JP5264153B2 - Color processing apparatus and color processing method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To solve the problem that a long time has been required for creating a color separation table because it is difficult to perform high-speed operation of color prediction in a printing device which uses many inks. <P>SOLUTION: In the color processing method, printing color information is obtained by performing a colorimetry upon a patch image output from a printing device. A prepared color separation table is then acquired as a reference table and from the reference table, ink values of frames constituting a color space are extracted. Based on the printing color information of the patch image, printing color information corresponding to the ink values of the frames is estimated by color prediction. At this point, a plurality of printer models 703 each of which may use inks less than that of the printing device, are assumed, color prediction is performed for each printer model, and prediction results are shared among the plurality of printer models, thereby reducing computational complexity. Then, the ink values of the frames are updated such that the printing color information estimated by color prediction can be closer to a predetermined target color, and the color separation table is created based on the updated ink values of the frames. <P>COPYRIGHT: (C)2009,JPO&amp;INPIT

Description

  The present invention relates to a color processing apparatus and a color processing method, and more particularly to a color processing apparatus and a color processing method for creating a color separation table for a printing apparatus that performs printing with a plurality of inks.

Conventionally, the Cellular Yule-Nielsen Neugebauer Model is known as a color prediction method for predicting the color output by a printer (for example, see Non-Patent Document 1). This model outputs color patches corresponding to preset grid points and estimates color information for an arbitrary input value according to the Neugebauer Model based on the color information obtained by measuring the color patches. It is.
However, not all grid points can be printed for reasons such as exceeding the total amount of ink. Therefore, in the said nonpatent literature 1, the color information of the grid point which cannot be printed is estimated by multiple regression from the color information of the grid point which printed and measured the color.
Y. Chen, RS Berns and LATaplin, “Six Color Printer Characterization Using an Optimized Cellular Yule-Nielsen Spectral Neugebauer Model”, J. Imaging Sci. Tech. Vol. 48, No. 6, pp. 519-528 (2004)

However, the above-described method for estimating color information of non-printable grid points by multiple regression has a problem that the amount of calculation increases exponentially as the number of inks increases. For example, in the method according to Non-Patent Document 1, an inverse matrix operation of 2 ink number times 2 ink number size is required. In general, the calculation amount of the inverse matrix is proportional to the cube of the matrix size. For example, when there are 11 colors of ink, compared to the case of 4 colors,
(2 ^{11/2 4)} also becomes the calculation amount ³ = 2097152 times. This has been a major obstacle in today's multi-color printers.

  The present invention has been made to solve the above-described problems, and creates a color separation table by performing high-speed color prediction in units of printer models with a smaller number of inks for a printing apparatus with a large number of inks to be used. An object of the present invention is to provide a color processing apparatus and a color processing method.

  As a means for achieving the above-described object, the color processing apparatus of the present invention has the following configuration.

A color processing apparatus for creating a color separation table for a printing apparatus that performs printing with a plurality of inks, an input unit that inputs print color information obtained by measuring colors of a patch image output from the printing apparatus; Reference table acquisition means for acquiring the prepared color separation table as a reference table, ink value extraction means for extracting ink values constituting a color space from the reference table, and print color information obtained by the input means Based on color prediction means for estimating print color information corresponding to the ink value by color prediction, and optimization for updating the ink value so that the estimated print color information approaches a predetermined target value And a color separation table creating means for creating the color separation table based on the updated ink value, and the color Measuring means, characterized in that on the line segment forms a frame constituting the color space, or printer model having only ink utilized in color of the color space performing the color prediction.

  For example, the color prediction unit includes a printer model selection unit that selects the printer model according to the ink value of the frame, and further acquires a peripheral grid point of the ink value of the frame for each printer model. Peripheral grid point acquisition means, grid point estimation means for estimating print color information for peripheral grid points for which print color information has not been obtained in the colorimetry step, and the peripheral grid points Frame information estimating means for estimating print color information corresponding to the ink value of the frame based on dot print color information.

  According to the present invention having the above-described configuration, a color separation table can be created by performing high-speed color prediction in units of a printer model having a smaller number of inks for a printing apparatus that uses a large number of inks.

  Hereinafter, the present invention will be described in detail based on preferred embodiments with reference to the accompanying drawings. The configurations shown in the following embodiments are merely examples, and the present invention is not limited to the illustrated configurations.

<First Embodiment>
Output Process In this embodiment, a color prediction method in the color separation table creation device will be described. In the color separation table creating apparatus of this embodiment, a color separation table for an ink jet printer equipped with 11 color inks of C, M, Y, K, Lc, Lm, Gy, R, G, B, and Lg. Shall be created.

  FIG. 1 shows a flow of printing processing of an ink jet printer assumed in the present embodiment. An image signal input unit 101 receives an image signal from a control unit such as an OS (Operation System) (not shown). This image signal is a color image signal composed of 8 bits for each of 3 channels of RGB, for example, and is defined in a color space such as sRGB or AdobeRGB which is a standard color space. An image signal input from the image signal input unit 101 is received by the color separation unit 102 at 11 of C, M, Y, K, Lc, Lm, Gy, R, G, B, and Lg corresponding to each ink of the printer. Each channel is converted into an 8-bit image signal.

  Here, FIG. 2 shows a look-up table (hereinafter referred to as a color separation table) which is referred to when the color separation unit 102 converts an 8-bit image signal for each of RGB 3 channels into an 8-bit image signal for each ink 11 channel. An example) is shown. As shown in the figure, the color separation table includes all combinations (for example, R, G, B channels sampled in 9 steps (0, 32, 64, 96, 128, 160, 192, 224, 255)). 9 × 9 × 9 = 729 colors), output values are described. That is, the output values are 8-bit output values (hereinafter referred to as ink values) of C, M, Y, K, Lc, Lm, Gy, R, G, B, and Lg. The color separation unit 102 refers to this color separation table and converts signals corresponding to all combinations of 8-bit RGB 3-channel input using tetrahedral interpolation, linear interpolation, etc., into 11-channel ink values. To do.

