KR20250007492A - Ink dosing for digital printing on reflective substrates - Google Patents

Abstract

A computer-implemented method for dosing ink in a digital printing device having a plurality of ink channels when printing on a reflective substrate is provided. A digital printing system configured according to the method is also provided. Target color data of a target color to be printed on a reflective substrate is captured, including both color data and spectral reflectance data. The captured color data is processed by a printer model to output a preliminary dosing ratio for each ink channel. A digital opacity value for the target color is computed from the dosing ratio for each ink channel. A natural opacity value for the target color is computed from the captured spectral reflectance data, and a difference between the digital opacity value and the natural opacity value is computed. A predicted dosing ratio for each ink channel is obtained by interpolating the difference for the ink phase of a diffuse ink component of the target color.

Description

Ink dosing for digital printing on reflective substrates

The present invention relates to a color data processing technique in a colorimetric space for dosing ink in a printing device or system having multiple ink channels when printing on a reflective substrate.

Digital printers are rapidly gaining adoption in the professional printing sector, as their ease of configuration via a computer terminal and associated programs significantly reduces the economic threshold for print volumes, allowing much more customized and much smaller production runs. Digital printers typically contain multiple ink channels, for example cyan, magenta, yellow and black ('CMYK') or more, for example additionally orange, green, violet and red ('CMYKOGVR'), regularly combined with diffusion inks such as white, silver or additional inks such as varnishes and primers.

In this context, the task of color matching is traditionally performed by digital print operators as an iterative process of attempting to match the printed output to the target color by concatenating transformations from the device-dependent chromatic space(s) associated with the printer through a reference device-independent chromatic space, a preferred example of which is the Commission on Illumination ('CIE') Brightness, Green-Red and Blue-Yellow ('L*a*b') color space ('CIELAB').

For example, color matching workflows have been improved in the context of digital printing by recent methods, as disclosed in the applicant's EP20178552.4. However, color matching and in particular opacity matching remain particularly difficult and burdensome tasks when printing on reflective substrates, referred to in the industry as 'RSM' (reflective, shiny and/or mirrored) substrates, regardless of the printer type and whether the print is printed directly onto the substrate or indirectly onto a film applied to the substrate. The amount of time and effort required to achieve the best match, if at all, is considerable.

A fundamental difficulty arises from specular reflection by highly reflective surfaces, such as metals, which cause significant light scattering compared to the diffuse reflection of more conventional printed substrates, such as white-coated surfaces, paper or cardboard.

The scattering of colors printed on a reflective substrate depends on the size and shape of the color pigments from the formulated ink. The pigments in conventional inks commonly used for analog printing on reflective substrates (e.g., aluminum cans, metal sheets) are more polydisperse and exhibit larger diameters than inkjet pigments used in digital printers, which allows for much greater volume to be expressed in the ink particles. The resulting mixture of reflectivity and scattering that appears in the printed layer can then lead to some involuntary selectivity in the orientation of the reflection. Therefore, printed articles on reflective substrates (e.g., flat surfaces or cylindrical objects) produced with conventional ink pigments frequently exhibit dull reflections, which result in increased scattering that is traded for loss of image density, clarity and sharpness. Conversely, products digitally printed with UV or latex inks applied directly onto a film, typically with a functional inkjet coating layer printed with water-based or eco-solvent inks that can be wrapped around or laminated onto a reflective substrate, or with toner or smaller inkjet pigments, exhibit more vivid reflections, sharpness and color clarity, resulting in a different color appearance when compared to conventional printing processes.

In certain printing applications, particularly when printing on white substrates, ink opacity (or density) is expressed by a measurement that defines the amount of light passing through an ink coating applied to the printed surface. This measurement is typically accomplished by reflectance or transmission densitometry. However, when other measurable methodologies are disclosed to define the opacity of a given color ink on a white substrate, the RSM substrate reflectance is a key factor that challenges existing methods, and therefore these methodologies will not provide satisfactory results. It is believed that no method exists for quantifying the opacity or transparency of a color on a RSM substrate. Currently, manufacturers of metallic decorative inks determine the opacity of their inks themselves, typically based on pigment selection to formulate color inks with or without a diffusion element, such as white ink.

As the printing of artwork and effects has become increasingly complex, correspondingly, the various inks specifically adapted for use in decorative applications present additional challenges in colour and opacity matching for printing on reflective substrates. Manufacturers formulating inks for reflective substrates have introduced inks with distinct opacity levels, which are semi-normalised into opaque, translucent and transparent inks, whereby opaque inks largely block the reflectivity of the underlying substrate, transparent inks allow the reflectivity of the underlying substrate to substantially pass through the colour, and translucent inks provide intermediate levels of opacity.

Expert knowledge highlights additional complexities in color matching for reflective substrate printing. For example, when printing with a wet screen process, inks must be matched as translucent variations, where bright, warm colors such as yellow, orange and red must be matched with slightly more opacity. Conversely, colors known as pastel colors, characterized by high brightness and low saturation, must be matched only as opaque variations to obtain the purest and best shade.

Color on a diffuse white surface is typically measured with a spectrophotometer under 45/0° or 0/45° geometry, and color on a reflective surface is typically measured with a multi-angle or spherical spectrophotometer under d/8° geometry, with a spherical spectrophotometer capturing specular reflection either including the specular component ('SPIN') or excluding the specular component ('SPEX') in the measurement. For example, a SPIN measurement is considered to capture total reflectance without variation regardless of surface type (matte, glossy, etc.) and is representative of 'true color', such that the target SPIN value should match for printed colors to exhibit satisfactory color representation. A SPEX measurement is typically considered to capture reflectance variation based on different surfaces, excluding the gloss component, and is representative of 'sample appearance', such that the SPEX value should match for printed colors to exhibit satisfactory color consistency. Measurement of color in SPIN or SPEX delivers a unique and individual value for the color, where the difference between the spectral reflectance and the respective converted L*a*b value for the color varies based on the degree of specular reflectance of the substrate, where non-linearity of the measurement difference over the spectral range can further vary the difference in the L*a*b values.

In a traditional design process defining the final print, colors may be selected from a given print reference or from a printed color book such as a Pantone® guide. However, the color selection may or may not be represented in a very different surface reflection, such as white paper, whereas the desired print is intended to be printed on an RSM substrate, which can result in errors in color selection due to a very different perception when compared to the final print on the RSM substrate.

Among the various proofing technologies (either digital or analog) that attempt to replicate the final print, some have been developed as digital imaging systems (e.g. Fuji FinalProof® or Kodak®Approval®) that combine image passes from specific digital donors and transfer film layers onto a production substrate using advanced laser imaging technology, allowing for adjustable color density.

Improvements have also been proposed to optimize color selection decisions in the design process. For example, PantoneLIVE® provides digital color information from colors available in pre-processed Pantone® guides on a variety of printing substrates, including reflective substrates such as metals or aluminum. However, the digital data available from the PantoneLIVE® library defined for metals or aluminum does not provide sufficient information to accurately and simultaneously reproduce the color according to its opacity concept in external conventional or digital systems, because the opacity of these surfaces is arbitrarily defined by a group of selectors from Pantone® or its preferred ink manufacturer partner, Sun Chemical, who do not disclose information or a given methodology for matching colors and opacity.

In other recent examples, physical color catalogs have been introduced that directly apply conventional inks to reflective surfaces such as aluminum. For example, the two-piece INX Color Perfection® catalog for metal decorative, offering a range of over 600 colors, claims to be the only color catalog available on metal rather than paper. However, this analog process of applying inks onto production substrates is labor intensive and difficult to maintain under strict color variances compared to different versions of the catalog. Color samples are available as physical color references, and the available digital data does not provide sufficient information to accurately replicate these colors, including opacity, with conventional or digital systems such as flat, roll-to-roll, high-speed or cylindrical printing devices.

