JP2880513B2 - Printing transmission primary color conversion method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial applications]

The present invention relates to a print transmission primary color conversion method for converting image data such as transmission, storage, and image processing, and more particularly to a print transmission primary color conversion method suitable for use in band compression of color image data.

2. Description of the Related Art Along with the advanced information society, attention has been paid to printing and color hard copy as means for providing a high-definition still image. Considering such printing or the like and transmission of image data between various image means, reduction of the data amount and matching of colors before and after conversion become problems. Various band compression techniques for compressing the band and reducing the amount of data to be transmitted have been practically used, and the NTSC band compression technique is well known in image transmission and reproduction means represented by television. I have. In this band compression technique, conventionally, the color information of red (R), green (G), and blue (B) in an image signal is converted into a luminance Y representing only a change in brightness of image data and a red to green (wavelength of 625 nm). ~Four
Chromaticity axis I over 92nm light, yellow to blue (wavelength 566n)
A YIQ conversion method of converting the chromaticity axis Q to light having a wavelength of m to 440 nm to perform band compression. The resolution of human vision is highest for the luminance Y, the chromaticity axis I is second, and the chromaticity axis Q is lowest. Therefore, in the band compression technique, the widest band is assigned to the luminance Y by utilizing this visual characteristic, and the chromaticity axis Q
The data is compressed by allocating the narrowest band to. That is, the band is assigned as follows: luminance Y> chromaticity axis I>
The relationship is a chromaticity axis Q. Recently, when image processing or communication is performed by a computer, data compression by digital image processing is often performed. At this time, the band compression is performed by an operation of thinning out pixels for the data of each chromaticity axis I and Q. Going. On the other hand, conventionally, with respect to a print image, when band compression is performed to transmit image data, cyan (C), magenta (M), and yellow ( Y ) of each separation color of the print image are used.
Is assumed to be a complementary color, YIQ conversion is performed using the inverted three primary colors (R '), green (G'), and blue (B ') using the same conversion formula as in the television, and the pixels are thinned out. Bandwidth compression was performed. However, the R, G, and B data of an image used in a television and the C, M, and Y data of a print image have some differences as follows. (1) R, G, B and C, M, Y do not have a complementary color relationship. Therefore, when the print image data is subjected to the YIQ conversion using the same conversion formula, an error occurs. (2) An image of a television is a system in which colors are generated by additive color mixture having a linear relationship of a linear expression, and linearity with a visual sensation can be assumed with a certain degree of accuracy or more. On the other hand, a print image is a system that develops a color by a complicated mechanism centered on subtractive color mixture, and the three primary colors R, G, and B have a non-linear relationship with the visual sensation. Therefore, despite the above-described differences between the three primary colors R, G, B of the television image and the separated colors C, M, Y of the image such as prints, the same YIQ conversion formula is used. It is an error to perform transmission primary color conversion. In fact, each separation color C as described above, M, 3 primary colors created by inverting the Y R ', G',
When the chromaticity axes I and Q are obtained from the B 'data, the chromaticity axes I and Q are not converted to the correct chromaticity axes, and the luminance Y
Is not the correct brightness. Therefore, since the wideband axis and the narrowband axis of color perception are not properly corresponded, there is a problem that band compression utilizing the visual effect cannot be performed. A typical model representing the relationship between the separation colors C, M, and Y and the three primary colors R, G, and B includes Neugebaue.
r) An equation is known, and this equation can be used for transmission primary color conversion. However, there is a problem that this equation cannot be used for transmission primary color conversion in its original form because it has no solution and therefore cannot be inversely converted.

[Problems to be solved by the invention]

SUMMARY OF THE INVENTION The present invention has been made to solve the above-mentioned conventional problems, and it is a first object of the present invention to provide a printing primary color conversion method capable of converting each separated color data of an image into YIQ data with high accuracy. It is a second object of the present invention to provide a transmission primary color conversion method capable of accurately obtaining print image data from YIQ-converted data.

