Identifying problematic nozzles in a digital printing system

The system uses image processing to evaluate nozzle performance in digital printing systems by computing scores from test marks, effectively identifying and addressing nozzle issues, thus enhancing print quality and productivity.

WO2024228074A1PCT designated stage expired Publication Date: 2024-11-07LANDA

Patent Information

Application Number
PCT/IB2024/053417
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2023-05-01
Filing Date
2024-04-08
Publication Date
2024-11-07

AI Technical Summary

Technical Problem

Digital printing systems face challenges in effectively identifying and diagnosing problematic nozzles, which can lead to inconsistent print quality and reduced productivity due to the lack of efficient methods for nozzle evaluation and malfunction detection.

Method used

A system and method utilizing a processor to print test marks on a substrate, processing images to compute performance scores for nozzles based on properties such as darkness consistency, positional accuracy, and relative intensity, and comparing these scores to predefined thresholds to identify and output problematic nozzles.

Benefits of technology

Enables accurate identification and diagnosis of problematic nozzles, improving print quality and productivity by providing a systematic approach to nozzle evaluation and malfunction detection.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure IB2024053417_07112024_PF_FP_ABST
    Figure IB2024053417_07112024_PF_FP_ABST
Patent Text Reader

Abstract

A system (10) includes, multiple print nozzles (99), and a processor (20), which is configured to: (i) using the print nozzles, print, on a substrate (50, 100), respective test marks (102, 102a) for use in evaluating the print nozzles by processing an image of the substrate, and (ii) using a subset of the print nozzles, print, on the substrate, respective orientation marks (112, 112a, 112b, 112c, 112d) arranged in one or more orientation lines (118), a distance between successive ones of the orientation marks varying across each of the orientation lines, such that the orientation marks facilitate identifying the print nozzle corresponding to any given one of the test marks.
Need to check novelty before this filing date? Find Prior Art

Description

[0001] IDENTIFYING PROBLEMATIC NOZZLES IN A DIGITAL PRINTING SYSTEM

[0002] CROSS-REFERENCE TO RELATED APPLICATIONS

[0003] This application claims the benefit of U.S. Provisional Patent Applications 63 / 499,266, and 63 / 499,267, both filed May 1, 2023, whose disclosure is incorporated herein by reference.

[0004] FIELD OF THE INVENTION

[0005] The present invention is related generally to the field of digital printing, and particularly to quality control in digital printing systems.

[0006] BACKGROUND OF THE INVENTION

[0007] U.S. Patent 11,001,077 describes a method that identifies at least one malfunctioning nozzle in a digital printing press, the digital printing press including a plurality of nozzles. The method includes printing a design on a substrate, acquiring at least one image of the printed design, and identifying at least one artifact in the acquired image. The method further includes identifying the malfunctioning nozzle and classifying the at least one malfunctioning nozzle according to the at least one of the acquired images of the printed design, at least a portion of a nozzle pattern and at least a portion of a uniformity pattern.

[0008] U.S. Patent Application Publication 2018 / 0086049 describes an image inspection device that analyzes data of a first read image obtained by reading a defective nozzle detection pattern, which is recorded in a first region of a recording medium by a single pass ink jet printing apparatus, to detect a defective nozzle of a line-type ink jet bead used to record the defective nozzle detection pattern, and stores a history of the detection result of the defective nozzle in a history information storage unit. The image inspection device analyzes data of a second read image of a printed image recorded in a second region of the recording medium to detect an image defect of the printed image and collates information about the detected image defect with history information stored in the history information storage unit to specify a defective nozzle causing the image defect.

[0009] SUMMARY OF THE INVENTION

[0010] An embodiment of the present invention that is described herein provides a system including multiple print nozzles and a processor. The processor is configured to print on a substrate, using the print nozzles, respective test marks for use in evaluating the print nozzles by processing an image of the substrate, and, using a subset of the print nozzles, print, on the substrate, respective orientation marks arranged in one or more orientation lines, a distance between successive ones of the orientation marks varying across each of the orientation lines, such that the orientation marks facilitate identifying the print nozzle corresponding to any given one of the test marks.

