JPH06326861A - Color conversion parameter setting method - Google Patents

Abstract

PURPOSE:To set a parameter center value of a color converter independently of fluctuation of printer gamma. CONSTITUTION:A Lab calculation section 6 calculates a Lab with respect to color conversion input data XYZ to obtain plural sets among the XYZ and Lab. A printer simulator 5 uses a neural net to convert a main density DyDmDc of a single color ink into a Lab value of a mixed color sample. The YMC subject to color conversion by a color conversion section 1 is converted into DyDmDc by an END gamma section 4 and the DyDmDc is converted into L'a'b' by the printer simulator 5. A DELTAE evaluation optimizing section 7 minimize an error square sum. The parameter for XYZ-YMC conversion is set by the nonlinear optimizing method.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method of setting color conversion parameters in a color printer, a color copying machine or the like.

[0002]

2. Description of the Related Art Conventionally, various methods of setting a masking coefficient for highly accurate color reproduction have been proposed. For example, assuming an ideal closed-loop model, there is a method of reading a printed color patch with a scanner of one's own system and setting a masking coefficient with a minimum density error (Journal of the Institute of Image Electronics Engineers of Japan, Vol. 14, No. 5, 1985, pp. 298-).
307).

Further, the Luv value of the printer sample is measured, and the inverse masking for minimizing the color difference between the Luv value and the Luv value obtained by the inverse masking output for converting the ink density signal into the three primary color density signals is calculated. , A method of setting a masking coefficient by an inverse function of this inverse masking (Patent Document 3)
-65877), based on the Neugebauer equation, RGB values are predicted from cmy, reproduced colors are simulated by virtual color patches, and a masking coefficient is set at the minimum color difference (Japanese Patent Laid-Open No. 61-28176).
No. 7, ibid., 61-281768), and further, integrated densities Dr, Dg, Db (measured by a scanner) are measured for mixed color samples, and main densities C,
M and Y are measured, and these sample data strings (Dr, D
g, Db, C, M, Y), and a method of setting a masking coefficient with the minimum error of the main density by the method of least squares (Journal of the Institute of Image Electronics Engineers of Japan, Vol. 16, No. 6, 1987 pp380-
386) has been proposed.

[0004]

Since the first method described above is an ideal loop model, it is impossible to set parameters corresponding to the document type. That is, only the masking coefficient for generation copy can be set. Also, since the masking coefficient is set with the minimum density error, the characteristics of the human eye are not taken into consideration. Furthermore, since the printer gamma fluctuation is not taken into consideration, it is difficult to set the parameter center value. Since it is based on and is a closed-loop model, it is not possible to set parameters as a device-independent color printer (that is, a color printer that interfaces with the outside by a standard signal).

The second method calculates inverse masking and sets parameters by an inverse function, so that it can be applied only to a specific color conversion method such as linear masking capable of obtaining an inverse function, and a printer. Since the gamma fluctuation is not taken into consideration, it is difficult to set the parameter center value,
Although it is possible to set parameters in a CRT hard copy, it is not possible to set parameters in the form of a copying machine.

The third method can be applied only to a specific printer for which the Neugebauer equation holds, and since the parameters are set based on modeling, it is difficult to set highly accurate parameters when applied to an actual machine. is there.

Since the fourth method sets the masking coefficient at the minimum error of the main density, it does not take human eye characteristics into consideration, and the scanner density (Dr, Dg, Db) of the color patch is not taken into consideration. ), It is not possible to set parameters corresponding to the document type (that is, only the masking coefficient for generation copy can be set), and since it is based on density, it is a parameter for device-independent color printer. Cannot be set.

It is an object of the present invention to provide a color conversion parameter setting method capable of setting a parameter center value of a color conversion device without being influenced by printer gamma fluctuation.

Another object of the present invention is to provide a color conversion parameter setting method capable of setting a parameter that minimizes color difference in a uniform color space.

Another object of the present invention is to provide a color conversion parameter setting method that can be applied to both a device-independent color printer and a copying machine.

Still another object of the present invention is to provide a color conversion parameter setting method applicable to any color conversion method and printer.

[0012]

In order to achieve the above object, according to the invention of claim 1, in a parameter setting method of a color conversion device, a measurement value 1 of a tristimulus value system of color patch mixture and the color patch. The measurement value 2 of each single color output constituting is measured with respect to a plurality of color patches, and the relationship for converting the signal system of the measurement value 2 into the signal system of the measurement value 1 is obtained,
The color conversion system output data is converted into the signal system of the measurement value 2,
A first output obtained by converting the converted signal system of the measured value 2 into a signal system of the measured value 1 according to the relationship, and a second output obtained by converting the color conversion system input data into the signal system of the measured value 1. It is characterized in that the parameter of the color conversion device is set with the error of 1 as the evaluation value.