[Means for achieving the object]

The present invention relates to a method for converting CMY data of cyan, magenta M, and yellow Y of each separation color of a print image into YIQ data of luminance Y, chromaticity axes I and Q in order to perform band compression and transmit. , Measuring the reflectance of the printed material having a CMY value passing through, thereby creating an equation that previously defines the relationship between the CMY data and the XYZ data of the visual tristimulus values X, Y, and Z; The first object has been achieved by including a step of converting the CMY data into XYZ data using the method and a step of converting the XYZ data into YIQ data. Further, the present invention converts the YIQ data converted into luminance Y and chromaticity axes I and Q, band-compressed and transmitted into CMY data of cyan C, magenta M and yellow Y of each separated color of a print image. In the print transmission primary color conversion method of reverse conversion, by measuring the reflectance of the printed matter that the CMY value is over,
A step of creating an expression that preliminarily defines a relationship between XYZ data of visual tristimulus values X, Y, and Z and CMY data;
The second object has been achieved by including a step of converting to XYZ data and a step of converting the XYZ data to CMY data using the relational expression.

Actions and effects of the present invention

It is known that the result of a visual experiment in which the relationship between the resolution and the optic nerve sensitivity was examined for the luminance Y and the chromaticity axes I and Q is as shown in FIG. As shown in FIG. 2, the change in luminance Y has the highest resolution, that is, the resolution. Further, regarding the chromaticity axis I and Q directions, the chromaticity axis I in the direction of 625 nm to 492 nm (red to green) at the main wavelength.
High sensitivity to resolution changes of 566 nm to 440 at the main wavelength
It is known that the sensitivity is low for a change in the resolution of the chromaticity axis Q in the nm (yellow to blue) direction. In YIQ conversion of television, the chromaticity axis having the higher resolution is defined as I, and the chromaticity axis orthogonal to the chromaticity axis I is defined as Q.
As a result, a matrix operation such as the following equation (1) is performed from R, G, and B data of an image to obtain YIQ data. On the other hand, the YIQ data and the three primary colors R, G, and B for the visual tristimulus values X, Y, and Z have linear relationships expressed by linear expressions, respectively, and are expressed by the following expressions (2) and (3). It is expressed as follows. This equation (3) is expressed by the following equation (4) using an inverse matrix. Therefore, by using the equations (4) and (2), it is possible to perform primary color conversion on the path of the RGB data XYZ data YIQ data. That is, RG via the visual tristimulus values X, Y, Z
B data can be converted to YIQ data or inversely. Therefore, conversion of CM Y data to XYZ data (hereinafter referred to as CM Y data
If realized referred to XYZ conversion), it is possible to convert the print image data expressed in CM Y to YIQ data. By the way, the conversion path of this CM Y XYZ conversion is C, M, Y
It shows visually what stimulus values X, Y, Z the printed matter having the dot percentage values c, m, y of the data have. For CM Y XYZ conversion, conversion by Neugebauer equation,
There are conversion by polynomial approximation, conversion by approximation other than polynomial using a halftone coloring model, and the like. The Neugebauer equation is an equation for calculating RGB reflectance when three colors C, M, and Y are mixed from halftone dot values c, m, and y of C, M, and Y data. Here, in order to explain the Neugebauer equation, the relationship of the halftone dot area according to the equation is shown in FIG. FIG. 3 shows the relationship between the two colors of yellow ink and magenta ink as an area occupied by a square having a dot area ratio of 1.0 × 1.0. In FIG. 3, if the reflectances Ry, Rm, and Rmy are the reflectances of the yellow ink, the magenta ink, and the magenta ink superimposed on the yellow ink, respectively, measured with red light, the total red light obtained by combining these is shown. (Total red light reflectance) can be calculated. For example, the contribution of the yellow-magenta overprint is m · y · Rmy, and the contribution of the magenta area is m · (1-y) · Rm.
Further, each of the reflectances Ry, Rm, and Rmy can be measured for blank paper. Accordingly, the total red light reflectance R is represented by the following equation (5). R = (1-y) · (1-m) + m · (1-y) · Rm +
