An intelligent optimization and color difference compensation process for printing parameters of an outer packaging carton

Abstract

The present application belongs to the technical field of computer vision and image data processing, and discloses an outer packaging carton printing parameter intelligent optimization and color difference compensation process, target printing image data and layout data are acquired, and printing result image data is collected after printing; pixel corresponding relationship is established by geometric registration under the layout coordinate reference, reference region set is determined and reference image pairs are extracted; imaging condition normalization and device-independent color representation conversion are performed on the reference image pairs, and regional color difference observation is obtained; global color transformation parameters and spatial color difference compensation field are obtained by solving under the constraints of total ink amount, step monotonicity, correction amplitude and spatial continuity, and compensated printing image data and operation level printing parameter correction amount files are output, and after printing, review printing result image data can be collected to calculate review color difference, when the review color difference exceeds the preset threshold, the printing parameter correction amount file is marked as invalid and blocked.

Description

Technical Field

[0001] This invention belongs to the field of computer vision and image data processing technology, specifically a process for intelligent optimization of printing parameters and color difference compensation for outer packaging paper boxes. Background Technology

[0002] Printing of outer packaging boxes typically includes various types of print objects such as text, solid blocks, logos, and machine-readable codes. Different objects have different sensitivities to color shift and registration errors. During production, it is required that the finished product color be consistent with the pre-press design, and that consistency be stable between batches. Therefore, the production site generally adopts a combination of pre-press color management and post-press inspection for control.

[0003] A common practice is to establish color management relationships based on equipment calibration data in the prepress stage, forming color characteristic files and equipment characteristic lookup tables that match the printing equipment. Color separation and output control are then performed in conjunction with process parameters such as ink volume, gray balance, dot gain, and tone curves. In some cases, cameras or scanning equipment are used to capture images of the printing results on-site. Geometric alignment and color reference relationships are established by using positioning marks in the color mark areas and reference color blocks. Based on this, color differences are calculated in several areas on the printing plate, and adjustment suggestions are made for ink volume, tone curves, or registration. Some processes involve repeated proofing or additional correction prints to verify the adjustment effects round by round before entering the stable production stage.

[0004] In actual printing operations, paper expansion and contraction, paper feed jitter, printing deformation, and angular distortion can cause local non-rigid displacement of the printing plate. Relying solely on global alignment often makes it difficult to obtain stable pixel correspondences, leading to drift of the sampling boundary of the object area and fluctuations in color difference statistics from batch to batch. At the same time, uneven acquisition lighting, white balance drift, and differences in imaging channels can introduce imaging deviations unrelated to printing, making it difficult to compare image-based color difference observations with prepress data under the same caliber. Existing solutions often disperse geometric alignment, white reference correction, and color calibration to be completed at different data sources or at different stages, making it difficult to form reusable regional observations and unified constraints under the same coordinate reference.

[0005] Furthermore, adjusting color deviations in spatial distribution using only global curves or single ink zones can easily lead to local overcompensation coexisting with undercompensation in other areas; while directly rewriting the color characteristic files and equipment characteristic lookup tables on the printing equipment side may affect the existing calibration system of other operations, increasing on-site changeover and traceability costs. Summary of the Invention

[0006] The purpose of this invention is to provide an intelligent optimization and color difference compensation process for printing parameters of outer packaging paper boxes, so as to solve the problems mentioned in the background art.

[0007] To achieve the above objectives, the present invention provides the following technical solution: an intelligent optimization and color difference compensation process for printing parameters of outer packaging paper boxes. This process takes the outer packaging paper box as the processing object, establishes a one-to-one correspondence between the target printed image and the printed result in the plate coordinates by unifying the plate coordinates between pre-press data and printing results, thereby converting printing deviations into calculable regional color difference measurements, and generating compensation data and parameter correction amounts that match the same printing operation.

[0008] Specifically, the present invention acquires target printed image data corresponding to the outer packaging paper box layout, and acquires layout data associated with the target printed image data; the layout data includes at least layout boundary information, layout positioning information, and outline information of layout content objects, which are used to establish a unified layout coordinate reference and determine the subsequent sampling area. After the printing operation corresponding to the target printed image data is completed, the printing result image data of the printing operation is acquired.

[0009] Subsequently, geometric registration is performed on the printed result image data and the target printed image data under the page coordinate reference, so that the two form a pixel correspondence under the same page coordinate. Based on the pixel correspondence and combined with the object contour information in the page data, a set of reference regions is determined, and reference image pairs corresponding one-to-one with the set of reference regions are extracted from the target printed image data and the printed result image data, so that the color difference calculation is limited to the object area related to the page printing quality.

[0010] After obtaining the reference image pair, the image pair is normalized to eliminate or correct the influence of non-printing factors caused by differences in illumination intensity distribution and channel response. After normalization, the reference image pair is converted to the device-independent color characterization space defined by the International Commission on Illumination, and the color difference is calculated within the reference region set to form a regional color difference observation. The regional color difference observation is used to characterize the color deviation of the printed result image data relative to the target printed image data in each reference region and serves as the input constraint for subsequent solutions.

[0011] Based on this, using the observed color difference measurements in different regions as constraints, a compensation solution is constructed and solved to obtain global color transformation parameters and a spatial color difference compensation field. The global color transformation parameters characterize the uniform color shift trend across the entire page, while the spatial color difference compensation field characterizes the non-uniform color shift related to the page's location. The compensation solution satisfies constraints on the upper limit of total ink volume, tonal monotonicity, the upper limit of printing parameter correction range, and spatial continuity to ensure that the solution results are consistent with the process constraints of the printing operation.

[0012] Finally, based on the global color transformation parameters and the spatial color difference compensation field, compensated printing image data is generated, and a printing parameter correction file is generated based on the same solution result. The compensated printing image data and the printing parameter correction file are used for the pre-press output and printing control of this printing operation, so that the deviation corresponding to the regional color difference observation can be specifically corrected and a reusable basis for operation parameter correction can be formed.

[0013] Furthermore, in the page layout data, the page layout cutting line information is used to determine the geometric boundary of the outer packaging paper box page. In order to make the target printing image data and the printing result image data geometrically registered and establish pixel correspondence under the same page layout coordinates, the present invention performs boundary normalization processing on the geometric boundary, determines the minimum bounding rectangle covering the geometric boundary based on the page layout cutting line, and establishes the page layout coordinate reference accordingly.

[0014] The page coordinate reference uses the preset corner point of the minimum bounding rectangle as the page coordinate origin, and makes the coordinate axis direction consistent with the direction of the two adjacent sides of the minimum bounding rectangle respectively; the preset corner point is agreed upon by the page data to ensure that the coordinate direction is consistent under different printing operations and different acquisition conditions.

[0015] By defining the unified coordinate reference of the page layout, any position within the page can be expressed in the same coordinate system. This provides a consistent coordinate reference for solving subsequent geometric registration, establishing pixel correspondence, and positioning the reference area set in the page layout coordinate system. It also reduces the registration error and area positioning deviation caused by coordinate inconsistencies introduced by changes in acquisition posture, differences in cropping size, or differences in page layout orientation.

[0016] Furthermore, after establishing a page coordinate reference and forming a pixel correspondence between the target printed image data and the printed result image data, this invention determines a set of reference areas based on prepress file data to limit the sampling range for color difference observation. Specifically, it acquires prepress file data corresponding to the target printed image data and parses it to obtain page object contour information. The page object contour information includes the boundary contour information and vector path information of the page content objects, which can determine the object area of ​​each page content object under the page coordinate reference.

[0017] Based on the outline information of the page objects, the page content objects are divided into text objects, solid objects, machine-readable coded objects, and identification objects according to object type, and the reference area set is formed by the object areas corresponding to each object type; under the pixel correspondence, reference image pairs corresponding to the object areas are extracted from the target printing image data and the printing result image data, and color difference is calculated in each object area to form a regional color difference observation.

[0018] To ensure that the color difference observations in different regions reflect the quality constraint priorities of different object types in the subsequent compensation solution, this invention sets weight coefficients for different object types and performs a weighted summary of the color difference observations in different regions corresponding to each object area. The weight coefficients are pre-set non-negative real numbers, and their setting is based on the process requirements of color consistency and readability reliability of text objects, machine-readable coded objects, identification objects, and solid objects in the printing operation. The weighted summary results serve as input constraints for the compensation solution, enabling the compensation solution to apply differentiated constraints to the color difference deviations of different object types while meeting the process constraints.

[0019] Furthermore, in order to ensure that the printed image data has usable positioning and color references under the same printing operation and acquisition conditions, the present invention sets a color mark area in the non-finished product retention area of ​​the page, and simultaneously acquires the image data of the color mark area when acquiring the printed image data. The non-finished product retention area of ​​the page is not used for the presentation of the finished product pattern, and is printed and acquired together with the page content, so that the color mark area provides acquisition condition reference consistent with the page content without affecting the finished product pattern.

[0020] The color mark area includes non-collinear positioning marks, a white reference block, and a preset color reference block. The non-collinear positioning marks are used to provide positioning feature points under the plate coordinate reference to establish the correspondence between the control points of the printed result image data and the target printed image data and support geometric registration. The white reference block is used to determine the brightness correction parameters and channel white balance correction parameters to perform imaging condition normalization on the printed result image data. The preset color reference block is used to provide known color reference samples to calibrate the color conversion relationship from the color representation of the acquired image to the device-independent color representation space and to provide comparable color representation input for subsequent color difference calculation.

