JP7255228B2 - Image processing device, image processing system and image processing program - Google Patents

Description

The present invention relates to an image processing device, an image processing system, and an image processing program.

Digital printing devices, such as electrophotographic or inkjet, are known today. Among such digital printing apparatuses, in the case of a printing apparatus that performs a large amount of printing, the stability of the output color is required during continuous output of hundreds or thousands of sheets. In particular, stable management of reproduced colors is important in use cases where the same type of document is repeated with only partial content replaced every few pages, or in use cases where distributed printing is performed at multiple sites.

However, unlike full-scale commercial printing, the operating environment in which these digital printers are utilized is often not strictly controlled. For this reason, digital printers are often used in temperature and humidity environments that affect printing results.

On the other hand, there are many unavoidable unstable factors such as changes in the state of the apparatus, such as the amount of toner supplied, due to mixed printing of various document types. For this reason, when stable management of output colors is required, it is necessary to frequently stop the machine and perform calibration to adjust tint and color shift. Regarding calibration, for example, Patent Document 1 (Japanese Patent Application Laid-Open No. 2017-187585) discloses a printing system or the like that enables calibration at an appropriate page position.

Determining the colors to be printed means that the reference characteristics, which are the correspondence standards of the colors to be printed with respect to the input data, are defined in advance. Normally, calibration is performed for the purpose of matching such reference characteristics and actual characteristics. In this sense, it is ideal to perform calibration immediately before each print job.

However, the more calibration operations, the more wasted paper, the more the job stops and the more man-hours. In some cases, it may be necessary to use less fresh calibration information.

In general, color differences are sensitively perceived when viewed side by side at the same time, and less perceived when compared individually. Therefore, it is required to manage the consistency of colors between prints in the same job, in which print results are often compared side by side, at a stricter level. However, if the actual characteristics at the start of the job deviate from the reference characteristics, the print color drifts until the actual characteristics match the reference characteristics due to feedback.

SUMMARY OF THE INVENTION The present invention has been made in view of the above-described problems, and particularly provides an image processing apparatus, an image processing system, and an image processing program capable of providing an image stabilization method that maintains an initial print state, and provides a color correction method in a print job. An object of the present invention is to provide an image processing device, an image processing system, and an image processing program capable of stabilizing reproducibility.

In order to solve the above-described problems and achieve the object, the present invention provides a print medium on which an output image corresponding to input image data is formed. The reflection characteristics of the output image are respectively detected, and the gradation control parameters for controlling the gradation characteristics of the basic colors of the output image formed on the print medium are set to the reflection characteristics of the output image formed on the print medium. a reflection characteristic detection unit having a gradation control parameter calculation unit that calculates based on the gradation control parameter calculation unit; and a gradation characteristic correcting section that corrects the gradation characteristic of the input image data based on the result of comparison with the second gradation control parameter newly calculated by the control parameter calculating section. Gradation control parameters are obtained by synthesizing a mode curve that approximates the color variation obtained from the measured values obtained by measuring the output image with a scanner, not only at the measurement points, but also at least in the entire gradation range for each page. is the coefficient.

According to the present invention, it is possible to provide an image stabilization method that maintains the initial printing state, and to stabilize color reproducibility in a print job.

FIG. 1 is a system configuration diagram of an image processing system according to the first embodiment. FIG. 2 is a block diagram showing the configuration of the color tone control section provided in the image processing system according to the first embodiment. FIG. 3 is a vertical sectional view of a laser printer provided in the image processing system of the first embodiment. FIG. 4 is a diagram for explaining mode curves. FIG. 5 is a diagram showing an example of a gradation correction curve formed by synthesizing a plurality of mode curves. FIG. 6 is a flow chart showing the flow of operations for estimating mode parameters in the initial state immediately after calibration in the image processing system according to the first embodiment. FIG. 7 is a flow chart showing the flow of operations for estimating normal mode parameters in the image processing system of the first embodiment. FIG. 8 is a block diagram showing the configuration of a color tone control section provided in the image processing system of the second embodiment. FIG. 9 is a flow chart showing the operation flow of tone correction processing of the image processing system according to the second embodiment. FIG. 10 is a diagram for explaining the in-plane deviation. FIG. 11 is a diagram showing density characteristics before and after tone correction processing in the image processing system of the second embodiment.

As an example, an image processing apparatus, an image processing system, and an image processing system according to an embodiment to which an image processing program is applied will be described below.

[First embodiment]
(System configuration of the first embodiment)
FIG. 1 is a system configuration diagram of main parts of an image processing system according to the first embodiment. As shown in FIG. 1, the image processing system of the first embodiment connects a user's personal computer device (user PC) 1, a server device 7, and an image forming device 8 to, for example, a LAN (Local Area Network) or the Internet. are formed by connecting to each other via a predetermined network 2 such as. At least one user PC 1 is connected to the network 2 and transmits image data and a print request to the image forming apparatus 8 . The server device 7 stores (accumulates) information necessary for color conversion performed by the image processing unit 3 .

The image forming apparatus 8 includes an image processing unit 3 that develops and processes document data 30 input via the network 2 , an electrophotographic printer engine 4 that executes printing, and an engine control unit that controls the printer engine 4 . has 9. The image forming apparatus 8 also includes a gradation processing unit 31 that converts the pixel array developed by the image processing unit 3 into the number of gradations that can be output by the printer engine 4, and outputs an output image 6 from the printer engine 4. It has an image inspection unit 5 for in-line inspection in front. Further, the image forming apparatus 8 detects color tone variation (density variation or hue variation, etc.) of the output image from the image detected by the image inspection section 5, and supplies correction parameters to the tone processing section 31 for color tone control. It has a portion 28 .

The engine control section 9 is provided in the same housing as the printer engine 4 . The engine control section 9 , the printer engine 4 , the image inspection section 5 , and the color tone control section 28 constitute a body unit group 32 . In the example of FIG. 1 , the gradation processing section 31 is illustrated as being provided outside the main body unit group 32 , but the gradation processing section 31 may be provided inside the main body unit group 32 .

The image processing unit 3 is composed of software and an expansion board. The image processing section 3 is separate from the main body unit group 32 and is replaceable with respect to the main body unit group 32 .

The image inspection unit 5 is composed of an RGB line sensor (RGB: red, green, and blue) serving as an image measuring unit, and a scanner 150 having a paper feeding mechanism. This image inspection unit 5 enables the colorimetry of the image on the plane.

The manuscript data 30 normally transmitted when the user PC 1 requests printing is, for example, bitmap data or text data in which colors are specified in RGB or CMYK (cyan, magenta, yellow, and black) as shown in FIG. , which is data in a complicated data format containing graphics drawing commands, and is transmitted to the image processing unit 3 via the network 2 .

The image processing unit 3 is also called a digital front end (DFE). The image processing unit 3 develops the received data and converts it into pixel array data (bitmap data or compressed data equivalent to the bitmap data) composed of the basic colors of the printer engine 4. 31. The gradation processing unit 31 converts, for example, each pixel data of bitmap data into pixel data of the number of gradations expressible by the printer engine 4 . The printer engine 4 forms an output image 6 on paper based on the bitmap data subjected to such conversion processing.

The image inspection unit 5 scans the output image of the printer engine 4 and supplies it to the color tone control unit 28 . The color tone control unit 28 uses, as a target value, a predicted value by a colorimetric prediction unit (reference numeral 21 in FIG. 2), which is a prediction unit to be described later, or a scanned image of an initial print output stored in the image memory 36, and this target value is A gradation correction parameter that minimizes the difference between the color and the color of the image scanned by the image inspection unit 5 is set in the gradation processing unit 31 . As a result, the gradation correction processing of the document data 30 can be performed in the gradation processing section 31, and the reproduced colors of the output image 6 can be stabilized.

(Details of color control section)
FIG. 2 is a block diagram showing each function of the color tone control section 28. As shown in FIG. As shown in FIG. 2, the tone control section 28 has a difference detection section 40 and a correction TRC calculation section 41 (TRC: Tone Reproduction Curve). The difference detection section 40 has a registration correction section 11 , a selector 14 , a subtractor 15 , a colorimetry prediction section 21 and an image memory 36 .

The corrected TRC calculator 41 includes a colorimetric region extractor 18, an address corrector 19, a sample extractor 20, a mode parameter calculator 22, a leveling processor 23, a subtractor 24, a TRC combiner 25, a calculation condition selector 35, It has a sample memory 37 , a first mode parameter memory 38 and a second mode parameter memory 39 .

The difference detection unit 40 and the corrected TRC calculation unit 41 may be partially or wholly realized by hardware. Also, the registration correction unit 11, the selector 14, the subtractor 15, and the colorimetry prediction unit 21 of the difference detection unit 40 may be realized by software. Similarly, the colorimetric region extraction unit 18, the address correction unit 19, the sample extraction unit 20, the mode parameter calculation unit 22, the leveling processing unit 23, the subtractor 24, the TRC synthesis unit 25, and the calculation Condition selector 35 may be implemented in software. An image processing program for realizing the difference detection unit 40 and the corrected TRC calculation unit 41 by software is stored in the ROM (Read Only Memory), RAM (Random Access Memory), HDD (Hard Disk Drive), etc. of the image forming apparatus 8. , and executed by a control unit such as a CPU (Central Processing Unit) of the image forming apparatus 8, thereby realizing each function described above.

