JP4650300B2 - Correction accuracy prediction apparatus, correction accuracy prediction method, and correction accuracy prediction program - Google Patents

Description

  The present invention relates to a correction accuracy prediction apparatus, a correction accuracy prediction method, and a correction accuracy prediction program for predicting correction accuracy when correcting color misregistration in a printing result.

  Conventionally, the color measurement result of a patch printed on a printer (reference printer) as a design reference is compared with the color measurement result of a patch printed on a printer (target printer) as a color misregistration correction target. Based on the comparison result, correction relation data for converting gradation data for determining the ink amount is generated. The correction-related data is printed with a color equivalent to the color obtained when the reference printer is printed based on the predetermined gradation data when the target printer performs printing based on the predetermined gradation data. As described above, the predetermined gradation data is corrected.

  About the said correction | amendment relation data, it is necessary to verify the precision of correction | amendment when this is used. That is, it is verified whether or not the result of printing with the corrected gradation data by the target printer is within a certain allowable range close to the ideal print result. Conventionally, this verification is performed by causing the target printer to print a patch again based on the gradation data after correcting the predetermined gradation data by the correction relation data and evaluating the colorimetric result of the patch. I was going.

  In addition, conventionally, the difference between the color measurement results of the patches printed on the reference printer and the color measurement results of the patches printed on the target printer is the difference between the common values of the gradation data at the time of printing. In some cases, a judgment was made as to whether or not the value was larger than a certain standard. In such a determination, if the difference is larger than a certain reference, the color reproduction capability of the target printer is too different from the color reproduction capability of the reference printer, and printing with an ideal color is possible even if correction based on correction-related data is performed. It was determined that it was not possible, and an error determination was made for adopting the correction-related data.

As a conventional technique, when a sample image data is supplied to an output device and printed, and a profile is created from colorimetric values of each color patch of the obtained sample image, a standard corresponding to the CMYK value of each color patch is used. A technique is known in which a colorimetric value (L ^* a ^* b ^* ) is compared with the colorimetric value of each color patch, and a warning mark is attached to the colorimetric result when the difference exceeds the tolerance ΔE. (See Patent Document 1).
JP 2002-94820 A

  The process of verifying the correction accuracy based on the correction relation data by printing the patch on the target printer again based on the gradation data corrected by the correction relation data as described above and measuring the color is performed by patch printing. As a result, the amount of work required to acquire correction-related data has been increased.

  Also, if the difference between the print result from the reference printer and the print result from the target printer is not a hue shift but simply a color density shift, this shift can be compensated for by using the correction relation data. There are many cases. However, if the above determination is made based on the difference between the color measurement result of the patch printed on the reference printer and the color measurement result of the patch printed on the target printer, the print result between the reference printer and the target printer Even when the deviation is actually caused by the deviation in density, the difference is somewhat large. For this reason, there has been an erroneous determination that the error determination is made despite the fact that the color misregistration can be compensated by the correction-related data.

  The present invention has been made in view of the above problems, and a correction accuracy prediction apparatus and a correction accuracy prediction capable of accurately and accurately determining the correction accuracy based on correction-related data for correcting color misregistration in a printing result. It is an object to provide a method and a correction accuracy prediction program.

  In order to achieve the above object, a correction accuracy prediction apparatus according to the present invention is obtained by colorimetry of a plurality of first patches, which are used as a reference for color correction, in which an input means is printed based on gradation data of a plurality of stages. In addition, the first data is input, and the second data obtained by the color measurement of the plurality of second patches printed on the printing apparatus to be color-corrected based on the plurality of gradation data is input. . The first data is acquired for the number of the first patch and corresponds to the gradation data of a plurality of stages, and similarly, the second data is acquired for the number of the second patch and each of the gradation data of the plurality of stages. It corresponds. The first and second data corresponding to each patch are represented by a plurality of color components in a predetermined color space.

  The correction relationship generating means associates the first data corresponding to arbitrary gradation data with the first data and the second data having the smallest color difference. Then, by correlating the arbitrary gradation data with the gradation data corresponding to the associated second data, correction relation data for each gradation data is generated. The prediction means is performed using the correction relation data using at least the color difference between the correction relations, which is the color difference between the first data and the second data associated in the correction relation generation means, as one determination factor. Predict the accuracy of color correction.

  In other words, according to the present invention, the difference between the color measurement results (color difference between correction relationships) between the gradation data forming the correction relationship generated by comparing the color measurement result of the first patch and the color measurement result of the second patch. Ask for. Therefore, when the color difference between the correction relations is small, it can be said that the effect of correcting the gradation data using the correction relation data is excellent, and conversely, when the color difference is too large, the correction relation data is used. Even if the gradation data is corrected, it can be said that the printing result by the printing apparatus does not become an ideal color like the color of the first patch. Therefore, the accuracy of color correction performed using the correction relationship data can be accurately predicted by using the color difference between correction relationships as an index for determination. In addition, unnecessary patch printing and colorimetry for verifying color correction accuracy, which has been conventionally performed, are no longer necessary, and the work efficiency for acquiring correction-related data increases.