According to a second aspect of the present invention, the color conversion device is a device-independent color conversion device in which input data of the color conversion system is a tristimulus value system signal.

According to a third aspect of the invention, the input data of the color conversion system is the halftone dot ratio of the printing process ink.

According to a fourth aspect of the present invention, a first output obtained by measuring a predetermined color chart with a colorimeter and the color chart are read, and the data after color conversion is converted into a second output by a printer simulator. , The first output and the second
It is characterized in that the parameters of the color conversion device are set so that the color difference of the output of 1 is minimized.

According to a fifth aspect of the present invention, the parameters of the color conversion device are obtained by using the predetermined color chart as a print document, the parameters are used as parameters for the print document, and the predetermined color chart is used as the printing paper. Parameter is obtained, the parameter is used as a parameter for printing paper, the parameter of the color conversion device is determined by using the predetermined color chart generated by the color conversion device, and the parameter is used as a parameter for generation copy. It is characterized in that the color chart is used as a plurality of document types to obtain the parameters of the color conversion device, and the parameters are used as the parameters for the plurality of document types.

According to a sixth aspect of the present invention, the signal system of the measurement value 1 of the tristimulus value system of the color patch mixture is a uniform color space system signal.

The invention according to claim 7 is characterized in that the relationship for converting the signal system of the measurement value 2 into the signal system of the measurement value 1 is constituted by a neural network.

According to the present invention, the parameters of the color conversion system are set based on the END gamma characteristic set in advance, and the parameters of the color conversion system are the parameters in the set END gamma characteristic, and the gamma of the device is set. The characteristics are measured, and the gamma parameter of the device is set based on the measured value and the set END gamma characteristic.

According to a ninth aspect of the present invention, the parameters of the color conversion system are set on the basis of the END gamma characteristic set in advance, and the parameters of the color conversion system are the parameters in the set END gamma characteristic, such as environmental fluctuations. Is estimated, and the gamma parameter of the device is set based on the estimated gamma characteristic and the set END gamma characteristic.

According to a tenth aspect of the present invention, in the method for setting the output gamma parameter of the color conversion device, the measurement value 1 of the tristimulus value system of color patch mixture and the measurement value 2 of each single color output composing the color patch. Is measured for a plurality of color patches, and the relationship of converting the signal system of the measurement value 2 into the signal system of the measurement value 1 is obtained under the achromatic condition, and the digital data by the gradation pattern output of each monochromatic output and the measurement are performed. It is characterized in that an output gamma parameter that satisfies an achromatic color condition is set based on the value 2 relationship.

According to the invention of claim 11, the measured value 1
It is characterized in that the relationship between the signal system of 1 and the signal system of the measured value 2 is configured by a neural network.

According to a twelfth aspect of the present invention, in the method of setting the input gamma conversion parameter of the color conversion device, the color conversion device has a reading system, and in the method, a signal value specific to the device when a predetermined color chart is read. And the measurement value 1 obtained by measuring the color chart, the relationship between the signal system of the measurement value 1 and the scanner signal value is obtained under the achromatic color condition, and the input gamma conversion parameter satisfying the achromatic color condition is set. .

According to a thirteenth aspect of the invention, the measured value 1
It is characterized in that the relation between the signal system of and the scanner signal value is constructed by a neural network.

According to a fourteenth aspect of the present invention, an input gamma conversion parameter is obtained by using the predetermined color chart as a print original, the parameter is set as an input gamma conversion parameter for the print original, and the predetermined color chart is input as photographic paper. A gamma conversion parameter is obtained, the parameter is used as an input gamma conversion parameter for printing paper, and the input gamma conversion parameter is obtained by using the predetermined color chart generated by a color conversion device. The present invention is characterized in that a parameter is obtained by using the predetermined color chart as a plurality of document types as a conversion parameter and the parameter is used as an input gamma conversion parameter for a plurality of document types.

According to a fifteenth aspect of the present invention, the signal system of the measurement value 1 is converted into the scanner signal value by the scanner signal value peculiar to the device when the predetermined color chart is read and the measurement value 1 obtained by measuring the color chart. The device-specific scanner when the conversion relationship is calculated under the achromatic condition, the RGB values after gamma conversion are calculated for each patch of the color chart, and the patch of a predetermined gray scale is read by each color conversion device. It is characterized in that an input gamma conversion parameter satisfying an achromatic color condition is set based on the signal value and the RGB value after gamma conversion in each patch.

According to the sixteenth aspect of the present invention, in the method of setting the input / output gamma, the achromatic condition is set to a predetermined hue.
It is characterized in that the input gamma conversion parameter is set or the achromatic color axis parameter of the color conversion system is set so as to have a predetermined chroma.