  The image signal (ink value) consisting of 8 bits for 11 channels converted by the color separation unit 102 is converted into 2-bit data for each 11 channels that can be printed by the image output unit 104 in the gradation correction unit 103. As this gradation correction method, for example, there is a method in which a Bayer type 16 × 16 matrix is applied to each of the input 11-channel images. That is, gradation correction is performed by setting 1 when the corresponding pixel value on the image is larger than the matrix element and 0 when the pixel value is equal to or less than the matrix element. An error diffusion method or the like can also be used as the halftoning method.

  In this embodiment, an apparatus for creating a color separation table as described above will be described. In the color separation table, the ink values of the 11 colors must be determined so that the color indicated by the input RGB3 channel signal and the color reproduced on the paper after printing are substantially equal. At this time, the color prediction method in this embodiment functions effectively.

Color Separation Table Creation Device FIG. 3 is a block diagram showing the configuration of a color separation table creation device that uses the color prediction method of this embodiment. In the figure, reference numeral 301 denotes a CPU that controls the entire system, 302 denotes a ROM that stores programs and parameters that do not need to be changed, and 303 denotes a RAM that temporarily stores programs and data supplied from each processing unit. A display unit 304 is a display device such as a monitor that displays processing results, instructions to the user, error messages, and the like. An operation unit 305 is an input device such as a pointing device or a keyboard for inputting data in response to a user operation.

  Reference numeral 306 denotes an 11-color ink jet printer (hereinafter referred to as a printer) that performs the output process as described above, and outputs a color patch created by the color prediction color patch creation unit 310 described later. Note that the color separation table created by the color separation table creation device of the present embodiment can be used by being uploaded to the printer 306.

  A color separation table creation unit 307 performs color separation table creation processing, which is a feature of the present embodiment, and details thereof will be described later.

  Reference numeral 308 denotes a reference table holding unit, which holds a plurality of color separation tables for typical paper prepared in advance. The color separation table creation apparatus according to the present embodiment refers to the color separation table held in the reference table holding unit 308, and automatically creates a color separation table for each of various papers or print quality settings. .

  A color prediction unit 309 calculates color information (CIE1976L * a * b *) for the ink value in response to a request from the color separation table creation unit 307, and details thereof will be described later.

  A color prediction color patch creation unit 310 creates a color patch for color prediction in the present embodiment. Details thereof will be described later.

  A colorimetric data acquisition unit 311 is connected to a colorimeter (not shown), and measures a color patch image created by the color prediction color patch creation unit 310 and printed by the printer 306. This color measurement result is input to the color prediction unit 309.

Color separation table creation processing Details of the color separation table creation processing in the color separation table creation unit 307 will be described below.

  FIG. 4 is a flowchart showing color separation table creation processing in the color separation table creation unit 307.

  First, in step S400, the colorimetric data acquisition unit performs colorimetry on the color patch image created by the color prediction color patch creation unit 310 and printed by the printer 306, and spectral reflectance data that is the acquired print color information is obtained. Input to the color prediction unit 309. Note that this color measurement step S400 is executed at any timing, for example, several days ago or immediately before, if it is performed prior to starting the color separation table creation processing in step S401 and subsequent steps. May be.

  Next, step S401 is a reading step for reading the reference table. Here, among the plurality of color separation tables held in the reference table holding unit 308, one color separation table closest to the paper quality or print quality setting corresponding to the color separation table created in the subsequent processing is selected. One is read out and stored in the RAM 303. For example, assume that one reference color separation table is registered in the reference table holding unit 308 for each type such as glossy paper, matte paper, coated paper, and plain paper. In this case, if a color separation table for paper belonging to glossy paper is created in the subsequent processing, the color separation table for representative paper for glossy paper is selected. The selection of the reference table may be performed by the user via the operation unit 305, or it may be configured to automatically select an appropriate reference table by discriminating paper attributes using some measurement means. May be.

  Next, step S402 is a frame extraction step for extracting frame information from the reference table selected in step S401. Here, the frame information will be described with reference to FIGS.

  FIG. 5 is a diagram schematically showing a color space. In the case of an RGB 8-bit image signal, W is (R, G, B) = (255,255,255), and K is similarly (0, 0, 0), R corresponds to (255, 0, 0), G corresponds to (0, 255, 0), and B corresponds to (0, 0, 255). Similarly, C corresponds to (0, 255, 255), M corresponds to (255, 0, 255), and Y corresponds to (255, 255, 0). The frame in this embodiment means a total of 19 lines connecting 18 points connecting these 8 points (W, K, R, G, B, C, M, Y) and a line connecting W and K. Means a line of books. That is, WC, WM, wy, WR, WG, WB, KK, KM, KY, KR, KG, KB, Y-R, Y-G, CG, CB, MB, MR, and W-K are 19 in total. That is, in step S402, the RGB values located on the 19 lines and the ink values of 11 colors corresponding to the RGB values are extracted from the reference table.

  This frame information will be described more specifically with reference to FIG. FIG. 6 is a diagram illustrating ink values in the WK frame. In the same figure, the grid numbers on the horizontal axis are obtained by numbering points on W-K in order from W, and there are 9 grids when the color separation table is in steps of 9 steps for each of RGB explained in FIG. . For example, the grid 1 corresponds to W and (R, G, B) = (255, 255, 255). Grid 2 corresponds to (R, G, B) = (224, 224, 224), and grid 3 corresponds to (192, 192, 192).

  In step S402, ink values corresponding to the RGB values of the grids 1 to 9 are extracted from the color separation table.