In the above context, the color and opacity matching workflow for printing on reflective substrates still largely corresponds to the existing workflow disclosed as prior art in EP20178552.4, where an operator initializes ink values, prints samples, checks color accuracy, and modifies ink values for matching as needed. WO2021/035105 to Sun Chemical Corporation discloses an example of such prior art in the specific field of reflective substrate printing.

Any color matching workflow begins with an operator obtaining a table of spot colors or target colors to be reproduced by a particular printer. Each printer has its own printer calibration, which depends on the ink channels, print head characteristics and printing substrate of the device, together with a unique color profile (e.g., an International Color Consortium 'ICC' profile) containing information for converting colors from the L*a*b color space or another color space to the device color space, where the obtained L*a*b (or other) values for each color in the table are input into a 'RIP' (raster image processor) computer program. The RIP program uses the printer's color profile to process the target L*a*b values and computes output values for each ink as a color-specific ink dose, also known as ink separation. For example, for a target color with L*a*b values of [14.191, 18.511, 4.177], a RIP program might suggest that the individual ink dose values for a CMYKOG printer should be [50.196, 100, 100, 50.196, 50.196, 0].

Since the device color space of a printer is defined by the inks used by the printer, in the context of reflective substrate printing this includes opaque, translucent and transparent variations. Depending on the nature of the artwork, for artworks requiring digital printing within the same workflow, targets with distinct opacity levels present a significant challenge when color matching is required. The operator faces the additional challenge that color matching in both SPIN and SPEX requires the use of individual color profiles, where each printer profile will compute separate and distinct ink separations, each of which is different.

Therefore, an improved method of dosing ink in a multichannel printer when printing works on a reflective substrate is desirable, which is suitable for providing ink separation that more closely matches the target color, including opacity, and with reduced user input requirements, compared to prior art techniques.

The present invention provides a method of dosing ink in a multi-channel printer that substantially automates the adjustment workflow inherent in the task of matching colors to actual and target colorimetric values when printing on a reflective substrate, wherein the opacity of a target color to be printed digitally or conventionally on a reflective substrate is determined.

According to one aspect of the present invention, a computer-implemented method of dosing ink in a printing device having a plurality of ink channels when printing on a reflective substrate is provided, the method comprising: capturing target color data of at least one target color to be printed on the reflective substrate, the target color data including both color data and spectral reflectance data including and excluding a specular component; generating a printer model; processing the captured color data with the printer model to output a preliminary dosing ratio for each ink channel representing an individual color component of the target color; computing a curve from each of the ratios and interpolating the computed curve to output a digital opacity value for the individual ink channels; combining the individual digital opacity values of all ink channels to output a digital opacity value for the at least one target color; computing a natural opacity value from the captured spectral reflectance data; calculating a difference between the computed digital opacity value and the computed natural opacity value; and interpolating the calculated difference for the ink phase of the diffusion ink component of at least one target color to obtain a predicted dosing ratio for each ink channel.

Embodiments of this method may include an additional step of calibrating the printer using both a specular component on the reflective surface and a specular component on the reflective surface having a layer of diffusing ink.

In an embodiment of the method, the step of generating a printer model may include an additional step of generating a printer calibration model based on a specular component on the reflective surface and a specular component on the reflective surface having a layer of diffuse ink.

Embodiments of the method may include an additional step of repeating the processing steps of classifying the target color or each target color according to a defined target opacity; and replacing the captured color data with data representing the classified color or each classified color, wherein the obtained predicted dosing ratio for each ink channel includes a dosing ratio for a diffuse ink, such as white ink.

Embodiments of the method may include the additional steps of adding a predicted ratio of the diffusing ink from the target natural opacity value to output a next digital opacity value; and calculating a difference between the natural opacity value and the next digital opacity value; the calculated difference being a value referred to as the delta opacity for the reflective substrate.

A variation of any of the preceding embodiments may further comprise an additional step of compensating for the calculated difference prior to the interpolation step, the compensation comprising setting the difference to zero when the natural opacity value is below a predefined threshold or when the calculated difference value is negative. A variation of this additional embodiment may further comprise an additional step of compensating the predicted dosing ratio for each ink channel with a composite function stored as a lookup table.

Variations of any of the preceding embodiments may further comprise the additional steps of generating a color chart from the predicted dosing ratio for each ink channel; or generating a color chart from the corrected dosing ratio for each ink channel; and printing the color chart with a printer on a reflective substrate. The substrate may be an object printed directly with a cylindrical printer, a flat surface, or an optically clear or transparent layer for placement on an intended reflective substrate.

Variations of any of the preceding embodiments may further comprise the additional step of determining the opacity of a formulated conventional ink wherein the colorant or base ink is characterized to specify the opacity of each ink step for the purpose of matching the particular opacity of the final conventional ink formulation.

Embodiments of the method may include an additional step of computing a diffusion ink percentage from a digital opacity value of a target color to obtain individual versions of a digital opacity ranging from transparent to opaque by maintaining saturation and hue of the target color (irrespective of the opacity level of the target color). The variation may include an additional step of predefining a threshold value for predicting opacity for pastel colors that typically match higher opacities.

Embodiments of the method may include the additional step of outputting, for each ink channel, a predicted dosing ratio for a lighter version and a darker version of the target color at a given distance in the L*a*b color space, e.g., a Delta-E measurement.

Embodiments of the method may optionally include the additional step of generating a digital ink drawdown or color card representing a digital color and opacity match deemed best for the target color, with lighter and darker versions thereof specified at a predetermined distance Delta-E in L*a*b color space.

In another aspect of the present invention, a digital printing system is provided, comprising a printer having a plurality of ink channels, means for capturing target color data of at least one target color to be printed on a reflective substrate by the printer, and a data processing terminal within the printer or operably interfaced with the printer and configured by a set of commands, the set of commands comprising: receiving the captured target color data, wherein the captured target color data includes both spectral reflectance data and color data, including or excluding a specular component; generating a printer model; processing the captured color data with the printer model to output a preliminary dosing ratio for each ink channel representing an individual color component of the target color; computing a curve from each of the ratios and interpolating the computed curve to output a digital opacity value for the individual ink channels; combining the individual digital opacity values of all the ink channels to output a digital opacity value for the at least one target color; computing a natural opacity value from the captured spectral reflectance data; calculating a difference between the computed digital opacity value and the computed natural opacity value; And to interpolate the calculated difference for the ink phase of at least one target color diffusive ink component to obtain the predicted dosing ratio for each ink channel.

Embodiments of the system may further include a network in which the data processing terminal and the remote terminal are operably interfaced, wherein the captured target color data is received from the remote terminal. Regardless of the network connection, the means for capturing is preferably a spherical or multi-angle spectrophotometer.

The reflective substrate may be selected from the group comprising 'RSM (reflective, shiny and/or mirrored)' substrates, including metals such as aluminum, plastics such as films, foils and vinyl, and glass.

In a networked embodiment, the system may further include bridging means for interfacing the data processing terminal and/or spectrophotometer with a remote storage means, e.g., a cloud-based data storage node, wherein the color data is optionally encrypted prior to being uploaded to the remote storage means.

In a networked embodiment, captured target color data can be uploaded to a database from a data processing terminal, stored in the database, and processed by a cloud-based data processing node operatively configured to replace the stored color data in response to a color data request from a remote data processing terminal.

According to another aspect, a method of observing a target color matched and applied to a reflective substrate according to the present invention is provided in a light booth, wherein the viewing angle is related to the opacity level of a given color.

According to another aspect of the present invention, there is also provided a set of instructions, recorded on a data transmission medium or stored in a network storage medium, which, when read and processed by a data processing terminal, configures said terminal to perform the steps of any of the embodiments described herein.

Other aspects are as set forth in the claims of the present application.