[0011] In some embodiments, the orientation lines include a first orientation line preceding the test marks and a second orientation line following the test marks.

[0012] In some embodiments, the test marks are arranged along N lines, each of which is parallel to an axis of the orientation lines, and the subset includes: a first series of the print nozzles corresponding to those of the test marks having first indices kl*dl for dl > 2N and kl = 1 . . ,floor(M / dl), M being a total number of the test marks, and, a second series of the print nozzles corresponding to those of the test marks having second indices k2*d2 for d2 > dl and k2 = 1 . . ,floor(M / d2), any one of the test marks being a (floor((j-l) / N)+l)t^1one of the test marks in a ((j- 1)%N+I)t^1one of the lines.

[0013] In some embodiments, dl and d2 are prime.

[0014] There is further provided, in accordance with some embodiments of the present invention, a system including multiple print nozzles and a processor. The processor is configured to print on a substrate, using the print nozzles, respective test marks for use in evaluating the print nozzles by processing an image of the substrate, the test marks being arranged in N lines, and using the print nozzles, print respective duplicates of at least some of the lines so as to facilitate the evaluation. The duplicates include a duplicate of an N^1one of the lines preceding a first one of the lines, and a duplicate of the first one of the lines following the N^1one of the lines.

[0015] In some embodiments, the duplicates further include: a duplicate of an (N-l)^ one of the lines preceding the duplicate of the N^1one of the lines, and a duplicate of a second one of the lines following the duplicate of the first one of the lines.

[0016] There is further provided, in accordance with some embodiments of the present invention, a system including a display and a processor. The processor is configured to receive an image of a substrate on which multiple test marks were printed using different respective print nozzles, to compute respective inconsistency scores for the print nozzles by, for each of the test marks, computing respective differences between pairs of adjacent pixels, in the image, corresponding to the test mark, and based on the differences, computing the inconsistency score for the print nozzle that was used to print the test mark, to compare the inconsistency scores to a predefined threshold, and to output on the display, based on the comparing, an output indicating those of the print nozzles whose inconsistency scores pass the predefined threshold.

[0017] In some embodiments, the image includes multiple color channels, and the processor is configured to compute the differences by, for each of the pairs of adjacent pixels: computing respective per-channel differences, for the color channels, between the pair of adjacent pixels, and computing the difference as a maximum of the per-channel differences.

[0018] In some embodiments, the processor is configured to compute the inconsistency score by: based on the differences, computing a representative difference, and computing the score based on the representative difference.

[0019] In some embodiments, the representative difference is an average of (i) a largest one of the differences, and (ii) a (k-1 )^- largest one of the differences.

[0020] There is further provided, in accordance with some embodiments of the present invention, a system including a display and a processor. The processor is configured to receive an image of a substrate on which multiple test marks were printed using different respective print nozzles, to compute respective locations of the test marks, along a cross-print axis, in the image, to compute, based on the locations, respective positional-deviation scores, which indicate respective positional deviations along the cross-print axis, for the test marks, to compare the positional-deviation scores to a predefined threshold, and to output on the display, based on the comparing, an output indicating those of the print nozzles that were used to print those of the test marks whose positional-deviation scores pass the predefined threshold.

[0021] In some embodiments, the test marks are arranged in multiple lines, each of which extends along the cross-print axis, and for any given one of the test marks, the processor is configured to compute the positional- deviation score for the given test mark based on: the location of the given test mark, and the respective locations of others of the test marks having the same relative position as the given test mark in s > 1 of the lines immediately preceding the line of the given test mark and s of the lines immediately following the line of the given test mark.

[0022] In some embodiments, s = 2.

[0023] In some embodiments, the processor is configured to compute the positional-deviation score for the given test mark by: computing multiple expected locations of the given test mark, based on the respective locations of the others of the test marks, computing respective absolute differences between the expected locations and the location of the given test mark, and computing the positional-deviation score based on the absolute differences.