  As another configuration of the present invention, the correction relationship generation means defines a first approximate curve on a predetermined color space with each first data as a reference point, and applies each gradation data on the first approximate curve. Specify the corresponding coordinate points. In addition, a second approximate curve is defined on the color space using each second data as a reference point, and coordinate points corresponding to each gradation data are defined on the second approximate curve. That is, the number of the first data and the second data, which are acquired only for the number of gradations when the first patch and the second patch are printed, is increased by interpolating each. Then, the correction relationship generating unit corresponds to the coordinate point defined on the second approximate curve having the smallest color difference from the coordinate point corresponding to the arbitrary gradation value defined on the first approximate curve. A gradation value is specified, and the specified gradation value is related to the arbitrary gradation value. As a result, it is possible to generate a correction relationship for more gradation data than the number of gradations on which the first and second patches are printed. Further, by interpolating the first data and the second data, it is possible to specify a corrected gradation value that is more optimal for an arbitrary gradation value (with a smaller color difference).

  The prediction unit may calculate an average value of color differences between correction relationships for each combination of the first data and the second data associated with each other in the correction relationship generation unit. That is, since there are as many color differences between correction relationships as the number of gradation data correction relationships defined in the correction relationship data, an average value thereof is calculated. This average value can be said to be one of the numerical values representing the degree of color misregistration with the ideal color in the printing result of the printing apparatus. Therefore, by performing the prediction according to the average value, the correction relation data is used. In this case, the accuracy of color correction performed in this case can be determined more accurately.

  The prediction means predicts that the accuracy of color correction performed using the correction relationship data is acceptable when the average value of the color difference between the correction relationships is below a predetermined threshold value. That is, since the prediction is performed based only on the average value, the prediction can be easily performed.

  The prediction may be performed by adding elements other than the average value of the color difference between the correction relationships. Therefore, the predicting unit calculates a color difference before correction that is a color difference for each combination of the first data and the second data having the same gradation data. Then, an average value of color differences before correction may be calculated, and the prediction may be performed based on the average value of color differences between correction relationships and the average value of color differences before correction. The prediction can be performed more accurately by adding the average value of the color differences before correction to the prediction determination element.

  If both the average value of the color difference between correction relationships and the average value of the color difference before correction exceed a preset threshold value for each, the color measurement results of the first patch and the second patch Since the overall tendency is considerably shifted, even if gradation data is corrected using the correction relation data in the printing apparatus, the printing result is an ideal color such as the color of the first patch. It will not be. Therefore, in this case, the prediction means predicts that the accuracy of color correction performed using the correction relation data is unacceptable. When it is predicted that the print job has failed in this way, the operator verifies whether the printing conditions of the first patch and the second patch are different from each other, and if necessary, the first patch and the second patch. You can also reprint patches.

On the other hand, when the average value of color differences between correction relationships exceeds a predetermined threshold value and the average value of color differences before correction is lower than the predetermined threshold value, colorimetry between the first patch and the second patch is performed. Although there are differences in the shape of the curves that each result draws in a predetermined color space, the two curves themselves are expected to be close. Therefore, in this case, the accuracy of color correction performed using the correction relation data is predicted as the attention level. Here, the prediction means stores the correction relationship data in a predetermined storage medium when it is predicted to pass, and does not save the correction relationship data when it is predicted to fail, If predicted, the correction related data is stored in the predetermined storage medium on the condition that an instruction to save the correction related data is received from the outside.
Note that the prediction unit may notify the result of the prediction performed as described above to the outside using a predetermined output device. That is, the prediction result is notified to an operator or the like by monitor display or audio output.

So far, the invention has been described in the category of an apparatus for performing color correction accuracy prediction, but the contents of the above-described correction accuracy prediction device can also be grasped as procedures executed by each means, and the correction accuracy prediction method It can also be grasped as an invention.
Similarly, the contents of the correction accuracy prediction apparatus can be grasped as an invention of a correction accuracy prediction program that causes a computer to execute the procedure executed by each means.

Embodiments of the present invention will be described in the following order.
(1) Schematic configuration of correction accuracy prediction apparatus (2) Correction accuracy prediction processing (2-1) Acquisition of reference color measurement data and test color measurement data (2-2) Details of generation of color correction table and correction accuracy prediction ( 2-3) Accuracy of prediction results (3) Summary

(1) Schematic Configuration of Correction Accuracy Prediction Device FIG. 1 is a block diagram showing a schematic configuration of a computer and the like corresponding to the correction accuracy prediction device of the present invention.
The computer 10 includes a CPU (not shown) that forms the center of arithmetic processing, a ROM, a RAM, and the like as storage media, and executes a predetermined program using peripheral devices such as the HDD 30. A display 40 as an output device is connected to the computer 10 via a video board (not shown). Printers 50 and 60 are connected to the computer 10 via an I / F 24a (for example, a serial I / F cable) as necessary. The printers 50 and 60 are printing devices controlled by the computer 10. In the present embodiment, a reference printer as a reference machine for color correction and a target printer (a mass-produced printer) to be subjected to color correction are assumed, but in the following, the reference printer 50 is used as necessary. The expression “target printer 60” is also used. Furthermore, although not shown, the computer 10 is connected to operation input devices such as a keyboard and a mouse via a serial communication I / O.

A colorimeter 70 is connected to the computer 10 via an I / F 24b (for example, USB I / F). The colorimeter 70 directs the color detection unit 70a to the object to be colorimetrically measured, thereby a plurality of color components L ^* , a based on the L ^* a ^* b ^* color system defined by the International Commission on Illumination (CIE). ^{* And} b ^* can be acquired as colorimetric data, and the acquired colorimetric data can be output to the computer 10. Of course, the color space to be measured may be a CIE-defined L ^* u ^* v ^* color space, a CIE-defined XYZ color space, an RGB color space, or the like. In the following, the display of “ ^* ” is omitted. In this embodiment, the color chart (a set of a plurality of patches) printed by the printers 50 and 60 is measured by the colorimeter 70. The color chart colorimetry may be performed using a scanner or the like in addition to the colorimeter 70 described above.