To better understand the present invention and to show how it may be practiced, specific embodiments, methods and processes according to the present invention will now be described, by way of example only, with reference to the accompanying drawings.
FIG. 1 is a logic diagram of a digital printing system in a configurable networking environment according to the present invention, including a local data processing terminal, a remote data processing terminal, and a spot color table interfacing with a digital multichannel color printer.
FIG. 2 is a hardware diagram of a typical hardware architecture of the data processing terminal illustrated in FIG. 1, including a processor and memory means for storing a data processing command set.
Figure 3 is a logical diagram illustrating matching of target colors for identical printing on a reflective substrate with the digital printing system of Figures 1 and 2.
FIG. 4 is a flowchart representation of a conventional method for matching target colors as exemplified in FIG. 3.
FIG. 5 is a flow diagram representation of a general ink dosing method according to the present invention that can be used to improve the target color matching method of FIGS. 3 and 4, including a step of predicting ink separation including diffusion ink and a step of correcting the predicted ink separation.
FIG. 6 is a logical diagram of the contents of the memory means of the terminal illustrated in FIGS. 1 and 2, including a set of commands implementing a method as illustrated in FIG. 5.
Figure 7 is a flowchart representation of a first embodiment of the step of predicting ink separation of Figure 5.
Figure 8 is a flowchart representation of a second embodiment of the step of predicting ink separation of Figure 5, where the target color is matched from transparent to translucent to opaque.
FIG. 9 is a flowchart representation of an embodiment of a step for correcting ink separation predicted in FIG. 5, FIG. 7, and/or FIG. 8.
FIG. 10 is a flowchart representation of a method for automatically computing lighter and darker versions of a target color ink separation output by the method described herein.
FIG. 11 illustrates a viewing booth through which a user can perform a visual screening of one or more target color(s) printed according to the methods described herein.

Specific modes contemplated by the inventors will now be described by way of example. In the following description, numerous specific details are set forth to provide a thorough understanding. However, it will be apparent to those skilled in the art that the invention may be practiced without being limited to these specific details. In other instances, well-known methods and structures have not been described in detail so as not to unnecessarily obscure the description.

Referring now to the drawings, and primarily to FIGS. 1 and 2, an example of a configurable digital printing system (100) according to an embodiment of the data processing method of the present invention is illustrated. The digital printing system (100) includes a data processing terminal (110) and a spectrophotometer (117) locally interfaced with a digital multichannel color printer (114), in this example an inkjet printer having eight color channels (1161-8) comprising cyan (c), magenta (m), yellow (y), black (k), orange (o), green (g), violet (v), and white (w) as diffusion inks ('CMYKOGV+W'), via one or more high bandwidth data connections (112).

The digital printing system (100) is located within a networking environment, where the data processing terminal (110) is a personal computer device that uploads and downloads data encoded as digital signals over a high-bandwidth wired or wireless data connection (112), where such signals are relayed to or from the computer (110) by a local router device (118) implementing a wired local network operating in accordance with the IEEE 802.3-2008 Gigabit Ethernet transmission protocol and/or a high-bandwidth wireless local network operating in accordance with the IEEE 802.11 Wi-Fi wireless transmission protocol, respectively.

A typical hardware architecture of a data processing terminal (110), i.e., a desktop computer, is illustrated in more detail in FIG. 2 by way of non-limiting example. The computer comprises a data processing unit (201), a data output means such as a video display unit (VDU) (202), a data input means such as an HID device, typically a keyboard (203) and a pointing device (mouse) (204), as well as the VDU (202) itself if the VDU (202) is a touch screen display, and a data input/output means such as a wired or wireless network connection (112) to a local and wide area network via a router (118), a magnetic data carrier medium reader/writer (206), and an optical data carrier medium reader/writer (207).

Within the data processing unit (201), a central processing unit (CPU) (208) provides task coordination and data processing functions. Data and instruction sets for the CPU (208) are stored in memory means (209), and a hard disk storage unit (210) facilitates non-volatile storage of instructions and data. A wireless network interface card (NIC) (211) provides an interface for a network connection (112) with the router (118). One or more universal serial bus (USB) input/output interfaces (212) facilitate connection to a keyboard and pointing devices (203, 204). Depending on the presence or absence of network connection features of the printer (114) and the spectrophotometer (117), data communication between the computer (110), the printer (114) and the spectrophotometer (117) may be routed via the router device (118), via a wired connection to the computer's USB interface (212), or via a combination thereof.

All of the above components are connected to a data input/output bus (213), and a magnetic data transfer medium reader/writer (206) and an optical data transfer medium reader/writer (207) are also connected to the data input/output bus (213). A video adapter (214) receives CPU commands via the bus (213) to output processed video data to the VDU (202). All components of the data processing unit (201) are powered by a power supply unit (215) that receives power from the local mains and converts it according to the ratings and requirements of the components.

The router (118) itself is connected to a wide area network, one example of which is the Internet (120), via a conventional ADSL or fiber optic connection, through which digital data can be downloaded from and uploaded to remote data processing terminals. The network connections and interoperable networking protocols of each data processing terminal allow the terminals to connect to each other and communicate data with and receive data from each other according to the specific embodiment methodologies described herein.

The remote data processing terminal may be a desktop computer, as described with reference to FIG. 2, or may be a portable variant, such as a laptop or tablet computer. The remote data processing terminal may also include a personal communication device (130), such as a smart phone, that broadcasts and receives data encoded as digital signals—including voice and/or alphanumeric data—via wireless data transmissions (132), wherein the signals are relayed to and from each device (130) by a plurality of geographically closest communication link relays (134). The plurality of communication link relays (134 _1-N ) allow the digital signals to be routed between the mobile devices (130 _1-N ) and their corresponding devices via a remote gateway (136). The gateway (136) is a communication network switch that couples digital signal traffic between a wireless telecommunications network, such as the network through which the wireless data transmissions (132) occur, and a WAN (120).

A digital printing system (100) outputs a printed matter (140) having a reflective substrate, a well-known example of which is a beverage can made of aluminum, where the accuracy of color reproduction by the printer (114) in the output printed matter (140) is of the utmost importance. Accordingly, a spot or target color table (150) is provided to assist in correcting the color reproduction by the printer (114) through a process known in the art as color matching to a data processing terminal (110).

Referring to FIGS. 3 and 4, color matching on a reflective substrate using the system (100) has typically been performed as an iterative trial-and-error process. In addition to turning on the computer (110), loading its operating system, and then loading a raster image processor (RIP) application, the prior art method begins by characterizing the printer (114) of the digital printing system (100) at step (401), wherein the color and white undergo individual linearizations at steps (401A, 401B).

In step (402), a forward printer model is defined at the operator terminal (110) based on a specular component that can be configured with a forward conversion function defined by the ICC profile of the printer.

In step (403), at least a first target color (300) is defined in a color library, e.g., a spot or target color table (150) for a print job, which requires matching of a digital printing system representing the targeted print color (300) for color accuracy and color opacity. Capturing of the target color data is accomplished by measuring L*a*b values (310) of the target color of the table (150) with a spectrophotometer (117), including SPIN or SPEX, and storing the measurements as individual sets of values in the terminal (110).

Each target color (300) is composed of multiple colors, each color corresponding to an ink available from each ink channel (116 _1-8 ) of the printer (114). For each target color, a conventional printer model provides a set of ink channel-wise dosing ratios (330), expressed as a percentage for each ink for all ink combinations for the target color, known in the art as color separation data.

A color accuracy and opacity matching step follows, where the forward printer model is initiated and the captured and stored target colors (310) or their entire library are read in step (404), the or each target color is classified based on its measured opacity in step (405), and the classified color(s) are processed with a color prediction algorithm to output individual ink separation(s) (330) in step (406), for example using the ink dose technique taught in EP20178552.4.

The color chart is generated as output ink separation(s) in step (407) - where the color separation (330) includes opacity - and is then processed by the forward printer model and printed as a color chart (150) to the printer (114) in step (408).