  The color separation table creation unit 307 optimizes each ink value for each of the 19 frames, and finally generates a color separation table in the entire color space other than the frames by interpolation processing.

  Returning to FIG. 4, after extracting ink values for each frame in step S <b> 402, the optimization processing in steps S <b> 405 to S <b> 408 is performed on each grid in each frame by the loop of steps S <b> 403 and S <b> 404. The ink value is determined. Hereinafter, this optimization process will be described.

  First, in step S405, a target color for each grid is set. Here, the target color is, for example, an L * a * b * value in the CIE 1976 L * a * b * color space, and a value determined in advance for each frame and each grid is recorded in the ROM 302. In step S405, a predetermined target value corresponding to the grid and frame is read from the ROM 302 and set in the RAM 303.

  In step S406, the color prediction unit 309 estimates (color prediction) the L * a * b * value in the CIE 1976 L * a * b * color space from the ink values of 11 colors in each grid. Details of the color prediction process will be described later. In step S407, the predicted L * a * b * is compared with the L * a * b * value of the target color set in step S405. If the color difference by the comparison does not satisfy the predetermined condition, the ink value of the corresponding grid is changed in step S408, and color prediction is performed again in step S406 for the updated value. The processes in steps S408 and S406 are repeated until a predetermined condition is satisfied in step S407.

  Here, the ink value is updated in step S408 by increasing or decreasing the ink value according to a predetermined algorithm, but the number of inks used in the corresponding frame is not increased. For example, when the ink used in the W-K frame is Lc, Lm, Gy, Lg, K, the ink value is changed among these four colors, and another ink (for example, Y) is newly used. It will never be done. That is, 0 is always set to ink values other than Lc, Lm, Gy, Lg, and K. In this embodiment, as described above, by limiting the number of inks by the frame, even in a printer having a large number of inks of 11 colors, the color prediction step (S406) can be executed at a relatively high speed.

  Note that the determination condition in step S407 may be different for each frame. Moreover, although the example which determines conditions by color difference was shown in this embodiment, this invention is not limited to this. For example, different conditions may be set for each frame, such as calculating and comparing hue differences in a certain frame and calculating and comparing brightness differences in another frame.

  When the ink value optimization processing for all frames and grids is completed by the loop of steps S403 and S404, finally, in step S409, the optimized ink values of each frame are converted into known algorithms such as tetrahedral interpolation and linear interpolation. Interpolate using. As a result, a color separation table for the entire color space is generated.

  As described above, the color separation table creation unit 307 creates a color separation table that realizes similar color reproduction on various papers with different color development characteristics by changing the ink value of each frame using color prediction. Can be created automatically. Although details will be described later, in this embodiment, color prediction can be performed at high speed by limiting and optimizing the ink value for each frame.

Color Prediction Processing Hereinafter, details of the color prediction processing executed by the color prediction unit 309 in step S406 described above will be described.

  FIG. 7 is a block diagram illustrating the configuration of the color prediction unit 309. In the figure, reference numeral 701 denotes an ink value input unit for inputting ink values of 11 colors (C, M, Y, K, Lc, Lm, Gy, R, G, B, and Lg). A printer model selection unit 702 selects an optimal printer model for color prediction based on the ink value input by the ink value input unit 701 and inputs the ink value to the printer model.

  A printer model 703 predicts a spectral reflectance as print color information that will be obtained when the printer 306 outputs an ink value. The color prediction unit 309 has a plurality of printer models with different numbers of inks and ink types, and each printer model has a peripheral primary acquisition unit 707, an unknown primary estimation unit 708, a Neugebauer that performs processing necessary for color prediction. Each has a processing unit 709. The primary in this embodiment corresponds to a grid point (grid) on the color space.

  FIG. 9 shows the number of inks and the ink type of each printer model when there are 15 different printer models. In FIG. 9, “0” in the ink type column means that the corresponding ink is not used, and “1” means that the ink is being used. As shown in FIG. 9, even a printer having 11 colors of ink is not considered to print using all 11 colors of ink at the same time. Division is possible. In this way, by dividing the 11-color printer into printer models with a smaller number of inks, it is possible to reduce the amount of calculation of color estimation, which will be described later, and to realize high-speed processing.

  Returning to FIG. 7, reference numeral 704 denotes a lattice point information holding unit for holding lattice point information. Here, the grid point information is a table as shown in FIG. 10, and the ID of all grids obtained by sampling each ink value at a certain interval, the ink value of each grid, and the spectral reflection corresponding to the ink value. Flag information indicating the presence or absence of rate data. Here, the flag information refers to whether or not spectral reflectance data obtained as a colorimetric value when a corresponding ink value is output from the printer 306 is stored in a primary spectral reflectance holding unit 705 described later. Is shown. FIG. 10 shows an example in which all inks are sampled with 4 grids of ink values 0, 85, 170, and 255. A combination of 4 to the 11th power (= 4194304) ink values and corresponding flag data are held. Yes. Here, the “presence / absence of spectral reflectance data” column in FIG. 10 corresponds to flag information, and “0” indicates that spectral reflectance data corresponding to the ink value is not stored in the primary spectral reflectance holding unit 705. “1” means stored.

  Returning to FIG. 7, reference numeral 705 denotes a primary spectral reflectance holding unit, and the spectral reflectance of the color prediction color patch acquired by the colorimetric data acquisition unit 311 in step S400 or the unknown primary estimation unit 708 in the printer model. The spectral reflectance estimated at is held. Here, FIG. 11 shows an example of data held by the primary spectral reflectance holding unit 705. As shown in FIG. 11, the primary spectral reflectance holding unit 705 holds the grid point IDs of the tables held in the grid point information holding unit 704 and the spectral reflectances corresponding to the respective grid points as a table. ing. In FIG. 11, the reflectance with respect to light having a wavelength of 400 to 700 [nm] is held at intervals of 10 [nm], but it goes without saying that the present invention is not limited to this example.