In the computer 10, a printer driver 21, a calibration module 22, and a display driver 23 are incorporated in the OS 20. The display driver 23 is a driver that controls image display on the display 40.
The printer driver 21 can perform predetermined image processing on image data for which a print instruction has been issued from an application program (not shown), generate raster data, and output the raster data to the printers 50 and 60 via the I / F 24a.

  That is, the printer driver 21 acquires image data from the HDD 30 or the like, and executes resolution conversion processing or color conversion processing on the image data as necessary to use ink types (for example, cyan) used by the printers 50 and 60. (C), magenta (M), yellow (Y), black (K), light cyan (Lc), light magenta (Lm)) ink quantity data defining the ink recording quantity is acquired, and the ink quantity data is obtained The halftone process is performed to generate halftone data, and the halftone data is subjected to a predetermined rasterization process and rearranged in the order used by the printers 50 and 60 to generate raster data for each ink color. The raster data is output to the printers 50 and 60.

  The resolution conversion process is a process of increasing or decreasing the number of vertical and horizontal pixels in order to secure the pixels necessary for printing when the number of pixels of the image data is excessive or insufficient. The color conversion process is a process for converting the image data into the ink color system in units of pixels when the image data is expressed in a color system different from the color system of the ink used for printing. is there. As an example, when the image data is dot matrix data in which each color component of RGB (red, green, blue) in accordance with the sRGB standard is expressed in gradation to define the color of each pixel, the printer driver 21 With reference to a color conversion lookup table (LUT) recorded in advance in the HDD 30, the RGB data of each pixel of the image data is converted into ink amount data defining the ink recording amount for each CMYKLcLm. The color conversion LUT is a table that expresses colors by using the sRGB color system and the CMYKLcLm color system, associates the colors, and describes the correspondence between a plurality of colors.

  The ink amount data is data expressed by gradation (for example, 256 gradations from 0 to 255) for CMYKLcLm for each pixel, and each gradation value corresponds to the ink recording amount of each ink type. For example, the dot recording rate (ratio of the number of formed dots to the total number of pixels in the unit area on the printing paper, which is a kind of ink recording amount) 0 to 100% linearly corresponds to the gradation value 0 to 255. For example, the ink recording amount having a gradation value is defined. In the halftone process, the printer driver 21 converts the tone value of each pixel in the ink amount data using a known method such as an error diffusion method or a dither method, and ejects / drops ink droplets for each ink type and each pixel. Halfton data that determines non-ejection is generated.

  The calibration module 22 includes a color correction table generation unit 22a and a correction accuracy prediction unit 22b in order to execute color correction table generation and correction accuracy prediction processing in the present embodiment. The color correction table generation and the correction accuracy prediction process will be described in detail later.

  The printers 50 and 60 eject dots of the corresponding types from the print heads provided corresponding to the ink types (CMYKLcLm in the above example), and form the dots by attaching them to the printing paper. Print an image corresponding to the image data. That is, when the printers 50 and 60 receive raster data for each color transmitted from the computer 10 via the I / F 24a, the printer 50 and 60 generate applied voltage data corresponding to the raster data. Then, from the applied voltage data, an applied voltage pattern (drive waveform) to a piezo element provided corresponding to the ink jet nozzle built in each print head is generated, and the piezo element is expanded and contracted by the drive waveform. Thus, each color ink is ejected in dot units from each print head.

(2) Correction Accuracy Prediction Processing FIGS. 2 and 3 are flowcharts showing the contents of the color correction table generation processing and the correction accuracy prediction processing in this embodiment. This process is executed by the calibration module 22. However, as a premise for performing the processing, the computer 10 measures the colorimetric data (first data) that is the colorimetric result of the color chart printed on the reference printer 50 and the colorimetric data of the color chart printed on the target printer 60. It is necessary to input and save test colorimetry data (second data) as a result.

(2-1) Acquisition of Reference Colorimetric Data and Test Colorimetric Data FIG. 4 shows an example of the color chart A. In the figure, a plurality of patches P are continuously printed in the vertical and horizontal directions of the printing paper M. The color chart A is printed on a predetermined printing paper (for example, glossy paper) by the reference printer 50 based on the patch image data 31 stored in the HDD 30 when the computer 10 is connected to the reference printer 50. Is. Here, the patch image data 31 is ink amount data expressing a plurality of patches P corresponding to a plurality of stages of ink recording amounts, and each patch P is expressed in gradation by a predetermined number of pixels. That is, the computer 10 acquires the patch image data 31 from the HDD 30, performs the above halftone processing and rasterization processing on the patch image data 31, and sends the generated raster data to the reference printer 50, thereby generating the patch. The color chart A corresponding to the image data 31 is printed.

  Each patch P constituting the color chart A is a solid image printed with any one color of each ink type, and each patch corresponding to one ink type is printed with a different gradation value. ing. In the present embodiment, since the patch P is printed with 33 gradation values selected from 0 to 255 gradations for each ink type, the color chart A has 33 × the number of ink patches P. It is constituted by. The operator measures the color chart A with the colorimeter 70, and the computer 10 inputs the colorimetric data of each patch P of the color chart A obtained by the colorimetry from the colorimeter 70. The color measurement data is stored in the HDD 30 as reference color measurement data 33.