Then, the printed color chart is measured with a spectrophotometer (117), including SPIN or SPEX, in step (409), and, depending on the magnitude of the difference observed with respect to the previously measured values for the target color in step (403), the color values in step (406) are typically corrected in step (410). This correction can be applied using a correction algorithm for ink color separation, e.g., an ink dose correction technique as taught in EP20178552.4, depending on the tolerance definition.

The iteration of the color accuracy and opacity matching steps (404 to 410) continues until the required values for the target color (300) are matched within an acceptable tolerance range or the color gamut limits of the printer for the solution are reached. In the example illustrated in FIG. 3, when the first color separation set (330 ₁ ) output in step (407) and printed on the color chart (150 _-3 ) is measured in step (409), it is determined that it does not correspond sufficiently closely to the target color (300), as exemplified by the reference number in the strikethrough font. In this example, the color values are corrected three times, whereby when the fourth color separation set (3304 ) output in step (407) and printed on the color chart (150 ) is measured in step (409), it is deemed to correspond sufficiently closely to the target color (300).

Then, the iterations of the printed color chart that are deemed to best match the target color(s) are exported as a digital library (330) of ink color separations in step (411). The exported digital library can then be used with a RIP application in a digital printing system (100) to print the artwork with the matching colors.

Through experience, the inventors have determined that the not insignificant number of iterations of the color accuracy and opacity matching steps (404 to 411) of the prior art is caused by the magnitude of the difference between the target color characteristics to be matched and the initial ink separation set (330 ₁ ) output in step (407). Through field knowledge and experimentation, the inventors have realized that a diffusion ink component of the color, such as white, silver, varnish, primer, etc. (herein, white (W) is used throughout this description to avoid unnecessary complexity, but may be replaced by any other diffusion ink) is used to compensate for the difference between the natural opacity of the target color (300 ) and the digital opacity in order to output an initial ink separation set (330 ₁ ) that is more accurate, i.e. corresponds more closely to the separations required to match the target color (300 ) and its opacity as compared to the prior art.

The present inventors have devised a technique for defining the opacity of a target color when printed conventionally or digitally on a reflective substrate based on measurements using a spherical or polygonal spectrophotometer (117), wherein the spectral reflectance with and without specular reflection (or a combination of different measurement angles) defines a target opacity value by, for example, subtracting a value representing the specular component excluded from a given color data of the target from a value representing the specular component included in a given color data of the target. The technique provides a method for matching the natural opacity of a target color with its associated digital opacity through the addition of a predetermined number of diffusing inks, such as white, silver, varnish, primer, etc.

Referring now to FIG. 5, a color matching process performed by the system (100) according to an embodiment of the ink dosing method according to the present invention is described, wherein similar reference numerals with respect to FIG. 3 represent similar steps.

In addition to turning on the computer (110) and loading a set of commands implementing its operating system and the associated data processing aspects of the color matching method, the printer (114) is initially re-characterized at step (401). The color is linearized at step (401A), and in the method of the present invention, the printer is subjected to correction using both the specular component and the specular reflection of, in this example, white ink, prior to the diffusion ink being linearized at step (401B) with a layer of diffusion ink applied to the substrate at step (501).

In step (402), a forward printer model is redefined at the operator terminal (110) based on the specular component. In step (502), an additional forward printer model is defined based on the specular component of the diffuse white layer applied to the substrate, and in step (503), the two forward printer models are combined to result in a single forward model combining the SPIN and SPEX spectral reflectances of each color.

In step (403), at least a first target color is defined in a color library, e.g., a spot or target color table (150), which requires matching with the digital printing system for color accuracy and color opacity. Capturing of the target color data is again accomplished by measuring the target color with a spectrophotometer (117), including SPIN and SPEX, and inputting the measurements as a set of individual values into the terminal (110). A target opacity is also defined in step (504), wherein the SPEX values are effectively subtracted from the SPIN values to define a degree of opacity therebetween as a percentage.

Then, the color accuracy and opacity matching step follows, where the forward printer model is initiated and in step (404), the stored target colors or their entire library are read and in step (405), each target color is classified according to its opacity level in step (504).

The classified color(s) are processed with a color prediction algorithm to output individual ink separation(s) at step (406), for example using the ink dose technique taught in EP20178552.4. At step (505), further described with reference to FIG. 8, or the or each classified color is further processed with an opacity prediction algorithm to output individual ink separation(s) comprising a diffuse ink. The opacity prediction algorithm relies on the natural opacity and the digital opacity values of the target colors. The opacity of each target color is computed using interpolation to the individual digital opacity curves of each component ink in the separation(s) of step (406) and then by summing the digital opacities of all the component inks. The natural opacity of each target color is computed from the SPIN/SPEX spectral reflectance data acquired at step (403). The opacity prediction algorithm computes the difference between the natural opacity value and the digital opacity value for the target color and interpolates this over the diffusion ink levels to output a prediction for the diffusion ink (e.g., white) ratio to impart its target opacity to the target color at print time.

Then, at step (407), a color chart is generated along with the output ink separation(s), which combines the color separation prediction of step (406) and the color opacity prediction of step (505). The generated color chart is processed by the forward printer model and printed at step (408) by the printer (114) onto a test substrate - which may be a reflective substrate of the intended application or, alternatively, an optically clear or transparent substrate that may be superimposed on the reflective substrate of the intended application.

Then, the printed color chart is measured with a spectrophotometer (117), including SPIN and SPEX, at step (410), and the color values are optionally corrected based on the observed differences with respect to a set of stored values for the target color. In this method, the color values are further corrected at step (506) based on the observed differences with respect to a set of stored values for the target opacity, the opacity correction algorithm being further described with reference to FIG. 9. In an embodiment of the method described below, the opacity correction algorithm relies on multiple interpolations of curves computed by the opacity prediction algorithm.

Iterations of the color accuracy and opacity matching steps (404 to 506) continue until the required values for the target color values are matched within an acceptable tolerance range or the color gamut limits of the printer for the solution are reached.

Then, the iterations of the printed color chart that are deemed to best match the target color(s) are exported as a digital library of ink color separations using diffusion inks in step (411), where the exported digital library generated by the method of the present invention can be used in conjunction with a RIP application in a digital printing system (100) to print the artwork on a reflective substrate in the matching colors.

Now, color data computing techniques are followed to implement steps (501 to 506) of the method according to the present invention as data processing commands, where the following notation is consistently used throughout for ease of understanding: the ink of the printer (114) of the example includes cyan (c), magenta (m), yellow (y), black (k), orange (o), green (g), violet (v) and white (w), and the color ink 'cmykog-v' is collectively represented by μ; the ink step is represented by α varying from 0 to 1, and the color wavelength is represented by λ, which is, for example, 400 ( ) to 700 ( ) and expresses values in nanometers (nm).

The spectral reflectance of the target color μ under specular reflection inclusion (SPIN) and specular reflection exclusion (SPEX) can be expressed as follows, respectively:

For example, there are 40 linear or non-linear interval ink steps ranging from 0 to 1, and for 10 nm increments, there are 31 linear interval wavelengths ranging from 400 to 700 nm. This is within μ. Total for each color of the dog dog size 2-D array of and It provides.

The natural opacity of the target color is computed from its SPIN/SPEX spectral reflectance data, where the areas under the SPIN and SPEX curves for the color ink are respectively given by:

The natural opacity of a cyan to violet colored ink is given by:

The area is computed numerically by element-wise division using the midpoint rule and the Hadamard product, but in an alternative embodiment, consider using Simpson's 1/3 ^rd , 3/8 ^th rules for greater accuracy.

