JP4506948B2 - Color conversion coefficient creation method, color conversion coefficient creation apparatus, color conversion coefficient creation program, and storage medium - Google Patents

Description

  The present invention relates to a technique for generating a color conversion coefficient of a multidimensional table used for conversion from a device color signal of a first image input / output device to a device color signal of a second image output device. It is.

  In recent years, printing simulations using a xerographic printer, an inkjet printer, or the like have become popular. In order to perform such a printing simulation, it is necessary to perform color conversion so that colors in printing can be reproduced by a printer that is actually used. As one of the color conversion techniques, there is a method of using a multidimensional lookup table (multidimensional LUT) and interpolation together.

  In color conversion in such a printing simulation, as described in Patent Document 1, for example, printing ink is used to maintain faithful color reproduction by colorimetric matching and the texture and contrast of the reproduced image. (K) It is important to preserve K so as to maintain the version information as much as possible. Furthermore, in order to maintain the texture and contrast of the reproduced image, not only the printing K plate information is maintained as much as possible, but also at the expense of fidelity of color reproduction, the printing K single color is the printer or K single color. It may be better to reproduce with. Furthermore, a phenomenon called “fogging” occurs due to the difference between the media used for printing and the printer (paper medium), and emphasis is placed on the fidelity of color reproduction for relatively bright colors such as yellow (Y) for printing. If the image is reproduced by a printer, other toners and inks are mixed on the printer side, which may be unfavorable because of turbid color reproduction.

  As a method for solving such a problem, as described in Patent Literature 1, Patent Literature 2, Patent Literature 3, and the like, white in printing is reproduced in white even with a printer, and a single color in printing is made monochromatic in a printer. There is a method for calculating such a color conversion coefficient. If the color conversion coefficient of the multi-dimensional LUT is determined using these methods, the white color in printing can be reproduced in white by the printer and the occurrence of “fogging” can be prevented. In addition, the K single color in printing can be reproduced with a K single color by a printer, and by maintaining the information of the black (K) plate of printing as much as possible, the texture and contrast of the reproduced image can be maintained.

  However, in any of the methods described in Patent Document 1, Patent Document 2, and Patent Document 3 described above, some or all of the output color signal components of the grid of the multidimensional LUT that satisfy the conditions are input. It is a method of equalizing the device color signal component, and whether or not a part or all of the input device color signal component of printing is equal to the output color signal component is independently determined for each grid of the multidimensional LUT. is doing. Therefore, the difference in characteristics between the printing that is the target and the printer that is the output device, that is, the white color difference between the two, the difference in chromaticity of the ink (toner) corresponding to both K, When the difference in hue and lightness of chromatic color ink (toner) is large, only the output color signal of a specific grid in the multidimensional LUT is largely corrected, and the output color signal in the grid adjacent to the specific grid It will not be possible to match with. When a printing simulation is performed using a multidimensional LUT in which such a situation has occurred, there has been a problem that defects such as pseudo contours and tone reversals occur in the reproduced image.

JP 2002-152543 A Special table 2003-501897 gazette JP 2002-330303 A

  The present invention has been made in view of the above-described circumstances, and reproduces the color reproduction in the first device such as the target printing as faithfully as possible with the second device such as a printer, as required. The reproduction-guaranteed color such as white or single color in the first device is reproduced in white or single color in the second device, and the boundary between the color that is faithfully reproduced and the color that is forcibly reproduced in white or single color. It is an object of the present invention to provide a color conversion coefficient creation method and a color conversion coefficient creation apparatus for determining a color conversion coefficient of a multidimensional LUT that performs color conversion so as not to cause gradation steps and gradation inversion. It is another object of the present invention to provide a color conversion coefficient creation program for executing such a color conversion coefficient creation method on a computer and a storage medium storing such a color conversion coefficient creation program.

  The present invention relates to a color conversion coefficient creating method and a color conversion coefficient creating apparatus for creating a color conversion coefficient of a multidimensional table used for converting a device color signal of a first device into a device color signal of a second device. As for the reproduction guarantee color consisting of any one or more of white, primary color, secondary color, or even 100% black, black single color in the first device, the corresponding white, primary color in the second device Color reproduction, secondary color, or even 100% black, black color reproduction guaranteed color, and color reproduction on the first device is reproduced as faithfully as possible on the second device and reproduction guarantee on the first device Multi-dimensional table color conversion so that no gap or gradation inversion occurs at the boundary with the portion where the color is reproduced with the corresponding guaranteed reproduction color in the second device It is intended to calculate the number.

  As one configuration example for creating such a color conversion coefficient of a multidimensional table, a device-independent color signal is used to reproduce a reproduction-guaranteed color in the first device with a corresponding reproduction-guaranteed color in the second device. A correction vector to be corrected is calculated by the reproduction guarantee correction data calculation means, and the device independent color signal corrected by applying the correction vector to the device independent color signal corresponding to the device color signal of the first device A device color signal of the second device is predicted by the inverse color prediction unit from the device model of the second device based on the corrected device-independent color signal obtained by the independent color signal correcting unit, and predicted. The color conversion coefficient output means outputs the device color signal of the second device as a color conversion coefficient of a multidimensional table, and the correction As a vector, a correction vector having a magnitude of 0 is associated with a color that is more than a predetermined distance away from the reproduction guaranteed color in the first device, and a color that is more than a predetermined distance away from the reproduction guaranteed color and the reproduction guaranteed color. A corrected device-independent color signal corresponding to the device color signal of the first device is obtained using a set of the device-independent color signal and the correction vector.