[0021] Furthermore, after establishing the page coordinate reference, the present invention performs geometric registration on the printed result image data and the target printed image data to establish a pixel correspondence between the two under the page coordinate reference. The geometric registration first establishes a correspondence between positioning points using non-collinear positioning marks in the color mark area, solves the global homography transformation based on the positioning point correspondence, and performs perspective correction on the printed result image data so that the perspective-corrected printed result image data achieves global geometric consistency with the target printed image data under the page coordinate reference.

[0022] Based on perspective correction, this invention selects control points in the perspective-corrected printing result image data and corresponding control points in the target printing image data. These control points are corner feature points and edge intersection feature points that can be repeatedly located in both images. The corresponding control points establish a correspondence through local feature matching. Based on this correspondence, a two-dimensional deformation field is calculated using the grid node displacement as a parameter under the page coordinate reference. Deformation compensation is then performed on the perspective-corrected printing result image data according to this two-dimensional deformation field, ensuring that the mapping relationship of the reference region set in the two images remains consistent with the pixel correspondence, thus supporting the subsequent extraction of reference image pairs and color difference calculation.

[0023] Furthermore, the printed result image data includes a color mark region. Before performing device-independent color characterization conversion and calculating color difference on the printed result image data, the present invention first performs imaging condition normalization processing to ensure that the input data used for color difference observation has a consistent brightness and channel response benchmark under the same acquisition conditions. The imaging condition normalization uses the white reference block in the color mark region as the correction benchmark. The white reference block and the page content form pixel data in the same acquisition process to determine the correction parameters for the printed result image data.

[0024] Specifically, based on the deviation between the pixel value of the white reference block in the printed result image data and the preset reference brightness, a brightness gain field distributed within the page coordinate range is established, and the brightness gain field is applied to the pixel brightness distribution of the printed result image data to perform brightness gain correction and brightness unevenness correction on the printed result image data; after completing the brightness correction, the channel correction coefficient is determined based on the pixel value of the white reference block in each color channel, and channel white balance correction is performed on the printed result image data to correct the color bias introduced by the channel response difference, so that the subsequent color difference calculation is based on the normalized color characterization data.

[0025] Furthermore, when solving for the global color transformation parameters and spatial color difference compensation field using the regional color difference observation as the observation input, the present invention sets the solution objective and constraints so that the obtained global color transformation parameters and spatial color difference compensation field can be used to generate compensated printed image data and to determine the printing parameter correction amount corresponding to the printing operation.

[0026] The solution objectives include a weighted summary term of color difference observations in different regions, a spatial continuity constraint term of the spatial color difference compensation field, and a printing parameter correction range constraint term. The weighted summary term of color difference observations in different regions is used to characterize the magnitude of color difference deviation within the reference region set. The spatial continuity constraint term is used to constrain the change in compensation amount of the spatial color difference compensation field at adjacent positions under the page coordinate reference, so that the compensation difference at adjacent positions remains under control. The printing parameter correction range constraint term is used to constrain the deviation of the printing parameter correction amount from the initial printing parameter set of the printing operation, which is given by the initial setting parameters of the printing operation.

[0027] Meanwhile, the solution process introduces process constraints related to printing output, including the upper limit constraint of total ink volume and the level monotonicity constraint. The upper limit constraint of total ink volume limits the sum of ink volume corresponding to each color separation channel to not exceed the upper limit determined by the process specification; the level monotonicity constraint limits the level relationship represented by the global color transformation parameters to remain monotonically unchanged with the input gray level, so that the generated compensation data and the printing parameter correction amount meet the level constraints of the printing output process.

[0028] Furthermore, the present invention represents the spatial color difference compensation field in a gridded manner under the page coordinate reference, and represents the spatial color difference compensation field as a set of two-dimensional grid nodes arranged within the page area. Each grid node stores the compensation vector at the corresponding position in the device-independent color representation space, which is used to represent the color compensation amount at that position.

[0029] When generating the compensated printed image data, for any pixel position under the plate coordinate reference, based on the spatial position relationship of the pixel position relative to the two-dimensional grid node, the compensation vector of the pixel position is calculated from the compensation vector of the neighboring grid node through bilinear interpolation, and the compensation vector is used to compensate the color representation of the target printed image data, so that the compensation amount can be continuously changed within the plate range and implemented at the pixel level.

[0030] Meanwhile, in order to ensure that the spatial color difference compensation field meets the spatial continuity constraint in the compensation solution, this invention sets an upper limit on the magnitude of the difference in compensation vectors between adjacent grid nodes to limit the excessive change in compensation amount at adjacent positions; the setting serves as a spatial constraint condition in the solution and participates in the compensation solution together with the observation input corresponding to the regional color difference observation.

[0031] Furthermore, after obtaining the global color transformation parameters and spatial color difference compensation field, and completing geometric registration to obtain the residual displacement, this invention outputs printing parameter correction data corresponding to the printing operation for pre-press output and printing control. The printing parameter correction data includes ink volume parameter correction, gray balance parameter correction, dot gain compensation parameter correction, tone curve parameter correction, and registration compensation parameter correction. Among them: the tone curve parameter correction and gray balance parameter correction are determined by the global color transformation parameters; the dot gain compensation parameter correction is derived from the tone curve parameter correction; the ink volume parameter correction is obtained by weighting the compensation vectors of the spatial color difference compensation field in the object area corresponding to the reference area set according to the object type weight coefficient. The object type weight coefficient is consistent with the weighting system used in the weighted summation of the color difference observations in different areas; the registration compensation parameter correction is determined by the residual displacement of the non-collinear positioning mark after geometric registration.

[0032] The printing parameter correction data is used as the output data of the printing job and is not rewritten in the process of application. This limits the parameter adjustment to the parameter level corresponding to the printing job and maintains the consistency of the calibration data of the equipment side in other printing jobs.

[0033] Furthermore, this invention uses the data collected from the same printing operation as the solution basis. The global color transformation parameters and spatial color difference compensation field are obtained from the printing result image data corresponding to the printing operation, and the compensated printing image data and printing parameter correction data are formed accordingly. The two together constitute the operation output data of the printing operation. In the process of forming the operation output data, the correction sample data obtained from reprinting is not introduced, so that the parameter solution and output are limited to the range of the data collected in the printing operation.

[0034] After printing is completed using the job output data, the image data of the printing result is obtained for verification. The verification color difference is calculated within the reference area set and in the same device-independent color representation space as the aforementioned color difference calculation. The verification color difference is compared with a preset threshold, which is determined by the printing quality specification or job acceptance standard. When the verification color difference exceeds the preset threshold, the printing parameter correction data is not provided as the output of subsequent printing jobs, and the corresponding printing parameter correction data is marked as invalid.

[0035] The beneficial effects of this invention are as follows:

[0036] 1. This invention establishes a page coordinate reference based on page data, performs geometric registration of the target printing image data and the printing result image data under the same coordinates, and provides stable control points with the help of non-collinear positioning marks in the color mark area; at the same time, it uses white reference blocks to perform brightness and white balance correction on the acquired image, and then converts it to the device-independent color representation space, so that the object area within the reference area set can form repeatable color difference observation under consistent sampling boundaries and imaging references, reducing the interference of paper deformation, registration residue and uneven lighting on color difference statistics.

[0037] 2. This invention introduces the regional color difference observation into a unified compensation solution framework, jointly solves the global color transformation parameters and spatial color difference compensation field, applies monotonicity constraints to the tone curve, and applies upper limits to the correction amounts of ink volume, gray balance, dot gain compensation, tone curve and registration compensation. At the same time, it uses hard constraints of the compensation difference between adjacent grid nodes to suppress spatial abrupt changes, so that the compensation can cover the color deviation of the page space without exceeding the execution range of the equipment and the process window, reducing the coexistence of local overcompensation and undercompensation.

[0038] 3. This invention outputs the solution results as a job-level printing parameter correction file. The file header is bound to the job identifier and establishes a reference relationship between the initial printing parameter set and the equipment-independent color representation space. The parameter correction section provides incremental expressions for ink volume, gray balance, dot gain compensation, tone curve, and registration compensation, which facilitates direct distribution and traceability. At the same time, this file does not write or modify the color characteristic file of the International Color Consortium or the equipment characteristic lookup table on the printing equipment side, thus avoiding disturbance to the calibration system of other jobs from the source. Furthermore, the release gate control and invalidation marking mechanism, which verify the color difference threshold, prevent abnormal corrections from entering the formal distribution process, reducing the risk of on-site changeover and the material consumption caused by repeated printing. Attached Figure Description

[0039] Figure 1 This is a flowchart of the core process of the present invention;

[0040] Figure 2 This is a flowchart of the geometric registration and imaging normalization process of the present invention;

[0041] Figure 3 This is a flowchart of the compensation solution and parameter generation process of the present invention. Detailed Implementation

[0042] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0043] like Figures 1 to 3 As shown, this embodiment of the invention provides an intelligent optimization and color difference compensation process for printing parameters of outer packaging paper boxes. This process takes a single printing job as the processing object, aligns the target printing image data before printing with the printing result image data after printing in the same page coordinate system, forms a calculable color difference observation in the area corresponding to the content object on the page, and solves for the global color transformation parameters and spatial color difference compensation field under the condition of satisfying the printing process constraints, and then outputs the compensated printing image data and printing parameter correction amount file.