The digital opacity of the target color is computed by interpolating the individual digital opacity curves of each component ink based on the individual ratio of each component ink in the target color, and then summing the digital opacity of all component inks. The digital opacity of a cyan to violet color ink is given by:

Here is the substrate opacity, calculated as the average of the opacity at 0% ink level for all colors except white, for example:

The ink opacity is curve-fitted. The composite curve is constructed from a weighted average of 5th and 6th degree polynomials, Fourier series, and sum-of-sines curve fits, where each component can be assigned an individual weight. The curve-fit equations are polynomial curve fit (degree 5) y ₁ , polynomial curve fit (degree 6) y ₂ , sum-of-sines curve fit (3 terms) y ₃ , and Fourier series sub-fit (3 terms) y ₄ , where the composite curve constructed from the above equations is given by:

Here w _k are the weights assigned to the individual curve fits: each ink has a different influence, and the formula here is averaged across inks, so the best fit is derived from the weighted average of the four.

Fitting the natural opacity curve according to equation (1) is given as follows:

The resulting digital opacity curve fit with the maximum function max applied to avoid negative digital opacity. is given as follows:

The best-fit opacity is obtained by subtracting the substrate opacity from the natural opacity of each color, with an opacity of 0 assigned whenever the calculation yields a negative value.

Now considering the diffuse ink component, the spectral reflectance of color white 'w' under specular reflection inclusion (SPIN) and specular reflection exclusion (SPEX) can be expressed as follows, where the subscripts (α and λ) represent the ink phase (from 0 to 1) and wavelength (typically in the range of 400 to 700 nm), respectively:

For example, there are 51 linear or non-linear interval ink steps ranging from 0 to 1, and for 10 nm increments, there are 31 linear interval wavelengths ranging from 400 to 700 nm. This is a size for the color white. 2-D array of and It provides.

To compute the natural opacity of white ink, the area under the SPIN/SPEX curve for white ink is given by:

In this example, the natural opacity of the white diffuse ink is given by the following equation, which uses the Hadamard product for element-wise division:

The digital opacity of a diffusion ink is given by subtracting the natural opacity of the white ink at 0% ink from the natural opacity, which is equivalent to:

Here, the minimum function min and the maximum function max are applied to avoid opacity percentages below 0% and above 100%:

Digital Opacity Wow ink step The functional relationship between the two is inverted for diffusion inks: the ink phase is matched to the opacity. Fitting the diffusion ink digital opacity curve according to equation (1) is given as follows:

Target color It consists of several colors from the color ink μ. The printer model uses the ink percentage of each color ink μ available from the ink separation data. Based on the ink separation, the digital opacity of each target color is provided by an individual digital opacity curve. Then, the digital opacity of all component inks can be found by interpolating the data. Meanwhile, SPIN/SPEX spectral reflectance data for each target color is also available, from which the natural opacity of the target color can be computed.

The underlying principle of the solution described herein is to compensate for the difference between the natural opacity of the target color and the digital opacity by using a diffusion or colorless ink, such as white, silver, primer, etc. The difference in opacity is thus interpolated for the diffusion ink step to obtain a first prediction for the diffusion ink (e.g. white in the example herein).

Based on the above, the prediction of the diffusion ink component ratio is computed as follows. Assuming there are N target colors, i.e. For each target color, About, SPIN and SPEX spectral reflectance and And:

Array here and are each am.

Each target color For , there is a color separation calculated from the printer model, which is displayed as follows:

Thus, for N target colors, this is the size Array of It provides.

The area under the target SPIN/SPEX curve is:

The natural opacity of the target color is given by:

The target color digital opacity is obtained from curve fitting in equation (2) as follows:

Difference between natural opacity and digital opacity of target color is the white ink percentage as follows: is compensated using:

In the embodiment, based on a threshold ε, the difference A preliminary correction can be applied between his calculations and his interpolations, where natural opacity Difference in opacity when less than ε is set to 0, and any negative opacity difference value is also set to 0, so that:

When the opacity difference is compensated for according to the above conditions, the function of the opacity difference is interpolated for the white ink step to obtain the first white ink prediction:

To be consistent with the notation below, the first white ink prediction for the i-th target color is is expressed by .

_The color matching techniques of the present invention described above are available in a variety of color matching processes that can be implemented as individual color data processing algorithms implemented in corresponding sets of commands that are processed by a computer (110) to dose ink across eight ink channels (116 1-8) of a printer (114), both of which are distributed and scalable across one or more nodes of a network.

Accordingly, with reference now to FIG. 6, when configured by a set of instructions implementing the color data processing techniques described herein, the contents of the memory means (209) of the computer (110) at runtime include an operating system as depicted in 601, for example, Windows 11™ distributed by Microsoft™ Inc. of Redmond, Washington, USA. The OS (601) includes instructions for managing basic data processing, interdependencies and interoperability of the computer hardware components as described with reference to FIG. 2, and communication subroutines (602) for configuring the computer (110) for bidirectional network communication via a NIC (211) interfacing with a wired connection (112) with a local router (118). The OS (601) also includes input subroutines for reading and processing input data, which comprises variously user direct inputs to human interface devices, namely a keyboard (203) and a computer mouse (204).

Next, a set of commands is illustrated at 604 for interfacing with the printer (114) and the spectrophotometer (117) via the OS (601) by means of one or more Application Programmer Interfaces (APIs) (605). The set of commands (604) includes and coordinates the data processing activities of additional functional data processing subroutines that implement the various functions and algorithms described herein, including a user interface (606) that is updated in real time and output to a display (202).

The command set (604) further maintains various data sets processed by its subroutines, including one or more ICC profiles of the printer (114), as shown in 607; a printer model based on the diffusion ink generated in step (502), as shown in 608; L*a*b data (310) for the or each target color (300) stored in step (403), as shown in 609; matrix, curve and interpolation data processed at runtime, as shown in 610; one or more ink separation sets (330 _1-N ) corresponding to individual dose amounts of the ink channels (116 _1-8 ) generated in steps (505 and 506), as shown in 611; and finally, at least one color library generated in step (411), as shown in 612, usable for printing the target color (300) on a reflective substrate in the printer (114).

Additional local data and network data that may be stored in the memory means (209) at runtime are also illustrated, some or all of which may be processed by or for the color matching application (604) and its subroutines or by other applications that are processed in parallel with the color matching application (604) at runtime. Examples of additional local data include spooling data for the printer (114) generated in response to a color chart print command prior to step (408); and/or L*a*b data (310) for the or each target color (300) acquired by the local spectrophotometer (117) and communicated to the terminal (110) via a USB connection to the USB module (212). Examples of network data include L*a*b data (310) for the or each target color (300) received from a remote device (130) over the WAN (120), to which the terminal (110) may respond with one or more ink separation set(s) (330) computed in accordance with the present invention; and/or remote application or OS update data communicated by a remote server over the WAN (120).

Referring now to FIGS. 7 and 8, each embodiment of the prediction step (505) is illustrated, each of which outputs an ink separation including a diffusion ink component for the or each target color.

The first embodiment illustrated in FIG. 7 is a simple implementation of the technique of the present invention, where the color printed on the diffusion substrate is set as the target color, and the method determines the amount of diffusion ink (e.g., white) required to match the target color when printed on the RSM substrate as previously described, and the technique uses the natural opacity of the target color. his digital opacity (via the addition of white/diffusion ink) The printed color is made as transparent (or opaque) as the target color by best matching it to the target color, so that when printed on an RSM substrate, the printed color will be perceived as the same as the target color printed on a white substrate.

The second embodiment illustrated in FIG. 8 refers to a 'TSO' (transparent - semi-opaque - opaque) color matching technique, where a color printed on a diffusion substrate is again set as a target color, and the method again determines the amount of diffusion ink (e.g., white) required to match the target color when printed on an RSM substrate as previously described, and the present embodiment determines the natural opacity of the target color according to a specified range. his digital opacity (via the addition of white/diffusion ink) The printed color is made more or less transparent (or opaque) by best matching it to the target color, so that when printed on the RSM substrate, the printed color will be perceived as being the same as, or more or less transparent (or opaque) than the target color printed on the white substrate.