  Alternatively, as another configuration example, the reproduction guarantee correction data calculation unit generates a correction vector for correcting the device-independent color signal in order to reproduce the reproduction guarantee color in the first device with the corresponding reproduction guarantee color in the second device. A device-independent color signal corrected and applied by applying a correction vector to the device-independent color signal constituting the device model of the first device is obtained by the target device base data correction unit, and corrected device-independent The device-independent color signal is predicted from the device color signal of the first device using the device color model of the first device based on the base data including the color signal, and the predicted device-independent color signal is predicted. To the device color signal of the second device using the device model of the second device. And the color conversion coefficient output means outputs the predicted device color signal of the second device as a color conversion coefficient of a multidimensional table, and further reproduces the correction vector as the correction vector in the first device. A correction vector having a magnitude of 0 is associated with a color that is separated from the guaranteed color by a predetermined distance or more, and the device-independent color signal of the reproduction guaranteed color and a color that is separated from the reproduction guaranteed color by a predetermined distance or more and the correction vector Using the set, a corrected device-independent color signal corresponding to the device-independent color signal constituting the device model of the first device is obtained.

  Note that the elements of guaranteed reproduction colors can be determined in advance or designated by the user.

  In the present invention, as a correction vector, a correction vector having a magnitude of 0 is associated with a color that is a predetermined distance or more away from a reproduction guarantee color in the first device, and the correction vector having a magnitude of 0 is associated with a device-independent color signal. Is applied so as to obtain a device-independent color signal corrected. As a result, it is possible to correct only the reproduction guarantee color and its neighboring colors.

  According to the present invention, the predetermined reproduction guarantee color (white, black, primary color, secondary color, or a combination thereof) in the first device is a reproduction guarantee color corresponding to the second device. (White, black, primary color, secondary color, or a combination thereof), and other colors are reproduced as closely as possible to the target device. It is possible to calculate a color conversion coefficient of a multidimensional LUT that is reproduced so that a gradation step or gradation inversion does not occur in the reproduction of the color corresponding to the boundary portion. Therefore, by using the multidimensional LUT set with the color conversion coefficient created according to the present invention and applying it to a color image in combination with an existing interpolation method, for example, a reproduction guaranteed color (for example, printing) in the first device ( White, black, primary color, secondary color, or a combination thereof is a guaranteed reproduction color (white, black, primary color, secondary color) that is also supported by a second device such as a printer. Or any combination thereof), no gradation step or reversal occurs near the boundary between the predetermined color and other colors. Thus, there is an effect that a good first device simulation can be realized by the second device.

  Further, the guaranteed reproduction colors (white, black, primary color, secondary color, or a combination thereof) in the first device also correspond to the reproduction guaranteed colors (white, black, 1 In the case of creating a multi-dimensional LUT having a large number of grids, the correction target for reproduction with either the secondary color, the secondary color, or a combination thereof is used as the base data of the first device. There is an effect that an increase in processing time for correction can be reduced.

  FIG. 1 is a block diagram showing a first embodiment of the present invention. In the figure, 1 is an address grid generation unit, 2 is a target device forward color prediction unit, 3 is a reproduction guarantee correction data calculation unit, 4 is a device-independent color signal correction unit, 5 is a color gamut compression unit, and 6 is an output device inverse unit. A color prediction unit 7 is a color conversion coefficient output unit. In the following description, the target device corresponding to the first device is a printing machine, and the output device corresponding to the second device is a CMYK printer. Therefore, the multi-dimensional LUT used for color conversion is assumed to be a four-dimensional LUT with CMYK input and CMYK output, and an example of determining the color conversion coefficient will be described. Further, when creating this color conversion coefficient, the device-independent color signal is processed, and the CIELAB color signal is used as the device-independent color signal.

  The address grid generation unit 1 generates CMYK addresses of the multidimensional LUT. Here, nine equally spaced grids are set on the C, M, Y, and K axes, and Ci, Mi, Yi, and Ki, which are CMYK addresses corresponding to the grids, are generated in order. Since the CMYK address has nine grids for each axis, i takes a value from 1 to 6561 (= 9 × 9 × 9 × 9). The number of grids on each axis can be freely set according to the purpose.

  The target device forward color prediction unit 2 converts the CMYK color signal that is the CMYK address generated by the address grid generation unit 1 into a CIELAB color signal that is a device-independent color signal. The conversion to the device-independent color signal can be performed using an existing method such as a method using a model using a neural network, a method using a regression model, or a method using matrix conversion. As a specific example, Li, which is a device-independent color signal, is obtained from a target device color signal Ci, Mi, Yi, Ki using a prediction model using weighted linear regression described in JP-A-10-262157. ai and bi can be predicted. The color conversion based on the prediction model requires target device base data that is a plurality of sets of device color signals CMYK and corresponding device-independent color signals CIELAB. This target device base data depends on the target device and is acquired in advance.