[0044] In this embodiment, firstly, target printed image data corresponding to the outer packaging paper box is acquired, and then the page data associated with the target printed image data is acquired. The page data is used to describe the spatial positional relationship between the page geometric boundary and the page content objects, including at least page cutting line information and page object outline information, so as to establish page coordinate reference and provide a basis for the positioning of the subsequent reference area set. The page coordinate reference can be jointly defined by the page boundary determined by the page cutting line, the page coordinate origin, and the coordinate axis direction, so that any position within the page can be expressed under a unified coordinate system.

[0045] After completing the printing operation corresponding to the target printed image data, the printing result image data of the printing operation is obtained. The printing result image data corresponds to the same page content as the target printed image data. It is used to characterize the color presentation of the printing operation and the geometric deviation caused by factors such as paper feed offset and paper expansion and contraction. In order to improve the stability of subsequent registration and normalization, the printing result image data may include color mark area image data located in the non-finished product retention area of ​​the page. The color mark area is set with positioning marks for registration and white reference blocks for imaging correction. A preset color reference block can also be set for calibration of color characterization conversion.

[0046] Subsequently, geometric registration is performed on the printed result image data and the target printed image data under the page coordinate reference to establish the pixel correspondence between the two. The geometric registration is performed in the order of global first and local second: first, the correspondence of control points is established based on the positioning marks in the color mark area, the transformation relationship used for overall alignment is solved, and perspective correction is performed on the printed result image data to align the overall shape of the page with the page coordinate reference; after perspective correction, in order to handle non-rigid deformation caused by paper expansion and contraction, repeatable control points are extracted from the perspective-corrected printed result image data and the target printed image data, and the correspondence of control points is established. The control points can be corner feature points or edge intersection feature points. Based on the two sets of control point correspondences, the two-dimensional deformation field with grid node displacement as the parameter is solved, and deformation compensation is performed on the perspective-corrected printed result image data. The pixel correspondence formed by the above processing is used to map the pixel position in the printed result image data to the corresponding position in the target printed image data, so that subsequent area sampling and comparison can be performed within the same page object area.

[0047] After obtaining the pixel correspondence, a set of reference regions is determined based on the outline information of the page objects. The set of reference regions consists of multiple object regions, each of which corresponds to a specific content object in the page, and is used to limit the sampling range for color difference observation. Subsequently, reference image pairs that correspond one-to-one with the set of reference regions are extracted from the target printing image data and the printing result image data, so that target image fragments and result image fragments are formed in the same object region respectively.

[0048] After obtaining the reference image pair, imaging condition normalization and device-independent color representation transformation are performed on the reference image pair. Imaging condition normalization uses a white reference block as the correction benchmark: a spatial correction amount for brightness correction is established based on the pixel value of the white reference block in the printed result image data, and brightness correction and brightness unevenness correction are performed on the printed result image data; at the same time, channel correction coefficients are determined based on the pixel value of the white reference block in each color channel, and channel white balance correction is performed to reduce the impact of differences in acquisition illumination and channel response on color comparison. Device-independent color representation transformation is used to transform the normalized color representation to a device-independent color representation space composed of luminance and chromaticity components; when setting a preset color reference block, a color conversion relationship can be established between the known color reference of the preset color reference block and its observed value in the image. After completing the above processing, the color difference is calculated in the reference area set to form a regional color difference observation. The regional color difference observation is used to characterize the color deviation of the printing result in each object area relative to the target printed image.

[0049] After obtaining the regional color difference observations, the regional color difference observations are used as the observation input to construct and solve a compensation problem, resulting in global color transformation parameters and a spatial color difference compensation field. The global color transformation parameters are used to characterize the uniform color shift and tone deviation across the entire page, while the spatial color difference compensation field is used to characterize the non-uniform color shift that varies with the page position. To ensure that the solution results meet the process boundaries of printing output, constraints consistent with the printing job configuration are introduced during the solution process: the total ink volume upper limit constraint is used to limit the sum of ink volumes in each color separation channel from the upper limit determined by the process specifications or job configuration.

[0050] The tonal monotonicity constraint is used to limit the tonal relationship characterized by global color transformation parameters to remain monotonically unchanged with the input gray level; the upper limit constraint of printing parameter correction range is used to limit the correction range relative to the initial set of printing parameters for this printing job to not exceed the range determined by the job configuration; the spatial continuity constraint is used to limit the compensation changes of the spatial color difference compensation field at adjacent positions under the page coordinate reference to keep them under control, so as to avoid abrupt changes in compensation in space.

[0051] Finally, based on the global color transformation parameters and the spatial color difference compensation field, the compensated printing image data is generated, and the printing parameter correction file is generated based on the same solution result. The compensated printing image data is used for image layer compensation in the pre-press output, and the printing parameter correction file is used for parameter layer correction of the printing control parameters. The compensated printing image data and the printing parameter correction file are used as the job-level output data of this printing job for compensation and parameter adjustment in this printing job.

[0052] In this embodiment, the layout data is obtained by parsing the prepress file, which is a data file used for layout production and output. It contains a two-dimensional plane coordinate definition, which includes the horizontal axis and the vertical axis, and measures the position in units of length. The layout data includes at least layout cutting line information and layout object outline information. The layout cutting line information is used to characterize the cutting boundary of the finished outer packaging paper box, and the layout object outline information is used to characterize the area range of layout content objects such as text objects, solid objects, machine-readable coded objects, and identification objects.

[0053] In the prepress file, the page trimming line is given in the form of a closed path. The closed path is composed of line segments and curve segments. In order for the trimming line to participate in subsequent geometric calculations, the closed path is discretized into a trimming line point set: for line segments, the endpoints are taken and equally spaced points are inserted as needed; for curve segments, sampling points are obtained by sampling according to the sampling interval, and the sampling points are incorporated into the trimming line point set. The sampling interval is determined by the job configuration and is not greater than the sampling step size corresponding to the minimum feature size of the page in the prepress file, so as to ensure that the discrete point set can represent the geometric boundary of the trimming line.

[0054] The circumscribed rectangle boundary is established based on the set of trimming line points. The circumscribed rectangle boundary is a rectangle parallel to the horizontal and vertical axes in the two-dimensional plane coordinates of the prepress document, and covers all points of the trimming line point set. Specifically, the minimum horizontal coordinate, maximum horizontal coordinate, minimum vertical coordinate, and maximum vertical coordinate of the trimming line point set are calculated and used as the left boundary, right boundary, lower boundary, and upper boundary of the circumscribed rectangle, respectively. The circumscribed rectangle boundary is determined by the rectangle enclosed by the four boundary lines.

[0055] After obtaining the boundary of the circumscribed rectangle, a layout coordinate reference is established. The origin of the layout coordinates is set at a specific corner point of the circumscribed rectangle. In this embodiment, a corner point that simultaneously satisfies the condition that the x-coordinate is equal to the minimum x-coordinate and the y-coordinate is equal to the minimum y-coordinate is selected as the origin of the layout coordinates. Starting from this corner point, the direction pointing to the right from the lower boundary of the circumscribed rectangle is set as the positive direction of the first coordinate axis, and the direction pointing upward from the left boundary of the circumscribed rectangle is set as the positive direction of the second coordinate axis. Thus, the layout coordinate reference is jointly determined by the boundary of the circumscribed rectangle, the origin of the layout coordinates, and the directions of the two coordinate axes.

[0056] After establishing the page coordinate reference, the outline information of the page object is expressed in the same two-dimensional plane coordinate and can be directly used for the positioning of the reference area set. Furthermore, when performing geometric registration on the printed result image data and the target printed image data, the page coordinate reference is used as a unified geometric reference, so that the geometric registration output can be given in the form of two-dimensional transformation relationship parameters. The two-dimensional transformation relationship parameters are used at least to convert the pixel coordinates of the printed result image data into page coordinates, and when necessary, the inverse transformation is obtained from the two-dimensional transformation relationship parameters to complete the conversion from page coordinates to pixel coordinates. Based on this coordinate transformation relationship, the pixel correspondence between the printed result image data and the target printed image data is constructed, so that the pixel set falling into the same object area can establish a correspondence between the two images.

[0057] In the expression and application of spatial color difference compensation field, the layout coordinates are also used as the position index. The two-dimensional grid nodes of the spatial color difference compensation field are arranged within the bounded rectangle. The number of grid rows and columns is determined by the job configuration. For any pixel position, the pixel position is first converted into layout coordinates according to the two-dimensional transformation relationship parameters. Then, the grid cell where it is located is determined according to the layout coordinates. The compensation vector of the pixel position is calculated by bilinear interpolation from the compensation vectors of the four corner nodes of the grid cell. This is used for subsequent pixel-level color compensation calculations. By uniformly using the layout coordinates as the index for the object area, pixel correspondence and compensation field grid, the cumulative deviation introduced by the conversion between multiple coordinate systems can be reduced.

[0058] In this embodiment, the reference region set is determined by the page object outline information obtained from the parsing of prepress file data, so that the color difference observation corresponds to the page content object and is consistent with the observation input for subsequent compensation solution. The prepress file data is the page file data used in the prepress plate making and output process, which contains geometric description information and object attribute information of the page object. The object attribute information includes at least one or more of the following: object type identifier, layer identifier, font resource identifier, fill attribute, and stroke attribute. By parsing the prepress file data, the object outline information of each page object under the page coordinate reference is obtained. The object outline information is represented by a closed outline or a set of closed paths, and an object region mask is generated accordingly. The spatial resolution of the object region mask is consistent with that of the target printed image data. The generation method is to rasterize the closed outline onto the pixel grid corresponding to the target printed image data under the page coordinate reference, so that the effective pixel set in the mask corresponds to the object region.