  Here, the lattice point information holding unit 704 and the primary spectral reflectance holding unit 705 are shared by the plurality of printer models described above. By this sharing, for example, the spectral reflectance estimated by the unknown primary estimation unit 708 of the printer model 01 is used for Neugebauer processing, which will be described later, without being redundantly estimated in other printer models such as the printer model 15. The In particular, since the unknown primary estimation processing described later requires a large amount of computation, the processing efficiency can be greatly improved by sharing the primary estimation result in this way.

  Returning to FIG. 7, reference numeral 706 denotes an L * a * b * output unit which converts the spectral reflectance predicted by each printer model into CIE 1976 L * a * b * and outputs it.

  Hereinafter, the color prediction processing in the color prediction unit 309 will be described in more detail with reference to the flowchart of FIG.

  First, in step S801, the ink value of each ink is input to the ink value input unit 701, and is input to an appropriate printer model selected by the printer model selection unit 702 in step S802. The printer model selection unit 702 selects a printer model having the smallest number of inks among printer models having all the inks whose input ink values are other than “0”. However, the printer model selection process is not limited to this example. For example, frame information may be input to the color prediction unit 309 by some means, and a printer model previously associated with the frame information may be selected.

  In step S <b> 803, each printer model acquires a peripheral primary (peripheral grid point) of the ink value in the peripheral primary acquisition unit 707. More specifically, IDs of all grids surrounding the input ink value are acquired from the lattice point information holding unit 704. Here, the grid that surrounds the ink values is a grid value that satisfies grid value A <ink value <grid value B in ink that is valid for the selected printer model (ink type other than “0” value shown in FIG. 9). A and B are meant. When there are a plurality of effective inks, values corresponding to the grid values A and B are obtained for each of the effective inks, and all the combinations thereof, that is, the number of neighboring primary primaries of 2 [number of effective inks]. Will exist.

  As described above, when the peripheral primaries are acquired in step S803, the processes in subsequent steps S805 to S808 are performed on all the peripheral primaries.

  First, in step S805, it is determined whether or not the spectral reflectance is set in the primary spectral reflectance holding unit 705. This determination is made by referring to flag data indicating the presence / absence of spectral reflectance data in the table illustrated in FIG. 10 held in the lattice point information holding unit 704.

  If it is determined in step S805 that there is spectral reflectance data, the process proceeds to step S808, and the corresponding spectral reflectance is stored in the RAM 303. Note that the corresponding spectral reflectance is acquired by searching the ID of the peripheral primary from the table of the primary spectral reflectance holding unit 705 illustrated in FIG.

  If it is determined in step S805 that there is no spectral reflectance data, the process proceeds to step S806, where the unknown primary estimation unit 708 estimates the spectral reflectance. This estimation is performed by performing multiple regression on the known spectral reflectance. Specifically, it follows the following formula.

R _j = F _j · P _j (1)
In Equation (1), R _j is an estimated spectral reflectance vector of the unknown peripheral primary, F _j is a coverage vector of the unknown peripheral primary, and P _j is an estimation matrix. Here, the coverage vector F _j is obtained by converting the input ink value vector into a ratio (coverage) that the paper surface is covered by 2 [number of effective ink] primaries. The term “primary” as used herein refers to all combinations when the coverage of each ink takes two states of 0% (ink value = 0) and 100% (ink value = 255), and therefore the number of primaries is 2. To the number of [effective inks]. The coverage depends on the tone correction unit 103 in the printer 306 and the dot gain generated when ink dots are formed on the paper. Conversion from the ink value to the coverage is performed by referring to a one-dimensional LUT set in advance for each ink, paper, and printing mode.

Further, the estimation matrix P _j is obtained by the following equation.

^{_{P j = (F p T w}} j T w j F p) -1 F p T w j T w j R p) ··· (2)
Here, w _j is a weight matrix, F _p is a known primary coverage matrix, and R _p is a known primary spectral reflectance matrix. The coverage matrix F _p is obtained by converting the ink value vector corresponding to the spectral reflectance data held in the primary spectral reflectance holding unit 705 into the coverage vector. Therefore, the coverage matrix F _p is a matrix having a size of (the number of primaries stored in the primary spectral reflectance holding unit 705) × (the number of effective inks of 2).

The spectral reflectance matrix R _p is a matrix composed of data obtained by raising the spectral reflectance corresponding to the coverage vector F _j held in the primary spectral reflectance holding unit 705 to the power of 1 / n. The spectral reflectance matrix R _p is (number of primaries stored in the primary spectral reflectance holding unit 705) × (the number of dimensions of spectral reflectance data, for example, 400 to 700 [nm], 31 dimensions in the case of 10 nm intervals. ) Size matrix. Note that n in the 1 / nth power is called a Yule-Neilsen Value, and a predetermined value is stored in the ROM 302.

The weight matrix w _j is a diagonal matrix composed of (2 to the effective ink power) × (2 to the effective ink power). As a diagonal element of the weight matrix w _j , for example, the reciprocal of the Euclidean distance between the coverage vector F _j and each row of the coverage matrix F _p is set. Thereby, the weight of the known primary whose distance is close to F _j can be increased. In the present embodiment, the estimation accuracy can be improved by adjusting the weights in this way.

Finally, the unknown peripheral primary spectral reflectance vector R _j estimated by the equation (1) is multiplied by n which is a Yule-Neilsen Value. Thereby, the unknown primary estimation process (grid point estimation process) in step S806 ends.