Further, the computer 10 causes the target printer 60 to print a color chart (represented as a color chart B for convenience) on the predetermined printing paper based on the patch image data 31 in a state of being connected to the target printer 60. . That is, the color chart is printed under the same conditions except that the connected printer is changed from the reference printer 50 to the target printer 60. It is assumed that the target printer 60 is the same type as the reference printer 50. Therefore, the color chart B to be printed has the same arrangement of patches as the color chart A shown in FIG. 4, and the gradation values when the patches having the same position are printed are also the same. However, since the target printer 60 may have a deviation from the reference printer 50 in its color reproduction capability, it is considered that there is a deviation in the actual color between patches whose positions are common in the color charts A and B.
The operator also measures the color chart B with the colorimeter 70, and the computer 10 inputs the colorimetric data of each patch obtained by the colorimetry and uses the colorimetric data as the test colorimetric data 34. Are stored in the HDD 30.

(2-2) Details of Color Correction Table Generation and Correction Accuracy Prediction The computer 10 executes the processing of FIGS. 2 and 3 in a state where the reference colorimetric data 33 and the test colorimetric data 34 are held.
In step S (hereinafter, step notation is omitted) 100, the calibration module 22 acquires the reference colorimetric data 33 from the HDD 30, and in S110 acquires the test colorimetric data 34 from the HDD 30.

In S <b> 120, the calibration module 22 specifies one ink type (for example, M ink) from among a plurality of ink types used in the target printer 60.
In S130, the calibration module 22 extracts the colorimetric data of each patch corresponding to the ink type specified in S120 from the standard colorimetric data 33, and uses the extracted colorimetric data as a reference point for a predetermined point. An approximate curve f1 is generated on the color space.
In S140, the calibration module 22 extracts the colorimetric data of each patch corresponding to the ink type specified in S120 from the test colorimetric data 34, and uses the extracted colorimetric data as a reference point for a predetermined point. An approximate curve f2 is generated on the color space.

FIG. 5 shows the approximate curve f1 and the approximate curve f2.
In the figure, the approximate curve f1 generated based on the reference colorimetric data 33 for the patch printed with M ink in the color chart A and the test measurement for the patch printed with M ink in the color chart B are shown. An appearance of the approximate curve f2 generated based on the color data 34 on the ab plane in the Lab space is shown. The approximate curve can be easily obtained by using an n-th order polynomial. The specific method for obtaining the approximate curve is not particularly limited, and various functions such as a spline function can be used.
In S150, the correspondence between each coordinate point corresponding to each gradation value on the approximate curve f1 and each coordinate point corresponding to each gradation value on the approximate curve f2 is specified, so that the ink type specified in S120 is determined. A color correction table (correction related data) is generated.

  FIG. 6 is an enlarged view of a part of the approximate curves f1 and f2 shown in FIG. In the figure, coordinate points (colors represented in the Lab color system) corresponding to each gradation value are indicated by black squares on the approximate curve f1, and each gradation value is indicated by a black circle on the recent curve f2. Coordinate points are shown. The white square is reference colorimetric data 33, and the white circle is test colorimetry data 34. That is, the calibration module 22 defines the coordinate points corresponding to all the gradation values including the gradation values of the patches printed on the color charts A and B on each approximate curve at predetermined intervals. Interpolate the reference color measurement data 33 and the test color measurement data 34.

  After performing the interpolation as described above, the calibration module 22 specifies one coordinate point on the approximate curve f2 that minimizes the color difference ΔE from the coordinate point corresponding to a certain gradation value i on the approximate curve f1, The gradation value j corresponding to the specified coordinate point is set as a corrected gradation value for the gradation value i. That is, when the target printer 60 prints a patch with M ink based on the gradation value i, it is ideal that a color equivalent to the color corresponding to the gradation value i on the approximate curve f1 is obtained. The tone value j for printing the color closest to the color corresponding to the tone value i on the approximate curve f1 among the colors that can be realized by the target printer 60 is the corrected value for the tone value i. is there.

The calibration module 22 executes the specification of the correction relationship between the gradation values as described above for each gradation value, so that the color correction table that uniquely defines the corrected gradation values for all gradation values. Is generated.
In S160, the calibration module 22 determines whether or not the color correction table has been generated for all ink types used by the target printer 60. If there is an ungenerated ink type, the process returns to S120 to perform color correction. After updating the ink type to be generated in the table, the processes from S130 are repeated. On the other hand, if color correction tables have been generated for all ink types, the process proceeds to S170.

In S170, the calibration module 22 calculates a color difference ΔE (color difference before correction) between the reference colorimetric data 33 and the test colorimetric data 34 having the same ink type and gradation value, and calculates the calculated color difference before correction. The average value CdV is calculated. The color difference before correction is calculated by the number of patches printed on the color chart A (or color chart B).
Each set of p1 and q1 and p2 and q2 shown in FIG. 6 indicates a combination of the reference colorimetric data 33 and the test colorimetric data 34 having the same gradation value at the time of printing. In S170, the color difference ΔE between the reference color measurement data 33 and the test color measurement data 34 having the same ink type and gradation value, such as the color difference ΔE between p1 and p2 and the color difference ΔE between q1 and q2. Average. Note that the reference colorimetric data 33 and the test colorimetric data 34 are not necessarily present on the approximate curve depending on the method used to obtain the approximate curve.