[0024] In some embodiments, the processor is configured to compute the positional-deviation score based on a minimum of the absolute differences.

[0025] In some embodiments, the test marks are arranged in multiple lines, each of which extends along the cross-print axis, and the processor is configured to compute the locations by: converting the image to a grayscale image, smoothing the grayscale image along a print axis, and for each of the lines in the smoothed grayscale image: identifying pixels whose values are local minima along the cross-print axis, and for each of the identified pixels: fitting a parabola to respective values of the identified pixel, at least one other pixel immediately preceding the identified pixel, and at least one other pixel immediately following the identified pixel, and computing one of the locations as a location of a vertex of the parabola.

[0026] In some embodiments, the processor is configured to compute the locations by, prior to identifying the pixels whose values are the local minima, smoothing the grayscale image along the cross-print axis.

[0027] There is further provided, in accordance with some embodiments of the present invention, a system including a display and a processor. The processor is configured to receive an image of a substrate on which multiple test marks were printed using different respective print nozzles, to compute respective intensities of the test marks, to compute respective relativeintensity scores for the test marks, each of the relative-intensity scores quantifying the intensity of the test mark relative to one or more others of the intensities, to compare the relative-intensity scores to a predefined threshold, and to output on the display, based on the comparing, an output indicating those of the print nozzles that were used to print those of the test marks whose relative-intensity scores pass the predefined threshold.

[0028] In some embodiments, for any given one of the test marks, the processor is configured to compute the relative-intensity score for the given test mark by: computing respective intensity ratios between the intensity of the given test mark and the intensities of others of the test marks, and computing the relative-intensity score based on the intensity ratios.

[0029] In some embodiments, the test marks are arranged in multiple lines, each of which extends along a cross-print axis, and the others of the test marks include: the test mark immediately preceding the given test mark in the line of the given test mark, and the test mark immediately following the given test mark in the line of the given test mark.

[0030] In some embodiments, the test marks are arranged in multiple lines, each of which extends along a cross-print axis, and the others of the test marks include: the test mark having the same relative position as the given test mark in one of the lines preceding the line of the given test mark, and the test mark having the same relative position as the given test mark in one of the lines following the line of the given test mark.

[0031] In some embodiments, the test marks are arranged in multiple lines, each of which extends along a cross-print axis, and the others of the test marks include those of the test marks that are most correlated, with the given test mark, with respect to a spread along the cross-print axis, the spread being a function of a resolution of a camera that acquired the image. In some embodiments, the test marks are arranged in multiple lines, each of which extends along a cross-print axis, and the processor is configured to compute the intensities by: converting the image to a grayscale image, smoothing the grayscale image along a print axis, and for each of the lines in the smoothed grayscale image: identifying pixels whose values are local minima along the cross-print axis, and computing the intensities of those of the test marks in the line based on the values.

[0032] In some embodiments, the processor is configured to compute the intensity of any given one of the test marks in the line based on an average of: the local minimum corresponding to the given test mark, respective values of one or more pixels immediately preceding the local minimum in the line of the given test mark, and respective values of one or more pixels immediately following the local minimum in the line of the given test mark.

[0033] There is further provided, in accordance with some embodiments of the present invention, a system including a display and a processor. The processor is configured to receive at least one image of at least one substrate on which multiple sets of test marks were printed, in different respective colors, using different respective sets of print nozzles, to compute respective performance scores for the test marks, to normalize the performance scores across the colors, to compare the normalized performance scores to a predefined threshold, and to output on the display, based on the comparing, an output indicating those of the print nozzles corresponding to those of the test marks whose performance scores pass the predefined threshold.

[0034] In some embodiments, the processor is configured to normalize the performance scores by: for any given one of the test marks in any given one of the sets of test marks, identifying the performance scores of those of the test marks having the same index as the given test mark in others of the sets of test marks, and normalizing the performance score of the given test mark based on the identified performance scores.

[0035] In some embodiments, the processor is configured to normalize the performance score of the given test mark by: identifying a percentile of a set of the performance scores that includes (i) the performance score of the given test mark and (ii) the identified performance scores, and setting the normalized performance score of the given test mark equal to ±d, d being a difference between the percentile and the performance score of the given test mark.