In FIG. 2, the image processing unit 3 develops the user document described in various input formats as described above into a CMYK image 10, which is a frame-sequential pixel array in which each CMYK value is 8 bits for each page. .

The gradation processing section 31 functioning as a gradation characteristic correction section has a gradation correction table 16 and a gradation conversion section 17 . A so-called Look Up Table (LUT) can be used as the tone correction table 16 . The gradation conversion unit 17 converts a pixel array of 8 bits for each color surface of CMYK by the area coverage gradation method into a pixel array of gradation of the number of bits (for example, 2 bits) that can be drawn by the printer engine 4 .

The gradation correction table 16 is an 8-bit input-8-bit output correction table for each CMYK color, and is a table for standardizing the density characteristics of the output image 6 corresponding to the input of the gradation processing unit 31 by pre-calibration. A value (tone correction data (TRC: Tone Reproduction Curve)) is set.

The scanner 150 is actually incorporated inside the image inspection section 5 that is directly connected (directly connected) to the printer engine 4 . The scanner 150 inline scans the output image 6 formed by the printing process of the printer engine 4 .

The color tone control unit 28 controls the change in print reproduction color based on the scanned image (RGB) formed by scanning the output image 6 from the printer engine 4 with the scanner 150 and the CMYK image 10 from the image processing unit 3. (difference) is detected. Then, the color tone control unit 28 corrects the gradation correction data (TRC) of the gradation correction table 16 so as to reduce the detected difference between the two. Thereby, the reproduced color of the output image 6 can be stabilized.

Prior to such a reproduction color stabilization operation, in the image processing system of the first embodiment, the scanned image data (RGB values) of the scanner 150 corresponding to the data of the CMYK image 10 are calibrated in advance. is set as a multidimensional LUT in the colorimetric prediction unit 21, which is a prediction unit. The colorimetry prediction unit 21 predicts the RGB measurement values scanned by the scanner 150 from the CMYK values of the CMYK image 10 using this multidimensional LUT and interpolation calculation. As an example of such a prediction model, for example, the device link model of the international standard ISO15076-1 can be used.

The registration correction unit 11 corrects slight image deviations (magnification, offset, rotation, distortion) between the predicted image (RGB) 13 of the colorimetric prediction unit 21 and the scanned image 12 . Auxiliary information (magnification, offset, rotation, distortion-related correction parameters) required for the correction at this time is shared as address correction information of the address correction unit 19, which will be described later.

Next, the value n of the repetition page number signal 33 shown in FIG. Specifically, when the actual number of repeated pages is greater than a predetermined specified number and when the repeated pages are not repeated, the repeated page number signal 33 of “n=0” is supplied from the image processing section 3 . When the repetition page number signal 33 of “n=0” is supplied, no image data is read from the image memory 36 , and the selector 14 selects the predicted image 13 as the reference image and supplies it to the subtractor 15 .

On the other hand, when the value of the repeated page number signal 33 is "n>0", the first n pages of the scanned image 12 are sequentially stored in the image memory 36 and supplied to the selector 14. . After the first n pages, the scanned image is not stored in the image memory 36, and the image set stored in the image memory 36 is repeatedly read and output at the timing when a new scanned image 12 is acquired. be done.

A subtractor 15 outputs differential image data obtained by subtracting the reference image from the scanned image 12 . In parallel with the processing of the subtractor 15, the colorimetric region extraction unit 18 extracts a colorimetric region of a predetermined size, such as a rectangular region of 2.5 mm square, from the CMYK image 10, An image area with relatively small density change is extracted as an image area suitable for colorimetry, and an address list indicating the position of each extracted image area (colorimetry suitable area position) is created.

Based on the address correction information from the registration correction unit 11 , the address correction unit 19 associates the colorimetry suitable area position registered in the address list with the image position of the differential image data output from the subtractor 15 . Based on the corrected address list, the sample extraction unit 20 extracts colorimetric area averages (c, m, y, k, Δr, Δg, Δb) from the output image of the subtractor 15 and the CMYK image 10 by random sampling. It is calculated and stored in a sample memory 37 which is a FIFO (First In First Out) provided as a measurement sample storage unit.

The mode parameter calculator 22, which is a gradation control parameter calculator, uses the samples accumulated in the sample memory 37 and the penalty values (first value and second value) given from the calculation condition selector 35, which is a switching unit. Based on this, the mode parameter update amount Δθ is determined by a calculation method to be described later.

Note that the initial mode parameter (θ0) and the mode parameter update amount (Δθ) are calculated by an equivalent algorithm (for example, Equation 8, which will be described later). For simplicity of description, the mode parameter update amount may also be simply referred to as a mode parameter hereinafter. In addition, since the tilde symbol (~) in Expression 8 is not essential from the point of view of the algorithm, it is omitted to simplify the description.

A mode parameter, which is a gradation control parameter, is a synthesis coefficient of a mode curve that best approximates the required TRC correction amount. In the case of 3 modes using 3 mode curves for each color of CMYK, the mode parameters consist of a total of 12 parameters. This mode parameter is stored in a first mode parameter memory 38, which is a storage unit such as a FIFO.

The leveling processing unit 23 removes samples showing abnormal values among the samples stored in the sample memory 37, and then averages a predetermined number of samples to determine the mode parameter update amount Δθ. .

On the other hand, the selection signal 34 for instructing switching of the mode parameter calculation condition is also a signal indicating one of the two states of "0" and "1". Specifically, the calculation condition corresponding to the value "0" of the selection signal 34 is associated with the setting operation of the mode parameter reference "θ0" corresponding to the initial printing state. Further, the calculation condition corresponding to the value "1" of the selection signal 34 is associated with the calculation operation of the mode parameter correction amount Δθ for correcting the print state change.

As an example, the operation of setting the reference θ0 is performed at the start of a continuous print job. However, even if it is difficult to obtain an image sample with sufficient reflection characteristics for a prolonged period of time, once it becomes possible to obtain an image sample with sufficient reflection characteristics, by resetting the reference θ0, it is possible to quickly recover by resuming control. It is possible to prevent serious density changes. For this reason, in the case of the image processing system of the first embodiment, when it is difficult to obtain image samples with sufficient reflection characteristics in a predetermined number of prints or at predetermined time intervals, the value of the selection signal 34 is set to It is designed to be reset to "0".

When the selection signal 34 is "0", the calculation condition selector 35 selects a parameter (a penalty value described later) corresponding to the first calculation condition for giving priority to the estimation accuracy. Then, according to the output from the leveling processing section 23, the value θ0 of the second mode parameter memory 39, which is a gradation control parameter holding section, is updated to "θ0=Δθ" (however, the initial value is "0"). be). When the selection signal 34 is "1", the calculation condition selector 35 selects a parameter corresponding to the second calculation condition for giving priority to the stability of the control result. In this case, the values in the mode parameter memory 39 are not updated.

The subtractor 24 subtracts the mode parameter θ0 held in the second mode parameter memory 39 from the output mode parameter Δθ of the leveling processing section 23 . Therefore, especially when the selection signal is "0", the subtraction value Δθ-θ0 is always "0".

The TRC synthesis unit 25 includes an accumulator 26 , a reference TRC storage unit 42 for each CMYK color, a mode curve storage unit 43 , and a synthesis TRC generation unit 44 . The mode curve storage unit 43 functions as a so-called LUT, and the value of the mode curve of the approximation basis of the TRC change difference is loaded from a predetermined storage unit. The reference TRC storage unit 42 is an LUT similar to the gradation correction table 16 . The accumulator 26 changes the polarity (sign) of "Δθ−θ0" from the subtractor 24 and accumulates it (corresponding to Equation 9, which will be described later). At the start of the job and when the selection signal 34 is "0", the gradation correction table 16 is copied and stored in the reference TRC storage unit 42, and all elements are initialized to "0".

The TRC synthesizing unit 25 sums products with the mode curve read from the mode curve storage unit 43 with the corresponding elements of the accumulator 26 as coefficients at the timing when the value of the accumulator 26 is updated for each color of CMYK. to the reference TRC stored in the reference TRC storage unit 42 to update the gradation correction table 16 with the composite TRC generated.

(Operation of the first embodiment)
Next, detailed operation of the image processing system of the first embodiment having such a configuration will be described. First, in the case of a job in which a set of documents having a predetermined number of pages or less is repeatedly printed, the image processing section 3 sets the value n (n>0) of the repetition page number signal 33 . In accordance with this, the value of the selection signal 34 is also set to "1".

As a result, the first n pages of scanned images 12 are stored in the image memory 36 . Further, as described above, the same scanned image 12 is also supplied to the selector 14, and the output result of the subtractor 15 (=the output of the difference detection section 40) is all "0". This means that the gradation correction data (TRC) is not updated by the correction TRC calculator 41 in the subsequent stage during the first n pages.

Next, after the (n+1)th page, the scan image 12 of the (n+1)th page and the corresponding scan image of the 1st page held in the image memory 36 are compared by the subtractor 15, and the n+1th page and the 1st page are compared. A difference (ΔRGB) between the scanned images is output from the difference detection unit 40 . Similarly, the n+2th page is compared with the scanned image 12 of the 2nd page. Similarly, the subtractor 15 of the difference detector 40 outputs the difference (variation: ΔRGB) between each scanned image 12 and the initial image.