  The reproduction guarantee correction data calculation unit 3 is a device-independent color signal in the target device so as to reproduce the reproduction guarantee color of the target device determined in advance or instructed by the user with the reproduction guarantee color of the output device. A correction vector for correcting is calculated. In this example, the correction vector is composed of a CIELAB color signal as a start point and a CIELAB color signal as an end point. The CIELAB color signal as the starting point corresponds to the corresponding target device color signal CMYK. The reproduction guaranteed colors are primary colors such as white (C = M = Y = K = 0%), solid black (C = M = Y = 0%, K = 100%), and black, and secondary colors. Or one of the pure colors.

  FIGS. 2 and 3 show the case where the black (K) single color is used as the guaranteed reproduction color, FIGS. 4 and 5 show the same case when the yellow (Y) single color is used as the guaranteed reproduction color, and FIGS. 6 and 7 show the same red color. (R) is an explanatory diagram of an outline of correction and a correction vector when a pure color is used as a guaranteed reproduction color. In FIG. 2, in the case where the K single color is used as the guaranteed reproduction color, the K single color locus in the target device is indicated by a broken line, and the K single color locus in the output device is indicated by a solid line. As described above, when the devices are different, the colorimetric values are often different even if reproduced with K single color. In the reproduction guarantee correction data calculation unit 3, when the K single color is set as the reproduction guarantee color (K single color guarantee), first, it corresponds to the K single color of the output device having the same brightness as the CIELAB color signal corresponding to the K single color of the target device. The CIELAB color signal to be calculated is calculated. For example, the CMYK color signal corresponding to the K single color of the target device as shown in FIG. 3 is created, and the CIELAB color signal corresponding to the K single color of the output device having the same brightness as the corresponding CIELAB color signal is calculated. Can do. Then, a vector having the CIELAB color signal corresponding to the K single color of the original target device as the start point and the CIELAB color signal corresponding to the K single color of the output device having the same calculated brightness as the end point is calculated as the correction vector. A plurality of such correction vectors are calculated by appropriately shaking K, and each correction vector is added to the reproduction guarantee correction data as shown in FIG.

  Note that the correction vectors shown in FIG. 3 and FIGS. 5 and 7 to be described later are examples in which the correction vector is composed of a CIELAB color signal as a start point and a CIELAB color signal as a corresponding end point. However, the present invention is not limited to this, and any format may be adopted as long as the reproduction guarantee correction data can apply a desired correction to the CIELAB color signal. For example, a difference vector between the CMYK color signal of the target device and the corresponding correction data can be used as the reproduction guarantee correction data.

  In this example, the CIELAB color signal corresponding to the K single color of the output device having the same lightness as the CIELAB color signal corresponding to the single K color of the target device is set as the end point, but the density may be used instead of the lightness. Using an appropriately weighted color difference or other existing color difference formula, etc., a correction vector is calculated with the CIELAB color signal corresponding to the K single color of the output device as the end point that minimizes the evaluation value. May be.

  In FIG. 4, in the case where the Y single color is set as the guaranteed reproduction color, the Y single color locus in the target device is indicated by a broken line, and the Y single color locus in the output device is indicated by a solid line. In this way, when the Y single color is used as the guaranteed reproduction color (Y single color guaranteed), first, the CIELAB color signal corresponding to the Y single color of the output device having the same saturation as the CIELAB color signal corresponding to the Y single color of the target device is first obtained. calculate. Then, a vector having the CIELAB color signal corresponding to the Y monochrome of the original target device as the start point and the CIELAB color signal corresponding to the Y monochrome of the output device having the same calculated saturation as the end point is calculated as the correction vector.

  A plurality of such correction vectors are calculated by appropriately shifting Y, and each correction vector is added to the reproduction guarantee correction data as shown in FIG.

  In this example, the CIELAB color signal corresponding to the Y monochromatic color of the output device having the same saturation as the CIELAB color signal corresponding to the Y monochromatic color of the target device is set as the end point. However, the lightness may be used instead of the saturation. In addition, by using appropriately weighted color differences or existing color difference formulas, etc., a correction vector whose end point is the CIELAB color signal corresponding to the Y color of the output device that minimizes the evaluation value is calculated. May be. 4 and 5 have taken up an example of Y single color, but other primary colors such as magenta (M) single color and cyan (C) single color also correspond to the CMYK color signal of the target device. A set of correction vectors can be calculated as reproduction guarantee correction data.

  In FIG. 6, in the case where the R pure color is used as a guaranteed reproduction color, the locus of the R pure color in the target device is indicated by a broken line, and the locus of the R pure color in the output device is indicated by a solid line. In addition, the outer surface of the WYRM including the R pure color locus in the output device is shown by hatching. In this way, when the R pure color is used as the guaranteed reproduction color (R pure color guarantee), first, the CIELAB color corresponding to the R pure color of the target device (a color signal sequence in which Y and M are equal and C is 0%). The CIELAB color signal on the color gamut outline surface (on the WYRM outline surface) of the output device having the smallest color difference from the signal is calculated. Then, a vector having the CIELAB color signal corresponding to the R pure color of the original target device as the start point and the CIELAB color signal on the color gamut outer surface of the output device having the smallest calculated color difference as the end point is calculated as a correction vector.