[0036] There is further provided, in accordance with some embodiments of the present invention, a method including, using multiple print nozzles, printing, on a substrate, respective test marks for use in evaluating the print nozzles by processing an image of the substrate, and using a subset of the print nozzles, printing, on the substrate, respective orientation marks arranged in one or more orientation lines, a distance between successive ones of the orientation marks varying across each of the orientation lines, such that the orientation marks facilitate identifying the print nozzle corresponding to any given one of the test marks.

[0037] There is further provided, in accordance with some embodiments of the present invention, a method including, using multiple print nozzles, printing, on a substrate, respective test marks for use in evaluating the print nozzles by processing an image of the substrate, the test marks being arranged in N lines, and using the print nozzles, printing respective duplicates of at least some of the lines so as to facilitate the evaluation, the duplicates including a duplicate of an N^1one of the lines preceding a first one of the lines, and a duplicate of the first one of the lines following the N^1one of the lines.

[0038] There is further provided, in accordance with some embodiments of the present invention, a method including receiving an image of a substrate on which multiple test marks were printed using different respective print nozzles, computing respective inconsistency scores for the print nozzles by, for each of the test marks, computing respective differences between pairs of adjacent pixels, in the image, corresponding to the test mark, and, based on the differences, computing the inconsistency score for the print nozzle that was used to print the test mark, comparing the inconsistency scores to a predefined threshold, and based on the comparing, outputting an output indicating those of the print nozzles whose inconsistency scores pass the predefined threshold.

[0039] There is further provided, in accordance with some embodiments of the present invention, a method including receiving an image of a substrate on which multiple test marks were printed using different respective print nozzles, computing respective locations of the test marks, along a cross-print axis, in the image, based on the locations, computing respective positional-deviation scores, which indicate respective positional deviations along the cross-print axis, for the test marks, comparing the positional-deviation scores to a predefined threshold, and based on the comparing, outputting an output indicating those of the print nozzles that were used to print those of the test marks whose positional-deviation scores pass the predefined threshold.

[0040] There is further provided, in accordance with some embodiments of the present invention, a method including receiving an image of a substrate on which multiple test marks were printed using different respective print nozzles, computing respective intensities of the test marks, computing respective relative-intensity scores for the test marks, each of the relativeintensity scores quantifying the intensity of the test mark relative to one or more others of the intensities, comparing the relative-intensity scores to a predefined threshold, and based on the comparing, outputting an output indicating those of the print nozzles that were used to print those of the test marks whose relative-intensity scores pass the predefined threshold.

[0041] There is further provided, in accordance with some embodiments of the present invention, a method including receiving at least one image of at least one substrate on which multiple sets of test marks were printed, in different respective colors, using different respective sets of print nozzles, computing respective performance scores for the test marks, normalizing the performance scores across the colors, comparing the normalized performance scores to a predefined threshold, and based on the comparing, outputting an output indicating those of the print nozzles corresponding to those of the test marks whose performance scores pass the predefined threshold.

[0042] The present invention will be more fully understood from the following detailed description of embodiments thereof, taken together with the drawings, in which:

[0043] BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Fig. 1 is a schematic side view of a digital printing system, in accordance with some embodiments of the present invention;

[0045] Fig. 2 is a schematic illustration of a substrate, in accordance with some embodiments of the present invention;

[0046] Fig. 3 is a schematic illustration of a portion of an image of a substrate corresponding to a test mark, in accordance with some embodiments of the present invention;

[0047] Fig. 4 is a schematic illustration of a portion of an image of a substrate, in accordance with some embodiments of the present invention;

[0048] Fig. 5 is a flow diagram for a method for computing the locations of test marks, in accordance with some embodiments of the present invention; and

[0049] Fig. 6 shows the values of a number of pixels along a line.

[0050] DETAILED DESCRIPTION

[0051] OVERVIEW

[0052] Embodiments of the present invention provide image-processing algorithms for identifying problematic print nozzles in a digital printing system.