Thus, in the second embodiment, the task is not to match the opacity of the printed color to the target color, but to compute the amount of diffusion ink, e.g., white, required to generate different versions of the target color given their different natural opacities. This advantageously allows the designer to evaluate how transparent or opaque different versions of the target color will appear when printed on the RSM substrate.

The range of natural opacity can be classified as follows: of the target color. The transparent version of the dog has a range of transparency can be expressed as my value; of the target color A translucent version of the dog can be expressed as a value within the semi-opaque range of the target color; The opaque version of the dog is the opaque range It can be expressed as my value. Target color Each of transparent, translucent and opaque colors It is expressed as a version of a dog (i.e., from transparent to opaque). (dog version).

The data required to predict TSO for N target colors can be taken from the previous section: , , , ,

The following sequence can be applied to obtain TSO from target color: Spectral Reflectance After this initial loading, the algorithm proceeds as follows:

spectral reflectance When this is loaded, the algorithm proceeds as follows:

Then, color separation data is loaded and the target color digital opacity is obtained from the curve fitting as previously defined in equation (5). The diffusion ink prediction is a total for each target color. There will be versions of each (T, S, O) (dog version). Therefore, the total number of diffusion ink predictions is am. Natural opacity for the dog version can be expressed as the difference between the natural opacity and the digital opacity of the kth version of the target color. is defined as . Then, the kth diffusion ink prediction is based on the following equations (6) and (7). is calculated as.

In general, an exception may be made for pastel colors, which are characterized by higher natural opacity. Pastel colors can be defined as a family of pale colors with high luminance values and low saturation. The terms 'high value' and 'low saturation' are intentionally relative terms. For these colors, the pastel flag It is used to classify pastel colors among other colors, where TSO is the total range from translucent to opaque, with all being located at a higher natural opacity or at a determinable threshold. It is expressed as a dog version.

A color's qualifier for whether it qualifies as a pastel is the Pastel Index, which ranges from 0 to 1 (or 0% to 100%). Based on the spectral reflectance of the color that can be scaled to , and under these conditions, constructs a degree of confidence in that qualification as a logic output function (true/1 or false/0). The following conditions C _n are provided as non-limiting examples and are considered suitable for providing reliable pastel indices.

Condition n ₁ may be related to the spectral reflectance of the color. One condition is and The color spectrum lying between the corresponding points of It may be that my multiple points n ₁ have a color spectrum close to the reference high brightness spectrum between the substrate minimum spectrum and the substrate maximum spectrum:

This condition is at least for the color spectrum. go and Forces that there must be an in-between.

Other conditions n ₂ is the color spectrum at The number of points that lie on the corresponding points of the color spectrum can be at least It forces the color to lie above the representative bright black (or gray) spectrum.

Other conditions A _up and A _down can represent the region between the color spectrum and the white spectrum and the region between the color spectrum and the light black spectrum, respectively. These conditions are that A _up and A _down are at least and It enforces that it must be a unit. Also, another condition can be non-A _up: A _down , which is the maximum It could be.

An additional set of conditions may be related to color space coordinates (L*a*b*, L*c*h*, L*u*v*, HSL, HSV). For example, in the CIE-L*a*b color space, the L* coordinate is at least In the HSL color space, the L ^hsl coordinate must be at least It should be in HSV color space. Coordinates are at least It should be in HSV color space. The coordinates are at most It should be in the L*c*h* color space. Coordinates are at least It should be in the Lch color space. The coordinates are at most It should be in the Luv color space. The coordinates are at most It should be in the Luv color space. The coordinates are at most It should be continued.

Another condition may be related to the average RGB ratio. This condition is based on the observation that the sRGB coordinates (R,G,B) of known pastel colors are similar in size. This condition requires that the average ratio κ be at least and up to It should be and can be defined as follows:

Additional criteria may be based on CMYKOGV or μ color separation. This criteria is based on the observation that known pastel colors have zero (or negligibly low) black ink in the μ color separation. Let μ be the % of black ink in the color separation. This condition is Go up to It forces you to do so.

If the index is 1 or 100%, all conditions are matched. However, the Pastel index threshold can be defined by the observer. The importance of conditions is not all the same, and different weights are assigned here. is assigned to each condition, and the pastel index Contains all constants used in conditions to determine the pastel-like quality of a color, according to:

Referring now to FIG. 9, an embodiment of a correction step (506) is illustrated, which outputs a corrected ink separation including a diffusion ink component for the or each target color. Under this technique, the predicted diffusion ink ratio is the difference between the natural opacity of the target color and the digital opacity. To compensate for this, the spectral reflectance within the SPIN and SPEX of the resulting colors is measured and color separation values are regenerated using the new SPIN and SPEX measurements, which provide the base data for calculating new natural opacity values.

Therefore, in step (506), the or each i-th target color is compensated based on the procedure outlined in the previous step, i.e., the diffusion/white ink percentage required for compensation between the natural opacity and the digital opacity of the target color. is calculated. The correction approach is based on the Euclidean distance between the printed color and the target color in any given color space (CIE-XYZ, L*a*b, xyz, etc.) as a value indicating how close the printed color is to the target color based on the first predicted diffuse ink component.

The International Commission on Illumination, commonly referred to as CIE, has defined a color difference that takes into account the effects of perceptual non-uniformity, commonly referred to as Delta-E, which expresses the difference between a printed color and a target color in the L*a*b* color space. However, the perception of color is also affected by the opacity of the color. Therefore, another useful measure of color difference is the difference between the natural opacity of the target color and the natural opacity of the printed color, which is referred to herein as Delta Opacity ('Delta-O'). The mathematical expressions for these two measurements are detailed below:

and Let be the CIE-L*a*b coordinates of two colors. The procedure for computing the color difference ΔE between two colors is given by the known Delta-E formula, such as Delta-E 2000.

Let be the natural opacity of the i-th target color (also called 'target opacity') according to equation (4) for calculating the natural opacity. Let be the natural opacity of the ith target color printed on the reflective substrate after the jth correction. This is also referred to as the 'measured opacity':

for example, When, is the natural opacity of the i-th target color after the 0th correction, i.e. in the first prediction. Then, Delta-O is defined as follows:

Predicted white ink Correction for Computing mainly relies on the opacity of the measured and printed colors. Then, the performance of the correction is evaluated through both the opacity difference and the color difference. The proposed equation is as follows:

Here κ is used to improve the equation and is the measured opacity and target opacity When they are numerically very close to each other, As approaches 1, the (j+1)th correction is defined in a way that is very close to the jth correction. The proposed equation is as follows:

Therefore, the embodiment of step (506) of FIG. 9 optimizes ink separation by including a diffusion ink prediction (330) output by any of the methods illustrated and described with reference to FIG. 5, FIG. 7 or FIG. 8.

The skilled reader will recognize that in analog printing systems, a target color is typically formulated using a variety of color base inks. These color base inks are similar to what μ defines for digital printing systems in conventional printing systems. The color base, represented as β _{c ,} is also known as a colorant or pigment, where the subscript 'c' represents the use of each colorant in a typical ink base system (e.g., 25 colorants). The techniques disclosed herein can be used to define the opacity of β _c , where μ can be simply replaced by β _c in the present description, which can advantageously enhance conventional ink formulation systems. The opacity prediction algorithm described herein can provide the user with opacity information along with the color to select the best matching recipe for the color. For example, step (401A) can be used to characterize β _c and step (401B) can be used to characterize a conventional diffusion ink. The natural opacity of β _c can be the curve fit according to equation (1), the resulting digital opacity curve according to equation (2), and the diffusion ink digital opacity according to equation (3). Equations (4) and (5) can be used to evaluate further corrections or selection of different recipes based on the measured Delta-E and Delta-0 differences.