According to a seventeenth aspect of the present invention, in an apparatus for making a hard copy with four colors including a black plate, the achromatic condition is determined based on the hue and chroma characteristics of the single black plate.

[0029]

In the Lab calculator, Lab for the color conversion input data XYZ is calculated to obtain a plurality of sets of XYZ and Lab. In the printer simulator, the main density DyDmDc of the single color ink is converted into the Lab value of the mixed color sample by the neural network. YMC color-converted by the color converter
Is converted to DyDmDc in the END gamma part, and DyDmD
Convert c into L'a'b 'with a printer simulator.
The ΔE evaluation optimizing unit sets a parameter of the XYZ-YMC conversion that minimizes the error sum of squares of Lab and L′ a′b ′ by a non-linear optimization method. As a result, the center value of the color correction parameter can be set without being affected by the gamma fluctuation of the device.

[0030]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be specifically described below with reference to the drawings. An example of color conversion parameter setting in the printer alone will be described first, and then an example of color conversion parameter setting in the copying machine mode will be described.

<Embodiment 1> Printer alone (color conversion from a standard signal); FIG. 1 is a block diagram of a first embodiment of the present invention, in which a tristimulus value signal XYZ is input and an output signal y is output. (Yellow), m
The setting of color conversion parameters in a device-independent color printer for conversion into (magenta) and c (cyan) will be described as an example. Here, the device-independent color printer is a printer that interfaces with the outside using standard signals (tristimulus values XYZ, L * a * b *, etc.) as described above.

In this embodiment, the tristimulus value signal XYZ is converted to YMC.
A color conversion unit 1 for performing color conversion into a color, a printer gamma unit including an END gamma unit 2 and a raw printer gamma unit 3 for correcting the output of the color conversion unit 1 according to the printer gamma characteristic, a printer 4, and a color conversion unit. END gamma unit 2 that corrects the output of 1 according to the printer simulator, printer simulator 5 that performs color prediction of print output configured by a neural network, and tristimulus value signal XYZ is L * a.
The color conversion parameter is set so as to minimize the sum of error squares of the output of the printer simulator 5 and the output of the L * a * b * calculating unit 6 and the L * a * b * calculating unit 6 converting into * b *. It is composed of the ΔE evaluation optimizing unit 7 (in the present invention, L * a * b * is represented as Lab).

The relation between Lab and tristimulus value signal XYZ is as follows (CIE 1976). L = 116 (Y / Yn) * (1/3) -16 a = 500 [(X / Xn) * (1/3)-(Y / Yn) * (1/3)] b = 200 [(Y / Yn) * (1/3)-(Z / Zn) * (1/3)] where * (1/3) represents the 1/3 power.

The color conversion unit 1 converts an input signal into an ink amount control signal for a recording system, and a masking method is generally used. In this embodiment, it is assumed that the input of the color conversion unit is XYZ and the output is a masking circuit that is a signal that is a linear signal to the Y signal at the time of achromatic color output. Note that this color conversion method is not limited to the masking circuit, and can be applied to various color conversion methods.

The printer gamma section corrects the assumed output of the color conversion section 1 in accordance with the actual printer gamma characteristic, and is composed of an END gamma section 2 and a raw printer gamma section 3 as will be described later. Execute with conversion. In the present embodiment, the reflectance is linear (linear to the Y signal), but it is also applicable to density linear, lightness linear, etc., and easily realized by adding an input gamma conversion circuit for changing the attribute to the input section. it can. Further, the present embodiment is not limited to the linear reflectance. Furthermore, although tristimulus values XYZ are assumed as inputs, Lab, Lu
Of course, the present invention can be applied to other standard signal forms such as v and NTSC RGB and a form that receives a relational expression of device-specific signal + standard signal as an input.

The parameter setting method of the color conversion unit according to the present invention will be described below. First step: Color prediction of print output (printer simulator) Printer simulator 5
Is for stably simulating color reproduction by an actual printer. The characteristics of an actual printer change due to environmental factors such as temperature. However, this characteristic variation is mostly due to the variation of the printer's gamma characteristic (relationship between the "printer signal" of each color and the "ink amount actually transferred to the paper"), and if the ink amount of each color is constant. It is considered that those colors are almost unchanged.

Therefore, the printer simulator receives the ink amount of each color as an input, and outputs the color when they are overlapped, the input is the average reflectance of the single color patch, and the output is the chromaticity (Lab) of the mixed color patch. And That is, a color patch (a set of mixed color samples and outputs of the single color inks forming the colors) as shown in FIG. 4 is output (hard copy). The Lab value of the mixed color sample and the main densities (Dy, Dm, Dc) of each single color ink are measured with a measuring instrument. The above process is performed on a large number of color patch samples. Using the measurement data set as learning data, a neural network arithmetic expression for converting the main density of the single color ink into the Lab value of the mixed color sample is obtained by back propagation (FIG. 5). In this embodiment, the neural network has three inputs,
It is composed of three outputs and two intermediate layers.