  In S180, the calibration module 22 realizes the color on the approximate curve f1 realized by the gradation value before correction and the gradation value after correction in the correction relationship of each gradation value defined in the color correction table. An average value CdS of the color difference Δ (color difference between correction relationships) with the color on the approximate curve f2 is calculated. There are basically as many color differences between correction relationships as the number of ink types × total gradation values. That is, the color difference ΔE between the coordinate point corresponding to the gradation value i on the approximate curve f1 and the coordinate point on the approximate curve f2 corresponding to the gradation value j that establishes the correction relationship with the gradation value i is corrected. The color difference between the correction relationships is averaged.

The processing up to S180 is mainly executed by the color correction table generation unit 22a. However, the order of calculating the average value CdV of the color difference before correction and the average value CdS of the color difference between correction relationships may be reversed.
After calculating CdS and CdV, in S190, the calibration module 22 compares CdS and CdV with predetermined threshold values, respectively, and predicts whether the correction accuracy based on the comparison result is good or not based on the comparison result. To do. The processes after S190 are mainly executed by the correction accuracy prediction unit 22b.

FIG. 7 shows a table of prediction branches according to the values of CdS and CdV.
In the present embodiment, a threshold value Th1 for CdS and a threshold value Th2 for CdV are obtained in advance by experiments, and a prediction branch as shown in FIG.
According to the figure, when (I) CdS ≦ Th1 and CdV ≦ Th2, and (III) CdS ≦ Th1 and CdV> Th2, the correction accuracy based on the color correction table is predicted to be passed. In addition, when (II) CdS> Th1 and CdV ≦ Th2, the prediction level is determined to be an attention level. In addition, when (IV) CdS> Th1 and CdV> Th2, it is predicted to be rejected.

  Each step of S200, S210, and S220 is selected according to the prediction result in S190. In other words, if it is determined that the result is acceptable, in S200, the calibration module 22 performs a process of displaying on the display 40 that the correction accuracy based on the color correction table is at the acceptable level (accepted message). If it is determined that the level is a caution level, it is determined in S210 that the correction accuracy based on the color correction table is at the caution level (caution message). Performs a process of displaying on the display 40 a message indicating a failure level (failure message). That is, the prediction result is notified to the worker. Of course, in S200 to S220, various displays can be performed in addition to the above messages, and the CdS and CdV values may be displayed alone or together with the above messages.

  After S200, the calibration module 22 stores the generated color correction table for each ink type in the HDD 30 as the color correction table 32 (S240). In addition, after S220, the calibration module 22 ends the flowchart without saving the color correction table in the HDD 30. Further, after S210, the calibration module 22 determines whether or not an instruction to save the color correction table is given from the outside (S230). If there is an instruction to save, the calibration module 22 saves the color correction table. If not, the flowchart is terminated without saving the color correction table.

That is, depending on the prediction result in S190, whether or not the color correction table is stored in the HHD 30 is branched, and if the correction accuracy based on the color correction table cannot be predicted as pass or fail, color The operator is inquired of whether or not the correction table needs to be stored, and left to the operator's judgment. However, regardless of the prediction result in S190, the operator may be inquired of whether or not to save the color correction table after each message.
The color correction table 32 for each ink type stored in the HDD 30 in this manner is used as appropriate in print control processing for the target printer 60 executed by the computer 10 (printer driver 21) thereafter. That is, the computer 10 uses the color correction table 32 to correct the ink amount data composed of the gradation values for each ink type as gradation values before correction.

Here, the basis for predicting the correction accuracy based on the color correction table based on the branch shown in FIG. 7 will be described.
First, in the case of (I), the color difference between the print result by the reference printer 50 and the print result by the target printer 60 is very small, and the print results tend to be very similar for each ink type and each gradation value. It can be said. Accordingly, if the ink amount data is corrected by the color correction table in the target printer 60, it can be predicted that a color equivalent to the printing result by the reference printer 50 can be sufficiently printed (correction accuracy is high). Make a pass judgment as follows.

  In the case of (IV), the color difference between the printing result by the reference printer 50 and the printing result by the target printer 60 is large, contrary to the case of (I). That is, since the color reproduction capability of the target printer 60 is largely deviated from that of the reference printer 50 as a whole, it is ideal that the reference printer 50 outputs regardless of how the ink amount data value is changed. Color cannot be output. Therefore, even if the ink amount data is corrected by the color correction table in the target printer 60, it is predicted that a color equivalent to the printing result by the reference printer 50 cannot be realized (correction accuracy is low), and a failure is determined. .

  Here, in the case of (III) above, that is, if CdS ≦ Th1, CdV> Th2 is judged as acceptable, and one feature of the present invention appears. CdS ≦ Th1 and CdV> Th2 are explained by taking the approximate curves f1 and f2 as an example, although there are large color differences such as between p1 and q1 and between p2 and q2 where the gradation values during printing are common. Both approximate curves f1, f2 themselves are close to each other over the whole, and it can be said that the curve shapes are very similar. That is, it can be said that the deviation between the printing result by the reference printer 50 and the printing result by the target printer 60 based on the common ink amount data is a density deviation. In such a situation, as long as the value of the ink amount data is changed in the target printer 60, it is possible to print a color equivalent to the printing result by the reference printer 50. Therefore, in the case of the above (III), it is predicted that the correction accuracy according to the color correction table is high, and the pass is judged.

  In the case of prediction using only the average value CdV of the color difference before correction, the case (IV) and the case (III) are not distinguished, and are rejected. That is, even if it should be accepted (when an ideal color can be output by the target printer 60 by correcting the ink amount data), it was also rejected. However, in the present invention, the average value CdS of the color differences between correction relationships is determined. By using it, it became possible to prevent such misjudgment.