  A plurality of such correction vectors are calculated by appropriately shifting Y (= M), and each correction vector is added to the reproduction guarantee correction data as shown in FIG.

  In this example, the CIELAB color signal on the color gamut outline of the output device that minimizes the color difference from the CIELAB color signal corresponding to the R pure color of the target device is set as the end point. By using a color difference formula or the like, a correction vector having the CIELAB color signal on the color gamut contour of the output device that minimizes the evaluation value as an end point may be calculated. 6 and 7, the example of the R pure color is taken up. However, other secondary colors such as green (G) pure color and blue (B) pure color correspond to the CMYK color signals of the target device in the same manner. A set of correction vectors can be calculated as reproduction guarantee correction data.

  Also, when the target device white color (C = M = Y = K = 0%) is used as the guaranteed reproduction color and it is desired to reproduce the output device white color (white reproduction guaranteed), the target device white color (WT) in FIG. A correction vector having a CIELAB color signal corresponding to the start point as a starting point and a CIELAB color signal corresponding to the white color (WO) of the output device as an end point may be calculated and added to the reproduction guarantee correction data. In FIG. 3, FIG. 5, and FIG. 7, this correction vector is also included.

  Further, when it is desired to reproduce the black of the target device (C = M = Y = 0%, K = 100%) with the black of the output device, the CIELAB color signal corresponding to the black of the target device in FIG. A correction vector starting from (KT) and ending with the CIELAB color signal (KO) corresponding to the black of the output device may be calculated and added to the reproduction guarantee correction data.

  For example, when “white reproduction guarantee”, “K single color guarantee”, and “Y single color guarantee” are always enabled, reproduction guarantee correction data including a set of correction vectors shown in FIGS. 3 and 5 may be generated. Further, for example, when determining an element for which reproduction is guaranteed according to an instruction from the user, reproduction guarantee correction data including a set of correction vectors corresponding to the reproduction guaranteed color instructed by the user may be created.

  A plurality of device-independent color signals that are separated from the correction vector in the reproduction guarantee correction data created as described above by a certain distance or more are extracted, and these device-independent color signals have a start point and an end point of size 0. Create a correction vector. The constant distance here is a distance in a device-independent color space. That is, in this example, it indicates the distance in the CIELAB color space. As a method for creating such a correction vector of size 0, for example, a CIELAB color signal corresponding to a CIELAB lattice point group (11 × 11 × 11 = 1331) that can be obtained by dividing each axis of CIELAB into 10 is generated. If each CIELAB color signal is at least a certain distance from a correction vector for realizing a single color (pure color) guarantee, in other words, a line connecting the start point and the end point of the correction vector, the CIELAB color signal is set to the start point. There is a method of adding a correction vector of zero size at the end point as reproduction guarantee correction data. Of course, other methods may be used. In this way, a zero correction vector is calculated and added to the reproduction guarantee correction data. In addition, a correction vector having a magnitude of 0 is added if the distance is more than a certain distance, and a correction vector having a magnitude corresponding to the distance is added from a certain distance to the distance 0. Also good.

  The reproduction guarantee correction data finally created in this way is a correction vector composed of a CIELAB color signal corresponding to a CMYK device color signal of a target device that is a correction source to be guaranteed reproduction and a CIELAB color signal that is a correction destination. And a set of zero correction vectors having a CIELAB color signal corresponding to the CMYK device color signal of the target device that is a correction source that does not need to be corrected at the start and end points.

  Returning to FIG. 1, the device-independent color signal correction unit 4 performs correction for reproduction guarantee on the device-independent color signal predicted by the target device forward color prediction unit 2. First, a correction vector to be applied to the predicted device-independent color signal Li, ai, bi is calculated from the reproduction guarantee correction data calculated by the reproduction guarantee correction data calculation unit 3 and Li, ai, bi. Here, the weighted linear regression described in, for example, Japanese Patent Application Laid-Open No. 10-262157 is performed using the start point CIELAB color signal and the corresponding end point CIELAB color signal constituting the correction data in the reproduction guarantee correction data as base data. Using the prediction model used, the prediction model may be applied to Li, ai, bi, and the corrected device-independent color signals L′ i, a′i, b′i may be calculated. As a result, it is possible to obtain corrected device-independent color signals L′ i, a′i, b′i that do not cause gradation inversion. At the same time, the corrected device-independent color signals L′ i, a′i, and b′i are the device-independent color signals that have been determined in advance or that have been guaranteed to be reproduced by the user. Become.

  The color gamut compression unit 5 determines whether or not the corrected device-independent color signals L′ i, a′i, and b′i are within the color reproduction range of the output device. Is a device-independent color signal L "i, a" i, b "that can be reproduced by the output device by clipping on the color gamut outline of the output device or mapping within the color gamut of the output device by existing methods. Any known method can be used as the color gamut compression method.