Next, based on the address list separately extracted from the CMYK image 10, the sample extraction unit 20 of the corrected TRC calculation unit 41 samples RGB variation amounts of the scanned image 12 from the difference image supplied from the difference detection unit 40. (Δrgb) is associated with the input value (cmyk) of the CMYK image 10 and stored in the sample memory 37 .

As described above, in the case of repetitive printing, the value "1" is set in the selection signal 34, so the mode parameter calculation unit 22 sets the second calculation condition that prioritizes the stability of the feedback result over the estimation accuracy. A mode parameter Δθ is calculated based on the penalty value, the integrated value integrated by the integrator 26 and the “θ0 value (where θ0=0)” stored in the second mode parameter memory 39 .

Here, the mode parameter Δθ calculated by the mode parameter calculator 22 is leveled (averaged) after abnormal values are removed by the leveling process 23 . On the other hand, since the value of the selection signal 34 is "1", the value of the second mode parameter memory 39 remains the initial value "0" and is not updated. Therefore, the mode parameter Δθ, which is the result of the smoothing process, is directly output from the subtractor 24 .

In the TRC synthesizing section 25, the table values of the gradation correction table 16 are loaded into the reference TRC storage section 42 prior to the start of the job. Data of three predefined curves are loaded into the mode curve storage unit 43, which is an LUT. Also, the accumulator 26 is initialized to "0".

The output value (.DELTA..theta.-.theta.0) from the subtractor 24 is reversed in polarity and added (=subtracted) by the integrator . The cumulative θ value resulting from this addition process is used as a coefficient corresponding to each of the three colors. The composite TRC generation unit 44 calculates the composite TRC, which is the tone correction data described above, by adding the product sum of the mode curve data read from the mode curve storage unit 43 for each color component to the reference TRC. . The synthesized TRC calculated in this manner is written back to the gradation correction table 16 . That is, the gradation correction data (TRC) stored in the gradation correction table 16 is updated to the synthesized TRC newly calculated by the TRC synthesizing section 25 .

For non-repeating printing or for repeated page units exceeding a predefined upper limit, the value n of the repeat page number signal 33 is set to "n=0". Further, the value of the selection signal 34 is set to "0" during the printing of a predetermined number of sheets after the start of the print job. In this case, the selector 14 of the difference detection section 40 selects the predicted image 13 . As a result, the difference detection unit 40 outputs the difference between the scanned image 12 and the predicted image 13 by the colorimetric prediction unit 21 . After that, the flow of processing up to accumulation in the sample memory 37 is the same as described above.

While the value of the selection signal 34 is "0", the mode parameter calculator 22 calculates the observed value accumulated in the sample memory 37, the penalty value which is the first calculation condition in which accuracy is prioritized, and the accumulator 26 A mode parameter is calculated and stored in the first mode parameter memory 38 based on the accumulated integrated value and the "θ0 value (where θ0=0 in this state)" held in the mode parameter memory. As described above, when the second mode parameter memory 39 is updated with the mode parameters leveled by the leveling processor 23, the value of the selection signal 34 is changed to "1". In subsequent processing, the tone correction data (TRC) of the tone correction table 16 is updated in the same manner as described above.

(printer engine)
FIG. 3 shows a vertical sectional view of a laser printer as an example of the printer engine 4. As shown in FIG. First, the configuration and operation of the developing unit 60k will be described. The photosensitive drum 50k rotates in the direction of arrow A in FIG. This rotational position is detected by a rotation detector 57 provided at the end of the photosensitive drum 50k. A charging device 52 uniformly charges the surface of the photosensitive drum 50k cleaned by the cleaning roller 51 for the photosensitive drum 50k.

Next, the laser beam 55 emitted from the laser unit 53 scans the surface of the photosensitive drum 50k while flickering in accordance with the signal from the exposure control device 65, thereby forming an electrostatic latent image on the photosensitive drum 50k. do. The scanning direction of the laser beam 55 at this time is the main scanning direction, and the rotating direction of the photosensitive drum 50k indicated by arrow A in FIG. 3 is the sub-scanning direction.

The formed electrostatic latent image is developed with black (K) toner charged to the opposite potential supplied by the developing roller 54 to form a toner image. A toner image obtained by development is transferred to the intermediate transfer belt 61 . The configuration and operation of the developing units 60c, 60m, and 60y are the same. That is, the developing units 60c, 60m, and 60y form respective toner images of cyan (C), magenta (M), and yellow (Y), which are sequentially transferred onto the intermediate transfer belt 61 in an overlapping manner. The belt cleaning mechanism 63 removes the colorimetry patches on the intermediate transfer belt 61 by contacting the intermediate transfer belt 61 in synchronization with the positions of the colorimetry patches.

The transfer roller 62 collectively transfers the C, M, Y, and K toner images superimposed on the intermediate transfer belt 61 onto the paper conveyed on the paper conveying path 59 . The fixing device 56 heats and presses the toner image on the paper to fix it on the paper.

(Operation to determine TRC correction amount)
Next, the TRC correction amount determination operation based on the image measurement value of the colorimetry area on the user image will be described. In the following description, variables for each sample are considered random variables, and subscripts of sample numbers are omitted.

First, it is assumed that initial values of tone correction data determined by calibration are set in the tone correction table 16 of FIG. 2 corresponding to the colorimetry area on the user image. When the CMYK gradation values on the input side of the gradation processing unit 31 at this time are (tilde c, tilde m, tilde y, tilde k) and each element is corrected by synthesizing two predetermined fluctuation modes. and put it like the following formula 1. It should be noted that the reason why the number of variation modes is two is for simplification of explanation. Therefore, the number of variation modes may be, for example, three or four. For example, in the configuration example of the mode curve to be described later, the case of 3 modes is illustrated.

Here, in Equation 1 above, c, m, y, and k are CMYK gradation values before correction, c0, m0, y0, and k0 are reference gradation characteristics, M1 is the first variation mode, and M2 is the second variation mode. The variation mode, 'θij', is the mode parameter (i={c,m,y,k}, j={1,2}). In particular, the mode parameter is a real scalar, and the reference tone characteristics and each variation mode M1, M2 are independent real vectors of the same dimension defined on the input range D of each c, m, y, k tone value is. Note that the superscript is not an exponent, but simply an identification suffix. (Mathematically, the above mode is the basis of the low-dimensional subspace of the vector space formed by real-valued functions on D, but in terms of implementation, it is an array or a curve implemented by a combination of array and interpolation. will be described later with reference to FIG.

Also, the input range D is usually implemented as a real number in the [0, 1] interval, an integer in the [0, 100] interval, or an integer in the [0, 255] interval. At this time, the relationship of the first-order terms related to "θ" in Equation 1 above is expressed as Equation 2 below in matrix notation.

This formula 2 becomes the conditions of the following formulas 2-1 and 2-2.

At this time, the mode parameter θ0 that approximates the initial printing state is calculated by the following Equation 4 as a solution of Equation 3 below.

Note that “S” in Equation 3 is S={(xn, cn)|n=1, 2, . A sample set extracted from the user image associated with the input tone value c=(c, m, y, k). Also, p0=p0(x) is the actual measurement value by the scanner 150 in the initial printing, and L=L(c) is the prediction value by the colorimetry prediction unit 21. FIG.

Also, J(c) in Equation 3 is the Jacobian matrix at the input value c=(c, m, y, k) of L shown in Equation 5 below.

Also, "E" in Equations 3 and 4 is an expected value (for example, sample average). Note that some of the obvious arguments are omitted in these formulas.

Further, small positive constants "w1" and "w2" in Equation 6 below are penalties for suppressing the magnitude of variation θi, and act as regularization weights that stabilize the solution of Equation 4. As this value, an appropriate value in relation to the singular values of the matrix JM is determined in advance based on experiments. Hereinafter, the minute positive constants "w1" and "w2" are also referred to as "penalty value w1" and "penalty value w2".

In the process after the initial state mode parameter θ0 is determined, the change in the printing state from the initial state is estimated by the following equation (6).

In Equation 6, "pnew=pnew(x)" is the newly observed user image actual measurement value, and "θold" is the previous mode parameter value used in this user image output. Further, in the two penalty terms in Equation 6, “w1” is a penalty value that suppresses the size of the updated mode parameter “θnew”, and “w2” is a change from the previous “θold”. It is a penalty value to suppress. In particular, if the value of the "penalty value w2" is set to a large value, it is possible to suppress rapid changes in the mode parameters and to obtain more gentle response characteristics. At this time, a new mode parameter θnew to be used in printing in the next step is given by “θnew=θold−Δθ”.

By the way, in Equation 6 above, the areas on the document for comparing pnew and p0 (for taking the difference) need to be the same area. Therefore, the application of Equation 6 is limited to repetitive printing of the same type of document. In order to avoid this restriction, using the approximation "p0≈L+JMθ0" by the above-mentioned formula 3 for the formula 6, and setting "Δtilde θ=Δθ+θ0", the mode parameter θ is determined.

At this time, the solution of Expression 7 is obtained by Expression 8 below.

As a result, mode parameters to be used for printing in the next step can be obtained as shown in the following equation (9).

Furthermore, since Equation 7 has the same structure as Equation 3, Equation 8 can be used to calculate the initial mode parameter θ0 instead of Equation 4. In particular, with θ0=0 as the initial value, by repeatedly using Equations 8 and 9 for the same initial image sample set S=S0, in estimating the initial mode parameter θ0, the solution due to the penalty is Can relieve contraction. A specific algorithm will be described later.