In general, due to the poor stability of the printer, it is difficult to repeatedly reproduce the same color due to the gamma fluctuation of the printer even if the same data is digitally printed out. That is, the relationship between the digital data and the Lab measurement is strongly influenced by the gamma fluctuation of the printer.
If you set parameters based on such data group,
It becomes difficult to set the center value of the parameter.

On the other hand, in this embodiment, the main density of each single color ink is converted into Lab, and basically the conversion system of the ink amount and Lab, so that the influence of the gamma fluctuation of the printer is affected. It is possible to predict the color of the print output that is not received, and the center value of the parameter can be set with high accuracy by setting the parameter by the method described later. As a result, it is possible to set a system that predicts a color at the time of printer output with high accuracy and does not depend on printer gamma fluctuation (converts the main density of a single color ink into the Lab value of a mixed color sample).

Further, according to the method of this embodiment, no specific modeling such as Neugebauer model is assumed, and the Lab value is converted into the Lab value by the neural network based on the measured value. It has the advantage of not limiting modeling. In the above description, the main density is used as the base, but the present invention is not limited to this, and any measurement value having a correlation with the ink amount may be used.

Second step: Setting of END gamma section In the END (equivalent neutral density) gamma section 2, an assumed output signal of the color conversion section (Y linear in this embodiment when achromatic color is output) is adjusted to the printer simulator 5. The output is the average reflectance (Dy, Dm, Dc) of a single color that is an input to the printer simulator 5. That is, a printer outputs a color patch, and the reflectance (Dy, Dm, Dc) of a single color and the Lab chromaticity are measured. Then, a Lab → Dy, Dm, Dc (B) conversion neural network is created (see FIG. 6).

For each Y (0-1) value, Y is converted into an L value (L = Y to the 1/3 power), and a = b = 0 (achromatic color condition), and the above (B) The Dy, Dm, and Dc values corresponding to the achromatic color are obtained from the relationship of the expressions.

In the END gamma section 2 set in this way, the achromatic color of the designated brightness is converted into the ink amount (Dy, Dm, Dc) of each color necessary for printing.

Third step: Generation of sample data for parameter setting of color conversion section XYZ values obtained by uniformly sampling the Lab space are used as sample data for parameter setting.

Fourth step: setting of parameters of the color conversion section The color conversion method in the color conversion section is set. An example of the masking method is as follows.

Y = a11X + a12Y + a13Z M = a21X + a22Y + a23Z C = a31X + a32Y + a33Z The Lab for the color conversion input data XYZ is calculated by the Lab calculating unit 6 to obtain a plurality of sets of correct answer data (XYZ, Lab). Also, the converted data is obtained. That is, X
YZ is color converted by the color conversion unit 1 (YMC), and YMC is converted to EN.
The D gamma unit 2 converts it into DyDmDc, and the printer simulator 5 converts it into L'a'b '. Then, in the ΔE evaluation optimizing unit 7, the error sum of squares Σ√ [(L
k-L'k) * 2 + (ak-a'k) * 2 + (bk-
b′k) * 2] (where k = 0 to the number of samples-1, *
The parameter (aij) of the XYZ → YMC conversion that minimizes the square) is set by a well-known nonlinear optimization method (for example, a method such as a hill climbing method).

The above error is due to the uniform color space (Lab
Since it corresponds to the color difference ΔE of the system), the optimum color conversion parameter can be set in consideration of the characteristics of human eyes.

As described above, the color correction system parameters (gamma conversion table, color correction coefficient) for the printer having the standard gamma characteristic can be set. In addition, in the above-mentioned example, END gamma was calculated as a = b = 0. Ideally, an achromatic color is reproduced as an achromatic color,
However, depending on the printer gamma fluctuation and the accuracy of the gamma setting, an achromatic color may be reproduced in a rainbow color (color difference), resulting in a significant deterioration in image quality. As a method of preventing this, a and b values are set to predetermined values (fixed value or a = f1
By intentionally removing the achromatic condition with (L), b = f2 (L), it is possible to suppress the hue deviation.

In addition, in the case of black plate processing, UCR / UCA
The formula is generally UCR / UCA when y = m = c and when the Bk amount is Bk = y = m = c, the UCR
After / UCA, the three colors of ymc are set to 0. In order to reproduce the same color in the case of three-color reproduction and the case of four-color reproduction, it is necessary to make the color when y = m = c the same as the Bk single color. In this case, a = f1 (L),
The expression (value) of b = f2 (L) may be set based on the guillescale patch for Bk single color, and END gamma may be set. Of course, it is also effective to intentionally combine the a and b values of the Bk version with Zlas.