Here, the size of the matrix shown in the above equations (1) and (2) increases exponentially with respect to the number of inks because the number of primaries is 2 [number of inks]. Further, since the inverse matrix operation is generally proportional to the cube of the matrix size, for example, when having 11 colors of ink, compared to the case of 4 colors,
(2 ^{11/2 4)} is ³ = 2097152 times becomes the calculation amount and predicted. In the present embodiment, the calculation amount is suppressed by dividing the 11-color printer into a plurality of printer models having a smaller number of inks, and the unknown primary estimation process can be performed at a higher speed.

  As described above, the spectral reflectance of the peripheral primary estimated in step S806 is added to the table of the primary spectral reflectance holding unit 705 in step S807. At this time, the flag data indicating the presence / absence of spectral reflectance data in the table of the lattice point information holding unit 704 is also updated at the same time.

  As described above, in the present embodiment, the same unknown primary estimation process is performed for a plurality of printer models by updating data in the primary spectral reflectance holding unit 705 and the lattice point information holding unit 704 shared by a plurality of printer models. It will not be duplicated. In particular, since the unknown primary estimation process requires a large amount of calculation, the processing time can be greatly shortened by improving efficiency in this way.

  If it is determined in step S804 that all the peripheral primary spectral reflectances have been prepared by the loop processing in steps S805 to S808, the process proceeds to step S809.

  Step S809 is a frame information estimation step in which the Neugebauer processing unit 709 estimates the spectral reflectance for the input ink value. Specifically, the spectral reflectance of the peripheral primaries is linearly interpolated based on the coverage of each peripheral primary and the coverage of the input ink value vector, and finally raised to the nth power again. Thus, the spectral reflectance is estimated.

  The estimated spectral reflectance is converted into an L * a * b * value by the L * a * b * output unit 706 in step S810 and then output in step S811.

Color Prediction Color Patch Creation Processing Details of the color prediction color patch creation processing in the color prediction color patch creation unit 310 will be described below using the flowchart of FIG. The color patch created here is output from the printer 306 and measured by a colorimeter (not shown). Then, the spectral reflectance data of each patch obtained by the color measurement is acquired by the color measurement data acquisition unit 311 and held in the primary spectral reflectance holding unit 705 of the color prediction unit 309.

  The color prediction process in the color prediction unit 109 described above has a feature that the accuracy increases as the number of color patches to be measured increases. However, as the number of inks increases, the number of colorimetric measurements becomes enormous and the printing load increases. For example, when each ink is sampled in 4 steps, there are 4 11 (= 4194304) grid points in an 11 color printer. Even if these grid points are excluded because of limitations due to ink overflow (hereinafter referred to as “printing amount limitation”), the load on printing / colorimetry becomes enormous. In addition, since the limit on the amount to be printed varies depending on the characteristics of the paper to be printed, the parameters of the gradation correction unit 103, and the like, the number of patches that must be printed and measured cannot be determined uniquely, and the operator may be confused.

  The color prediction color patch creation unit 310 according to the present embodiment controls the number of color patches to be output while maintaining the accuracy as much as possible by considering these problems and optimizing the color separation of the frame.

  In FIG. 12, first, as a pre-process for creating a multi-color patch, a primary color patch is generated in step S <b> 1201 and output from the printer 306. Here, the primary color patch refers to, for example, ink values = 0, 16, 32, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224 in each ink of the printer 306. , 240, and 255. In this case, a color patch of the number of inks × the number of gradations = 11 × 16 = 172 colors is output. The color patch output in step S1201 is colorimetrically measured in step S1202, and the colorimetric value is stored in the RAM 303 via the colorimetric data acquisition unit 311.

  In step S1203, the number of patches actually printed is acquired. This may be configured to be designated by the user via the operation unit 305 or may be configured to read data stored in advance in the ROM 302.

  Step S1204 is a step of setting an initial value of the lattice point interval, and three gradations of 0, α, and 255 are set for each ink. Here, α is the maximum ink value that falls within the printing limit when all the inks are simultaneously printed on the paper with the same ink value. By setting three gradations including α in this way, spectral reflectances are acquired at all lattice points formed by combinations of 0 and α, and thus the inverse matrix in the above-described unknown primary estimation process (S805). Operations can be executed.

  Step S1205 is a step of creating, for example, the lattice point information shown in FIG. 10 based on the lattice point interval set in step S1204. Since three gradation levels are set for each ink at the initial setting stage, the number of gradation levels is the power of the number of inks, that is, 3 to the 11th power = 177147 grid point data. At this time, 0 is set in the flag indicating the presence / absence of spectral reflectance data.

  Step S1206 is a step of calculating the placement amount of the lattice points created in step S1205 and counting the lattice points (that is, printable patches) that are within the placement amount limit. Here, the hit amount is given by, for example, the sum of the ink values of the respective inks at the lattice points. Here, the number of grid points whose total sum is equal to or less than a threshold (imprinting amount limit) prepared in advance for each parameter of the paper and the gradation correction unit 103 is counted.

  In step S1207, the number of grid points counted in step S1206, that is, the number of printable patches, is compared with the number of print execution patches acquired in step S1204, and the process branches according to the comparison result. That is, if the number of printable patches is smaller than the number of print execution patches, the process proceeds to step S1208. If the number of printable patches is equal to or greater than the number of print execution patches, the process proceeds to step S1210.

  Step S1208 is a step of calculating the accuracy of the primary color. Here, using the spectral reflectance of the primary color acquired in step S1202 as a reference, an RMS error with respect to the spectral reflectance estimated from the lattice point is taken. For example, in step S1202, the spectral reflectance of the primary color of each ink is set to the ink value = 0, 16, 32, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240. , 255, 16 gradations. In addition, the lattice points that are initially set in step S1204 are three gradations of 0, α (assumed to be 80), and 255. Therefore, here, from the spectral reflectances of 0, 80, and 255, the spectral reflectances of 16, 32, 64, 96, 112, 128, 144, 160, 176, 192, 208, 224, and 240 which are other lattice points. And the RMS error with the measured spectral reflectance is calculated. This estimation process is performed by converting the ink value into a coverage, linearly interpolating a spectral reflectance of 1 / nth power in the coverage space, and again raising it to the nth power.