y (1−m) Ry + m · y · Rmy (5) The area of the three colors C, M, and Y measured with red light, that is, the total red light reflectance is represented by the following equation for the same reason as described above. It is. R = (1-c) · (1-m) · (1-y) + c · (1-
m) · (1-y) Rc + m · (1-c) · (1-y) · Rm
+ Y ・ (1-c) ・ (1-m) ・ Ry + m ・ y ・ (1-
c) · Rmy + cy · (1-m) · Rcy + cm · (1-
y) · Rcm + c · m · y · Rcmy (6) where Rc is the reflectance measured for the cyan ink, Rcy is the reflectance measured for the cyan ink superimposed on the yellow ink,
Rcm indicates the reflectance measured for the cyan ink overlaid on the magenta ink, and Rcmy indicates the reflectance when yellow, magenta and cyan are overlaid. The reflectance by blue light and green light can be similarly obtained by substituting G and B for R in equation (6) (Literature: “Color” by AC Cole published by Printing Academic Press. Reproduction theory "). From the equation (6), if the stimulus values Xc and Xcm are used instead of the reflectances Rc and Rcm in the equation (6), the CM from the separated color data cmy by halftone dot% to the three stimulus values X, Y, and Z It can be seen that Y XYZ conversion can be performed. From equation (6), it can be said that the feature of the conversion by the Neugebauer equation is to use a cubic equation. However, since the Neugebauer equation has no solution and no inverse transformation as described above,
In the case of the inverse transformation, another approximation formula must be used.
Therefore, it is difficult to use in data compression applications. For this reason, many approximations and degenerations have been attempted in order to compress data, but since the conditions are limited to RGB filter densities, they cannot be used for XYZ conversion for printing. The Neugebauer equation has a large error and is disadvantageous in terms of accuracy. As a result of various studies, the inventors measured the chromaticity of a color patch printed matter in which the dot% values c, m, and y of the respective colors C, M, and Y were previously determined using a chromaticity meter, and obtained the resulting data. It was found that a least-squares method was used to calculate a polynomial for CMY → XYZ conversion using the calculated coefficients. Here, the configuration of this polynomial will be described. At the time of forward conversion, X of the three stimulus values is represented by the following equation (7). _{X = a 10 + a 11 c} + a 12 m + a 13 y + a 14 cm + a 15 cy + a 16 my + a
₁₇ c ² + a ₁₈ m ² + a ₁₉ y ² + a _{1 10} c ² m + a _{1 11} m ² y + a _{1 12} y ² c +
a _{1 13} c ² y + a _{1 14} m ² c + a _{1 15} y ² m + a _{1 16} cmy (7) where a _ij is a coefficient, and its subscript i = 1, 2, 3,
It takes a value of j = 0 to 16. The coefficient a _ij can be determined in various ways, for example, as follows. That is, many print Karapatsuchi the dot% values are pre-divide (e.g. CM Y 1728 colors each 12 steps), to measure the XYZ values of all Karapatsuchi like color colorimeter. Using these data, the coefficient a _ij is determined by the least squares method. The other stimulus values Y and Z also form the same polynomial as in equation (7), and the coefficients a _ij are determined in the same procedure. Also, from the three stimulus values XYZ calculated from the equation (7) and the like, 3
The formula for the inverse transform to the primary data CM Y, can be obtained as the following equation (8). C = b ₁₀ + b ₁₁ X + b ₁₂ Y + b ₁₃ Z +... B _{1 16} XYZ (8) where b _ij is a coefficient and i is 1, 2, 3, j
Takes a value from 0 to 16. The other primary color data M and Y can be similarly obtained. In this inverse transformation, similarly to the forward transformation, the coefficient b _{ij is} obtained by the least squares method using the measured data of the color patches.
Can be determined. The present invention has been made based on the above viewpoints. According to the present invention, since YIQ data can be created via the tristimulus values XYZ from CM Y data, it can be corrected non-complementary with the RGB. Therefore, unlike the conventional method, it is possible to convert the color image information for printing into the correct luminance Y and chromaticity axes I and Q, and apply this YIQ conversion to the image for printing by the compression method described in the preceding paragraph. Since compression can be performed, the compression efficiency can be increased. In addition, intuitive color adjustment, which was impossible in the past, can be easily performed by adjusting the data of the chromaticity axes I and Q. Furthermore, since the conversion into YIQ data is not limited to the type of image data of the television and the image data for printing, it is possible to directly exchange (interface) between the television and the printing through the YIQ data. Become. Note that the conversion is not limited to using a polynomial as in equations (7) and (8). Other suitable techniques can be used.