[0059] After obtaining the object outline information, the page objects are divided according to object type. The object type division is based on the object attributes in the prepress file data, and a certain priority rule is adopted when type conflicts occur. Specifically, objects that meet the font resource identifier or text drawing instructions are determined as text objects; objects that meet the coded object identifier or are located in the coded layer are determined as machine-readable coded objects; objects that meet the identifier object identifier or are located in the identifier layer are determined as identifier objects; and other objects represented by fill paths and whose fill attributes are consistent within the object area are determined as solid objects. If the same object meets multiple determination conditions at the same time, it is first determined as a machine-readable coded object, then as a text object, then as an identifier object, and finally as a solid object. Thus, the page object outline information is divided into object outline information corresponding to text objects, object outline information corresponding to solid objects, object outline information corresponding to machine-readable coded objects, and object outline information corresponding to identifier objects.

[0060] The reference region set consists of object regions corresponding to each object type. To reduce the impact of geometric registration residual errors on the sampling of object edge pixels, an effective reference region is generated for each object region. The effective reference region is obtained by shrinking the object region mask inward. The shrinkage margin is given by the job configuration and converted into the number of pixels according to the spatial resolution of the target printing image data. An equidistant shrinkage process is performed on the object region mask to form the effective reference region mask. The area of ​​the object region is determined by the number of effective pixels in the object region mask. When the area of ​​the object region is smaller than the area threshold given by the job configuration, the object region is not included in the reference region set to avoid instability in the regional color difference statistics caused by small-area objects. In the case of overlapping between different object regions, overlapping pixels are assigned according to the above object type priority rule so that overlapping pixels are assigned to only one type of effective reference region, and the regions corresponding to machine-readable coded objects and text objects are kept intact.

[0061] After establishing pixel correspondence through geometric registration, reference image pairs corresponding one-to-one with the reference region set are extracted from the target printing image data and the printing result image data based on the effective reference region mask. The pixel correspondence is used to map the pixel coordinates of the printing result image data to the corresponding positions under the plate coordinate reference, and further establishes a consistent sampling range within the same object area with the target printing image data. Subsequently, imaging condition normalization and device-independent color representation transformation are performed on the reference image pairs, and the regional color difference observation is calculated in the device-independent color representation space. The regional color difference observation is calculated separately for each object area: for each object area, the regional statistical values ​​of the pixel color representation of the target image segment and the result image segment within the effective reference area are taken, and the regional statistical values ​​adopt the mean vector of the pixel color representation within the area; then, based on the two sets of mean vectors, the regional color difference observation of the object area is calculated according to the color difference metric specified in the job configuration.

[0062] To ensure differentiated observation weights for different object types in the compensation solution, this embodiment performs weighted aggregation of regional color difference observations by object type. The job configuration pre-sets a weight coefficient for each object type, with the weight coefficient being a non-negative real number. For multiple object regions belonging to the same object type, the regional color difference observations are first aggregated by the same type, using a weighted average of the region area to reflect the coverage of that object type on the page. Then, the aggregated result is multiplied by the weight coefficient corresponding to that object type. Finally, the weighted results of different object types are summed to obtain the weighted color difference observations used for the compensation solution, ensuring that the observation input is consistent with the compensation solution at the object semantic level.

[0063] In this embodiment, the printed result image data includes image data of the color mark area located within the non-finished product retention area on the printing plate. The non-finished product retention area is the edge material reserved area outside the plate cutting line, or located within the reserved edge that is removed after the finished product is cut. Its plate position is defined by the plate data. The plate data records the plate cutting line and the plate coordinate range of the color mark area. The color mark area is set outside the cutting line and maintains a safe margin with the finished product content area. The safe margin is determined by the margin parameter in the plate data to ensure that the color mark area does not enter the finished product area after the finished product is cut, and at the same time, the color mark area remains stably visible within the acquisition field of view.

[0064] The color mark area has a fixed layout in the page data and includes non-collinear positioning marks, white reference blocks and preset color reference blocks. The theoretical position and size of the above elements under the page coordinate reference are given by the page data. The theoretical position includes at least the theoretical center coordinates of each positioning mark and the boundary range information of each reference block, so as to determine its theoretical position in the target printing image data and to detect and sample it in the printing result image data.

[0065] The non-collinear positioning markers include at least three non-collinear positioning marker units. Each positioning marker unit adopts a closed contour structure with obvious contrast and contains a center positioning feature with a definite geometric center. After acquiring the printed result image data, the positioning marker units are detected in the color mark area. The detection process includes binarization segmentation, connected component filtering, and geometric fitting. The control points are taken as the coordinates of the geometric center point of each positioning marker unit. At the same time, the theoretical center coordinate set of each positioning marker unit under the page coordinate reference is read from the page data. Based on the relative geometric layout of the positioning marker units in the color mark area and the layout order given in the page data, a one-to-one correspondence between the observation center point and the theoretical center point is established to form a control point correspondence. Based on the control point correspondence, the pixel coordinates of the printed result image data are converted into two-dimensional transformation relationship parameters of the page coordinate reference, and a pixel correspondence is established accordingly. When it is necessary to reverse map the page coordinates to the pixel coordinates of the printed result image data, the inverse transformation is obtained from the two-dimensional transformation relationship parameters for coordinate inverse calculation.

[0066] The white reference block is positioned at a fixed location within the color mark area, its boundary defined by the layout data. The white reference block provides a benchmark for brightness and white balance correction. To ensure feasible sampling for brightness unevenness correction, in one implementation, the white reference block extends along the length of the color mark area to form a strip, or it consists of multiple white sub-blocks distributed along the length. During imaging condition normalization, the white reference block area is first located in the printed image data, and boundary culling is performed on this area to obtain the internal effective sampling area. This reduces the impact of registration residual error and edge ink diffusion on sampling. Based on the internal effective sampling area, the average pixel value of each color channel is calculated separately. A correspondence is established with the white reference target value recorded in the job configuration to obtain the channel gain coefficient of each color channel. The channel gain coefficient is used to correct the white balance so that the white reference block area meets the white reference target value after correction. For brightness unevenness correction, multiple sampling positions are selected along the direction of the white reference strip, the brightness deviation of each sampling position is calculated, and a low-order fitting is performed on the change of brightness deviation with position to obtain the distribution of brightness correction amount in the direction of the color mark area. The brightness correction amount is extended to the content area of ​​the page according to the positional relationship of the page coordinate reference to realize brightness correction and brightness unevenness correction, so that the image entering the color representation conversion and color difference calculation has a consistent imaging reference.

[0067] The preset color reference block consists of multiple color blocks. The position and boundary range of each color block are given by the layout data. The layout data or the work standard library stores the target color reference value corresponding to each color block. The target color reference value is defined in a device-independent color representation space. The device-independent color representation space adopts a three-dimensional color coordinate system composed of a luminance component and two chromaticity components. It is used to compare colors under different acquisition and device conditions. After normalizing the imaging conditions, the regions of each preset color reference block are located in the printed image data. Boundary culling is performed on each color block to obtain the internal effective sampling area, and the color channels of each block are calculated. The average pixel value is used to form the observed color vector of the color block. The observed color vector is paired with the corresponding target color reference value to construct a color corresponding sample set. The parameters of the device-independent color representation conversion model are calibrated by least squares solution. In one implementation, the conversion model is a polynomial mapping model from normalized color channel values ​​to device-independent color coordinates. After calibration, the conversion model is used to perform device-independent color representation conversion on the reference image pairs in the reference area set, and the regional color difference observation is calculated in the device-independent color representation space so that the color difference observation and the subsequent compensation solution adopt the same color representation caliber.

[0068] In the image data of the printing result of the same printing job, the control points required for geometric registration, the reference benchmarks required for imaging condition normalization, and the calibration basis required for device-independent color characterization conversion are obtained simultaneously through the color mark area, so that the establishment of pixel correspondence, reference area sampling and color difference observation calculation are based on the same data source and are completed under the same page coordinate benchmark.

[0069] Establishing two-stage geometric registration and pixel correspondence

[0070] In this embodiment, in order to establish the pixel correspondence between the printed result image data and the target printed image data under the page coordinate reference, the geometric registration adopts a two-stage process. The first stage uses the non-collinear positioning marks in the color mark area to solve the global homography transformation and perform perspective correction on the printed result image data to align its overall geometric shape with the page coordinate reference. The second stage solves the two-dimensional deformation field represented by the displacement of the grid control points based on the perspective correction and performs deformation compensation on the perspective-corrected printed result image data to eliminate local non-rigid deformation caused by paper expansion and contraction, paper feed offset, etc.

[0071] Global homography transformation and perspective correction based on non-collinear positioning markers

[0072] At least three non-collinear positioning mark units are set within the color mark area. In one implementation, four positioning mark units are set. The theoretical center coordinates of each positioning mark unit under the page coordinate reference are recorded in the page layout data. After obtaining the printing result image data, each positioning mark unit is detected within the color mark area and its observation center coordinates are extracted. The observation center coordinates are used as control points. A one-to-one correspondence between the observation center coordinates and the theoretical center coordinates is established based on the layout relationship of the positioning mark units to form a control point correspondence.