  In step S1209, the ink having the maximum RMS error calculated in step S1208 and its ink value are detected, and the value is added to the grid point for the ink. For example, if the RMS value of the C ink ink value 192 is the maximum, the lattice point spacing of the C ink is set to 0, 80 (= α), 192, 255. As described above, in step S1209, the accuracy is improved by adding grid points to the places where the primary color estimation accuracy is the worst.

  After the lattice point interval is changed in step S1209, the process returns to step S1205 again to reconstruct the lattice point information. For example, in all 11 inks, when C is 4 gradations and the other inks are 3 gradations, the total number of grid points is 4 × 3 × 3 × 3 × 3 × 3 × 3 × 3 × 3 × 3 × 3 × 3 = 236196.

  Thereafter, the processes in steps S1205 to S1209 are repeated until the number of printable patches is equal to or greater than the number of print execution patches in step S1207.

  If the number of printable patches is equal to or greater than the number of print execution patches in step S1207, the process proceeds to step S1210, and the distance D between each grid point and the frame information of the reference table is calculated. Here, as the reference table, a table referred to by the color separation table creation unit 307 is read from the reference table holding unit 308. The distance D between the grid point and the reference table is calculated by first calculating the Euclidean distance between each ink value vector corresponding to the frame on the reference table and the ink value vector of the grid point, and the smallest one is the distance. Determine as D.

  Subsequently, in step S <b> 1211, patches corresponding to the number of print execution patches are selected in order from the one with the smallest distance D, and image data in a format that can be output by the printer 306 is created. In the case of creating a color patch, the printer 306 performs processing after the gradation correction unit 103 shown in FIG.

  As described above, since the color separation table creation unit 307 creates a color separation table with reference to the reference table, the color prediction unit 309 is also expected to input many colors close to the reference table. Therefore, the color prediction color patch creation unit 310 can control the number of output color patches without degrading the required accuracy by determining the priority order of the patches based on the Euclidean distance from the reference table. .

  As described above, according to the present embodiment, the color information of grid points that cannot be printed is estimated by performing color prediction by dividing a printer having a large number of inks into a plurality of printer models having a smaller number of inks. Therefore, the amount of calculation for performing the operation is suppressed, and faster color prediction is possible. In addition, since the estimation result of the non-printable grid point color information having a relatively large calculation amount is shared by a plurality of printer models, duplicate processing is not performed in each printer model, and the processing becomes efficient.

Second Embodiment
Hereinafter, a second embodiment according to the present invention will be described. In the first embodiment described above, an example in which color prediction is performed in a plurality of printer models according to the ink type and the number of inks has been shown. However, in the second embodiment, when the ink type and the number of inks are limited, A simpler configuration is shown. The configuration of the color separation table creation device in the second embodiment is substantially the same as that of FIG. 1 in the first embodiment described above, but the color prediction unit 309 is unnecessary and the color prediction function is included in the color separation table creation unit 307. It is.

  FIG. 13 is a block diagram illustrating a configuration of a color separation table creation unit 307 that creates a color separation table for each frame in the second embodiment. Also in the color separation table creation unit 307 in the second embodiment, 18 lines connecting adjacent points of 8 points (W, K, R, G, B, C, M, Y) on the color space. A color separation table is created for each of a total of 19 frames of lines connecting W and K.

  In FIG. 13, reference numeral 1301 denotes a reference table acquisition unit, which acquires a reference table from the reference table holding unit 308. A frame information extraction unit 1302 extracts frame information from the reference table, and passes the frame information of the corresponding reference table to a frame color separation unit 1303 described later. A frame color separation unit 1303 determines the color separation value of the frame by updating and optimizing the ink value of the reference table for each frame, as in the first embodiment. In the second embodiment, there are 19 frame color separation units 1303 corresponding to the frames. Reference numeral 1304 denotes an interpolation processing unit which creates a color separation table for the entire color space by interpolation based on the color separation values optimized by the respective frame color separation units 1303.

  Each frame color separation unit 1303 has a printer model color prediction unit 1305 corresponding to the ink type and the number of inks used by each frame. Each color predicting unit 1305 can access a common primary spectral reflectance holding unit 1306 and a lattice point information holding unit 1307. Each color prediction unit 1305 performs color prediction processing using the primary spectral reflectance holding unit 1306 and the lattice point information holding unit 1307, as with each printer model 703 in the color prediction unit 309 of the first embodiment described above. .

  As described above, according to the second embodiment, when the ink type and the number of inks are limited in advance, color prediction is performed for each frame with the simple configuration shown in FIG. 13 without providing a printer model selection unit. be able to.

  In addition, according to the configuration shown in FIG. 13, color prediction processing can be performed for each frame. Therefore, for example, when color separation table creation processing including color prediction processing is executed in a computer capable of parallel processing, parallelization is facilitated. .

<Third Embodiment>
The third embodiment according to the present invention will be described below. The configuration of the color separation table creation device in the third embodiment is substantially the same as that of FIG. 1 in the first embodiment described above. In the third embodiment, color prediction is performed so that color prediction processing at the time of color separation table creation can be executed in time parallel in an arithmetic device having a plurality of CPUs 301 such as a cluster supercomputer and capable of performing parallel operation. The unit 309 is configured.

  FIG. 14 is a block diagram illustrating a configuration of the color prediction unit 309 according to the third embodiment. In the same figure, the same number is attached | subjected to the structure similar to FIG. 7 of 1st Embodiment mentioned above, and description is abbreviate | omitted. That is, the printer model selection unit 702, the printer model 703, the lattice point information holding unit 704, the primary spectral reflectance holding unit 705, and the L * a * b * output unit 706 perform the same processing as in the first embodiment. .