【Example】

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. This embodiment is an apparatus having a configuration as shown in FIG. 1 for performing YIQ conversion of C, M, and Y data of an image. This YIQ
The conversion device, as shown in FIG. 1, C ² from CM Y data input values of the ^{image, M 2, Y 2, CM} , M Y, and higher-order terms generator 10 which calculates the Y C ... CM Y , And each generated higher-order term is described below (1
It is mainly composed of a matrix operation unit 12 for multiplying by a coefficient matrix S as in equations (1) and (12) and performing YIQ conversion, and a coefficient storage means 14 for storing the coefficient matrix S. The high-order term generation unit 10 and the matrix calculation unit 12 can be configured by a software calculation procedure stored in a computer, or can be configured by an analog calculator in hardware. Next, the calculation procedure of the YIQ conversion device will be described. Halftone% of C, M, Y input as CMY data of image
c, m, and y change in the range of 0 to 255 gradations. In the case where the CM Y data is converted into YIQ data in advance, if the tristimulus values XYZ are calculated by the above equation (7), the YIQ is calculated by the above equation (2) using the tristimulus values X, Y, and Z. Is obtained, but these equations (7) and (2) can be expressed as one equation as shown in the following equation (9). Also, from the equation (2) and the equation (9), the luminance Y, the chromaticity axis I,
Q can be expressed as in the following equation (10). Here, if the coefficient matrix S in the expression (10) is placed as in the following expression (11), the expression (10) becomes as in the following expression (12). When performing the forward conversion, each coefficient a in the coefficient matrix is obtained in advance as follows. First, the total of 12 levels of each color C, M, Y
Printed on the 12 ^3-color, halftone dot% c, m, y is a known Karapatsuchi measured using a color difference meter each stimulus values X, Y, seek Z. Next, the tristimulus value and the halftone dot% c, m,
y is substituted into equation (9), and each coefficient a ₁₀ -a
_{3 Ask} for ₁₆ . In this case, since the tone value of the dot percentage is 0 to 255, the dynamic range can be sufficiently obtained and the error can be reduced. During forward transform inputs the CM Y data of the image, C from CM Y data input values of the input image with higher-order terms generator 10
² , M ² , Y ² , CM, MY , Y C... CM Y are calculated and input to the matrix calculator 12. In this matrix computation section 12, (11) supra coefficient matrix S in (3) and (9) is calculated from the equation, the coefficient matrix using the S (12) below the line operations of connexion CM Y Find YIQ data from data. Note that the coefficient matrix S obtained as described above is stored in the coefficient storage means 14 as appropriate. Further, if the inverse transformation to the CM Y data from the YIQ theta obtains the inverse transformation matrix S in the same procedure, to determine a CM Y data subjected to calculation using the YIQ data in matrix calculator 12. Next, an example in which compression processing is performed on the YIQ-converted image data as described above will be described. First, an image data transmission system to which thinning interpolation is applied will be described. This image data transmission system is, as shown in FIG.
A Q conversion device 16 and an inverse YIQ conversion device 18 are provided. The YIQ
The conversion device 16 and the inverse YIQ conversion device are the same as those shown in FIG.
The conversion device is used for forward conversion and inverse conversion. A quantizing unit 20 for linearly quantizing the data of the chromaticity axes I and Q is provided between the conversion units 16 and 18, and a thinning unit for thinning out.
22, each operation unit of the interpolation unit 24 for interpolating the data after transmission is provided. The luminance Y data and the thinned chromaticity axis I and Q data are transmitted to the interpolation unit 24 and the inverse YIQ converter 18 via the communication unit 26. In the decimation interpolation apparatus first converts the CM Y value of the print image data to YIQ values YIQ converter 16, 0 chromaticity axis I, Q data of which the linear quantizer 20
