches uniformly selected in a uniform color space, parameter storage means for storing nine color masking parameters, and all of the target image. or color conversion means for converting the three primary color values of some pixels into color coordinate values in a uniform color space; and color distribution calculation means for calculating statistics of the distribution of the color coordinate values in the color space; weight calculation means for calculating weights for each color patch from the color coordinate values of the color patches; weight storage means for storing the calculated weights; and density value calculation means for calculating three primary color density values from the color coordinate values of the color patches. and a black amount calculation means for calculating a black main density from the three primary color density values and a given black generation threshold value, and a UCR calculation means for calculating a three primary color density value after undercolor removal from the black main density and the three primary color density values. and a masking calculation means for calculating the three primary color main densities of the ink by performing a masking calculation on the three primary color density values after removing the undercolor using the color masking parameters, and a masking calculation means for calculating the three primary color main densities of the ink based on the three primary color main densities and the black main density of the ink. Obtain reproduction color coordinate values in a uniform color space based on the combination, obtain color differences with color coordinate values stored in the color patch storage means, and further use weights for each color patch stored in the weight storage means. It consists of an evaluation value calculation means for calculating an evaluation value, and a control means for updating the color masking parameter so as to minimize the evaluation value, and determines the optimal color masking parameter for the target image by convergence calculation. Color masking parameter determination device. (2) Color patch storage means for storing color coordinate values of a plurality of virtual color patches uniformly selected in a uniform color space, parameter storage means for storing nine color masking parameters, and generating black ink. a black threshold storage means for storing a black threshold value which is a threshold value for coloring the target image; a color conversion means for converting the three primary color values of all or some pixels of the target image into color coordinate values in a uniform color space; Color distribution calculation means for calculating distribution statistics in color space; Weight calculation means for calculating weights for each color patch from the statistics and color coordinate values of the color patches; Weight storage for storing the calculated weights. means, density value calculation means for calculating three primary color density values from the color coordinate values of the color patch, black amount calculation means for calculating a black main density from the three primary color density values and the black threshold value, a UCR calculation means for calculating three primary color density values after removing the undercolor from the three primary color density values; and a masking calculation for calculating the three primary color main densities by performing masking calculation using the color masking parameters on the three primary color density values after removing the undercolor. means, from the three primary color main densities of the inks and the black main density, obtain a reproduced color coordinate value in a uniform color space by this combination of inks, and obtain a color difference from the color coordinate value stored in the color patch storage means; Further, an evaluation value calculation means for calculating an evaluation value using the weight for each color patch stored in the weight storage means, and control for updating the color masking parameter and the black threshold so as to minimize the evaluation value. A color masking parameter determination device for determining optimal color masking parameters and black thresholds for a target image by means of convergence calculation.