Referring now to FIG. 10 , an operator running a conventional printer may require a physical representation (1000) of a target color that defines the production tolerance of a print assignment. The target color is typically a sample of an ink formulation, also known as an ink drawdown (1010), that is applied directly to a printing substrate (1000). The target ink drawdown (1010) is often accompanied by a darker version (1020) and a lighter version (1030), which correspond to thinner and heavier weights, respectively, of ink films of the target color applied with the same ink formulation to the same substrate (1000). In this context, the embodiment of the color matching method illustrated in FIG. 10 automatically computes, from the best matching output of step (506), lighter and darker representations of the target color at a particular distance ΔE in the L*a*b color space.

The L*a*b values of the target color and the substrate are known, together with their spectral reflectances, ideally of a black color printed on the same substrate on which the target color is measured. Tolerance thresholds for lighter and darker versions, e.g., by entering preset ΔE values and including potentially different values for each of the lighter and darker versions, will result in a spectral curve that is typically above (positively offset from) the spectral curve of the target color (higher luminance at coordinate L). A darker color (lower luminance at coordinate L) will typically result in a spectral curve that is below (negatively offset from) the spectral reflectance curve of the target color.

To obtain a brighter color (1030), the spectral reflectance curve of the target color can be increased in positive incremental steps, resulting in a spectral XYZ, XYZ Monitor ΔE through L*a*b and then finally is computed. If the difference between the target (desired) ΔE and the theoretical ΔE is within a specified tolerance, the processing loop can be terminated. The computed lighter color ink separation is very likely to be different from the target color. Therefore, the opacity of the lighter color generally requires a compensation according to equations (6) and (7). The darker color (1020) can be obtained by following the same methodology, but instead applying a negative offset increment.

However, shifting the spectral reflectance curve up or down by a constant offset across the curve can result in perceptually different colors that deviate significantly not only in the L coordinate but also in the a and b coordinates. To address this issue, the offset itself can be made a function of wavelength. A strategy is proposed in which the offset varies along the spectrum based on the distance of the spectral curve from the corresponding substrate and ultimately on the black spectral reflectance.

Due to the varying printing conditions such as ink, printing speed, indoor humidity and temperature, and tolerance of the measuring device, the theoretical L*a*b value of the target color is expected to be different from the theoretical L*a*b value of the target color printed on the substrate. Most importantly, in digital printing technology, in order to print a color using the μ-ink process, the target L*a*b coordinates are converted to the corresponding μ-ink color separation based on a given printer model or other printer calibration profile which is not unique to the color separation itself as previously described. Therefore, the theoretical target color and the printed target color are expected to have a non-zero ΔE between them.

This ΔE is automatically converted to the corresponding lighter and darker versions of the theoretical target color. Therefore, the ΔE obtained between the printed target and the printed lighter version is generally not exactly the same as the desired ΔE. To compensate for this deviation, an iterative correction procedure is required, as provided in this embodiment.

Target : To obtain the target spectrum for, the CIE-XYZ tristimulus coordinates of the theoretical target color T ^th are the spectral reflectance is computed using. For example, D50 is the light source spectral power distribution. It is used for conversion from CXYZ tristimulus values to CIE-L*a*b coordinates, along with the CIE-1931 standard observer color matching function. is used for. The L*a*b coordinates of the theoretical target color T ^{th are} Let us say that it is represented as . So the behavior of this section can be written as:

The spectral reflectances of brighter and darker colors can be expressed respectively as:

and Let be the spectral reflectances of the target, white, and black colors, respectively. The difference in the spectral curves is defined as:

is to raise or lower the spectrum curve. Let's say my percentage increment is between 0 and 1. Then, the offset function is defined as follows:

Spectral reflectance of the brighter/darker color spectrum is obtained by adding an offset corresponding to the target spectrum:

The conversion from spectrum to L*a*b uses the same procedure and utility functions detailed above. Expressing the brighter/darker spectrum as brighter/darker is as follows:

The library function Delta-E 2000 is used to compute the color difference between the target and the light/dark shade. The target vs. light/dark color difference is expressed as:

Color difference obtained by raising or lowering the target spectral curve within a specified tolerance Make sure it is the same color as the one you want:

, After it is saved, Using a conversion function such as , we obtain RGB coordinates of light/dark versions for plotting color samples for visual overview on a monitor display. The solution is finally printed to the system (100), which includes darker and lighter colors (1020, 1030) and a target color (1010) for evaluating color matching.

Viewing conditions are important components for a good evaluation of color matching and opacity matching of printed target colors. The light source, whether flat or cylindrical, has a strong influence on the angular scattering of the colors printed on the reflective substrate. Rotating the colors while maintaining the same lighting conditions provides the impression that the colors change when visualized from different viewing angles. In this context, normalized lighting and viewing systems that provide fairly standardized viewing conditions are developed and adopted by experts.

Figure 11 illustrates an example of a color observation station (1100) for observing printed colors or color printed objects (flat, cylindrical or shaped) under normalized direct or indirect light (1120) within a light booth (1130). The light booth is typically configured with a shielding wall (1140) to prevent light contamination of the normalized light (1120) by other adjacent light sources such as office lamps, windows, etc.

A color or object placeholder (1150) positionable within a light booth under normalized illumination (1120) comprises a support plane (1152) having an upper surface facing the light source (1120), the support plane being rotatable by a manual or automatic crank member (1154) to facilitate observation of the color under different viewing angles, reflected or non-reflected conditions.

The observation station (1100) preferably has a color matching opacity (e.g., or ) and a viewing angle θ are configured to correlate under normalized lighting conditions (e.g., D50 or D65), where θ is the positive angle of the placeholder support plane (1152) measured from the light booth axis plane, which is ideally perpendicular to the light direction (1122) of the light source (1120). Repeated observation tests have shown that the range of opacity values between transparent and opaque is typically 15 to 90, respectively. Therefore, preferably, the same value is used to define the range of θ, where the range is substantially 90°. For observation conditions suitable for transparent colors evaluated under specular observation conditions, as opposed to opaque colors observable under diffuse lighting conditions, for example, at 15°, it is practically Starting from, for example, 0° provides similar observation conditions as 90°.

Since the observer (1160) is positioned at a certain distance from the placeholder, the color to be evaluated is visually evaluated from above at a viewing angle according to transparency, where the support surface (1152) is then rotated so that each successive color of the sample can be evaluated.

The opacity range of the color can be classified according to the output of the transparent-translucent-opaque embodiment, thereby simplifying the color evaluation procedure within the range. For example, a transparent color is Opacity of the range, Translucency from, and have opaque colors from , and they are It falls within the range.

Accordingly, the present invention provides a computer-implemented method for dosing ink in a digital multichannel printer in a single workflow when printing on a reflective substrate, embodiments of which further provide a method for easily calibrating and adjusting a value representing an initially computed spread ink dose.

The present disclosure advantageously simplifies and accelerates the color matching process for professional users, including through the diffusion effect, the first ink separation set (330), which obviously matches the target color (300) much more closely than prior art techniques. The benefits of this technique extend to optical checks by professional users; a color chart printed using the first ink separation set (330) computed according to the present invention and/or the subsequent ink separation set corrected according to the present invention can be printed on an optically clear or transparent layer substrate, which can be superimposed on a printable blank of a reflective substrate, for evaluation by the user, e.g., in a viewing booth (1100).

It has been observed that overlapping a transparent layer (plastic of a certain thickness) affects the angular scattering of incident light. Therefore, the transparent layer can be considered to be somewhat of a part of the diffusion factor based on the total amount of ink deposited in the printing process. Additionally, the ink deposited on the transparent layer can be applied to the front or back of the substrate, enabling the differential gloss effect characteristic of the transparent layer.