Performs linear quantization to 8-bit data of 255 gradations. Next, the thinning unit 22 thins out data of appropriate pixels from the linearly quantized data, and compresses the data. The compression ratio differs depending on the thinning ratio (thinning ratio). The thinned chromaticity IQ data and luminance Y data are transmitted via the communication unit 26. The transmitted luminance Y data directly reaches the inverse YIQ conversion unit 18, but the transmitted chromaticity axis I and Q data is interpolated by the interpolation unit 24 to obtain the original data. And reaches the inverse YIQ converter 18. The luminance Y data and the interpolated chromaticity axis I and Q data are subjected to inverse YIQ conversion by the inverse YIQ conversion unit 18 and output as image data C ′, M ′, and Y ′. As described above, even if the data of the chromaticity axes I and Q are thinned out, the effect on the image quality is small due to the resolution characteristics of the color sensation as shown in FIG. Therefore, data compression without lowering image quality,
Transmission can be realized. Next, referring to FIG. 5, discrete cosine transform (DCT)
An image data transmission system to which is applied will be described. In this case, a DCT unit 28 for performing DCT is connected to the YIQ conversion device 16 as shown in FIG.
32 is connected to reach the transmission unit 26. On the output side of the transmission unit 26, an inverse DCT unit 30 for performing an inverse DCT on the transmitted data is provided, and reaches the inverse YIQ conversion device 18. Here, the DCT unit 28 divides the image into blocks of N × N pixels (for example, 16 × 16 pixels) for each YIQ data, and performs DCT on the divided image data.
That is, in FIG. 5 the image data transmission system and YIQ converting the CM Y data YIQ conversion section 16, the converted YIQ each image is decomposed into blocks of N × N pixels and performs DCT.
Since this DCT is a kind of spatial frequency analysis (a kind of orthogonal transform like FFT), it is necessary to reduce (cut) high frequency component information and compress the data amount according to the resolution (spatial frequency characteristics) of the color vision system. Can be. The amount of information that can be reduced is YI
Since the values become larger in the order of Q, the chromaticity axis Q can be compressed most and the luminance Y cannot be compressed most.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a detailed configuration of a YIQ conversion device according to an embodiment of the present invention, and FIG. 2 is a line showing an example of a change in sensitivity to visual resolution for explaining the principle of the present invention. FIG. 3 is a plan view for explaining the reflection of the halftone dot area ratio, and FIG. 4 is a block diagram showing a configuration example of an image data transmission system using thinning-out interpolation to which the above embodiment is applied. FIG. 5 is a block diagram showing a configuration example of an image data transmission system using the discrete cosine transform. 10 high-order term generation unit, 12 matrix operation unit, 14 coefficient storage means, 16 YIQ conversion device, 18 inverse YIQ conversion device, 20
…… Linear quantizer, 22… Thinner, 24… Interpolator, 26
... Communication unit, 28 DCT unit, 30 inverse DCT unit, 32 quantization unit.

Claims (2)

(57) [Claims] 1. A method for converting CMY data of cyan, magenta, and yellow Y of each separation color of a print image into YIQ data of luminance Y and chromaticity axes I and Q for band compression and transmission. Then, by measuring the reflectance of the printed material having the CMY value passing, the CMY data and the X values of the visual tristimulus values X, Y and Z are determined in advance.
Creating an equation that defines the relationship between the YZ data; converting the CMY data into XYZ data using the relational expression; and converting the XYZ data into YIQ data. Print transmission primary color conversion method. 2. Converting YIQ data converted to luminance Y and chromaticity axes I and Q, band-compressed and transmitted to CMY data of cyan C, magenta M and yellow Y of each separation color of a print image. This is a method of converting a print transmission primary color to convert, by measuring the reflectance of a printed material having a CMY value passing through, and previously determining the XYZ data of visual tristimulus values X, Y, Z and CMY
Creating an expression that defines the relationship between the data, converting the YIQ data to XYZ data, and converting the XYZ data to CMY data using the relational expression. Print transmission primary color conversion method.