  The output device inverse color prediction unit 6 holds the K amount of the target device as much as possible from the device independent color signals L ″ i, a ″ i, b ″ i that can be reproduced by the output device and the device color signal Ki of the target device. C′i, M′i, Y′i, and K′i, which is a combination of coverages that reproduce the colors of the color signals Ci, Mi, Yi, and Ki in the target device with the output device as accurately as possible. The prediction method used sometimes is not particularly limited, and for example, the K-preserving reverse color prediction method described in Patent Document 1 may be used, or described in Japanese Patent Application No. 2002-271322. A LABK → CMYK model corresponding to the total amount regulation may be used.

  The color conversion coefficient output unit 7 outputs C′i, M′i, Y′i, and K′i, which are device color signals of the output device, in a format that can be used as a multidimensional table. For example, it can be output as a de facto standard profile such as a device link / ICC profile. Alternatively, for example, when used by being incorporated in an output device, it may be stored as a parameter of a multidimensional LUT in the output device or may be directly set.

  As described above, according to the first embodiment, the guaranteed reproduction color (white, solid, primary color, secondary color, or a combination thereof) in the target device is also supported by the output device. Reproduction guarantee colors (white, solid, primary color, secondary color, or combinations thereof) are reproduced, and other colors are reproduced as closely as possible to the target device. It is possible to calculate a color conversion coefficient of a multi-dimensional LUT that reproduces a color that corresponds to a boundary portion with a color of a color that does not cause a gradation step or gradation inversion. Therefore, by performing color conversion of a color image in combination with an existing interpolation method using a multidimensional LUT to which a color conversion coefficient created according to the present invention is applied, reproduction guaranteed colors (white, solid, (Primary color, secondary color, or a combination thereof) is reproduced with a reproduction guaranteed color (white, black, primary color, secondary color, or a combination thereof) that is supported by the output device. On the other hand, it is possible to realize an excellent target device simulation in which no gradation step or inversion occurs in the vicinity of the boundary between the guaranteed reproduction color and other colors.

  In the above description, a printer is assumed as the target device and a printer is assumed as the output device. However, the present invention is not limited to this, and both the target device and the output device may be arbitrary apparatuses. In the above description, the color conversion coefficient for performing the conversion from CMYK to CMYK is obtained. However, for example, the target device side may be another color signal such as RGB or LAB, and the output device side is the same. The device-independent color signal to which reproduction guarantee correction data is applied is not limited to CIELAB, and other device-independent color signals such as CIELV, CIECAM02, and IPT may be used.

  FIG. 8 is a block diagram showing a second embodiment of the present invention. In the figure, the same parts as those in FIG. Reference numeral 11 denotes a target device base data correction unit. Also in the second embodiment, the target device, which is the first device, is a printing machine, the output device, which is the second device, is a CMYK printer, and the CMYK input and CMYK output four-dimensional LUT color conversion coefficients are set. An example of determination will be described.

  The reproduction guarantee correction data calculation unit 3 calculates a correction vector by the method described in the first embodiment and creates reproduction guarantee correction data. The created reproduction guarantee correction data is a set of correction vectors having a CIELAB color signal corresponding to the CMYK device color signal of the target device as a correction source to be guaranteed reproduction as a start point and a CIELAB color signal as a correction destination as an end point. And a set of zero correction vectors having CIELAB color signals that do not need to be corrected at the start and end points.

  The target device base data correction unit 11 performs correction for guaranteeing reproduction on the device-independent color signal constituting the base data of the target device. For the correction processing for guaranteeing reproduction, for example, a prediction model using weighted linear regression described in JP-A-10-262157 is used, and the CIELAB color signal constituting the base data used for the prediction model is used. Can be done. The base data of the target device is data used for a prediction model for converting from CMYK to LAB by forward color prediction. For example, a CMYK color signal distributed over the entire CMYK color space of the target device and a CIELAB color signal obtained by outputting the CMYK color signal to a medium by the target device and measuring the color with a measuring instrument. be able to. In the second embodiment, the CIELAB color signal in the base data is corrected by the reproduction guarantee correction data.

  However, the present invention can also be applied to a device model based on other methods, for example, a prediction model based on a neural network. In this case, a correction for guaranteeing reproduction may be performed on the device-independent color signal which is teacher data for learning the coupling coefficient of the synapse coupling.

  As a procedure for correcting the base data of the target device, for each set of CMYK color signals and CIELAB color signals, a set of a plurality of CMYK color signals and CIELAB color signals constituting the base data of the target device. The correction process may be performed to correct the CIELAB color signal to be corrected.