  Further, in the case of the above (II), the approximate curves f1 and f2 will be described as an example. The results of printing by the reference printer 50 based on common tone values such as the color difference between p1-q1 and p2-q2 Since the difference from the print result by the target printer 60 is small, the two approximate curves f1 and f2 are considered to be close to each other to some extent, but the color itself on the approximate curve f2 realized by the corrected gradation value is ideal. The color has shifted to some extent. Such a situation mainly corresponds to a case where there is a hue shift between the printing result by the reference printer 50 and the printing result by the target printer 60. In such a situation, when the value of the ink amount data is changed in the target printer 60, the print result may not be brought close to the range that the operator can accept as being equivalent to the print result by the reference printer 50. . Therefore, it cannot be said that the correction accuracy according to the color correction table is high or low, and a determination of the attention level is made.

In this embodiment, not only CdS but also the value of CdV is used for predicting the correction accuracy in that a branch between failure or attention level can be made.
However, if the prediction result of the attention level is not required, the correction accuracy prediction determination by the color correction table may be performed based only on the comparison result between CdS and the threshold value T1. In this case, the process of S170 is not necessary, the determination is simplified even after S190, and S210 and S230 can be omitted.

  In the flowchart, after the color correction table generation process (S150), CdS and CdV are obtained and compared with the threshold values Th1 and Th2. However, CdS and CdV are obtained and the threshold value Th1 is obtained. The process of comparing with Th2 may be performed prior to the color correction table generation process. In this case, a color correction table may be generated and stored in the HHD 30 only when a pass (or attention level) determination is made.

  Further, the CdS and CdV may be obtained for each ink type. That is, in S170, an average value is calculated for each group of color differences before correction, which is calculated for the number of patches printed on the color chart A (or color chart B), and which has the same ink type of patch. May be. In S180, an average value may be calculated for each group of inter-correction relationship color differences that have the same ink type for the inter-correction relationship color differences calculated for the number of ink types × all gradation values. As a result, both CdS and CdV are obtained for each ink type. A threshold value to be compared with CdS and CdV is also prepared for each ink type, and if each CdS and CdV is compared with each threshold value, correction by the color correction table for each ink type is performed based on various comparison results. It becomes possible to predict the accuracy more precisely.

(2-3) Prediction Result Accuracy FIG. 8 is a table showing experimental results for verifying the accuracy of color correction accuracy prediction according to the present invention.
In the figure, “printing conditions” including items “voltage”, “type”, “media”, and “resolution” are shown from 1 to 9. This printing condition means a printing condition in the target printer. “Voltage” indicates the degree of increase / decrease in the voltage applied to the piezo element of the print head. In this figure, “0” (unit) means the voltage that should be applied in order to eject a predetermined amount of ink. V), “+1.5” obtained by adding 1.5 V to the voltage to be originally applied, and “−1.5” obtained by reducing 1.5 V from the voltage to be originally applied. . The “type” indicates a model of the printer, and in the figure, a model “a” and a model “s” are used.

“Media” indicates the type of printing paper used for printing the color chart. In FIG. 5, five types of media “M1” to “M5” are used. “Resolution” indicates a print resolution, and a plurality of print resolutions such as 1440 × 1440 (dpi) are used.
Also, CdV in the figure is 1 based on the colorimetric data of each patch of the color chart (color chart A) printed on the reference printer based on the patch image data 31 and the patch image data 31. The average value of the color difference before correction calculated by comparing the color measurement data of each patch of the color chart (color charts B1 to B9) printed under the printing conditions of .about.9.

Note that the reference printer in the experiment printed a color chart A and a color chart C described later under the conditions that the voltage was 0, the type was S, the medium was M5, and the resolution was 1440 × 720.
CdS in the figure is the color correction for each printing condition 1 to 9 by comparing the colorimetric data of each patch of the color chart A with the colorimetric data of each patch of the color charts B1 to B9. It is an average value of color differences between correction relationships calculated in the process of calculating a table.

  Further, the “average value of corrected color differences” in the same figure was obtained as follows. First, the computer 10 applies a verification color chart (color chart C) to the reference printer based on verification patch image data (data different from the patch image data 31) representing a predetermined verification color chart. Let it print. Next, the computer 10 performs correction using the color correction table for each of the printing conditions 1 to 9 in the process of printing the color chart based on the printing conditions 1 to 9 based on the verification patch image data. The verification color chart (color charts D1 to D9) is printed. Then, the computer 10 calculates the average value of the color difference of the colorimetric data between patches having the same gradation value at the time of patch printing for each combination of the color chart C and the color charts D1 to D9. The calculated average values are set as “average values of corrected color differences” for the printing conditions 1 to 9.

  Here, the verification patch image data is image data for printing a plurality of patches expressing each color scattered substantially evenly over a predetermined color space. For example, when a color is expressed by gradation values for each RGB, each gradation value for each RGB is obtained by changing each gradation value for each RGB from the lowest gradation to the highest gradation in a predetermined number of steps. Define image data. Therefore, the “average value of color differences after correction” is a color shift between the print result of the target printer and the print result of the reference printer after color correction according to the color correction table, and covers almost the entire color space. It can be said that this is an average of the deviations seen.