The parameter setting method for each printer will be described below. Fifth Step: Parameter Setting for Individual Printer The printer gamma has a double structure as shown in FIG. 1 in order to avoid the influence of gamma fluctuation of the printer itself and to use the printer simulator. In other words, for masked CMY data, E
Through the ND gamma unit 2, the END gamma unit 2 converts the ink amount of each color required for printing an achromatic color of the designated brightness as described above. The output of the END gamma unit 2 is the average reflectance (Dy, Dm, Dc) of a single color that is input to the printer simulator 5.

Further, in the raw printer gamma section 3,
This monochromatic reflectance (Dy, Dm, Dc) is converted into a printer signal based on the printer signal-monochromatic reflectance conversion characteristic of the printer itself in order to be used for printer output.
It defines the relationship between (Dy, Dm, Dc) and (y, m, c). With this double structure, the color correction coefficient can be set without being affected by the gamma fluctuation of the printer itself. Further, therefore, the printer characteristic variation can be corrected only by the printer gamma.

The printer outputs a color patch of a YMC monotone gradation pattern, and the reflectance (Dy, Dm, D
c) is measured, and the raw printer gamma table is set based on the relationship between the printer signal used for patch output and the reflectance.

Relationship (EN) between YMC digital data and main ink densities (Dy, Dm, Dc) when gamma is set
D gamma) and the above relationship (raw printer gamma)
A table of the printer gamma unit (END gamma + raw printer gamma) for each printer is set based on the above (the color conversion parameters are fixed).

As described above, in this embodiment, the color conversion parameter is fixed and the difference between the devices is absorbed by adjusting the printer gamma section.

Next, a parameter setting method that follows environmental changes (temperature, humidity, etc.) will be described. Similar to the parameter setting for each printer, the color conversion parameters are fixed and the parameters are set by adjusting the printer gamma section.

For example, the change of the printer gamma characteristic with respect to the temperature change is experimentally obtained in advance. For example, the change in solid density is obtained with temperature as a parameter (FIG. 8). From the change in the solid density, the conversion at the intermediate density is estimated (for example, linear prediction), and the relationship between the digital data and each ink main density is estimated. Then, the above-described fifth step is performed thereafter to set the gamma parameter of the printer.

<Embodiment 2> Printer alone (color conversion from process ink halftone dot ratio);
FIG. 2 is a block configuration diagram of a second embodiment of the present invention, and a method of setting parameters when a halftone dot ratio of process ink (yellow, magenta, cyan, black) is input as a standard signal will be described. . Since the spectral reflectance characteristics of the color material are slightly different between the process ink and the general printer, it is necessary to perform color conversion to correct the difference in these characteristics.

First step: Color prediction of print output (printer simulator) The same as described above. Second step: setting of the END gamma part The same as described above. Third Step: Generation of Sample Data for Parameter Setting of Color Converting Section Lab values of a large number of color patch samples (printed color patches) are measured. Halftone dot rate (Y%, M of the sample process ink)
%, C%) value and (Lab) measured value are made into a table (6 in FIG. 2) as sample data for parameter setting.

Fourth Step: Setting of Parameters of Color Converting Section Same as the above. However, the input portion is the halftone dot ratio (Y%, M%, C%) of the process ink.

Y = a11Y% + a12M% + a13C% M = a21Y% + a22M% + a23C% C = a31Y% + a32M% + a33C% A plurality of correct data sets ((Y%, M%, C%), Lab) are obtained. In addition, Y%, M% and C% are set in the color conversion unit 1.
Color conversion with (YMC), and YMC is D in the END gamma unit 2
yDmDc is converted, and DyDmDc is converted into L'a'b 'by the printer simulator 5. Then, in the ΔE evaluation optimizing unit 7, the error sum of squares Σ√ [(Lk−L′k) * 2
+ (Ak-a'k) * 2 + (bk-b'k) * 2] (where k = 0 to the number of samples-1, * is the square) Y% M% C% → YMC Conversion parameter (aij)
Is set by a non-linear optimization method.

In the above description, Y% M% C%-
The color conversion → YMC is matrix conversion, but the color conversion method is not limited to this and can be applied to various color conversion methods.

<Embodiment 3> Copier mode (color conversion by scanner + printer) FIG. 3 is a block diagram of a third embodiment of the present invention, which is a copier mode (one of scanner 8 and printer 4). How to set parameters for body type will be described. Basically, the parameters are set by the same method as the setting of the color conversion parameters in the device-independent color printer described above, and the difference is that the input is an RGB signal unique to the device.

First Step: Color Prediction of Print Output (Printer Simulator) The same as described above. Second Step: Setting of END Gamma Unit The printer gamma unit performs the same as described above. Details of the input gamma unit 9 will be described later.