[0053] In particular, the processor controlling the print system causes at least some of the nozzles to print respective test marks on a substrate. Subsequently, the processor receives an image of the substrate, and by processing the image, computes one or more performance scores for each of the nozzles that printed a test mark, based on the properties of the test mark. In response to the performance score of a nozzle exceeding or failing to exceed (depending on the convention) a predefined threshold, the processor may output a suitable warning indicating that the nozzle is problematic.

[0054] For example, one type of performance score may quantify the darkness consistency (i.e., uniformity) with which the nozzles print. To calculate this performance score for any given nozzle, the processor may compare adjacent pixels along the length of the corresponding test mark (i.e., the test mark printed by the given nozzle), checking the extent to which these pixels vary in their values.

[0055] Another type of performance score may quantify the positional accuracy with which the nozzles print. To calculate this performance score for any given nozzle, the processor may calculate an expected location, in the image, of the corresponding test mark, based on the respective locations of nearby test marks. Subsequently, the processor may compare the actual location of the test mark to the expected location.

[0056] Another type of performance score may quantify the relative darkness with which the nozzles print. (A relatively low darkness level may indicate, for example, that the nozzle lacks sufficient ink.) To calculate this performance score for any given nozzle, the processor may compare the darkness of the corresponding test mark to the darkness of nearby test marks.

[0057] Typically, the printing system comprises multiple arrangements of nozzles, each arrangement being configured to print in a different respective color. In such embodiments, a different respective set of test marks may be printed by each arrangement of nozzles. Prior to comparing the performance score of any given nozzle to the relevant threshold, the processor may normalize the score relative to nozzles of other colors having the same relative position, in their respective arrangements, as the given nozzle.

[0058] SYSTEM DESCRIPTION

[0059] Reference is initially made to Fig. 1, which is a schematic side view of a digital printing system 10, in accordance with some embodiments of the present invention.

[0060] System 10 is configured to print any content, such as a digital image 42, on a target substrate, which may be in the form of individual sheets 50 or a continuous web. The target substrate may comprise, for example, paper, a folding carton, or a multilayered polymer.

[0061] System 10 comprises a flexible, looped intermediate transfer member (ITM) 44, which cycles through a fluid- applying station 60, a drying station 64, an impression station 84, and a treatment station 52.

[0062] ITM 44 may comprise any suitable material and any suitable number of layers. During the installation of system 10, the edges of the material are adhered to one another, at a seam 45, so as to form a loop. An example of a method and a system for the formation of seam 45 is described in detail in U.S. Patent Application Publication 2020 / 0171813, whose disclosure is incorporated herein by reference.

[0063] System 10 further comprises an ITM module 70, which is configured to guide the cyclic movement of the ITM.

[0064] Typically, ITM module 70 comprises a roller 76 and one or more rollers 78, the configuration and operation of which are described, for example, in U.S. Patent Application Publication 2017 / 0008272 and in PCT International Publication WO / 2013 / 132424, whose respective disclosures are incorporated herein by reference. Typically, ITM module 70 further comprises a powered tensioning roller, also referred to herein as a dancer assembly 74, whose movement is schematically represented in Fig. 1 by a double-sided arrow. Dancer assembly 74 is configured to control the slack in ITM 44 so as to compensate for any stretching of ITM 44 with age. In some embodiments, dancer assembly 74 is motorized.

[0065] In some embodiments, ITM module 70 further comprises a tension drive roller 98 and a control drive roller 79, which are powered by respective motors, which are typically electric. Tension drive roller 98 and control drive roller 79 are configured to rotate about their respective axes.

[0066] In some embodiments, ITM module 70 further comprises one or more tension sensors (not shown), which may be integrated with ITM 44 at various positions along the ITM, or may be external to the ITM. The tension sensors are configured to generate signals indicative of the mechanical tension applied to the ITM.

[0067] In some embodiments, at least one of rollers 78 comprises a motion encoder (not shown) configured to record the position of ITM 44 so as to facilitate control of the cycling. For example, the motion encoder may comprise a rotary encoder configured to produce rotary-based position signals indicative of an angular displacement of the roller.