(Configuration example of fluctuation mode curve (basis function))
Next, FIG. 4 shows an example of the variation mode curve described above. In the above description, for the sake of simplicity, the case of two modes has been described as an example, but FIG. 4 shows a selection example of mode curves (basis functions) in the case of three modes. In FIG. 4, the horizontal axis is the input gradation value (%), and the vertical axis is the gain with the maximum value normalized to "1". The curve 47 is the first mode curve M1, the curve 48 is the second mode curve M2, and the curve 49 is the third mode curve M3. As these first to third mode curves M1 to M3, it is preferable to prepare a combination that can effectively approximate the gradation characteristic change with as few as possible, but it is necessary to stick to the main component. No need.

FIG. 5 shows an example of a gradation correction curve formed by synthesizing these mode curves. A thick curve 44 shown in FIG. 5 indicates a gradation correction curve (composite TRC) formed by synthesizing the three mode curves of the first to third mode curves M1 to M3 of FIG. By synthesizing the first to third mode curves M1 to M3 in this manner, a sufficiently smooth synthetic TRC can be obtained. In the case of the mode curves M1 to M3, with respect to the constant reference gradation characteristics (y=x), the assumed variation of the gradation characteristics is approximately in the range of -10 to +10. By doing so, it is possible to obtain a synthesized TRC approximated as a sufficiently smooth curve that is difficult to break.

Thus, in this example, a common mode curve can be used for all C, M, Y, and K colors. In particular, by using the above-described composite TRC formed by combining different mode curves specialized for the gradation characteristic change tendency of each color, it is possible to approximate the gradation characteristic change with a smaller number of modes. By approximating the gradation characteristic change with such a small number of modes, it is possible to estimate the gradation characteristic change from a more biased small number of samples.

(Flow of θ estimation processing)
Flowcharts of FIGS. 6 and 7 show the flow of calculation of the mode parameter θ in the corrected TRC calculator 41 of FIG. For simplification of description, description of some functions such as the leveling processing unit 23 is omitted.

First, the flowchart of FIG. 6 is the flow of operation for estimating the mode parameter θ0 in the initial state immediately after calibration (when the selection signal 34 is "0") when the number of repeated pages is "0". In the flowchart of FIG. 6, in step S100, as described above, the calibration TRC is set in the gradation correction table 16, and output of user images over multiple pages is started.

Next, in step S101, the sample memory 37 accumulates image samples. Note that the gradation correction table 16 and the sample memory 37 may be implemented as software on a personal computer device. The image samples stored in the sample memory 37 are stored as pairs of difference ΔRGB values between the input CMYK values and the predicted image by the colorimetric prediction unit 21 . In particular, the sample memory 37 stores a mixed color sample set including a single color sufficiently containing the cyan component (C) as "S0,C" and a mixed color sample set including a single color sufficiently containing the magenta component (M) as "S0 ,m”, a mixed color sample set including a single color sufficiently containing the yellow component (Y) as “S0,y”, and a mixed color sample set including a single color sufficiently containing the black component (K) as “S0,k”. accumulate.

Next, each process between steps S102 to S106 is executed for each mixed color sample set "S0,C", "S0,m", "S0,y", and "S0,k". Specifically, in step S103, a first calculation condition is set for the penalty values rw1 and rw2, and both "θnew" and "θ0" are initialized to "0".

In step S104, the process of updating "θnew" is repeatedly executed a specified number of times, for example, N times (N is a natural number), according to the above-described Equations 8 and 9. The number of iterations may be one (N=1). However, at this time, "θ0" in Expression 8 is "0".

In step S105, ".theta.new" obtained as a result of the iteration is held as ".theta.0, x" (x=c, m, y, k).

When the processing of steps S102 to S106 is completed for all "S0, x" (x=c, m, y, k), in step S107, the cyan component is extracted from "θ0, c", and " A magenta component is extracted from "θ0,m", a yellow component is extracted from "θ0,y", and a black component is extracted from "θ0,k". Then, by recombining the extracted components, the initial mode parameter θ0 is constructed, and all the processing in the flowchart of FIG. 6 is completed.

In the description of the flow chart of FIG. 6, the number of repeated pages is assumed to be "0". Since it is a difference, the value of the initial mode parameter θ0 is fixed at "0".

Next, the flow chart of FIG. 7 is the flow of operation for estimating the normal mode parameter θ (when the selection signal 34 is "1"). In this case, printing continues based on the TRC determined based on some ".theta.new" (step S110). In step S111, image samples are accumulated in the sample memory 37, as in the description of FIG. At this time, the mixed color sample set including a single color sufficiently containing the cyan component (C) is "SC", the mixed color sample set including a single color sufficiently containing the magenta component (M) is "Sm", and the yellow component (Y) A mixed color sample set including a single color sufficiently containing the black component (K) is stored as "Sy", and a mixed color sample set including a single color sufficiently containing the black component (K) is stored as "Sk".

Next, each process of steps S112 to S116 is executed for each mixed color sample set "SC", "Sm", "Sy" and "Sk". Specifically, in step S113, the second calculation conditions are set for the penalty values w1 and w2, θ0 is set to the value obtained in the flow of the flow chart of FIG. 6, and "θold" is changed to "θnew" ( θold=θnew).

In step S114, "θnew" is updated according to Equations 8 and 9. Then, "θnew" is held as "θx" (x=c, m, y, k).

When the processes of steps S112 to S116 are completed for Sx (x=c, m, y, k), the process proceeds to step S117. In step S117, the cyan component is extracted from .theta.c, the magenta component is extracted from .theta.m, the yellow component is extracted from .theta.y, and the black component is extracted from .theta.k. Then, by recombining each component, the update mode parameter θnew is constructed, and all the processing of the flowchart of FIG. 7 is completed.

(Effect of the first embodiment)
As is clear from the above description, the image processing system according to the first embodiment can maintain the initial printing state, and in a print job based on a large number of primary colors, printed matter with stable color reproducibility can be obtained. can be obtained. In particular, even if the actual characteristics at the start of continuous printing deviate slightly from the calibration, priority is given to color stability, and the initial printing characteristics can be maintained (image stabilization method for color images). can be provided).

[Second embodiment]
Next, the image processing system of the second embodiment will be explained. If the gradation correction is performed in real time based on the printed user image, there is a possibility that the accuracy of gradation correction may deteriorate due to insufficient information. Specifically, there are various tendencies of user images to be printed, and among them, there are images in which the amount of information is insufficient due to the structure of the image, and it is difficult to obtain sufficient gradation correction accuracy. . If the amount of information is insufficient, the calculation error of the gradation correction amount will increase, and the correction will be stronger than necessary. When this overcorrection occurs, the correction accuracy of the gradation correction deteriorates, making it difficult to maintain the reliability of printing.

The image processing system of the second embodiment treats a plurality of images as one set and supplements the amount of information that is lacking in a single image, thereby preventing overcorrection and improving the accuracy of gradation correction. This is an example.

In the image system of the first embodiment described above, a second embodiment will be described below in which tone correction is performed using a set of a plurality of images as an information amount. However, the gradation correction operation described in the second embodiment can also be applied to a printing system that performs conventional calibration, and in this case as well, effects similar to those described later can be obtained. See the discussion below for details.

(Configuration of color tone control section)
FIG. 8 is a block diagram showing each function of the color tone control section 28. As shown in FIG. As shown in FIG. 8, the tone control section 28 has a difference detection section 40 and a correction TRC calculation section 41 (TRC: Tone Reproduction Curve). The difference detection section 40 has an RGB conversion section 70 , alignment sections 71 and 72 , and a subtractor 73 .

The corrected TRC calculator 41 includes a region extractor 81 , a main scanning deviation correction processor 82 , a local θ calculator 83 , a storage (buffer memory) 84 , a subscanning deviation correction processor 85 , a validity determination processor 86 , a transfer It has a section 87 , a map generation section 88 , a colorimetry list generation section 89 and a timer 90 . The validity determination processing section 86 has a validity determination section 91 and a θ correction section 92 .

The difference detection unit 40 and the corrected TRC calculation unit 41 may be partially or wholly realized by hardware, or may be partially or wholly realized by software. An image processing program for realizing the difference detection unit 40 and the corrected TRC calculation unit 41 by software is stored in the ROM (Read Only Memory), RAM (Random Access Memory), HDD (Hard Disk Drive), etc. of the image forming apparatus 8. , and executed by a control unit such as a CPU (Central Processing Unit) of the image forming apparatus 8, thereby realizing each function described above.

In FIG. 8, the image processing unit 3 converts the document data 30 described in various input formats as described above into a CMYK master image in which each CMYK value is an 8-bit frame-sequential pixel array for each page. It is supplied to the detection section 40 , the correction TRC calculation section 41 and the gradation processing section 31 .

The gradation processing unit 31 converts a pixel array of 8 bits for each color surface of CMYK by, for example, an area gradation method into a pixel array of gradation of the number of bits (for example, 2 bits) that can be drawn by the printer engine 4. Output. The image inspection unit 5 is actually directly connected (directly connected) to the printer engine 4 and scans the output image 6 formed by the printing process of the printer engine 4 inline.