Third Step: Generation of Sample Data for Parameter Setting of Color Converting Section A large number of color patch samples (eg, printed color patches) are read by the scanner 8 (r, g, b). The Lab value of the sample is measured by the Lab measuring unit 6. The scanner read value of the sample is gamma-converted by the input gamma unit 9 and (R, G, B) and (Lab) measured values are used as sample data for parameter setting.

In the above example, sample data for parameter setting is generated based on the printed color patch. The parameters set in this way are used as parameters for the print document. Similarly, sample data for parameter setting is generated based on the color patch on the photographic paper, and this parameter is used as the parameter for the photographic paper original. In addition, sample data for parameter setting is generated based on both color patches for printing and printing paper, and this parameter is used as a parameter for printing / printing paper original.

Further, sample data for parameter setting is generated based on the color patch generated by the color conversion device,
This parameter is used as a parameter for generation copy. In this way, by setting the parameters for each document type, it is possible to perform highly accurate color conversion adapted to the document type.

The reason for setting the parameters for each document type is as follows.
When scanning various originals with a scanner, the spectral reflectance characteristics of the color materials differ for each original type and the sensitivity characteristics of the scanner differ from those of the human eye, so even if the original colors are the same (meta color), the RGB values are different. Because. Further, for the same reason, it is effective to set the input gamma conversion unit for each document type.

Fourth step: setting of parameters of the color conversion section The same as described above. However, the input section is RGB
(Gamma converted data).

Y = a11R + a12G + a13B M = a21R + a22G + a23B C = a31R + a32G + a33B Multiple sets of correct data ((R, G, B), L'a'b ') are obtained. Further, R, G, B are color-converted by the color conversion unit 1 (YMC), and YMC is DyDm by the END gamma unit 2.
Convert to Dc, DyDmDc printer simulator 5
Is converted to L'a'b '. Then, in the ΔE evaluation optimizing unit 7, the error sum of squares Σ√ [(Lk−L′k) * 2 + (a
k−a′k) * 2 + (bk−b′k) * 2] (however,
k = 0 to the number of samples-1, * is the square)
A parameter (aij) for GB → YMC conversion is set by a non-linear optimization method.

Next, a method of setting the input gamma section 9 will be described. In the third step, (r, g, b) and La obtained in
Using the b value as data, a neural network arithmetic expression for converting the Lab value into the (r, g, b) value is obtained by back propagation (FIG. 7).

Lab → (r, g, b) Equation (C) For each Y (0-1) value, Y is converted into an L value (L = Y to the 1/3 power), and a = b = 0. Under the condition of (achromatic color), the (r, g, b) value is obtained from the relationship of the above formula (C), and the input gamma unit 9 is calculated in accordance with the relationship between this Y value and (r, g, b).
The gamma conversion table (r → R, g → G, b → B) is created. The (r, g, b) values are obtained for the Y values of several points by the above method, and r → Y (R), g → Y (G), b → Y (B)
Is obtained by a method such as least squares assuming a spline or an nth-order (for example, a cubic) curve, and a gamma conversion table (r → R, g → G, b → B) of the input gamma part is created.

As a result, for an achromatic original (a = b = 0), a gamma conversion table with R = G = B and Y linear can be generated. Similarly, in the printer gamma unit, the response to several points is similarly obtained by neural, and the gamma conversion table is obtained by a method such as least square assuming a spline or an nth-order (eg, cubic) curve based on this data. There is a way to set. According to this method, smoothness of conversion can be guaranteed.

In the above example, the input gamma conversion table is generated based on the printed color patch. The parameters set in this way are used as parameters for the print document. Similarly, an input gamma conversion table is generated based on the color patch on the photographic paper, and this parameter is used as a parameter for the photographic paper original. Also, an input gamma conversion table is generated based on both color patches for printing and printing paper, and this parameter is used as a parameter for printing / printing paper document.

An input gamma conversion table is generated based on the color patch generated by the color conversion device, and this parameter is used as a parameter for generation copy.
As described above, by setting the input gamma conversion table for each document type, it is possible to perform highly accurate color conversion adapted to the document type.

As yet another embodiment, a simple setting method of input gamma conversion table parameters will be described.
The RGB value of the color patch of the predetermined achromatic chart is determined in advance by reading a predetermined achromatic chart with a scanner and converting the chart based on the gamma conversion table set by the method such as the least square method described above. And
When setting the scanner gamma of each device, the achromatic color chart is read and an input gamma conversion table suitable for the device is set based on the relationship between the rgb value and the RGB value. Thereby, the input gamma conversion table can be set in a short time.

[0076]

As described above, according to the first aspect of the present invention, the relationship between the measured value of the single color ink sample and the measured value of the mixed color sample is the relationship between the ink amount and the color. Since the parameter is set by the measured value without being affected by the gamma fluctuation, the center value of the color correction parameter can be set.