[0068] Alternatively or additionally, one or more motion encoders may be integrated with other components of ITM module 70 or with the ITM itself. Examples of a motion encoder integrated with the ITM are described, for example, in PCT International Publications WO / 2021 / 044303 and WO / 2020 / 003088, whose respective disclosures are incorporated herein by reference.

[0069] Fluid- applying station 60 comprises multiple print bars 66, a simplified example of which is shown in an inset portion 11 of Fig. 1. Each print bar 66 is mounted onto a frame (not shown) positioned at a fixed distance from the outer surface of ITM 44, and comprises multiple individually-controllable print nozzles 99, each of which is configured to apply a printing fluid (e.g., an aqueous ink) onto ITM 44. Typically, each print bar comprises multiple adjacent print heads 46.

[0070] Typically, each print bar 66 contains, and thus applies, a printing fluid of a different respective color. For example, for CMYK printing, which uses cyan (C), magenta (M), yellow (Y), and key / black (K), fluid-applying station 60 may comprise four print bars 66, one for each of these four colors. Alternatively, as shown in Fig. 1, for a seven-color printing scheme such as CMYKOBG, which includes the additional colors orange (O), blue (B), and green (G), fluidapplying station 60 may comprise seven print bars 66, one for each color.

[0071] Alternatively, two print bars may print in the same color.

[0072] The coordinate system 13 shown in Fig. 1 includes an x-axis, also referred to herein as a “print axis,” along which ITM 44 passes through fluid-applying station 60. Coordinate system 13 further includes a y-axis, referred to herein as a “cross-print axis,” and a z-axis, which is typically vertical. Print bars 66 are arranged along the print axis, each print bar being displaced from the ITM along the z-axis. The length of each print bar, which is approximately equal to the width of the ITM, is typically aligned with the cross-print axis.

[0073] Typically, the number of nozzles 99 is much greater, and the configuration of each print bar is more complex, than is depicted in Fig. 1. For example, each print head 46 may comprise multiple (e.g., two) two-dimensional arrays of nozzles spaced slightly apart from one another, each of the arrays lying in a plane that is perpendicular to the z-axis. Each of these arrays may comprise, for example, 16 lines of 64 nozzles each, the lines typically being slightly offset, with respect to each other, along the cross-print axis. Each print bar may comprise, for example, 24 print heads. Thus, for example, assuming two arrays per print head, each print bar may comprise 32 x 64 x 24 = 49,152 nozzles.

[0074] By applying the printing fluid to the ITM, fluid- applying station 60 forms a mirror image of the content (e.g., digital image 42) on the ITM. Subsequently, as described below, this mirror image is mirrored onto the target substrate.

[0075] In some embodiments, fluid-applying station 60 further comprises drying elements 62 positioned between print bars 66. Drying elements 62, which may comprise, for example, blowers of a gas (e.g., air), such as a hot gas, and / or infrared-based heaters, are configured to partially dry the printing -fluid droplets deposited on the surface of ITM 44 by the print bars. Furthermore, the flow of air from drying elements 62 may help reduce condensation at the surface of the print heads, remove printing-fluid residue from the area around the main droplets, prevent clogging of the nozzles, and / or prevent droplets of different colors from undesirably merging into one another.

[0076] In the example of Fig. 1, rollers 78 are positioned at the upper run of ITM 44 and are configured to keep ITM 44 taut, and to help control the speed of the ITM, as the ITM passes through fluid- applying station 60.

[0077] After being formed at fluid- applying station 60, the mirror image of the content arrives at drying station 64. Drying station 64 is configured to cause most or all of the liquid carrier in the printing fluid to evaporate, leaving behind, for example, only a layer of resin and a coloring agent, which may be heated to the point of being rendered a tacky ink film. Drying station 64 may comprise air blowers 68 and / or any other suitable drying elements, which are configured to direct hot or cool air (or another gas), and / or to direct infrared radiation, at the printing fluid.