  Specifically, the target CMYK color signal and CIELAB color signal set are Cj, Mj, Yj, Kj and Lj, aj, bj, respectively. At this time, the reproduction guarantee correction data calculated by the reproduction guarantee correction data calculation unit 3 is applied to Lj, aj, bj to correct Lj, aj, bj. For example, weighted linear regression described in Japanese Patent Laid-Open No. 10-262157 is used with CIELAB color signals that are the start points of correction vectors in the reproduction guarantee correction data and CIELAB color signals that are corresponding end points as base data. The prediction model is applied to Lj, aj, bj using a prediction model or the like, and corrected device-independent color signals L′ j, a′j, b′j are calculated. Then, the CIELAB color signals Lj, aj, bj of the set of the CMYK color signals and the CIELAB color signals that are the object may be replaced with L'j, a'j, b'j. As a result, the reproduction guarantee correction data can be applied to the target base data, and the corrected target base data can be calculated.

  The correction target base data created in this way has the characteristics of the target device, but the characteristics of the output device for the guaranteed reproduction colors (white, black, primary color, secondary color, or a combination thereof). In the vicinity of the boundary between the reproduction guarantee color and the other colors, the characteristics of the two continuously change.

  The address grid generation unit 1 generates CMYK addresses of the multidimensional LUT. Here, as in the first embodiment, an equally spaced grid is set for each CMYK axis. The number is set to 17, and Ck, Mk, Yk, Kk which are CMYK addresses corresponding to the grid. Are generated in order. Since 17 grids are provided for each axis, k takes a value from 1 to 83521 (= 17 × 17 × 17 × 17). The number of grids on each axis can be freely set according to the purpose.

  The target device forward color prediction unit 2 converts the CMYK color signal that is the CMYK address generated by the address grid generation unit 1 into a device-independent color signal. Here, the CIELAB color signal is used as the device-independent color signal, but other device-independent color signals such as CIELV, CIECAM02, and IPT may be used as in the first embodiment.

  In addition, the conversion to the device-independent color signal can be performed using an existing method such as a method using a model using a neural network, a method using a regression model, or a method using matrix conversion. As a specific example, using a prediction model using weighted linear regression described in Japanese Patent Laid-Open No. 10-262157, Lk which is a device-independent color signal from the target device color signals Ck, Mk, Yk, Kk. , Ak, bk can be predicted. At this time, the color conversion based on the prediction model in the target device forward color prediction unit 2 is performed based on the corrected target device base data created by the target device base data correction unit 11.

  The color gamut compression unit 5 determines whether or not the device-independent color signals Lk, ak, and bk predicted by the target device forward color prediction unit 2 are within the color reproduction range of the output device. If not, a device-independent color signal L′ k, which can be reproduced by the output device by clipping to a color on the color gamut outline of the output device or mapping to a color inside the color gamut of the output device by an existing method. Convert to a′k, b′k. Any known method may be used as the color gamut compression method. Note that, depending on the color gamut compression method to be employed, the color gamut outline information of the target device may be required. In this case, it is necessary to create the color gamut outline information based on the corrected target device base data. As a method for calculating the color gamut outline information based on the base data, for example, an arbitrary method such as a method described in JP-A-2003-8912 can be used.

  The output device inverse color prediction unit 6 holds the K amount of the target device as much as possible from the device independent color signals L′ k, a′k, b′k that can be reproduced by the output device and the device color signal Kk of the target device. C′k, M′k, Y′k, K′k, which is a combination of coverages for reproducing the colors of the color signals Ck, Mk, Yk, Kk in the target device with the output device as accurately as possible. This prediction method is also arbitrary and is not particularly limited. For example, the K-preserving reverse color prediction method described in Patent Document 1 may be used, or the LABK → CMYK model corresponding to the total amount restriction described in Japanese Patent Application No. 2002-271322 may be used. .

  The color conversion coefficient output unit 7 outputs C′i, M′i, Y′i, and K′i, which are device color signals of the output device, in a format that can be used as a multidimensional table. For example, it can be output as a de facto standard profile such as a device link / ICC profile. Alternatively, for example, when used by being incorporated in an output device, it may be stored as a parameter of a multidimensional LUT in the output device or may be directly set.

  As described above, also in the second embodiment, the reproduction guaranteed color in the target device is reproduced with the corresponding reproduction guaranteed color in the output device, and the other colors as in the first embodiment. Is reproduced as faithfully as possible on the target device, and the color conversion coefficient of the multidimensional LUT that reproduces the color corresponding to the boundary between the guaranteed reproduction color and other colors without causing a gradation step or gradation inversion. Can be calculated. Therefore, by using the multi-dimensional LUT to which the color conversion coefficient created by the present invention is applied and performing color conversion of the color image in combination with the existing interpolation method, the reproduction guaranteed color in the target device can be supported by the output device. The output device can realize a good simulation of the target device in which no gradation step or reversal occurs in the vicinity of the boundary between the guaranteed reproduction color and other colors while reproducing with the guaranteed reproduction color.

  Note that the second embodiment can be modified in the same manner as the first embodiment described above.

  In the first and second embodiments described above, the method for determining the color conversion coefficient of the multidimensional LUT has been described. By applying the color conversion coefficient created according to the present invention to the multidimensional LUT, it is possible to realize good color reproduction. This multidimensional LUT is used together with an interpolation process, and any interpolation method such as tetrahydra interpolation or cubic interpolation can be used as an interpolation method.