[0073] The global homography transformation parameters are solved based on control point correspondences. These parameters characterize the projective transformation on a two-dimensional plane and can be obtained through linear least squares. To reduce the impact of detection errors or abnormal correspondences on the solution results, the reprojection error of the control points before and after the transformation is used as a consistency measure. The control point correspondences are then screened for consistency, and abnormal correspondences with reprojection errors exceeding a threshold are removed. The threshold is given by the job configuration or determined by empirical parameters from similar printing jobs.

[0074] After obtaining the global homography transformation parameters, perspective correction is performed on the printed image data. The output image of the perspective correction uses the pixel grid corresponding to the page coordinate reference as the resampling grid, so that the perspective-corrected printed image data and the target printed image data are in the same page coordinate frame. Perspective correction adopts inverse mapping: for each pixel position in the output image, the sampling position in the original printed image data is calculated using the inverse transformation of the global homography transformation, and the pixel value at the sampling position is obtained by bilinear interpolation. When the sampling position exceeds the boundary of the original printed image data, the corresponding pixel is marked as an invalid pixel or filled with a preset background value for subsequent masking.

[0075] Two-dimensional deformation field solution and deformation compensation based on grid control point displacement

[0076] Even after perspective correction, local non-rigid deformations may still exist. To compensate for these local deformations, a set of control points is extracted from the perspective-corrected printing result image data, and a corresponding set of control points is extracted from the target printing image data. The control points are selected as stable geometric feature points, which in this embodiment are corner feature points or edge intersection feature points. Corner feature points are obtained through corner detection, and edge intersection feature points are obtained through edge extraction, line segment fitting, and intersection calculation. Control point matching is limited to the vicinity of the object area covered by the page object area or the reference area set, and the initial alignment relationship provided by perspective correction is used as the matching constraint to ensure that the matching is performed within the local neighborhood. After the matching is completed, the reprojection error or neighborhood consistency of the control point pair is used as the screening condition to remove abnormal matching point pairs that do not meet the error threshold.

[0077] After establishing the correspondence between two sets of control points, a two-dimensional deformation field represented by the displacement of grid control points is constructed. The grid is generated under the coordinate reference of the printing plate and covers the bounding rectangle range determined by the cutting line of the printing plate. The number of grid rows and columns is determined by the job configuration. For each pair of control points, its residual displacement vector is calculated. The residual displacement vector is the difference between the position of the control point in the target printing image data and the corresponding position of the control point in the perspective-corrected printing result image data. Using the residual displacement vector of the control point as the observation, the displacement vector of each grid node is solved by weighted fitting: for any grid node, the weight is determined according to the distance of the node to the surrounding control points, so that the control points that are closer to the node contribute more to the displacement of the node; at the same time, a smoothing constraint is added to the displacement difference of adjacent grid nodes, so that the displacement of the grid node is continuously controlled as the position changes, avoiding local abrupt changes.

[0078] After obtaining the displacement vector of the grid node, the displacement vector of any pixel position is calculated by bilinear interpolation, and deformation compensation is performed on the perspective-corrected printed image data. The deformation compensation adopts inverse mapping: for any pixel position in the target printed image data, its grid cell is first determined and the displacement vector is obtained by interpolation. Then, the sampling position in the perspective-corrected printed image data is calculated based on the displacement vector, and the pixel value is obtained by bilinear interpolation, thus obtaining the deformation-compensated printed image data. By combining global homography transformation and local two-dimensional deformation field, a pixel mapping relationship from the printed image data to the target printed image data can be established under the plate coordinate reference. This is used for subsequent object area sampling of reference area set, calculation of regional color difference observation, and interpolation application of spatial color difference compensation field.

[0079] Realization of imaging condition normalization

[0080] In this embodiment, to reduce measurement deviations caused by uneven illumination, lens vignetting, and differences in imaging channel response during image acquisition, imaging condition normalization is performed on the printed image data before performing device-independent color representation conversion on the reference image. The input for imaging condition normalization is the printed image data after perspective correction and deformation compensation. The white reference block is located within the color mark area, and its position and boundary range are given by the page data. The job configuration records at least the following parameters: white reference target brightness value, white reference target channel ratio, white reference block edge removal width, brightness gain coefficient limiting range, channel white balance gain coefficient limiting range, and the number of rows and columns of the brightness gain field grid.

[0081] First, based on the boundary range of the white reference block given by the layout data, the white reference block area is determined in the registered printing result image data. In order to reduce the influence of geometric registration residual error and edge ink diffusion on white reference sampling, the edge removal width given by the job configuration is removed inward along the boundary within the white reference block area to obtain the effective sampling area inside the white reference block.

[0082] Establishment of luminance gain field, luminance gain correction and shadow compensation

[0083] To ensure sufficient spatial observation of the luminance gain field, the white reference block in this embodiment consists of multiple white reference sub-blocks, which are distributed in a two-dimensional array within the color mark area. The layout position of each white reference sub-block is given by the layout data. For each white reference sub-block, a luminance statistical value is calculated within its effective sampling area. The luminance statistical value is taken as the average luminance of the effective pixels within the sub-block. The pixel luminance is calculated by the color channel pixel values ​​according to the linear combination relationship given by the job configuration. The linear combination relationship converts each color channel pixel value into a single luminance value.

[0084] A correspondence is established between the brightness value of the white reference target and the brightness statistics of each white reference sub-block to obtain the brightness gain coefficient of each white reference sub-block. The brightness gain coefficient is the ratio of the brightness value of the white reference target to the brightness statistics of the corresponding sub-block. The brightness gain coefficient is then subjected to a limiting process. The limiting range is given by the job configuration and is used to suppress excessive amplification caused by noise and local abnormal sampling.

[0085] After obtaining discrete luminance gain coefficients, a luminance gain field covering the entire page area is constructed. The luminance gain field is represented under the page coordinate reference and stored in the form of two-dimensional grid nodes. The grid covers the bounding rectangle range defined by the page clipping line. The number of grid rows and columns is given by the job configuration. The luminance gain coefficient of each white reference sub-block is used as an observation. The observation is propagated to the grid nodes through interpolation, and the gain value of the grid nodes is subjected to low-frequency smoothing processing to make the luminance gain change continuously with spatial position and characterize the low-frequency luminance attenuation distribution. The low-frequency luminance attenuation includes the shadow effect caused by uneven illumination and lens vignetting. In this embodiment, shadow compensation is achieved by correcting the low-frequency attenuation through the luminance gain field.

[0086] Luminance gain correction is performed on the registered printed image data based on the luminance gain field. For any valid pixel location in the printed image data, the corresponding grid cell is first determined in the luminance gain field grid, and bilinear interpolation is used to obtain the luminance gain value corresponding to that pixel location. Then, this luminance gain value is applied to the pixel values ​​of each color channel of that pixel to obtain the luminance-corrected color channel pixel values. The luminance-corrected pixel values ​​are truncated to ensure they fall within the range of valid pixel values. Pixel locations marked as invalid during the geometric registration stage are not included in the luminance gain correction and remain as invalid pixels for masking during subsequent color difference calculations.

[0087] Channel white balance correction

[0088] After completing luminance gain correction and shadow compensation, the effective sampling area inside the white reference block is used as the white balance correction reference area. The channel statistics of each color channel in the reference area are calculated respectively. The channel statistics are taken as the average value of the corresponding channel pixel value. According to the white reference target channel ratio given by the job configuration, the channel white balance gain coefficient is solved so that the white reference block area meets the target channel ratio after correction. The channel white balance gain coefficient is the ratio of the target channel statistics to the current channel statistics. The channel white balance gain coefficient is then subjected to limiting processing. The limiting range is given by the job configuration.

[0089] The channel white balance gain coefficient is applied to the corresponding color channel of the printed result image data. Channel white balance correction is performed on the entire printed result image data, and the corrected pixel values ​​are truncated to fall within the effective pixel value range. After the above processing, the normalized printed result image data is obtained, which serves as the input data for subsequent device-independent color representation conversion and regional color difference calculation.

[0090] Construction and solution of compensation problems

[0091] In this embodiment, based on the pixel correspondence established by geometric registration and the imaging condition normalization and device-independent color characterization conversion processing results, the regional color difference observation is calculated within the reference area set. In order to form job-level output data that can be distributed within the same printing job, a compensation problem is constructed and solved. The solution output includes global color transformation parameters and spatial color difference compensation field, and based on this, the compensated printing image data and printing parameter correction amount file are generated.

[0092] The solution variables for the compensation problem include: global color transformation parameters, spatial color difference compensation field, and printing parameter correction amount. Among them, the global color transformation parameters are used to characterize the global mapping relationship of the target printing image data in the device-independent color representation space, and include the parameterized representation of the tone curve; the spatial color difference compensation field is used to characterize the local compensation vector at different positions under the plate coordinate reference; the printing parameter correction amount is used to characterize the correction amount relative to the initial printing parameter set of the printing operation. The initial printing parameter set is obtained from the process settings before the start of the printing operation, and includes at least the ink volume parameters, gray balance parameters, dot gain compensation parameters, tone curve parameters, and registration compensation parameters of each color separation channel.

[0093] Composition of the objective function

[0094] The objective function for solving the compensation problem consists of three parts: a weighted sum of regional color difference observations, a spatial continuity penalty term for the spatial color difference compensation field, and a penalty term for the correction magnitude of printing parameters.