According to the second aspect of the present invention, since the measured value 1 is a tristimulus value system signal, a parameter center value from the standard signal (tristimulus value system signal) to YMC conversion can be set, and the device Independent color processing is possible.

According to the third aspect of the invention, it is possible to set the center value of the color correction parameter using the process ink% as an input.

According to the fourth aspect of the invention, since the parameter is set based on the relationship between the scanner reading value and the signal of the tristimulus value system, the center value of the parameter in the form of a copying machine can be set.

According to the fifth aspect of the invention, since the parameters are set for each document type, it is possible to perform highly accurate color conversion corresponding to the document type.

According to the sixth aspect of the invention, since the color conversion parameter is set with the color difference (ΔE) in the uniform color space coordinates as the evaluation value, the parameter setting with the minimum ΔE can be performed, and the human eye can see. The system can be identified in consideration of its characteristics.

According to the seventh and eleventh aspects of the present invention, the signal conversion from the color conversion output data to the tristimulus value system has a considerable non-linearity. Although it is difficult to predict, the neural network has the property of being a universal nonlinear converter, and is optimal as such a nonlinear converter. In the present invention,
Since the neural network configures the signal conversion from the color conversion output data to the tristimulus value system, it is possible to perform conversion with high accuracy and without limiting the model.

According to the invention described in claim 8, the color conversion system parameters are set with high accuracy assuming a virtual device gamma (that is, the device gamma when the color conversion system parameters are set). Since the assumptions are clear), the parameters of the color conversion system can be fixed and the individual device gamma can be corrected (by correcting the individual device gamma characteristics to the virtual device gamma). Can be set.

According to the ninth aspect of the present invention, the parameter of the color conversion system is set with high accuracy assuming the virtual device gamma (that is, the device gamma when the parameter of the color conversion system is set). (Because the assumption is clear), it is possible to predict the device gamma characteristic fluctuation due to environmental changes and correct the device gamma (by correcting the predicted device gamma characteristic to a virtual device gamma) to follow the environmental fluctuations. Highly accurate identification of can be easily achieved.

According to the tenth aspect of the invention, it is possible to set the output gamma with high accuracy.

According to the twelfth aspect of the invention, since the parameter is set in consideration of the deviation from the achromatic color of the original, it is possible to perform highly accurate color conversion for the achromatic color input.

According to the thirteenth aspect of the present invention, highly accurate input gamma setting can be performed.

According to the fourteenth aspect of the invention, it is possible to set the input gamma corresponding to the document type.

According to the fifteenth aspect of the invention, the input gamma can be easily set.

According to the sixteenth aspect of the present invention, since the parameters are set under the conditions of predetermined hue and chroma, it is possible to prevent the achromatic color from being reproduced with various colors even if there is a gamma fluctuation. It

According to the seventeenth aspect, the same color can be reproduced in the three-color reproduction and the four-color reproduction.

[Brief description of drawings]

FIG. 1 is a block configuration diagram of a first embodiment of the present invention.

FIG. 2 is a block diagram of a second embodiment of the present invention.

FIG. 3 is a block configuration diagram of a third embodiment of the present invention.

FIG. 4 shows a color patch that is a set of mixed color samples and outputs of single-color inks that form the colors.

FIG. 5 is a neural network that converts the main density of a single color ink into Lab values of mixed color samples.

FIG. 6 is a neural network that converts Lab values of mixed color samples into main densities of single color inks.

FIG. 7 is a neural network for converting Lab values of mixed color samples into rgb.

FIG. 8 shows changes in solid density with temperature.

[Explanation of symbols]

 1 Color conversion unit 2 END gamma unit 3 raw printer gamma unit 4 Printer 5 Printer simulator 6 Lab calculation unit 7 ΔE evaluation optimization unit

Claims (17)