The difference detection unit 40 of the color tone control unit 28 converts the scanned image (RGB) formed by the image inspection unit 5 scanning the output image 6 from the printer engine 4 and the CMYK master image from the image processing unit 3 into RGB. A change (difference) in print reproduction color is detected based on the RGB master image converted into an image. Then, the correction TRC calculation unit 41 of the color tone control unit 28 sets tone correction data (TRC) for reducing the detected difference between the two to the tone processing unit 31 . Thereby, the reproduced color of the output image 6 can be stabilized.

Specifically, the corrected TRC calculator 41 performs tone correction processing in real time based on data obtained from the CMYK master image and the difference between the RGB master image and the scanned image. RGB difference data (ΔRGB), which is the difference between the RGB master image and the scanned image obtained from the difference detection unit 40, and coefficients for the color variation model from the CMYK master image obtained from the image processing unit 3 (hereinafter referred to as θ ) to update the gradation correction data (TRC) and supply it to the gradation processing unit 31 .

(Operation of difference detector)
That is, the operation of each section of the difference detection section 40 is as follows. The RGB conversion unit 70 has a data table in which the CMYK values of the manuscript data 30 and the RGB values obtained from the printed manuscript data 30 are associated with each other. The values in this data table are updated for each calibration. The RGB conversion unit 70 converts the CMYK values (CMYK master image) of the document data 30 into RGB values to form an RGB master image by referring to the data table.

The alignment unit 71 and the alignment unit 72 correct the positional deviation between the scanned image formed by the image inspection unit 5 scanning the printed material 6 inline and the RGB master image described above (position processing). The subtractor 73 detects the difference between the scanned image scanned by the image inspection section 5 and the RGB master image, and supplies the difference data (ΔRGB) to the area extraction section 81 of the corrected TRC calculation section 41 .

(Operation of corrected TRC calculator)
The operation of each part of the corrected TRC calculator 41 is as follows. The map generator 88 extracts small areas that can be used as colorimetry areas for each CMYK color from the CMYK master image, and forms a map that summarizes all the extracted colorimetry areas. As an example, the map generator 88 extracts small areas based on the following conditions.

1. The size of the small area is 20 x 202 pixels (can be changed),
2. The change in density of each color in the extended area obtained by enlarging the small area is small,
3. The size of the expansion area is 50 x 502 pixels (can be changed),
4. the subregion is farther than the default value from the edge of the paper,
5. the average density of each color in the subregion is within a given range,
6. the total amount of toner in the small area is less than or equal to a certain value;
7. For CMY plates, the density of K being mixed is less than a predetermined threshold;
8. In the case of the K plate, the density is at a certain ratio or more with respect to the density of the mixed colors CMY. A map formed based on such conditions is used to generate a colorimetry list.

The colorimetry list generator 89 randomly selects a specified number of colorimetry regions from each segment based on the above map to form a "colorimetry list". A segment is each area obtained by equally dividing an image in the sub-scanning direction. The number of image divisions is, for example, 16 segments per page (changeable). The above-mentioned "θ value" is calculated based on the information of the colorimetry area registered in the colorimetry list.

Next, the region extraction unit 81 extracts the RGB difference data corresponding to the colorimetry region registered in the colorimetry list, and supplies it to the main scanning deviation correction processing unit 82 along with the coordinate information and the CMYK gradation information. .

Density variations in the main scanning direction may occur due to developer, toner bias, charging unevenness, or the like, which may cause variation bias (main scanning deviation) in printed colors along the main scanning direction. The main scanning deviation correction processing unit 82 calculates a main scanning deviation amount for the extracted colorimetry area based on the coordinate information and the CMYK gradation information. Then, the main scanning deviation correction processing unit 82 subtracts the calculated main scanning deviation amount from the RGB difference in the colorimetry area, thereby forming RGB difference data from which the influence of the main scanning deviation is removed. It is supplied to the subsequent local θ calculation processing unit 83 .

The local θ calculator 83 calculates the local θ for each segment based on the coordinate information of the colorimetry area registered in the colorimetry list, the CMYK information, and the RGB difference information after main scanning deviation correction processing. . The calculated local θ of each segment is stored in a storage unit 84 such as a buffer memory. A segment for which it is difficult to obtain a sufficient number of colorimetry regions becomes an invalid segment, and the local θ calculated from this invalid segment is not used.

Next, the photosensitive drum (reference numeral 50k in FIG. 3, etc.) is normally positioned by strict control, but due to variations in the control precision of parts, slight eccentricity between the drum rotation axis and the drum center axis can occur. occur. When such eccentricity occurs, a periodic variation (sub-scanning deviation) corresponding to the phase of the photosensitive drum occurs. Sub-scanning deviations appear periodically. Therefore, the influence of the sub-scanning deviation can be canceled out and reduced between the colors at positions where the drum period is out of phase by half.

A sub-scanning deviation correction processing unit 85 detects a pair of local θ calculated from a region in which the drum period is half phase-shifted among the local θ calculated by the local θ calculator 83 . At this time, if more than a predetermined number of pairs are detected, the sub-scanning deviation correction processing unit 85 averages the pairs of local θ and treats them as one local θ, and further calculates θ calculated from the pairs. It is averaged to obtain the final θ value (leveled θ). Further, when the number of pairs is less than the predetermined number, the sub-scanning deviation correction processing unit 85 averages all the local θ values to obtain a final θ value (leveled θ). This leveling θ is a coefficient for gradation correction that is commonly used for input images of a plurality of pages forming one set.

Based on the timer information from the timer 90, the number of drum phase half-cycle pairs detected by the sub-scanning deviation correction processing unit 85, and the colorimetry region information from the colorimetry list generation unit 89, the validity determination processing unit 86 performs gradation. The validity of the value of the leveling .theta. for tone correction is determined, and the value of the leveling .theta. is corrected based on the determination result. The synthesizing unit 93 generates (updates) tone correction data (TRC) based on the corrected θ value and supplies it to the transfer unit 87 . The transfer unit 87 supplies the generated gradation correction data (TRC) to the gradation processing unit 31 . As a result, it is possible to print a CMYK image that has undergone gradation correction processing, maintain the initial printing characteristics, and obtain a printed matter with stable color reproducibility.

(Overall print process flow)
Next, the overall flow of print processing will be described using the flowchart of FIG. Note that an area extraction unit 81, a main scanning deviation correction processing unit 82, a local θ calculation unit 83, a sub-scanning deviation correction processing unit 85 (including smoothing θ calculation), a validity determination processing unit 86, and a map generation unit, which will be described below. Each process of the unit 88, the colorimetry list generation unit 89, and the synthesis unit 93 is executed independently for each CMYK color plane.

First, in step S121, the difference detection unit 40 and the corrected TRC calculation unit 41 acquire the image data (CMYK image data) of the print image from the image processing unit. In step S122, the RGB conversion section 70 of the difference detection section 40 converts the CMYK image data into RGB master data.

In step S123, the subtractor 73 of the difference detection unit 40 calculates the difference (ΔRGB) between the RGB image data of the print image scanned by the image inspection unit 5 (scanner) and the RGB master data, and the corrected TRC calculation unit 41.

In step S124, the map generation unit 88 of the correction TRC calculation unit 41 extracts small areas that can be used as colorimetry areas for each CMYK color from the CMYK master image based on the conditions 1 to 8 described above. A map that summarizes all the measured colorimetric regions is formed. Further, in step S124, the colorimetry list generation unit 89 of the corrected TRC calculation unit 41 randomly selects a prescribed number of colorimetry regions from each segment based on the above map to form a "colorimetry list". do.

In step S125, the region extraction unit 81 of the correction TRC calculation unit 41 extracts the RGB difference data corresponding to the colorimetry region registered in the colorimetry list, and extracts the main scanning deviation along with the coordinate information and the CMYK gradation information. It is supplied to the correction processing section 82 . The main scanning deviation correction processing unit 82 calculates a main scanning deviation amount for the extracted colorimetry area based on the coordinate information and the CMYK gradation information. Further, the main scanning deviation correction processing unit 82 subtracts the calculated main scanning deviation amount from the RGB difference in the colorimetry area, thereby forming RGB difference data from which the influence of the main scanning deviation is removed.

Further, in step S125, the local θ calculation processing unit 83 calculates the above-described segment θ based on the coordinate information of the colorimetry area registered in the colorimetry list, the CMYK information, and the RGB difference information after the main scanning deviation correction processing. The local θ is calculated for each time and stored in the storage unit 84 .

In step S126, the color tone control unit 28 determines whether or not the processing of steps S121 to S125 has been completed for all images of one set, which is a predetermined number of images. If the processing of steps S121 to S125 has not been completed for all images of one set, the processing returns to step S121, and the processing of steps S121 to S125 is repeatedly executed. On the other hand, when the processing of steps S121 to S125 has been completed for all images of one set, the processing proceeds to step S127. When the processing of steps S121 to S125 for all images of one set is completed, the storage unit 84 stores local θ for each segment of all images of one set.

Next, when the processing of steps S121 to S125 for all images of one set is completed and the processing proceeds to step S127, the sub-scanning deviation correction processing unit 85 of the correction TRC calculation unit 41 performs the local θ calculation unit Among the local θ calculated in 83, a pair of local θ calculated from a region where the drum cycle of the photosensitive drum (for example, the photosensitive drum 50k) is half phase shifted is detected and averaged to obtain the final θ value ( Calculate the smoothing θ).