  Hereinafter, the processing in the configuration unique to the third embodiment in FIG. 14 will be mainly described.

  First, the ink value file input unit 1401 inputs a file in which a set of predicted ink value vectors is recorded. This file may be any file as long as it is composed of a plurality of predicted ink value vectors. Further, a plurality of predicted ink value vectors may be acquired from a pointer or the like in which an array of ink value vectors is stored.

  Next, a plurality of ink value vectors input from the ink value file input unit 1401 are stored in an appropriate queue memory 1403 so that the load control unit 1402 performs color prediction using an appropriate printer model 703 for each ink value vector. Sorted. The queue memory 1403 temporarily stores the input ink value vector according to the processing status of the subsequent printer model selection unit 702 and the plurality of printer models 703. For example, in the queue memory 1403 followed by the printer models 01 to 03, if no processing is performed in the printer models 01 to 03, the ink value vector is input to the subsequent printer model selection unit 702. On the other hand, if the process is being executed in any of the printer models 01 to 03, the input ink value vector is held until the process is completed.

  Then, the printer model selection unit 702 and the printer model 703 perform color prediction processing in the same manner as in the first embodiment described above to estimate the primary spectral reflectance.

  The spectral reflectance estimated by each printer model 703 is converted into an L * a * b * value by the L * a * b * output unit 706. Here, since the L * a * b * values output from the L * a * b * output unit 706 are output in the order in which the processes of each printer model 703 are completed, they are input by the ink value file input unit 1401. The order of the ink value files is different. Therefore, the L * a * b * file output unit 1407 temporarily stores the output result of the L * a * b * output unit 706. Then, when the processing for all the ink value vectors in the input ink value file is completed, the output results are rearranged in the order of the input ink value file to create an output file.

  As described above, one printer model selection unit 702 and one or more printer models 703 are connected to one queue memory 1403. When an arithmetic device having a plurality of CPUs such as a cluster supercomputer is applied, the processing of the printer model selection unit 702 and the printer model 703 connected to one queue memory 1403 is performed by one CPU. It is desirable to configure. In addition, it is desirable that the processing load executed by each CPU (a plurality of printer models 703 connected to one queue memory 1403) be equalized. This is realized by adjusting the distribution of the printer model 703 connected to the queue memory 1403 in accordance with the number of inks because the amount of computation depends on the number of inks of the printer model 703.

  Here, an adjustment example in the case where the 15 types of printer models shown in FIG. 9 are processed in parallel using 14 CPUs will be specifically described.

  First, three 8-color ink printer models (07) with the largest amount of computation are created, and each of them is computed by different CPUs. That is, the created three printer models are assigned to different queue memories. Configure to be connected.

  Then, two printer models of the next seven-color ink printer model (No. 08) having the largest amount of calculation are created and connected to different queue memories so that each is calculated by another CPU.

  For the six-color ink printer models (01, 02, 05, 06, 09, 10), one printer model is created and connected to one queue memory so that it can be calculated by one CPU. To do.

  Since the 5-color ink printer model has a smaller amount of calculation than the 6-color ink printer model, it is connected to one queue memory in combination with the 4-color ink printer model. That is, the printer models No. 03 and No. 13, and No. 14 and No. 15 are combined so that each is calculated by one CPU.

  Finally, the remaining four or less printer models are connected to one queue memory so that all are processed by one CPU.

  As described above, according to the third embodiment, when color prediction processing is executed in parallel by a plurality of CPUs, CPUs that perform calculations according to the number of ink colors of the printer model are allocated to each CPU. The amount of computation is equalized. Therefore, there is no waiting time for waiting for the processing in the CPU having a heavy load to end, and the color prediction processing can be executed at a higher speed.

<Modification>
In each embodiment described above, an example in which color prediction is performed by preparing a plurality of printer models in advance has been described, but the present invention is not necessarily limited to this example. For example, each time color prediction for one color is performed, a new printer model corresponding to the ink value may be newly generated and deleted when estimation is completed. In this case, the primary spectral reflectance holding unit and the grid point information holding unit are generated in advance, and each holding unit is used in common regardless of the printer model that is generated / erased halfway. Can be executed automatically.

  In each embodiment, an example is shown in which color estimation is performed using spectral reflectance. However, the present invention is not limited to this example, and color estimation may be performed using, for example, an XYZ color system. . In this case, since the number of dimensions is three dimensions of XYZ, the calculation amount is reduced as compared with the 31-dimensional spectral reflectance data at intervals of 10 [nm] at 400 to 700 [nm] mentioned in each embodiment. High-speed color prediction processing can be expected.

  Further, it goes without saying that the number of inks of the printer, the number of bits of image data, ink values, grid point information, and the like described in each embodiment are merely examples, and are not limited thereto. Further, the color prediction of the present invention is not intended only to create a color separation table.

<Other embodiments>
Although the embodiment has been described in detail above, the present invention can take an embodiment as a system, apparatus, method, program, recording medium (storage medium), or the like. Specifically, the present invention may be applied to a system composed of a plurality of devices (for example, a host computer, an interface device, an imaging device, a web application, etc.), or may be applied to a device composed of a single device. good.

  The present invention also provides a software program that realizes the functions of the above-described embodiments directly or remotely to a system or apparatus, and the system or apparatus computer reads out and executes the supplied program code. Achieved. The program in this case is a computer-readable program that corresponds to the flowchart shown in the drawing in the embodiment.

  Accordingly, since the functions of the present invention are implemented by computer, the program code installed in the computer also implements the present invention. In other words, the present invention includes a computer program itself for realizing the functional processing of the present invention.

  In that case, as long as it has the function of a program, it may be in the form of object code, a program executed by an interpreter, script data supplied to the OS, or the like.