A network distributed embodiment of the computer implemented method disclosed herein is contemplated, wherein the predictive aspect of the method is augmented by machine learning algorithm(s) trained to detect changes in the printing system and compensate for predicted ink separation (330) based on the detected changes.

Such machine learning techniques can be developed from a collection of data collected over time that represents color and opacity variations in a printing system. This data can be based on variables such as ink or substrate variation from production batch to production batch, variation in the calibration of a spectrophotometer, wear and tear on digital printing system components such as print heads, etc., all of which are known to affect the consistency of final printed color from batch to batch over time.

In a specific example, particularly relevant to RSM substrate printing, the use of colorless inks in printing systems introduces larger and heavier particles compared to the pigment particles of other color inks: white inks use titanium dioxide particles which are difficult to keep suspended in the ink formulation vehicle, ultimately causing loss of ink density or clogging of the print head, and where these colorless inks require permanent or semi-permanent stirring.

Using machine learning techniques in conjunction with this method is expected to continuously optimize predictions, because while printing systems are regularly re-linearized and/or recalibrated, the corresponding data is only used to perform the job; therefore, capturing, aggregating and auditing this data to find patterns can continuously improve the predictive model for the color and opacity matching process, for example by using a gradient descent algorithm, a first-order iterative optimization to find local minima of differentiable functions.

In this specification, the terms "comprise, comprise, comprised and comprising" or variations thereof and the terms "include, includes, included and including" or variations thereof are to be considered entirely interchangeable and are all to be accorded the broadest possible interpretation, and vice versa.

The present invention is not limited to the embodiments described above, but may vary both in configuration and details.

Claims (23)

A computer-implemented method for dosing ink in a printing device having multiple ink channels when printing on a reflective substrate, the method comprising:
A step of capturing target color data of at least one target color to be printed on the reflective substrate, wherein the target color data includes both color data and spectral reflectance data including and excluding a specular component.
Steps to create a printer model;
Processing the captured color data with the printer model to output a preliminary dosing ratio for each ink channel representing an individual color component of the target color;
A step of computing a curve from each ratio and interpolating the computed curve to output a digital opacity value for each ink channel;
A step of combining individual digital opacity values of all ink channels to output a digital opacity value for at least one target color;
A step of computing a natural opacity value from the captured spectral reflectance data;
a step of calculating the difference between the computed digital opacity value and the computed natural opacity value; and
interpolating said calculated difference for ink steps of said at least one target color diffusion ink component to output a predicted dosing ratio for each ink channel;
Computer implementation method. In the first paragraph,
Comprising an additional step of calibrating the printer using both specular and specular components on the layer of diffusion ink applied to the substrate;
Computer implementation method. In the second paragraph,
The step of generating the printer model further includes the step of generating the printer model based on the specular component and the specular component on the diffusion ink layer applied to the substrate.
Computer implementation method. In the third paragraph,
A further step of defining a target opacity value by subtracting a value representing the exclusion of a specular component of the target color data from a value representing the inclusion of a specular component of the target color data,
Computer implementation method. In the fourth paragraph,
an additional step of classifying the target color or each target color according to the target opacity defined above; and
A further step of repeating a processing step of replacing the captured color data with data representing the classified color or each classified color,
The output predicted dosing ratio for each ink channel includes a dosing ratio for the diffusion ink.
Computer implementation method. In clause 5,
An additional step of adding the predicted diffusion ink ratio to the natural opacity value to output the next natural opacity value; and
comprising an additional step of calculating the difference between said natural opacity value and said next natural opacity value;
The above calculated difference is a value referred to as delta opacity for the reflective substrate.
Computer implementation method. In Article 6,
An additional step of correcting the predicted dosing ratio for each ink channel by a composite function stored as a lookup table is included.
Computer implementation method. In any one of paragraphs 1 to 7,
An additional step of compensating said calculated difference prior to said interpolation step, wherein said compensating step comprises a step of setting said calculated difference to 0 when said natural opacity value is below a predefined threshold or when said calculated difference value is negative.
Computer implementation method. In any one of claims 1 to 8,
The above target color is formulated by a conventional ink formulation system having a specialized colorant, a base ink set and a conventional diffusion ink, wherein the method includes an additional step of specifying an opacity level at each ink step.
Computer implementation method. In any one of claims 1 to 9,
An additional step of inputting a range of natural opacity values, wherein each output predicted dosing ratio comprises a diffusion ink percentage corresponding to a target color having a digital opacity selected from a range from transparent to opaque, including semi-opaque.
Computer implementation method. In Article 10,
An additional step of defining a set of conditions based on color characteristics associated with pastel colors;
An additional step of filtering the captured target color data by the above defined set of conditions; and
An additional step of classifying the target color data as a pastel color according to the above filtering is included.
Computer implementation method. In any one of claims 1 to 11,
an additional step of setting at least one tolerance threshold representing a distance from the coordinates of the target color in the L*a*b color space; and
An additional step of computing a predicted dosing ratio for each ink channel for a lighter version and a darker version of the target color, respectively, based on the tolerance threshold.
Computer implementation method. In any one of claims 1 to 12,
an additional step of generating a color chart from the predicted dosing ratio for each ink channel; or
An additional step of generating a color chart from the above-mentioned corrected dosing ratio for each ink channel; and
Comprising an additional step of printing said color chart with said printer on a test substrate;
Computer implementation method. In Article 12,
The above test substrate is transparent,
Computer implementation method. In clause 13 or 14,
An additional step of constructing a color observation station having a normalized light source and a support plane with a variable field of view;
an additional step of positioning a printed substrate on the support plane; and
comprising an additional step of orienting the viewing angle of the support plane according to the computed opacity level of the target color;
Computer implementation method. In Article 15,
The computed opacity level of the target color is in the range of 15° representing a substantially transparent target color to 90° representing a substantially opaque target color.
Computer implementation method. As a digital printing system,
A printing device having a plurality of ink channels, means for capturing target color data of at least one target color to be printed on a reflective substrate by the printing device, and a data processing terminal within the printing device or operably interfaced with the printing device and configured by a set of commands, the set of commands comprising:
Receiving captured target color data, wherein the captured target color data includes both spectral reflectance data and color data including or excluding a specular component;
Create a printer model;
Processing the captured color data with the printer model to output a preliminary dosing ratio for each ink channel representing an individual color component of the target color;
Computing a curve from each ratio and interpolating the computed curve to output a digital opacity value for each ink channel;
Combining the individual digital opacity values of all ink channels to output a digital opacity value for at least one target color;
Computing natural opacity values from the captured spectral reflectance data;
Computing the difference between the computed digital opacity value and the computed natural opacity value; and
For interpolating said calculated difference for the ink phase of said at least one target color diffusive ink component to output a predicted dosing ratio for each ink channel,
Digital printing system. In Article 17,
The means for capturing the above is a spherical or multi-angle spectrophotometer,
Digital printing system. In clause 17 or 18,
The system further comprises a network through which the data processing terminal and the remote terminal are operably interfaced, wherein the captured target color data is received from the remote terminal.
Digital printing system. In Article 19,
Further comprising a bridging means for interfacing said data processing terminal and/or said spectrophotometer with a remote storage means; optionally, said data processing terminal is further configured to encrypt said captured color data before uploading said captured color data to said remote storage means.
Digital printing system. In Article 20,
The above remote storage means is a data processing node configured to process the stored color data and output a predicted dosing ratio for each ink channel according to a color data request from the remote data processing terminal or other device.
Digital printing system. In any one of Articles 17 to 21,
The above reflective substrate is selected from the group including 'RSM (reflective, shiny and/or mirrored)' substrates, including metals such as aluminum, plastics such as films, foils and vinyl, and glass.
Digital printing system. A set of commands recorded on a data transmission medium or stored in a network storage medium, which configures a data processing terminal to perform the steps of a method according to any one of claims 1 to 16 when read and processed by said terminal.