[Claims] 1. In a parameter setting method of a color conversion device, a measurement value 1 of a tristimulus value system of color patch mixing and a measurement value 2 of each monochromatic output forming the color patch are measured for a plurality of color patches. Then, the relation for converting the signal system of the measurement value 2 into the signal system of the measurement value 1 is obtained, and the color conversion system output data is converted into the signal system of the measurement value 2,
A first output obtained by converting the converted measurement value 2 signal system into a measurement value 1 signal system according to the above relationship; and a second output obtained by converting color conversion system input data into the measurement value 1 signal system. A color conversion parameter setting method, characterized in that the parameter of the color conversion device is set using the error of 1. 2. The color conversion parameter setting according to claim 1, wherein the color conversion device is a device-independent color conversion device in which input data of the color conversion system is a signal of a tristimulus value system. Method. 3. The color conversion parameter setting method according to claim 1, wherein the input data of the color conversion system is a halftone dot ratio of a printing process ink. 4. A first output obtained by measuring a predetermined color chart with a colorimeter, and reading the color chart, converting the data after color conversion into a second output by a printer simulator, and outputting the first output. And a parameter of the color conversion device is set such that the color difference between the second output and the second output is minimized. 5. A parameter of a color conversion device is obtained by using the predetermined color chart as a print document, the parameter is used as a parameter for a print document, and a parameter of the color conversion device is obtained by using the predetermined color chart as photographic paper. The parameter is used as a parameter for printing paper, the parameter of the color conversion device is obtained by generating the predetermined color chart by a color conversion device, and the parameter is used as a parameter for general copying, and the predetermined color chart is divided into a plurality of colors. 5. The color conversion parameter setting method according to claim 4, wherein a parameter of the color conversion device is obtained as a document type, and the parameter is set as a parameter for a plurality of document types. 6. The color conversion parameter setting method according to claim 1, wherein the signal system of the measurement value 1 of the tristimulus system of the color patch mixture is a signal of the uniform color space system. 7. The color conversion parameter setting method according to claim 1, wherein the relationship for converting the signal system of the measurement value 2 into the signal system of the measurement value 1 is configured by a neural network. 8. A color conversion system parameter is set on the basis of a preset END gamma characteristic, and the color conversion system parameter is set to the parameter in the set END gamma characteristic, and the gamma characteristic of the device is measured. The color conversion parameter setting method according to claim 1, wherein a gamma parameter of the device is set based on a measured value and the set END gamma characteristic. 9. A parameter of a color conversion system is set on the basis of a preset END gamma characteristic, and the parameter of the color conversion system is set to the parameter of the set END gamma characteristic, and the gamma characteristic associated with environmental changes is estimated. The color conversion parameter setting method according to claim 1, wherein the gamma parameter of the device is set on the basis of the estimated gamma characteristic and the set END gamma characteristic. 10. A method for setting an output gamma parameter of a color conversion device, wherein a measurement value 1 of a tristimulus value system of color patch mixture and a measurement value 2 of each single color output forming the color patch are converted into a plurality of color patches. On the basis of the relationship between the measured value 2 and the digital data by the gradation pattern output of each monochromatic output, the relationship of measuring the measured value 2 and converting the measured value 2 signal system into the measured value 1 signal system is obtained under achromatic conditions. A color conversion parameter setting method characterized by setting an output gamma parameter that satisfies an achromatic color condition. 11. The color conversion parameter setting method according to claim 10, wherein the relationship between the signal system of the measured value 1 and the signal system of the measured value 2 is configured by a neural network. 12. A color conversion device has a reading system, and in a method of setting an input gamma conversion parameter of the color conversion device, a device-specific signal value when a predetermined color chart is read and the color chart is measured. The method of setting a color conversion parameter, characterized in that the relationship between the signal system of the measurement value 1 and the scanner signal value is obtained from the measured value 1 and the input gamma conversion parameter that satisfies the achromatic color condition is set. 13. The color conversion parameter setting method according to claim 12, wherein the relationship between the signal system of the measurement value 1 and the scanner signal value is configured by a neural network. 14. An input gamma conversion parameter is obtained using the predetermined color chart as a print document, the parameter is used as an input gamma conversion parameter for the print document, and the input gamma conversion parameter is obtained using the predetermined color chart as photographic paper. The input gamma conversion parameter for photographic paper is used as the parameter, the input gamma conversion parameter is obtained by using the color conversion device to generate the predetermined color chart, and the parameter is set as the input gamma conversion parameter for general copying. 13. The color conversion parameter setting method according to claim 12, wherein the color chart is used as a plurality of document types to obtain a parameter, and the parameter is used as an input gamma conversion parameter for the plurality of document types. 15. The relationship for converting the signal system of the measurement value 1 into the scanner signal value by the scanner signal value specific to the device when the predetermined color chart is read and the measurement value 1 obtained by measuring the color chart is achromatic. The RGB value after gamma conversion is obtained for each patch of the color chart in advance, and the scanner signal value unique to the device when reading a patch of a predetermined gray scale by each color conversion device and the patch signal in each patch 13. The color conversion parameter setting method according to claim 12, wherein an input gamma conversion parameter that satisfies an achromatic color condition is set based on the RGB value after gamma conversion. 16. A method for setting an input / output gamma, wherein an achromatic condition is set to have a predetermined hue and a predetermined chroma, and an input gamma conversion parameter is set or an achromatic color axis parameter of a color conversion system is set. A parameter setting method characterized by: 17. The parameter setting method according to claim 16, wherein in an apparatus for performing hard copy with four colors including a black plate, the achromatic condition is determined based on the hue and chroma characteristics of a single black plate. .