[0078] At impression station 84, ITM 44 passes between an impression cylinder 82 and a pressure cylinder 90. (In some embodiments, a motion encoder, such as a rotary encoder, is integrated with impression cylinder 82 and / or pressure cylinder 90.) Impression cylinder 82 and pressure cylinder 90 are configured to impress the content onto the target substrate, which is conveyed by a substrate transport module 80 from an input stack 86 to an output stack 88 via impression station 84.

[0079] In particular, while both the ITM and the target substrate pass between the impression cylinder and the pressure cylinder, the pressure cylinder presses both of these elements against the impression cylinder, thereby forming the impression. Subsequently, the target substrate is transported to output stack 88.

[0080] In some embodiments, for simplex printing (in which content is printed on only one side of the substrate), only one impression station 84 is used, as shown in Fig. 1. In other embodiments, the system comprises two or more impression cylinders, so as to permit duplex printing and / or to enable faster simplex printing. In addition, mixed lots of single- sided and double-sided prints can also be printed.

[0081] In alternative embodiments, a different configuration of transport module 80 is used for printing on a continuous web substrate. Detailed descriptions of duplex printing systems and systems for printing on continuous web substrates are provided, for example, in U.S. patents 9,914,316 and 9,186,884, in PCT International Publication WO 2013 / 132424, in U.S. Patent Application Publication 2015 / 0054865, and in U.S. Provisional Application 62 / 596,926, whose respective disclosures are incorporated herein by reference.

[0082] Typically, impression cylinder 82 is periodically engaged with, and disengaged from, ITM 44, so as to facilitate transferring the content from the ITM to the target substrate. Nonetheless, the aforementioned rollers and dancer assembly may isolate the upper run of the ITM, on which the printing fluid is applied, from the mechanical vibrations occurring at the lower run of the ITM.

[0083] Treatment station 52, which may be located, for example, between impression station 84 and fluid-applying station 60, is configured to treat the ITM by, for example, cooling the ITM, applying a treatment fluid to the ITM, and / or cleaning the ITM. At treatment station 52, the temperature of the ITM may be reduced to a desired temperature level before the ITM enters fluid- applying station 60. The treatment may be carried out, for example, by passing ITM 44 over one or more rollers or blades. Treatment station 52 may further comprise one or more bars (not shown), which are adjacent to print bars 66 and are configured to jet the treatment fluid onto ITM 44. System 10 further comprises a processor 20 configured to control other modules of system 10 such as ITM module 70, fluid- applying station 60, and substrate transport module 80. Thus, for example, processor 20 may synchronize the application of the printing fluid with the movement of ITM 44.

[0084] Typically, processor 20 performs this control functionality by exchanging signals with multiple sensors and controllers belonging to system 10. By way of example, Fig. 1 shows one such controller 54. Other controllers may include, for example, a controller for quality-control station 55, which is described below, and a controller for dancer assembly 74. For example, processor 20, together with the latter controller, may control dancer assembly 74 responsively to signals from the aforementioned tension sensors.

[0085] System 10 further comprises a wired and / or wireless communication interface 22, which may comprise any suitable circuitry. Via communication interface 22, processor 20 may exchange signals with any component of system 10 and, optionally, with entities external to system 10. In some embodiments, processor 20 and communication interface 22 are contained in a control console 12.

[0086] For example, communication interface 22 may be connected to a cable 57, which is connected to controller 54.

[0087] In some embodiments, processor 20 is configured to receive, from temperature sensors (not shown), signals indicative of the surface temperature of ITM 44, and to control the operation of treatment station 52 responsively thereto. Such functionality is described, for example, in PCT International Publications WO / 2013 / 132424 and WO / 2017 / 208152, whose respective disclosures are incorporated herein by reference.

[0088] In some embodiments, console 12 comprises a display 34, which is configured to display data from processor 20, such as data input to the processor, by a user, using input devices 40. Such data may include, for example, digital image 42 or the test marks described below with reference to Fig. 2. An example configuration of console 12 and display 34 is described in detail in U.S. Patent 9,229,664, whose disclosure is incorporated herein by reference.