  As can be seen from the figure, in this experiment, only the printing condition 7 has the same printing condition as that of the reference printer, and the printing conditions other than that are different in some items. In general, when calculating the color correction table, it is necessary to compare the color measurement results of the printed matter printed on the reference printer and the target printer that share all the printing conditions. In this experiment, however, there is a difference in the type. By preparing various printing conditions, a plurality of target printers having a difference in color reproducibility from the reference printer are conveniently created.

Here, when the same figure is seen in detail, in the printing conditions 6 to 9, the “average value of corrected color differences” shows a somewhat low value. That is, in the printing conditions 6 to 9, by using the generated color correction table, the color of the printing result by the target printer becomes a value close to the color of the printed matter by the reference printer (high-precision correction can be performed by the color correction table). It was done).
More specifically, when it is assumed that the average color difference after correction must be 0.8 or less in order to evaluate that the color of the print result of the target printer is equivalent to the color of the printed matter of the reference printer In this experiment, printing conditions 7 to 9 correspond.

Here, according to the prediction criterion of FIG. 7, if CdS is equal to or less than the threshold value Th1, the correction accuracy based on the color correction table is determined to be an acceptable level. It can be seen that it may be set to a predetermined value around 5 to 0.9.
In the printing condition 9, only the “voltage” item is changed as compared with the printing condition of the reference printer. The purpose of changing only the item “voltage” is to slightly change the ink discharge amount in each print head to cause a deviation in the density of the print result. In this printing condition 9, although CdV is as high as 3.54, CdS is as low as 0.47, and the average value of corrected color differences is as low as 0.72.

In other words, the printing condition 9 is a typical case where an ideal print color can be realized by correction using the color correction table if CdV is large but CdS is small. That is, if it is a determination using only CdV as in the prior art, the color correction table generated under the printing condition 9 is likely to have been rejected even though its correction accuracy is actually high. However, in the present invention, such erroneous determination is eliminated. Based on the results of this experiment, the threshold value Th2 can be set to a predetermined value around 3.0. Also, since the printing conditions 1 to 6 are different from the standard printer in “type”, “media”, and “resolution”, both CdS and CdV are high values, and as a result, the correction accuracy by the color correction table is also high. It is low.
That is, according to this experiment, it can be seen that the correction accuracy of the color correction table can be accurately predicted by using CdS and CdV.

  Here, referring to FIG. 8 (particularly, the columns of printing conditions 7 to 9), the relationship between CdS and “average value of corrected color difference” is a predetermined coefficient (for example, coefficient = 1.8) with respect to CdS. ) Is approximately equivalent to “average color difference after correction”. Therefore, the computer 10 multiplies CdS calculated based on the colorimetric results of the patches (color charts A and B) printed with each ink single color by a predetermined coefficient determined in advance based on the above-described experiment. Thus, the predicted value of “average value of corrected color differences” for the target printer can be calculated. In S200 to S220, the computer 10 may display not only each message, CdS, and CdV but also a predicted value of “average value of corrected color difference” calculated as described above. By performing such display, the operator can easily know an average deviation over almost the entire color space of the printing result of the target printer after color correction.

(3) Summary As described above, according to the present invention, the color measurement result of the patch printed for each ink type and for each of the plurality of gradations by the reference printer 50 used as the reference for color correction, and the target printer for which color correction is performed. When generating a color correction table for correcting color misregistration of the printing result by the target printer 60 based on the comparison with the color measurement result of the patch printed for each ink type and for each of the plurality of gradations at 60, at least between correction relations An average value CdS of color differences is obtained, an average value CdV of color differences before correction is further obtained, and the correction accuracy based on the color correction table is predicted according to the value of the CdS (and CdV). As a result, it is possible to accurately determine whether or not appropriate correction is performed by the color correction table without performing verification operations such as reprinting and colorimetry of the patch after correcting the image data by the color correction table. I can know.

The block diagram which shows the computer etc. which function as a correction accuracy prediction apparatus. The flowchart which showed the content of the color correction table production | generation process and the correction precision prediction process. The flowchart which showed the content of the color correction table production | generation process and the correction precision prediction process. The figure which showed an example of the color chart. The figure which showed the approximated curve produced | generated on ab plane. The figure which expanded and showed a part of approximate curve. The figure which showed the branch of the prediction according to the value of CdS and CdV. The figure which showed the experimental result which verified the precision of correction | amendment precision prediction.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 ... Computer, 21 ... Printer driver, 22 ... Calibration module, 22a ... Color correction table production | generation part, 22b ... Correction precision prediction part, 30 ... HDD, 31 ... Patch image data, 32 ... Color correction table, 33 ... Standard measurement Color data 34 ... Test colorimetric data 50 ... Printer (reference printer) 60 ... Printer (target printer) 70 ... Colorimeter

Claims (5)