(Details of calculation method of leveling θ)
A method for calculating such a leveling θ will be described in detail. First, in FIG. 10(a), no in-plane deviation occurs in either the main operation direction, which is the moving direction of the ink head relative to the paper, or the sub-scanning direction, which is the transport direction of the paper. It shows an image in an ideal state in which colors are printed uniformly on paper.

However, in practice, due to factors such as variations in component accuracy of the image forming apparatus 8, in many cases, periodic in-plane deviations occur in the main scanning direction and the sub-scanning direction, respectively, as shown in FIG. 10(b). do. Such an in-plane deviation becomes noise when calculating the θ value and causes an error.

For example, the CMYK master image including the object 100 shown in FIG. 10C is printed over the image with in-plane deviation shown in FIG. 10B. In this case, as shown in FIG. 10(c), the CMYK master image including the object 100 does not show color variation as a whole, but the in-plane deviation shown in FIG. 10(b) occurs. 10(b), the influence of the in-plane deviation shown in FIG. 10(b) appears strongly in the area of the object 100 where the color is printed as shown in FIG. 10(d). For this reason, although the θ value should originally be "0", the θ value is affected by an error corresponding to the in-plane deviation. In addition, since the same image does not always appear on the next and subsequent pages, and the appearance of in-plane deviation varies depending on the page, even if tone correction is performed while being affected by in-plane deviation error, the correct scale will not be obtained. Tone correction becomes difficult.

For this reason, the image forming system according to the second embodiment evenly extracts an area where a deviation of a dark color appears and an area where a deviation of a light color appears, and cancels out the effects of both. It reduces the influence of in-plane deviation.

Specifically, in the case of the main scanning deviation, as shown in FIG. 10(e), the influence of the sub-scanning direction is small. Therefore, the influence of the deviation can be eliminated to some extent by calculating the amount of deviation appearing in the coordinates in the main scanning direction in advance and removing it from the RGB difference information of each colorimetry area. However, in order to obtain even higher gradation correction accuracy, it is desirable that the colorimetric regions are uniformly extracted in the main scanning direction.

In the case of the sub-scanning deviation, as shown in FIG. 10(f), the effect in the main scanning direction is small. It is difficult to calculate the amount of deviation in advance.

For this reason, the image forming system of the second embodiment regards segments that are shifted by half the phase of the drum cycle as a pair, and uses the average of the local θ calculated in those segments to obtain the deviation offset the effects of Note that since deviations of segments that are not selected as a pair remain, the local θ of the segments that are not selected are not used for calculating the leveling θ. Further, when it is difficult to detect a pair of segments due to the layout of the image, the sub-scanning deviation correction processing unit 85 does not calculate the average of local θ between pairs, but calculates the total average of local θ of valid segments. calculate. This makes it possible to reduce the influence of the overall sub-scanning deviation.

As an example, the sub-scanning deviation correction processing unit 85 detects a pair of segments in which deviations appear in opposite phases, such as segment A and segment B shown in FIG. 10(f). Further, the sub-scanning deviation correction processing unit 85 also detects segments C and D with weak deviations as a pair because they are positioned in opposite phases.

Although eight segments are shown in the example of FIG. 10(f), the number of segments per page is arbitrary. For example, one page may be divided into 16 segments as described above. Also, segments of other pages may be detected as pairs within the set.

(Appropriateness judgment operation)
Next, in step S128 of the flowchart of FIG. 9, the validity determination unit 91 of the validity determination processing unit 86 uses the value of the leveling θ calculated by the sub-scanning deviation correction processing unit 85 for tone correction processing. Validity determination processing is performed to determine whether or not the value is appropriate for . Further, in step S128, the θ correction unit 92 of the validity determination processing unit 86 corrects the value of the leveling θ based on the determination result of the validity determination unit 91. FIG.

Specifically, when judging validity, the amount of information required to calculate the θ value (the amount of information used to calculate the θ value) and the possibility of an error in calculating the θ value are considered. There is a need. In this example, the amount of information required to calculate the θ value (the amount of information used to calculate the θ value) and the possibility of errors in calculating the θ value are used to determine validity. We will, but the determination of validity may be based on either factor. In this case, it is possible to reduce the computational complexity of the validity determination computation and obtain the determination result at high speed.

The amount of information required to calculate the θ value is related to the number of colorimetric regions, the number of segment pairs, and the comprehensiveness of gradation colors. As the number of colorimetry regions increases, the amount of information that can be used increases, so the validity improves. The same applies to the number of pairs of segments, the more the number, the better the relevance. The numerical value "25 pairs" shown in Table 1 below is a threshold value for determining whether or not there are a sufficient number of pairs. If there are, the local θ is averaged for each pair to calculate the leveling θ, and if there are less than 25 pairs, the average for each pair is not calculated, but all the segments are averaged to calculate the leveling θ.

Regarding the comprehensiveness of gradation colors, if the gradation of the colorimetry area on the color plate is within the specified range, the information is insufficient for the gradation outside the specified range, and the color variation in that range is calculated correctly. may not be possible. Therefore, by obtaining information on a wide range of gradations, it is possible to improve the above-described validity determination accuracy.

The in-plane deviation, the processing time, and the like are factors that may cause a calculation error in the θ value. As described above, if the in-plane deviation is included, the accuracy of the θ value calculation decreases. Therefore, it is preferable to obtain a sufficient amount of information from a wide range of colorimetric regions to suppress the influence of in-plane deviation. The validity judging section 91 divides one page into three in the main scanning direction and detects the ratio of the number of colorimetry areas existing in each divided range, so that the colorimetry points are widely distributed. determine whether or not Further, if the sub-scanning deviation is widely distributed, it is expected that the number of pairs of segments will increase according to this distribution. Therefore, the validity determination unit 91 determines the sub-scanning deviation based on the number of pairs of segments.

As for the processing time, since the user image is printed in parallel with the calculation of the θ value, the colors printed by the printer engine 4 may change over time. Therefore, there is a high possibility that the calculated θ value does not follow the actual color change that fluctuates over time. Therefore, the validity determination unit 91 determines that the validity of the processing time is lost unless the printing process is completed within a certain period of time.

Hereinafter, each condition used for such validity determination processing will be described using Tables 1 to 5 shown as examples. Tables 1 to 5 are specific conditions for judging the validity of the leveling θ calculated for each set of images. Numerical values in each table are tabulated values for each color plate and each set. Also, this numerical value is an example, and may be arbitrarily changed according to the design or the like.

First, the validity determination unit 91 determines that the calculated value of the leveling θ is “highly valid” when the conditions shown in Table 1 below are met.

As shown in Table 1, when the colorimetry area is classified into three gradation areas of highlight, middle, and shadow by the color plate, there are 187 or more colorimetry areas in each gradation area (basic color distribution of gradation), and when the image is divided into three in the main scanning direction, the left side . Central. 1/6 or more of the total number of colorimetry areas are present in the three areas on the right, and 25 or more effective pairs of segments located in opposite phases of the sub-scanning deviation are established. If the condition that the printing process is completed within θ is satisfied, a sufficient amount of information can be acquired and the error is reduced. is determined to be "highly valid".

Under the conditions shown in Table 1, it is considered possible to sufficiently suppress errors in both the main scanning deviation and the sub scanning deviation. In addition, it is considered that the distribution is wide even when looking at each gradation. In Table 1, the condition for the number of colorimetry areas is left blank, but this is because some colorimetry areas are not used due to sub-scan segment pair processing. The reason for this is that there is no such color, and that the specific number is specified by the gradation color instead.

Next, the validity determination unit 91 determines that the calculated value of the leveling θ is “highly valid” when the conditions shown in Table 2 below are met.

As shown in Table 2, when the colorimetry area is classified into three gradation areas of highlight, middle, and shadow by the color plate, each gradation area has 1/6 or more of the total colorimetry area. exists, and when the image is divided into three in the main scanning direction, the left side . Central. In the three areas on the right, more than 1/6 of the total number of colorimetry areas exists, the total number of colorimetry areas is 560 or more, and the printing process is completed within the set time. When the condition is satisfied, the validity determination unit 91 determines that the calculated value of the leveling θ is “highly valid”.

Under the conditions of Table 2, although the number of sub-scanning segment pairs could not be obtained sufficiently, there is a sufficient colorimetric area, so the influence of the sub-scanning deviation can be covered. be. Under the conditions of Table 2, specific numerical values are set for determining the number of colorimetry areas, so the gradation color is determined by the ratio of the number of colorimetry areas.

Next, the validity determination unit 91 determines that the calculated value of the leveling θ is “medium validity” when the conditions shown in Table 3 below are met.

As shown in Table 3, when the colorimetry area is classified into three tone areas of highlight, middle, and shadow by the color plate, there are one or more tone areas with 187 or more colorimetry areas. When the image is divided into three in the main scanning direction, the left, center, and right three areas each have 1/8 or more of the total number of colorimetry areas, and the total number of colorimetry areas is 360 or more. However, if the printing process is completed within the set time and the conditions in Tables 1 and 2 are not satisfied, the validity determination unit 91 determines that the calculated value of the leveling θ is , is judged to be “moderately valid”.

The conditions in Table 3 indicate that there is a possibility that any of the main scanning, sub-scanning, and gradation colors cannot cover the error occurring in the leveling θ.

Next, the validity determination unit 91 determines that the calculated value of the leveling θ is "low validity" when the conditions shown in Table 4 below are met.