  In color conversion, color conversion may be performed in combination with a multi-dimensional LUT and a one-dimensional lookup table (one-dimensional LUT). Even in such a case, it is possible to determine the color conversion coefficient of the multidimensional LUT according to the present invention.

  FIG. 9 is an explanatory diagram of an example of a computer program and a storage medium storing the computer program when the functions of the embodiments of the present invention are implemented by the computer program. In the figure, 21 is a program, 22 is a computer, 31 is a magneto-optical disk, 32 is an optical disk, 33 is a magnetic disk, 34 is a memory, 41 is a magneto-optical disk apparatus, 42 is an optical disk apparatus, and 43 is a magnetic disk apparatus.

  Part or all of the processing of each unit described in each embodiment described above can be realized by a program 21 that can be executed by a computer. In this case, the program 21 and data used by the program can be stored in a computer-readable storage medium. A storage medium is a signal format that causes a state of change in energy such as magnetism, light, electricity, etc. according to the description of a program to a reader provided in the hardware resources of a computer. Thus, the description content of the program can be transmitted to the reading device. For example, there are a magneto-optical disk 31, an optical disk 32 (including a CD and a DVD), a magnetic disk 33, a memory 34 (including an IC card, a memory card, and the like). Of course, these storage media are not limited to portable types.

  By storing the program 21 in these storage media and mounting these storage media in, for example, the magneto-optical disk device 41, the optical disk device 42, the magnetic disk device 43, or a memory slot (not shown) of the computer 22, It is possible to read the program 21 and execute the functions described in the embodiments of the present invention. Alternatively, a storage medium may be attached to the computer 22 in advance, and the program 21 may be transferred to the computer 22 via a network, for example, and the program 21 may be stored and executed on the storage medium.

  Of course, some functions may be configured by hardware, or all may be configured by hardware. Alternatively, it can be incorporated as part of other software. Further, it can be incorporated as a part of another apparatus such as an output device.

  Further, the color conversion coefficient generated by the function shown in each embodiment of the present invention can be stored in the same storage medium. The stored color conversion coefficient can be read by a computer when in use or in advance, and color conversion processing can be performed by a multidimensional LUT to which the color conversion coefficient is applied.

It is a block diagram which shows the 1st Embodiment of this invention. It is explanatory drawing of the outline | summary of a correction | amendment in case black (K) single color is used as a reproduction guarantee color. It is explanatory drawing of an example of a correction vector in case black (K) single color is used as a reproduction guarantee color. It is explanatory drawing of the outline | summary of a correction | amendment in case yellow (Y) single color is used as a reproduction guarantee color. It is explanatory drawing of an example of the correction vector in case yellow (Y) single color is used as a reproduction guarantee color. It is explanatory drawing of the outline | summary of a correction | amendment in case red (R) pure color is used as a reproduction guarantee color. It is explanatory drawing of an example of the correction vector in case red (R) pure color is used as a reproduction guarantee color. It is a block diagram which shows the 2nd Embodiment of this invention. It is explanatory drawing of an example of the computer program in the case of implement | achieving the function of each embodiment of this invention with a computer program, and the storage medium which stored the computer program.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 ... Address grid production | generation part, 2 ... Target device forward color prediction part, 3 ... Reproduction guarantee correction data calculation part, 4 ... Device independent color signal correction part, 5 ... Color gamut compression part, 6 ... Output device reverse color prediction part , 7 ... Color conversion coefficient output unit, 11 ... Target device base data correction unit, 21 ... Program, 22 ... Computer, 31 ... Magneto-optical disk, 32 ... Optical disk, 33 ... Magnetic disk, 34 ... Memory, 41 ... Magneto-optical disk 42, an optical disk device, 43 ... a magnetic disk device.

Claims (12)