[0095] Weighted sum of color difference observations by region

[0096] The reference region set consists of multiple object regions. For each object region, the target color statistical vector of the region corresponding to the target printed image data and the result color statistical vector of the region corresponding to the printed result image data are obtained in the device-independent color representation space. Based on this, the regional color difference observation of the object region is obtained. A color difference residual is established on an object region basis. The color difference residual is the difference between the predicted compensation amount of the object region calculated by the current global color transformation parameters and the spatial color difference compensation field and the regional color difference observation of the object region. Weight coefficients are assigned to object regions according to object type, and the color difference residuals of each object region are weighted and summed according to the weight coefficients to form the observation cost.

[0097] Spatial continuity penalty term of spatial color difference compensation field

[0098] The spatial color difference compensation field is represented by two-dimensional grid nodes under the page coordinate reference. The grid covers the bounding rectangle range determined by the page clipping line. Each grid node stores the compensation vector under the device-independent color representation space. Adjacent grid nodes are determined according to the four-adjacency relationship. The difference vector of the compensation vector of adjacent grid nodes is taken with the Euclidean norm, and a penalty is applied to the difference norm to form a spatial continuity penalty term.

[0099] Printing parameter correction range penalty item

[0100] The printing parameter correction amount is represented as a correction vector relative to the initial printing parameter set. The norm of the correction vector is taken and a penalty is applied to form the printing parameter correction magnitude penalty term. The printing parameter correction magnitude penalty term is used to preferentially select the solution with the smaller correction magnitude within the solution set that satisfies the constraint conditions. Together with the upper limit constraint of the printing parameter correction magnitude described later, it limits the executable range of the job-level parameter output.

[0101] Setting constraints

[0102] Total ink volume limit constraint

[0103] The total ink volume limit constraint restricts the sum of ink volumes corresponding to each color separation channel to not exceed a preset limit. The total ink volume limit is determined by the process specifications of the printing materials and ink system and recorded in the job configuration. To make the constraint have a discretization caliber, discrete constraint points are established in the reference area set on the basis of object regions: For each object region, the regional statistical value of the pixel set in the object region is used as the representative input of the object region, and the channel coverage of the object region in each color separation channel is calculated; the channel coverage is determined by the mapping relationship of the tone curve parameters in the corresponding channels, and the ink volume value of each color separation channel is obtained by converting the channel coverage and the ink volume parameter of the corresponding channel. The ink volume values ​​of each color separation channel are summed and limited to not exceeding the total ink volume limit.

[0104] Tonal Monotonicity Constraint

[0105] The monotonicity constraint of the tone curve ensures that the tone curve remains monotonically constant with the input gray level. The tone curve is parameterized at discrete gray level nodes, and the gray level nodes cover the entire range from low gray to high gray. For adjacent gray level nodes, the output coverage of the next node is limited to be no less than the output coverage of the previous node. When the tone curve uses piecewise linear interpolation or spline interpolation, the monotonicity constraint is applied to the interpolation control points.

[0106] Upper limit constraint on printing parameter correction range

[0107] Set upper limits for the correction range of ink volume parameters, gray balance parameters, dot gain compensation parameters, tone curve parameters, and registration compensation parameters. The upper limits of the correction range are determined by the executable range of the printing equipment and the operation process window and recorded in the operation configuration. During the solution process, the correction amount of the printing parameters must fall within the corresponding upper limit boundary.

[0108] Spatial continuity constraints

[0109] An upper limit is set for the difference norm of the compensation vector between adjacent grid nodes to restrict the compensation difference between adjacent locations in the form of hard constraints. The upper limit value is given by the job configuration and is matched with the grid resolution.

[0110] Solution process and constraint handling

[0111] The compensation problem is solved iteratively. During initialization, the global color transformation parameters are set to the initial parameters obtained by fitting the color difference observations of the sub-regions. The spatial color difference compensation field is initialized to a zero field, and the printing parameter correction is initialized to a zero vector. In each iteration, the predicted compensation amount of the object region is calculated based on the current global color transformation parameters and the spatial color difference compensation field to form the color difference residual. At the same time, the spatial continuity penalty term and the printing parameter correction magnitude penalty term are calculated. Subsequently, the global color transformation parameters, the spatial color difference compensation field, and the printing parameter correction are updated to reduce the objective function.

[0112] The constraints are handled using projection and truncation: When the upper limit constraint of total ink volume is violated, the channel coverage or ink volume parameters of the relevant color separation channels are scaled to bring the sum of channel ink volume at the input of the object area back to within the upper limit; when the monotonicity constraint of the tonal curve is violated, the discrete control points of the tonal curve are monotonically corrected to ensure that adjacent grayscale nodes satisfy the monotonically non-decreasing relationship; when the upper limit constraint of the printing parameter correction range is violated, the excess correction amount is truncated to fall within the boundary range; when the spatial continuity constraint is violated, the excess part of the difference of the compensation vector of adjacent grid nodes is restricted and the grid node compensation vector is written back. The iteration termination condition is given by the job configuration and is determined based on the objective function's decrease being less than the threshold or the number of iterations reaching the upper limit.

[0113] After the solution is completed, the global color transformation parameters and spatial color difference compensation field are output. The target printed image data is compensated based on the global color transformation parameters and spatial color difference compensation field to generate the compensated printed image data. At the same time, the printed parameter correction file is exported from the same solution result, so that the compensated printed image data and the printed parameter correction file together constitute the job-level output data of the printing operation.

[0114] Mesh representation, interpolation calculation and spatial continuity constraints of spatial chromatic aberration compensation field

[0115] In this embodiment, the spatial color difference compensation field is used to characterize the local color difference compensation amount at different positions on the printing plate. The spatial color difference compensation field is defined under the device-independent color representation space, which is a three-dimensional color coordinate system composed of luminance component, first chromaticity component and second chromaticity component. The compensation vector of the spatial color difference compensation field has the same dimension as the three-dimensional color coordinate system. The compensation vector is used to perform position-related incremental correction on the color coordinates under the device-independent color representation space, and is used to calculate the predicted compensation amount of the object area during the compensation solution process and to generate the compensated printed image data.

[0116] Two-dimensional mesh node creation and compensation vector storage

[0117] The spatial color difference compensation field is discretized by two-dimensional grid nodes under the page coordinate reference. The two-dimensional grid covers the bounding rectangle boundary range determined by the page clipping line. The boundary of the bounding rectangle and the coordinate axis direction are determined by the page coordinate reference. The number of grid rows and columns are recorded in the job configuration, and the grid node spacing is determined by the boundary range of the bounding rectangle and the number of rows and columns, so that the grid nodes are evenly distributed within the bounding rectangle range.

[0118] For any grid node, its two-dimensional coordinate position is determined based on the page coordinate reference, and a compensation vector is stored at that node. The compensation vector is a three-dimensional vector in the device-independent color representation space. The storage structure of the grid node adopts the form of a two-dimensional array, where the array index corresponds to the grid row and column, and the array element is the compensation vector. In this way, the spatial color difference compensation field is discretely represented by a finite number of grid nodes, which facilitates the updating and constraint processing of the compensation vector during the compensation solution iteration.

[0119] Bilinear interpolation calculation of pixel position compensation vector

[0120] For any pixel position within the page, the pixel correspondence established by geometric registration is first used to determine the two-dimensional coordinates of the pixel position under the page coordinate reference. Based on the position of the two-dimensional coordinates in the bounding rectangle, the grid cell in which it is located is determined. The grid cell is surrounded by four adjacent grid nodes, namely the top left node, the top right node, the bottom left node, and the bottom right node.

[0121] After determining the grid cells, the horizontal and vertical proportions of the pixel position within the grid cell are calculated: the horizontal proportion is the ratio of the distance from the pixel position to the left boundary of the grid cell to the width of the grid cell, and the vertical proportion is the ratio of the distance from the pixel position to the top boundary of the grid cell to the height of the grid cell. The compensation vectors of the four corner grid nodes are weighted based on these horizontal and vertical proportions to obtain the compensation vector for the pixel position. Bilinear interpolation is used for weighting; first, linear interpolation is performed horizontally on the compensation vectors of the two upper and two lower nodes respectively, and then linear interpolation is performed vertically on the two horizontal interpolation results to obtain the final compensation vector. If the pixel position falls exactly at the grid node coordinates, the compensation vector stored in that grid node is directly taken as the pixel position compensation vector.

[0122] When a pixel position falls on or exceeds the boundary of the bounding rectangle, the two-dimensional coordinates of the pixel position under the layout coordinate reference are clamped to the inside of the bounding rectangle boundary. The bilinear interpolation calculation is then performed at the clamped coordinate position to avoid out-of-bounds access. For pixel positions marked as invalid in the geometric registration stage, no compensation vector is calculated and the pixel is kept as invalid so that it can be masked in subsequent color difference calculation and compensation applications.

[0123] Definition and Implementation of Spatial Continuity Constraints

[0124] To limit abrupt changes in the spatial color difference compensation field, a spatial continuity constraint threshold is set in the operation configuration. The spatial continuity constraint sets an upper limit on the difference norm of the compensation vectors of adjacent grid nodes. Adjacent grid nodes are determined by a four-adjacency relationship, that is, for any grid node, it only forms an adjacent pair with its adjacent nodes in the upper, lower, left, and right directions. For any adjacent pair, the Euclidean norm of the difference between the compensation vectors of the two nodes is calculated, and the Euclidean norm is limited to not exceeding the spatial continuity constraint threshold.