As shown in Table 4, the total number of colorimetry areas is one or more, the printing process is completed within the set time, and the conditions in Tables 1 to 3 above are not satisfied. Furthermore, the validity determination unit 91 determines that the calculated value of the leveling θ is “low validity”. That is, the conditions in Table 4 are conditions indicating that the minimum number of colorimetric regions exist. Under this condition, there is a high possibility that the error will be considerably large, so the validity determination unit 91 determines that the calculated value of the leveling θ is "low validity".

Next, the validity determination unit 91 determines that the calculated leveling θ is “invalid” when the conditions shown in Table 5 below are met.

Table 5 is a condition indicating that none of the conditions in Tables 1 to 4 is satisfied, and that there is no colorimetry area. In this case, the calculated leveling θ is discarded because there is no basis for judging the color variation.

Since the amount of information for calculating the θ value with high accuracy is often insufficient with only the colorimetry area obtained for each page, in the case of the image forming system of the second embodiment, several pages are grouped together. A θ value is calculated for each set and fed back to the gradation processing unit 31 . This makes it easier to secure the number of colorimetric regions and the number of sub-scanning segment pairs. In the above-described image forming system, the above-described validity determination processing is also performed for each set, so that determination can be performed in accordance with the θ value calculation mechanism of the image forming system of the second embodiment. .

The number of pages in one set is the number of pages that can be secured empirically enough to secure the number of colorimetry areas and the number of pairs of sub-scanning segments for general catalogs and other manuscripts (for example, about 8 sheets of A3 size) ) is set in advance. Alternatively, without setting the number of pages in one set in advance, the document is printed and scanned repeatedly, and the number of colorimetry areas and the number of pairs of sub-scanning segments that meet the conditions in Table 1 or Table 2 above are A dynamic determination method such as setting the number of pages at the time of acquisition as one set may be used.

(correction operation of θ value)
Next, in step S128 of the flowchart of FIG. 9, the θ correction unit 92 of the validity determination processing unit 86 corrects the value of the leveling θ based on the determination result of the validity determination unit 91, thereby reducing the error reduce the impact of Tables 6 to 9 below show thresholds and correction values for determining whether or not to correct the θ value for each of the conditions described in Tables 1 to 5 above. Although it is only an example, the magnitude relationships shown in Tables 6 to 9 are "numerical value A>numerical value B>numerical value C>numerical value D" and "coefficient F>coefficient G>coefficient H". Such numerical values and coefficients may be set based on how to obtain a variation model.

First, Table 6 is a table showing thresholds and correction values for judging whether or not to correct a leveling θ value whose validity is determined to be “high” based on the conditions in Tables 1 and 2 above. is. As shown in Table 6, the θ correction unit 92 notifies the occurrence of density abnormality via the notification unit, for example, when the leveling θ value for which the validity is determined to be “high” is equal to or greater than the numerical value D. Notification (audio message, text error message, electronic sound, etc.) is made, and the leveling θ value is corrected to "0". In this case, tone correction data (TRC) is not updated. Further, when the leveled θ value for which the validity is determined to be “high” is less than the numerical value D and equal to or greater than the numerical value C, the θ correction unit 92 multiplies the leveled θ value by the coefficient H, for example. to correct. Further, when the leveled θ value for which the validity is determined to be “high” is less than the numerical value C, the θ correction unit 92 does not correct the leveled θ value.

Next, Table 7 is a table showing thresholds and correction values for judging whether or not to correct a leveling θ value whose validity is determined to be “medium” based on the conditions in Table 3 above. . As shown in Table 7, the θ correction unit 92 notifies the occurrence of density abnormality via the notification unit, for example, when the leveling θ value for which the validity is determined to be “medium” is equal to or greater than the numerical value D. Notification (audio message, text error message, electronic sound, etc.) is made, and the leveling θ value is corrected to "0". In this case, tone correction data (TRC) is not updated. Further, when the leveled θ value whose validity is determined to be “medium” is less than the numerical value D and greater than or equal to the numerical value B, the θ correction unit 92 multiplies the leveled θ value by the coefficient G as an example. to correct. Further, when the leveled θ value whose validity is determined to be “medium” is less than the numerical value B, the θ correction unit 92 does not correct the leveled θ value.

Next, Table 8 is a table showing thresholds and correction values for judging whether or not to correct a leveled θ value whose validity is determined to be “low” based on the conditions in Table 4 above. . As shown in Table 8, the θ correction unit 92 notifies the occurrence of density abnormality via the notification unit, for example, when the leveling θ value whose validity is determined to be “low” is equal to or greater than the numerical value D. Notification (audio message, text error message, electronic sound, etc.) is made, and the leveling θ value is corrected to "0". In this case, tone correction data (TRC) is not updated. Further, when the leveled θ value whose validity is determined to be “low” is less than the numerical value D and equal to or greater than the numerical value A, the θ correction unit 92 multiplies the leveled θ value by the coefficient F as an example. to correct. Further, when the leveled θ value whose validity is determined to be “low” is less than the numerical value A, the θ correction unit 92 does not correct the leveled θ value.

Next, Table 9 is a table showing thresholds and correction values for judging whether or not to correct a leveling θ value whose validity is determined to be “invalid” based on the conditions in Table 5 above. . As shown in Table 9, the θ correction unit 92 corrects the leveled θ value to “0” when the validity is determined to be “invalid”. In this case, tone correction data (TRC) is not updated.

A large θ value of the leveling θ means that the color varies accordingly. Therefore, if the θ value is equal to or greater than the predetermined threshold value, the θ correction unit 92 determines that the influence of the error is likely to increase. The impact is judged to be weak. When the θ value is large due to an error, if the gradation correction data (TRC) is generated as it is, there is a high possibility of overcorrection. Therefore, the θ correction unit 92 corrects the θ value to a small value. .

On the other hand, when the value of θ is small, there is a low possibility of over-correction even if there is an error, and there is a high possibility that correction will lead to insufficient control, making it difficult to eliminate color variations. In this case, the θ correction unit 92 does not correct the θ value.

The θ value also increases when there is a large color variation, not due to the effect of error, and correction is applied to the θ value. Color variation can be gradually restored while preventing correction. Also, in the process control of the engine, in addition to the function of performing real-time gradation correction processing in the image forming system of the second embodiment, if another function for correcting color fluctuations is provided, this By performing double correction together with the function correction process, the inconvenience of overcorrection can be minimized. If the θ value is too large, there is a high possibility that the printer engine 4 is malfunctioning. (message, electronic sound, etc.), and processing for resolving the abnormality of the printer engine 4 is performed.

(Effect of correction of θ value)
FIG. 11 is a graph showing the relationship between density variation and the number of printed sets. The thick line graph shows the color variation when the θ value is not corrected by the validity judgment, and the thin line graph shows the color variation when the θ value correction is applied by the validity judgment. Also, the dotted line indicates the reference density. In the real-time gradation correction process, it is preferable to perform the gradation correction process so that the color variation is close to the reference density indicated by the dotted line.

When the influence of the error becomes large at a certain point, the change in density becomes large with respect to the reference density as shown in the thick line graph unless the θ value is corrected. If the change in density becomes large, even if it is restored in the next set, the error will become large, and there is a possibility that the correction will be overdone again.

On the other hand, when the influence of the error becomes large at a certain point, if the above-described correction processing is performed on the θ value, the change in density can be suppressed to a small level, as indicated by the thin line graph. Also, before the influence of the error on the .theta. value becomes large, the control is the same as when the .theta. value is not corrected, and there is no shortage of control.

(printing process)
Next, when the correction process corresponding to the validity of the leveled θ value is performed on the leveled θ value in this way, the process proceeds to step S129 in the flowchart of FIG. 9 . In step S<b>129 , the synthesizer 93 updates the gradation correction data (corrected TRC) to the value supplied from the θ corrector 92 . This gradation correction data (correction TRC) is transferred to the gradation processing section 31 by the transfer section of the correction TRC calculation section 41 in step S130. In step S<b>131 , the gradation processing unit 31 performs gradation correction processing based on the updated correction TRC on the CMYK master image acquired by the image processing unit 3 and supplies the result to the printer engine 4 . As a result, printing reflecting the corrected TRC can be executed.

Finally, in step S132, the printer engine 4 determines whether printing of all pages specified by the user has been completed. If printing of all pages has not been completed (step S132: No), the process returns to step S122, and the processes after step S122 are repeatedly executed. On the other hand, if printing of all pages is completed (step S132: Yes), the process of the flowchart of FIG. 9 ends.

(Effect of Second Embodiment)
As is clear from the above description, the image processing system according to the second embodiment can make up for the insufficient amount of information in a single image by making a set of a plurality of images. In addition, considering the amount of information in the set (the number of colorimetric regions for each density distribution and the amount of in-plane variation assumed to be given to the image), the validity of the set (the above-mentioned smoothed θ value validity). A threshold value or an adjustment magnification for adjusting the correction amount is determined according to the determination result.

Specifically, when the correction amount (leveled θ value) exceeds a predetermined threshold, it is assumed that there is a high possibility of the influence of error, and the correction amount (leveled θ value) is multiplied by the adjustment magnification. , the amount of correction can be suppressed, and the inconvenience of overcorrection can be prevented. For example, if a large amount of correction is calculated not because of an error but because of an actual sudden change in color, it is difficult to perform sufficient correction within that set. It is possible to perform the gradation correction process while gradually approaching the correct amount of correction while preventing this.