  In a color conversion coefficient creating method for calculating a color conversion coefficient of a multidimensional table used for conversion from a device color signal of a first device to a device color signal of a second device, a reproduction guaranteed color in the first device A correction vector for correcting the device-independent color signal for reproduction with the corresponding reproduction guarantee color in the second device is calculated by the reproduction guarantee correction data calculation means, and corresponds to the device color signal of the first device. A device-independent color signal corrected by applying the correction vector to the device-independent color signal is obtained by a device-independent color signal correcting unit, and the second color signal is corrected based on the corrected device-independent color signal. The device color signal of the second device is predicted by the inverse color prediction means from the device model of the device, and the predicted second device A device color signal is output by a color conversion coefficient output means as a color conversion coefficient of a multidimensional table, and as the correction vector, for a color that is further away from a reproduction guaranteed color in the first device by a predetermined distance or more. And a correction vector having a magnitude of 0 is associated, and the device of the first device is set using the combination of the reproduction guarantee color and the device independent color signal of the color that is separated from the reproduction guarantee color by a predetermined distance or more and the correction vector. A method for creating a color conversion coefficient, comprising: obtaining a corrected device-independent color signal corresponding to a color signal.   In a color conversion coefficient creating method for calculating a color conversion coefficient of a multidimensional table used for conversion from a device color signal of a first device to a device color signal of a second device, a reproduction guaranteed color in the first device A correction vector that corrects a device-independent color signal for reproduction with a corresponding reproduction-guaranteed color in the second device is calculated by a reproduction-guaranteed correction data calculating unit, and forms a device model of the first device. A device-independent color signal corrected by applying the correction vector to the color signal is obtained by a target device base data correction unit, and the first device based on the base data including the corrected device-independent color signal is obtained. A device-independent color signal is converted from the device color signal of the first device using a device model. The device color signal of the second device is predicted by the inverse color prediction unit using the device model of the second device from the predicted device-independent color signal, and the predicted second device. The device color signal is output by the color conversion coefficient output means as the color conversion coefficient of the multi-dimensional table, and further, the correction vector is a color that is more than a predetermined distance away from the reproduction guarantee color in the first device. A correction vector having a size of 0 is associated with the reproduction guaranteed color and a set of the device independent color signal of the color that is separated from the reproduction guaranteed color by a predetermined distance or more and the correction vector. A color conversion coefficient product characterized by obtaining a corrected device-independent color signal corresponding to the device-independent color signal constituting the device model. Method.   3. The color conversion coefficient creation method according to claim 1, wherein the reproduction guarantee color is one or more of white, primary color, and secondary color.   The color conversion coefficient creation according to claim 1 or 2, wherein the reproduction guarantee color is one or more of white, black 100%, black single color, primary color, and secondary color. Method.   5. The color conversion coefficient creating method according to claim 1, wherein the element of the reproduction guarantee color is predetermined or designated by a user.   In a color conversion coefficient creating apparatus for calculating a color conversion coefficient of a multidimensional table used for conversion from a device color signal of a first device to a device color signal of a second device, a reproduction guaranteed color in the first device Reproduction guarantee correction data calculation means for calculating a correction vector for correcting a device-independent color signal for reproduction with the corresponding reproduction guarantee color in the second device, and the correction vector calculated by the reproduction guarantee correction data calculation means A device-independent color signal correction unit applied to a device-independent color signal corresponding to a device color signal of the first device, and a device-independent color signal corrected by the device-independent color signal correction unit An inverse color prediction means for predicting a device color signal of the second device from a device model of the second device; Color conversion coefficient output means for outputting the device color signal of the second device predicted by the inverse prediction means as a color conversion coefficient of a multidimensional table, and the reproduction guarantee correction data calculation means further includes the first device. A correction vector having a magnitude of 0 is associated with a color that is more than a predetermined distance away from the reproduction guarantee color in the device, and the device-independent color signal correction means is separated from the reproduction guarantee color and the reproduction guarantee color by a predetermined distance or more. An apparatus for generating a color conversion coefficient, wherein a corrected device-independent color signal corresponding to a device color signal of the first device is obtained using a set of a color device-independent color signal and the correction vector.   In a color conversion coefficient creating apparatus for calculating a color conversion coefficient of a multidimensional table used for conversion from a device color signal of a first device to a device color signal of a second device, a reproduction guaranteed color in the first device Reproduction guarantee correction data calculation means for calculating a correction vector for correcting a device-independent color signal for reproduction with the corresponding reproduction guarantee color in the second device, and the correction vector calculated by the reproduction guarantee correction data calculation means A target device base data correction unit applied to a device independent color signal constituting the device model of the first device; and the base data including the device independent color signal corrected by the target device base data correction unit. Using the device model of the first device, Forward color predicting means for predicting a device independent color signal from the vice color signal, and using the device model of the second device from the device independent color signal predicted by the forward color predicting means. An inverse color prediction unit that predicts a device color signal; and a color conversion coefficient output unit that outputs the device color signal of the second device predicted by the inverse color prediction unit as a color conversion coefficient of a multidimensional table, and the reproduction The guarantee correction data calculation means further associates a correction vector having a magnitude of 0 with respect to a color that is a predetermined distance or more away from the reproduction guarantee color in the first device, and the target device base data correction means is the reproduction guarantee color And a set of a device-independent color signal of a color that is a predetermined distance or more away from the reproduction guarantee color and the correction vector, and Color conversion coefficient creating apparatus characterized by obtaining a corrected device-independent color signal corresponding to the device-independent color signals constituting the device model of the device.   8. The color conversion coefficient creating apparatus according to claim 6, wherein the reproduction guarantee color is one or more of white, primary color, and secondary color.   8. The color conversion coefficient generation according to claim 6, wherein the reproduction guarantee color is one or more of white, black 100%, black single color, primary color, and secondary color. apparatus.   10. The color conversion coefficient creating apparatus according to claim 6, wherein the reproduction guarantee color element is predetermined or designated by a user.   6. A color conversion coefficient creating program for creating a color conversion coefficient of a multidimensional table used for converting a device color signal of a first device into a device color signal of a second device. A color conversion coefficient creating program that causes a computer to execute the color conversion coefficient creating method according to claim 1.   In a computer-readable storage medium storing a color conversion coefficient creating program for creating a color conversion coefficient of a multidimensional table used for converting a device color signal of a first device into a device color signal of a second device A computer-readable storage medium storing a color conversion coefficient creation program that causes a computer to execute the color conversion coefficient creation method according to any one of claims 1 to 5.

Images