  Recording media for supplying the program include the following media. For example, floppy disk, hard disk, optical disk, magneto-optical disk, MO, CD-ROM, CD-R, CD-RW, magnetic tape, nonvolatile memory card, ROM, DVD (DVD-ROM, DVD- R).

  As a program supply method, the following method is also possible. That is, the browser of the client computer is connected to a homepage on the Internet, and the computer program itself (or a compressed file including an automatic installation function) of the present invention is downloaded to a recording medium such as a hard disk. It can also be realized by dividing the program code constituting the program of the present invention into a plurality of files and downloading each file from a different homepage. That is, a WWW server that allows a plurality of users to download a program file for realizing the functional processing of the present invention on a computer is also included in the present invention.

  In addition, the program of the present invention is encrypted, stored in a storage medium such as a CD-ROM, distributed to users, and key information for decryption is downloaded from a homepage via the Internet to users who have cleared predetermined conditions. It is also possible to make it. That is, the user can execute the encrypted program by using the key information and install it on the computer.

  Further, the functions of the above-described embodiments are realized by the computer executing the read program. Furthermore, based on the instructions of the program, an OS or the like running on the computer performs part or all of the actual processing, and the functions of the above-described embodiments can also be realized by the processing.

  Further, the program read from the recording medium is written in a memory provided in a function expansion board inserted into the computer or a function expansion unit connected to the computer, and then executed, so that the program of the above-described embodiment can be obtained. Function is realized. That is, based on the instructions of the program, the CPU provided in the function expansion board or function expansion unit can perform part or all of the actual processing.

It is a figure which shows the flow of the printing process of the printer in one Embodiment which concerns on this invention. It is a figure which shows an example of the color separation table in this embodiment. It is a block diagram which shows the structure of the color separation table preparation apparatus in this embodiment. It is a flowchart which shows the color separation table creation process in this embodiment. It is a figure for demonstrating the color separation frame in this embodiment. It is a figure for demonstrating the ink value of the color separation frame in this embodiment. It is a block diagram which shows the structure of the color estimation part in this embodiment. It is a flowchart which shows the color prediction process in this embodiment. It is a figure explaining the number of inks and the kind of ink of each printer model in this embodiment. It is a figure which shows the example of data hold | maintained at the lattice point information holding part in this embodiment. It is a figure which shows the example of data hold | maintained at the primary spectral reflectance holding | maintenance part in this embodiment. It is a flowchart which shows the color patch creation process in this embodiment. It is a block diagram which shows the structure of the color separation table preparation part in 2nd Embodiment. It is a block diagram which shows the structure of the color estimation part in 3rd Embodiment.

Claims (12)

A color processing device that creates a color separation table for a printing device that performs printing with a plurality of inks,
Input means for inputting print color information obtained by measuring the color of the patch image output from the printing apparatus;
Reference table acquisition means for acquiring a color separation table prepared in advance as a reference table;
Ink value extracting means for extracting ink values constituting the color space from the reference table;
Color prediction means for estimating print color information corresponding to the ink value by color prediction based on the print color information obtained by the input means;
Optimization means for updating the ink value so that the estimated print color information approaches a predetermined target value;
Color separation table creating means for creating the color separation table based on the updated ink value;
The color predicting means includes
A color processing apparatus , wherein the color prediction is performed by a printer model having only ink used in a line segment constituting the color space or in one color of the color space . The color predicting unit includes a printer model selecting unit that selects the printer model according to the ink value, and for each printer model,
A peripheral grid point acquiring means for acquiring a peripheral grid point of the ink value;
Among the acquired peripheral grid points, grid point estimation means for estimating print color information for peripheral grid points for which print color information has not been obtained in the input step;
Print color information estimation means for estimating print color information corresponding to the ink value based on the print color information of the peripheral grid points;
The color processing apparatus according to claim 1, further comprising: The color predicting unit includes a printer model selecting unit that selects the printer model according to the ink value,
Before SL printer model selecting means, color processing according to claim 1, characterized in that in the printer model using all use ink showing the ink value of the frame, to select the most ink small number of printer model apparatus. The color predicting unit includes a printer model selecting unit that selects the printer model according to the ink value,
Before SL printer model selection unit, the color processing apparatus according to claim 1, characterized by selecting the associated in advance with the printer model to the frame.   The color prediction unit further includes a holding unit that holds the print color information, and the holding unit is shared by the plurality of printer models. The color processing apparatus as described.   The color processing apparatus according to claim 1, wherein the print color information is spectral reflectance information.   7. The patch image according to claim 1, wherein a predetermined number of patches are set in the patch image according to a priority order based on a reference table acquired by the reference table acquisition unit. The color processing apparatus as described.   The color processing apparatus according to claim 1, wherein the color prediction unit performs the color prediction in units of the printer model in time parallel.   The color processing apparatus according to claim 8, wherein the color prediction unit performs the color prediction in time parallel in units of combinations according to the number of inks used in the printer model. A color processing method for creating a color separation table for a printing apparatus that performs printing with a plurality of inks,
An input step of inputting print color information obtained by measuring the patch image output from the printing apparatus;
A reference table acquisition step of acquiring a color separation table prepared in advance as a reference table;
An ink value extracting step of extracting ink values constituting a color space from the reference table;
A color prediction step of estimating print color information corresponding to the ink value by color prediction based on the print color information obtained in the input step;
An optimization step of updating the ink value so that the estimated print color information approaches a predetermined target value;
A color separation table creating step for creating the color separation table based on the updated ink value;
In the color prediction step,
A color processing method , wherein the color prediction is performed by a printer model having only ink used in a line segment forming a frame constituting the color space or one color of the color space .   A computer program that causes a computer to function as each unit of the color processing apparatus according to any one of claims 1 to 9 by being read and executed by the computer.   A computer-readable recording medium having recorded thereon the computer program according to claim 11.

Images