Therefore, it is possible to prevent the inconvenience of a sudden increase in the amount of correction, and to approach the correct amount of correction with a smaller number of sets than the method of determining the upper limit of the correction amount and the method of skipping the correction of that set. . Therefore, it is possible to improve the gradation correction accuracy.

Finally, each embodiment described above is presented as an example and is not intended to limit the scope of the present invention. For example, in the description of each of the above-described embodiments, the user image colorimetry RGB by the scanner 150 is used as the evaluation value to simplify the description. L: lightness, a: green/red, b: blue/yellow) or the like may be used. By using colorimetric values of a uniform color space such as Lab, more faithful control of color differences becomes possible. As an example, when Lab values are used, the RGB values in the above formulas 1 to 8 are replaced with Lab values, and the Jacobian matrix for RGB → Lab conversion is set to "J'", and the Jacobian matrix "J" is Replace with "J'J". Also in this case, the same effect as described above can be obtained.

In addition, in the description of the above-described embodiment, in order to simplify the description, the user image colorimetry RGB by the image inspection unit 5 is used as the evaluation value. (L: lightness, a: green/red, b: blue/yellow) or the like may be used. By using colorimetric values of a uniform color space such as Lab, more faithful control of color differences becomes possible. Also in this case, the same effect as described above can be obtained.

The novel embodiments described above can be embodied in various other forms, and various omissions, replacements, and modifications can be made without departing from the scope of the invention. In addition, the embodiments and modifications of the embodiments are included in the scope and gist of the invention, and are included in the scope of the invention described in the claims and equivalents thereof.

1 User's personal computer device (user PC)
2 network 3 image processing unit 4 printer engine 5 image inspection unit 6 output image 7 server device 8 image forming device 9 engine control unit 10 CMYK image 11 registration correction unit 12 scanned image 13 predicted image 14 selector 15 subtractor 16 gradation correction table 17 Gradation Conversion Section 18 Colorimetric Area Extraction Section 19 Address Correction Section 20 Sample Extraction Section 21 Colorimetry Prediction Section 22 Mode Parameter Calculation Section 23 Leveling Processing Section 24 Subtractor 25 TRC Synthesis Section 26 Integrator 28 Color Tone Control Section 30 Document Data 31 Gradation processing section 32 Main unit group 33 Repetition page number signal 34 Selection signal 35 Calculation condition selector 36 Image memory 37 Sample memory 38 First mode parameter memory 39 Second mode parameter memory 40 Difference detection section 41 Correction TRC calculation Unit 42 Reference TRC storage unit 43 Mode curve storage unit 44 Composite TRC generation unit 70 RGB conversion unit 71 Alignment unit 72 Alignment unit 73 Subtractor 81 Area extraction unit 82 Main scanning deviation correction processing unit 83 Local θ calculation unit 84 Storage unit 85 sub-scanning deviation correction processing unit 86 validity determination processing unit 87 transfer unit 88 map generation unit 89 colorimetry list generation unit 90 timer 91 validity determination unit 92 θ correction unit 93 synthesis unit 100 object 150 scanner

JP 2017-187585 A JP 2017-64979 A JP 2012-205124 A

Claims (17)

Among print media on which an output image corresponding to input image data is formed, the reflection characteristics of each output image formed on each print medium before and after each other are detected and formed on the print medium. a gradation control parameter calculator that calculates a gradation control parameter for controlling the gradation characteristics of the basic colors of the printed output image based on the reflection characteristics of the output image formed on the printing medium; Department and
A first gradation control parameter that is a gradation control parameter that is a reference value calculated by the gradation control parameter calculation unit, and a second gradation control parameter that is newly calculated by the gradation control parameter calculation unit. a gradation characteristic correction unit that corrects the gradation characteristic of the input image data based on the result of comparison with the
The gradation control parameters are composed of a mode curve that approximates the color variations of the measured values obtained by measuring the output image with a scanner, not only at the measurement points, but also at least in the entire range of gradations on a page-by-page basis. An image processor that is a coefficient. The gradation control parameter calculation unit calculates the first gradation control parameter and the 2. The image processing apparatus according to claim 1, wherein said second gradation control parameter is calculated. The gradation control parameter calculator,
When the same image is formed on each of the print media in sequence, the tone control parameter is the measured value of the reflectance properties of the initial image set and the measured value of the reflectance properties of the image sets other than the initial image set. The image processing apparatus according to claim 1, wherein the calculation is performed based on the difference. The image processing apparatus according to claim 1, wherein the gradation control parameter calculator calculates three or less gradation control parameters for each basic color. further comprising a prediction unit that forms a predicted value obtained by predicting reflection characteristics of the output image;
The gradation control parameter calculator,
When different images are sequentially formed on each of the print media, the predicted value is used as the reference value, and the measured value of the reflection characteristics of the initial image is compared with the predicted value to perform the first gradation control. calculating a parameter, comparing images subsequent to the initial image with the predicted value, and calculating the second gradation control parameter;
When the same image is formed on each of the print media in sequence, calculating the first tone control parameter from the measured reflectance properties of the initial image set, and based on the input image data other than the initial image set, 2. The image processing apparatus according to claim 1, wherein said second gradation control parameter is calculated. The gradation control parameter calculated by the gradation control parameter calculation unit based on a predetermined amount of information is a reasonable value for use in correcting the gradation characteristics of the input image data in the gradation characteristics correction unit. A determination unit that determines whether there is
a correction unit that corrects the value of the gradation control parameter based on the determination result of the determination unit;
The image processing apparatus according to claim 1, further comprising: 7. The image processing apparatus according to claim 6, wherein the determination section uses the number of colorimetric regions of the input image data as the predetermined amount of information to perform the determination. 7. The image processing apparatus according to claim 6, wherein the determining unit performs the determination using the number of colorimetric regions corresponding to a predetermined gradation of the input image data as the predetermined amount of information. 7. The image processing apparatus according to claim 6, wherein the determination unit performs the determination using a colorimetric area distribution of the output image. The determination unit performs the determination with a predetermined plurality of pages as one set,
7. The image processing apparatus according to claim 6, wherein the correction unit performs the correction with a plurality of predetermined pages as one set. 7. The image processing apparatus according to claim 6, wherein the determination unit performs the determination with a set of page numbers obtained when a predetermined number of colorimetric regions of the output image are obtained. having two levels of threshold values, a first threshold value for the value of the gradation control parameter and a second threshold value higher than the first threshold value;
When the value of the gradation control parameter is greater than or equal to the first threshold value and less than the second threshold value, the correction unit corrects the gradation control parameter using a correction value obtained by multiplying a predetermined coefficient. 7. The image processing apparatus according to claim 6, wherein the value is corrected. having two levels of threshold values, a first threshold value for the value of the gradation control parameter and a second threshold value higher than the first threshold value;
The image processing apparatus according to claim 6, wherein the correction unit does not correct the value of the gradation control parameter when the value of the gradation control parameter is less than the first threshold value. having two levels of threshold values, a first threshold value for the value of the gradation control parameter and a second threshold value higher than the first threshold value;
7. The image processing apparatus according to claim 6, wherein the correction unit corrects the value of the gradation control parameter to "0" when the value of the gradation control parameter is equal to or greater than the second threshold. . The gradation characteristic correction unit corrects the gradation characteristic of the input image data using a gradation correction table in which a table value for standardizing the density characteristic of the output image is set,
Initial values of the table values determined by calibration are set in the gradation correction table,
A case where the elements of the CMYK gradation values, which are the gradation values of the input image data, are tilde c, tilde m, tilde y, and tilde k, and each element is corrected by synthesizing predetermined first and second variation modes. , put it like the following formula 1,

Here, in the above equation (1), c, m, y, and k are CMYK gradation values before correction, c0, m0, y0, and k0 are reference gradation characteristics, M1 is the first variation mode, and M2 is the Second variation mode, "θij" is the gradation control parameter (i = {c, m, y, k}, j = {1, 2}) , and the gradation control parameter is a real scalar wherein the reference gradation characteristics and the variation modes M1 and M2 are independent real vectors of the same dimension defined on the input ranges of the c, m, y, and k gradation values;
The image processing apparatus according to claim 1, characterized by: a gradation processing unit that performs gradation correction of an input image;
a printing unit that prints the input image;
an image reading unit for reading a printed image of the input image printed by the printing unit;
An image processing system comprising: the image processing apparatus according to any one of claims 1 to 15. the computer,
Among print media on which an output image corresponding to input image data is formed, the reflection characteristics of each output image formed on each print medium before and after each other are detected and formed on the print medium. a gradation control parameter calculator that calculates a gradation control parameter for controlling the gradation characteristics of the basic colors of the printed output image based on the reflection characteristics of the output image formed on the printing medium; Department and
A first gradation control parameter that is a gradation control parameter that is a reference value calculated by the gradation control parameter calculation unit, and a second gradation control parameter that is newly calculated by the gradation control parameter calculation unit. functioning as a gradation characteristic correction unit that corrects the gradation characteristic of the input image data based on the result of comparison with the
The gradation control parameter is a mode curve that approximates the color variation obtained from the measurement values obtained by measuring the output image with a scanner, not only at the measurement point but also at least in the entire gradation range for each page. An image processing program that is a composite coefficient of .

Images