First data obtained by colorimetry of a plurality of first patches serving as a reference for color correction printed based on a plurality of levels of gradation data, and the above-described multiple levels of gradation on a printing apparatus to be subjected to color correction Input means for inputting second data obtained by color measurement of a plurality of second patches printed based on the data;
The first data corresponding to the arbitrary gradation data is associated with the second data having the smallest color difference with the first data, and the gradation data corresponding to the arbitrary gradation data and the associated second data. Correction relationship generating means for generating correction relationship data for each gradation data by relating
It calculates an average value of the correction relationship between the color difference to calculate the respective correction relationship between the color difference which is the color difference for each combination of the first data and the second data associated with the above SL correction relation creating means, and floor The color difference before correction, which is the color difference for each combination of the first data and the second data having the same tone data, is calculated, the average value of the color difference before correction is calculated, and the average value and correction of the color difference between correction relationships are calculated. Prediction means for predicting the accuracy of color correction performed using the correction relation data based on the average value of previous color differences ,
The prediction means predicts that the accuracy of color correction performed using the correction relationship data is acceptable when the average value of the color difference between correction relationships is below a first threshold, and the color difference between the correction relationships. When the average value of the color difference exceeds the first threshold value and the average value of the color difference before correction exceeds the second threshold value, the accuracy of color correction performed using the correction relationship data is rejected. When the average value of the color difference between the correction relationships is higher than the first threshold value and the average value of the color difference before correction is lower than the second threshold value, the correction relationship data is used. If the accuracy of color correction is predicted as a caution level, the correction related data is stored in a predetermined storage medium when predicted to pass, and the correction related data is not stored when predicted to fail, When it is predicted to be the attention level, an instruction to save correction-related data Correction accuracy prediction apparatus the correction relationship data on condition that received from the outside, characterized in that stored in the predetermined storage medium.   The correction relationship generating means defines a first approximate curve on a predetermined color space with each first data as a reference point, and defines a coordinate point corresponding to each gradation data on the first approximate curve, A second approximate curve is defined on the color space using the second data as a reference point, coordinate points corresponding to each gradation data are defined on the second approximate curve, and defined on the first approximate curve. By associating the gradation value corresponding to the coordinate point defined on the second approximate curve having the smallest color difference from the coordinate point corresponding to the arbitrary gradation value with the arbitrary gradation value, The correction accuracy prediction apparatus according to claim 1, wherein the correction relation data is generated. 3. The correction accuracy prediction apparatus according to claim 1, wherein the prediction means notifies the prediction result to the outside using a predetermined output device. First data obtained by colorimetry of a plurality of first patches serving as a reference for color correction printed based on a plurality of levels of gradation data, and the above-described multiple levels of gradation on a printing apparatus to be subjected to color correction An input step of inputting second data obtained by color measurement of a plurality of second patches printed based on the data;
The first data corresponding to the arbitrary gradation data is associated with the second data having the smallest color difference with the first data, and the gradation data corresponding to the arbitrary gradation data and the associated second data. A correction relationship generation step of generating correction relationship data for each gradation data by relating
It calculates an average value of the correction relationship between the color difference to calculate the respective correction relationship between the color difference which is the color difference for each combination of the first data and the second data associated with the above SL correction relation generation step, and floor The color difference before correction, which is the color difference for each combination of the first data and the second data having the same tone data, is calculated, the average value of the color difference before correction is calculated, and the average value and correction of the color difference between correction relationships are calculated. A prediction step of predicting the accuracy of color correction performed using the correction relation data based on the average value of the previous color difference ,
In the prediction step, when the average value of the color difference between the correction relationships is lower than the first threshold value, the accuracy of the color correction performed using the correction relationship data is predicted to be acceptable, and the color difference between the correction relationships is determined. When the average value of the color difference exceeds the first threshold value and the average value of the color difference before correction exceeds the second threshold value, the accuracy of color correction performed using the correction relationship data is rejected. When the average value of the color difference between the correction relationships is higher than the first threshold value and the average value of the color difference before correction is lower than the second threshold value, the correction relationship data is used. If the accuracy of color correction is predicted as a caution level, the correction related data is stored in a predetermined storage medium when predicted to pass, and the correction related data is not stored when predicted to fail, Save correction-related data if predicted to be attention level Correction accuracy prediction method characterized by the condition that has been accepted shown externally storing the correction relationship data to the predetermined storage medium. A correction accuracy prediction program,
First data obtained by colorimetry of a plurality of first patches serving as a reference for color correction printed based on a plurality of levels of gradation data, and the above-described multiple levels of gradation on a printing apparatus to be subjected to color correction An input function for inputting second data obtained by colorimetry of a plurality of second patches printed based on the data;
The first data corresponding to the arbitrary gradation data is associated with the second data having the smallest color difference with the first data, and the gradation data corresponding to the arbitrary gradation data and the associated second data. A correction relationship generation function for generating correction relationship data for each gradation data by relating
It calculates an average value of the correction relationship between the color difference to calculate the respective correction relationship between the color difference which is the color difference for each combination of the first data and the second data associated with the above SL correction relation generation function, and floor The color difference before correction, which is the color difference for each combination of the first data and the second data having the same tone data, is calculated, the average value of the color difference before correction is calculated, and the average value and correction of the color difference between correction relationships are calculated. Based on the average value of the previous color difference, the computer is caused to execute a prediction function that predicts the accuracy of color correction performed using the correction-related data ,
The prediction function predicts that the accuracy of the color correction performed using the correction relationship data is acceptable when the average value of the color difference between the correction relationships is lower than a first threshold, and the color difference between the correction relationships. When the average value of the color difference exceeds the first threshold value and the average value of the color difference before correction exceeds the second threshold value, the accuracy of color correction performed using the correction relationship data is rejected. When the average value of the color difference between the correction relationships is higher than the first threshold value and the average value of the color difference before correction is lower than the second threshold value, the correction relationship data is used. If the accuracy of color correction is predicted as a caution level, the correction related data is stored in a predetermined storage medium when predicted to pass, and the correction related data is not stored when predicted to fail, When it is predicted to be the attention level, an instruction to save correction-related data Correction accuracy prediction program the correction relationship data on condition that received from the outside, characterized in that stored in the predetermined storage medium.

Images