[0125] During the compensation solution iteration process, after the grid node compensation vector is updated, all adjacent pairs are traversed according to a fixed scanning order and the difference norm is checked. The fixed scanning order is to first traverse the grid rows from top to bottom, and then traverse the grid columns from left to right. For each grid node, the adjacent pairs formed by its right and lower adjacent nodes are checked. If the difference norm of an adjacent pair exceeds the spatial continuity constraint threshold, the compensation vector of the currently scanned grid node remains unchanged, and the compensation vector of its adjacent nodes is written back with a limit: the difference vector is scaled proportionally to the spatial continuity constraint threshold, and the compensation vector of the adjacent node is updated with the scaled difference vector so that the adjacent pair satisfies the upper limit of the difference norm. Through the scanning and writing back rules determined above, the spatial color difference compensation field keeps changing continuously under the page coordinate reference, and works together with the regional color difference observation constraints, the total ink volume upper limit constraints, and the tonal monotonicity constraints to form a spatial compensation result that can be used to generate job-level output data.

[0126] Generation, content structure, and distribution of printing parameter correction files

[0127] In this embodiment, after the compensation solution is completed, the global color transformation parameters and spatial color difference compensation field are obtained. In order to enable the solution results to be directly applied to the current printing job without changing the existing calibration system of the printing equipment, a printing parameter correction file is generated. The printing parameter correction file is a job-level output file that corresponds one-to-one with the current printing job. It is used to describe the incremental correction relative to the initial printing parameter set of the current printing job. The initial printing parameter set is obtained from the process settings before the start of the current printing job and is recorded in the job configuration.

[0128] Structure and fields of printing parameter correction file

[0129] The printing parameter correction file includes a header and a parameter correction section. The header is used to bind the job and establish reference relationships. The header records the job identifier, generation time, reference identifier of the initial printing parameter set, and the definition of the device-independent color representation space used. The parameter correction section includes the correction amount of ink volume parameter, gray balance parameter, dot gain compensation parameter, tone curve parameter, and registration compensation parameter.

[0130] Among them, the correction amount of the tone curve parameter is represented by the incremental sequence of channel coverage on the discrete gray level node; the correction amount of the gray balance parameter is represented by the incremental sequence of the color separation channel ratio on the same gray level node; the correction amount of the dot gain compensation parameter is represented by the incremental sequence of compensation parameters corresponding to the channel coverage; the correction amount of the ink volume parameter is represented by the increment of the ink volume coefficient of the color separation channel; and the correction amount of the registration compensation parameter is represented by the two-dimensional displacement increment under the page coordinate reference. The discrete gray level node, parameter unit, value range and limiting boundary are all given by the job configuration to ensure that the parameter output can be directly sent and meets the process window and the equipment's executable range.

[0131] To avoid affecting other printing operations, the printing parameter correction file is only used as the job-level output file for this printing operation for distribution and application. It does not write or modify the International Color Consortium color characteristic file and equipment characteristic lookup table on the printing equipment side. In this embodiment, the International Color Consortium color characteristic file and equipment characteristic lookup table are used in read-only mode for parameter conversion and sensitivity estimation to derive the job-level correction amount.

[0132] Determination of correction amounts for tone curve parameters and gray balance parameters

[0133] The global color transformation parameters include the job-level incremental expression of the initial tone curve and the job-level correction expression of the neutral direction. To generate the tone curve parameter correction amount, the channel coverage of the initial tone curve and the channel coverage of the solved tone curve are calculated on the discrete gray-level nodes given in the job configuration, and the difference between the two is used as the tone curve parameter correction amount. The coverage between discrete gray-level nodes is reconstructed in a piecewise linear manner to ensure the continuity of tone change and to maintain consistency with the tone monotonicity constraint.

[0134] The gray balance parameter is used to constrain the proportion of color channels in neutral gray, so that the output color is stable along the neutral direction at different gray levels. In this embodiment, the gray balance target is determined by the target channel proportion given by the job configuration. Based on the correction expression for the neutral direction in the global color transformation parameters, the deviation between the current color channel proportion and the target channel proportion is calculated at the discrete gray level nodes. The increment of this deviation at each gray level node is represented as the gray balance parameter correction amount. The gray balance parameter correction amount and the tonal curve parameter correction amount work together to calculate the subsequent channel coverage amount, so that the gray level is continuous and satisfies the tonal monotonicity constraint.

[0135] Determination of the correction amount for network expansion compensation parameters

[0136] The dot gain compensation parameter is used to compensate for the dot gain effect during the printing process. In this embodiment, the dot gain compensation parameter and the tone curve are described using the same set of discrete gray-level nodes. To generate the dot gain compensation parameter correction amount, the coverage increment before and after tone curve correction is calculated for each color separation channel at the discrete gray-level nodes. According to the dot gain compensation parameter format specified in the job configuration, the coverage increment is converted into the dot gain compensation parameter increment, thereby forming the dot gain compensation parameter correction amount. By generating the dot gain compensation parameter correction amount and the tone curve parameter correction amount in a linked manner under the same gray-level node system, the two types of corrections have the same direction of action on the same channel, avoiding duplicate compensation or offset compensation caused by inconsistency between parameters.

[0137] Determination of ink quantity parameter correction amount

[0138] The ink volume parameter is used to limit the ink intensity of each color separation channel and the total ink volume level. The ink volume parameter correction amount must be consistent with the upper limit constraint of the total ink volume. In this embodiment, the ink volume parameter correction amount is determined by the weighted average of the compensation vectors of the spatial color difference compensation field in the object region of the reference region set.

[0139] Specifically, for each object region in the reference region set, the pixel set corresponding to the object region under the page coordinate reference is determined based on the pixel correspondence established by geometric registration. Then, a compensation vector is calculated for each pixel position within the pixel set using bilinear interpolation. The weighted average of the compensation vectors within the object region is calculated to obtain the region compensation vector of the object region. The weighting coefficient is determined by the object type weight coefficient and the number of valid pixels within the object region. Invalid pixels are not included in the statistics. The region compensation vectors of each object region are summarized according to the object type weight coefficient to obtain the job-level compensation vector used for ink volume correction.

[0140] Since the compensation vector is defined in the device-independent color representation space, in order to convert the job-level compensation vector into a correction of the ink volume direction of the color separation channel, this embodiment reads the device characteristic lookup table and establishes a local sensitivity relationship in read-only mode. The local sensitivity relationship is estimated by the finite difference method: under the initial printing parameters of the current job, the coverage of a single color separation channel is incrementally perturbed according to the channel coverage step given by the job configuration, and the color change in the device-independent color representation space is obtained through the device characteristic lookup table; the above perturbation is performed on each color separation channel to obtain the local influence relationship of the channel coverage change to the three-dimensional color coordinate change; based on the local influence relationship, the job-level compensation vector is converted into an incremental correction of the color separation channel coverage; and under the condition of satisfying the upper limit constraint of the total ink volume and the upper limit constraint of the printing parameter correction range, the incremental correction of the color separation channel coverage is converted into the ink volume parameter correction amount and written into the printing parameter correction amount file. The above process only reads the device characteristic lookup table for conversion and does not write or modify the device characteristic lookup table or the color characteristic file of the International Color Consortium.

[0141] Determination of the correction amount of the registration compensation parameter

[0142] The registration compensation parameter is used to compensate for the relative displacement error between color separation channels. In this embodiment, the registration compensation parameter correction amount is determined by the residual displacement of the non-collinear positioning marks after geometric registration. Specifically, after completing the global homography transformation and local deformation compensation, for the non-collinear positioning marks in the color mark area, the residual displacement vector between the observed position of the positioning mark in the printed result image data and its theoretical position in the page data is calculated. The residual displacement vectors of each positioning mark are summarized using a fixed statistical method to form the job-level residual displacement. The statistical method uses the median of the vector as the central quantity, and the residual displacement vectors that deviate too much from the central quantity are removed according to the residual threshold given by the job configuration. Finally, the average value of the retained residual displacement vectors is taken to obtain the job-level residual displacement. The job-level residual displacement is written into the printing parameter correction amount file as the registration compensation parameter correction amount, and is used for job-level correction of the color separation channel alignment in the subsequent execution of this printing job.

[0143] Implementation methods for issuing calibration files at the job level and not writing them back to the device side

[0144] The printing parameter correction file and the compensated printing image data together serve as the job-level output data for this printing job. During job execution, the printing data processing link reads the printing parameter correction file and overlays various correction values ​​onto the initial printing parameter set for this printing job to obtain the effective printing parameters for this job. The overlay process only applies to this printing job and does not write the overlaid effective printing parameters back to the International Color Consortium color characteristic file and equipment characteristic lookup table on the printing equipment side. This ensures that the equipment-side calibration system remains unchanged for other jobs and allows the correction values ​​for this job to be independently saved and reused in the job management record.

[0145] Solution boundaries, verification color difference gating, and invalidation markers within the same task

[0146] In this embodiment, the global color transformation parameters and spatial color difference compensation field are obtained by solving the printing result image data collected by the printing batch under the same job number. The printing batch is a batch completed under the same page data, the same target printing image data, the same initial printing parameter set, and the same printing equipment and material configuration conditions. The printing equipment and material configuration conditions are recorded by the job configuration and remain unchanged during the execution of the batch.

[0147] To ensure clear data boundaries for the solution input, when generating the compensated printed image data and the printed parameter correction file, no additional correction sample image data printed to correct the color status of the equipment or materials is included. The input for the solution stage only includes the target printed image data, the layout data associated with the target printed image data, and the printed result image data collected after the printing batch is solved. The printed parameter correction file generated by the solution stage is in a pending release state before it is approved. It is only allowed to be used for the review printing batch under this job number and will not enter the formal issuance process.

[0148] Verification of image data acquisition and alignment for printing results

[0149] In the verification printing batch, the compensated printing image data and the printing parameter correction file in the waiting-to-release state are used for printing, and the verification printing result image data is collected. The acquisition equipment, imaging channel configuration and exposure strategy of the verification printing result image data are consistent with those of the solved printing batch, and are recorded by the job configuration to reduce the impact of differences in acquisition conditions on the verification color difference. The verification printing result image data includes the color mark area image data located in the non-finished product retention area of ​​the plate. The color mark area includes non-collinear positioning marks, white reference blocks and preset color reference blocks, so that the verification stage uses the same geometric registration and imaging condition normalization caliber as the solution stage.

[0150] For the image data of the verification printing result, geometric registration is performed with the target printing image data under the page coordinate reference defined by the page data to establish the pixel correspondence. Under the pixel correspondence, the corresponding pixel set of the reference area set in the verification printing result image data is determined. Then, imaging condition normalization and device-independent color representation transformation are performed on the verification printing result image data to make the color representation caliber of the verification stage consistent with that of the solution stage. Pixels marked as invalid in geometric registration are not included in the subsequent verification color difference statistics.

[0151] Verification of color difference calculation methods and threshold settings

[0152] Within the reference region set, the verification color difference is calculated. For each object region in the reference region set, the target color statistical vector of the region corresponding to the target printed image data and the verification color statistical vector of the region corresponding to the verification printed result image data are calculated in the device-independent color representation space. The regional verification color difference of the object region is calculated in the same color difference measurement method as in the solution stage. The effective pixel set of the object region is determined by geometric registration and the reference region set. Invalid pixels are not included in the statistics.

[0153] To form the job-level verification criteria, the regional verification color difference of each object area is weighted and summarized according to the object type weight coefficient to obtain the job-level verification color difference. The object type weight coefficient is given by the job configuration and is consistent with the weight coefficient of the regional color difference observation in the solution stage. The verification color difference threshold is given by the job configuration. The threshold is derived from the process specifications of the corresponding material system, ink system and printing equipment, and is used to limit the acceptable color difference range of the job-level output data.

[0154] Gating and invalidation marking

[0155] When the color difference verified at the job level is not greater than the verification color difference threshold, the printing parameter correction file is switched from the pending release state to the valid state, and the printing parameter correction file is allowed to enter the formal issuance process. After formal issuance, the printing data processing link will add the correction amount to the initial printing parameter set of this job, which will only be effective for subsequent printing batches within the range of this job number.

[0156] When the color difference during job-level verification exceeds the verification threshold, the printing parameter correction document is kept in a pending release state and marked as invalid. This prevents the printing parameter correction document from entering the formal issuance process. The invalidation mark is written into the header of the printing parameter correction document, including at least a validity flag field and a verification failure record field. The verification failure record field records the job-level verification color difference value and the verification color difference threshold for subsequent traceability. The invalidation mark only affects the issuance and application of the job-level document under this job number. It does not write or modify the International Color Consortium color characteristic file and equipment characteristic lookup table on the printing equipment side, nor does it change the applicability of the equipment calibration system to other jobs.

[0157] After the printing parameter correction file is marked as invalid, the compensation solution is retried based on the printing result image data already collected under this job number, or the verification failure record is output to the job management record for manual intervention. The above processing does not use the image data of the additional printed correction sample as the solution input, thereby reducing job interruption and material consumption caused by repeated printing while maintaining clear data boundaries.

[0158] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0159] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. An outer packaging carton printing parameter intelligent optimization and color difference compensation method, characterized in that: The method comprises: ​ acquiring target printing image data corresponding to a layout of an outer carton and layout data associated with the target printing image data; after a printing job corresponding to the target printing image data is completed, acquiring printing result image data of the printing job; performing geometric registration on the printing result image data and the target printing image data under a layout coordinate reference defined by the layout data to obtain pixel correspondence; determining a reference region set under the pixel correspondence and extracting a reference image pair corresponding to the reference region set from the target printing image data and the printing result image data; performing imaging condition normalization and device-independent color representation conversion on the reference image pair and calculating color difference in the reference region set to form a sub-region color difference observation; using the sub-region color difference observation as an observation constraint, constructing and solving a compensation solution problem to obtain global color transformation parameters and a spatial color difference compensation field, and the compensation solution problem satisfies total ink amount upper limit constraint, step monotonicity constraint, printing parameter modification amplitude upper limit constraint and spatial continuity constraint; generating compensated printing image data according to the global color transformation parameters and the spatial color difference compensation field, and generating a printing parameter modification amount file according to the same solution result, and using the compensated printing image data and the printing parameter modification amount file as job-level output data of the printing job.

2. The method for intelligent optimization of printing parameters and color difference compensation of an outer packaging carton according to claim 1, characterized in that: The layout coordinate reference is jointly defined by an outer rectangle boundary determined by layout cutting lines in the layout data, a layout coordinate origin and a coordinate axis direction, the layout coordinate origin is arranged at a corner point of the outer rectangle, and the coordinate axis direction is consistent with two adjacent sides of the outer rectangle.

3. The method according to claim 2, wherein the method is characterized by: The determination of the reference region set comprises: acquiring pre-printing file data corresponding to the target printing image data and parsing to obtain layout object contour information, dividing the layout object contour information into text objects, solid objects, machine-readable code objects and identification objects according to object types, and the reference region set is composed of object regions corresponding to the object types; the sub-region color difference observation is weighted and summarized according to weight coefficients corresponding to the object types, and the weight coefficients are pre-set non-negative real numbers.

4. The method according to claim 3, wherein the method is characterized in that: The printing result image data includes color patch region image data located in a layout non-product reserved region, and the color patch region includes non-collinear positioning marks, white reference blocks and pre-set color reference blocks; the non-collinear positioning marks provide control points for the geometric registration, the white reference blocks provide brightness and white balance correction references for the imaging condition normalization, and the pre-set color reference blocks are used to calibrate a conversion model corresponding to the device-independent color representation conversion.

5. The method according to claim 4, wherein the method is characterized by: The geometric registration comprises: solving global homographic transformation parameters based on the non-collinear positioning marks and performing perspective correction on the printing result image data based on the global homographic transformation parameters; A first set of control points is extracted from the perspective-corrected printed result image data, and a second set of control points corresponding to the first set of control points is extracted from the target print image data, a two-dimensional deformation field expressed in terms of grid control point displacements is solved based on the first set of control points and the second set of control points, and a deformation compensation is performed on the perspective-corrected printed result image data, wherein the control points are corner feature points or edge intersection feature points.

6. The method according to claim 5, wherein the method is characterized in that: The imaging condition normalization comprises: A luminance gain field is established according to pixel values of the white reference block in the printed result image data, and a luminance gain correction and shading compensation are performed on the printed result image data; Channel white balance correction is performed on the printed result image data according to pixel values of the white reference block in each color channel.

7. The method according to claim 6, wherein the method is characterized by: An objective function of the compensation solving problem is composed of a weighted sum term of sub-region color difference observation, a spatial continuity penalty term of the spatial color difference compensation field, and a print parameter modification amplitude penalty term; wherein: The spatial continuity penalty term imposes a penalty on a difference norm of compensation vectors of adjacent positions of the spatial color difference compensation field under a layout coordinate reference; The print parameter modification amplitude penalty term imposes a penalty on a modification norm relative to an initial print parameter set of the print job; The total ink amount upper limit constraint limits the sum of ink amounts corresponding to each color separation channel to be less than a pre-set upper limit; The tone scale monotonicity constraint limits a tone scale curve represented by the global color transformation parameter to be monotonically non-decreasing with respect to an input gray scale.

8. The method according to claim 7, wherein the method is characterized in that: The spatial color difference compensation field is represented by two-dimensional grid nodes under a layout coordinate reference, each grid node stores a compensation vector in a device-independent color representation space; a compensation vector of any pixel position is obtained by bilinear interpolation from the two-dimensional grid nodes; The spatial continuity constraint sets an upper limit on a difference norm of compensation vectors of adjacent grid nodes.

9. The method according to claim 8, wherein the method is characterized by: The print parameter modification amount file includes ink amount parameter modification amount, gray balance parameter modification amount, dot gain compensation parameter modification amount, tone scale curve parameter modification amount, and register compensation parameter modification amount; wherein: The tone scale curve parameter modification amount and the gray balance parameter modification amount are determined by the global color transformation parameter; The dot gain compensation parameter modification amount is determined by the tone scale curve parameter modification amount; The ink amount parameter modification amount is determined by a weighted average of compensation vectors of the spatial color difference compensation field in the object regions corresponding to the reference region set; The register compensation parameter modification amount is determined by a residual displacement of the non-collinear positioning marks after the geometric registration; The print parameter modification amount file is a job-level output file, and does not rewrite the International Color Consortium color characteristic file and the device characteristic lookup table on the printing device side.

10. The method according to claim 9, wherein the method is characterized in that: The global color transformation parameter and the spatial color difference compensation field are solved from the printed result image data corresponding to the same print job, and do not introduce a re-printed proof data when generating the compensated print image data and the print parameter modification amount file; After the printing is completed using the job-level output data, review printing result image data is acquired, and review color difference is calculated in the reference region set, and when the review color difference is greater than a pre-set threshold, the printing parameter correction amount file is not issued and is marked as invalid.

Images