[0089] System 10 further comprises a quality-control station 55, which may be positioned adjacent to impression cylinder 82, as shown in Fig. 1, or at any other suitable location. Qualitycontrol station 55 comprises a camera (not shown), which is configured to acquire one or more digital images of a substrate, including the target substrate and / or the ITM, on which test marks or other content is printed. The camera may comprise any suitable image sensor, such as a contact image sensor (CIS) or a complementary metal oxide semiconductor (CMOS) image sensor, together with a scanning aperture of any suitable size.

[0090] Quality-control station 55 further comprises a wired and / or wireless communication interface (not shown) and a controller (not shown). The digital images acquired by the camera are communicated by the controller, via this communication interface and communication interface 22, to processor 20. Processor 20 processes the images so as to ascertain the quality of the printing. For example, as described in detail below with reference to the subsequent figures, the processor may identify problematic nozzles that adversely affect the printing quality. Typically, upon identifying one or more problematic nozzles, processor 20 outputs an output (e.g., on display 34) indicating the problematic nozzles. Alternatively or additionally, the processor may stop the operation of system 10.

[0091] The configuration of system 10 shown in Fig. 1 is provided purely by way of example. Additional components and configurations are described in detail, for example, in U.S. Patents 9,327,496 and 9,186,884, in PCT International Publications WO / 2013 / 132438, WO / 2013 / 132424, and WO / 2017 / 208152, and in U.S. Patent Application Publications 2015 / 0118503 and 2017 / 0008272, whose respective disclosures are incorporated herein by reference.

[0092] Processor 20 may be embodied as a single processor, or by a cooperatively networked or clustered set of processors. Alternatively or additionally, at least some of the functionality of processor 20, as described herein, may alternatively be performed by a controller belonging to system 10, such as the controller of quality-control station 55. Thus, it should be understood that “the processor,” as used in the context of the present application (including the claims), may refer to processor 20, to any one or more controllers belonging to system 10, or to a cooperative processing unit comprising processor 20 together with one or more controllers.

[0093] The functionality of any of the processing units (i.e., any of the processors and controllers) described herein may be implemented solely in hardware, e.g., using one or more fixed-function or general-purpose integrated circuits, Application-Specific Integrated Circuits (ASICs), and / or Field-Programmable Gate Arrays (FPGAs). Alternatively, this functionality may be implemented at least partly in software. For example, the processing unit may be embodied as a programmed processing unit comprising, for example, a central processing unit (CPU), a graphics processing unit (GPU), a tensor processing unit (TPU), and / or a digital signal processor (DSP). Program code, including software programs, and / or data may be loaded for execution and processing by the CPU, GPU, TPU, and / or DSP. The program code and / or data may be downloaded to the processing unit in electronic form, over a network, for example. Alternatively or additionally, the program code and / or data may be provided and / or stored on non-transitory tangible media, such as magnetic, optical, or electronic memory. Such program code and / or data, when provided to the processing unit, produce a machine or special-purpose computer configured to perform the tasks described herein.

[0094] QUALITY CONTROL

[0095] Reference is now made to Fig. 2, which is a schematic illustration of a substrate 100, such as ITM 44 or a target substrate such as a sheet 50 (Fig. 1), in accordance with some embodiments of the present invention.

[0096] Using print nozzles 99 (Fig. 1), the processor prints, on substrate 100, respective test marks 102. As described in detail below with reference to the subsequent figures, the processor uses test marks 102 in evaluating the print nozzles, by processing an image of substrate 100. In other words, the processor identifies any abnormal or missing test marks by processing the image, and identifies any problematic nozzles in response to the abnormal or missing test marks. Thus, for example, in response to identifying a blank space 121 where a test mark was expected, the processor may ascertain that the corresponding nozzle is not printing, e.g., by virtue of being blocked. Other examples of problematic-nozzle identification are provided below, with reference to the subsequent figures.

[0097] Typically, the processor uses each print bar 66 (Fig. 1) to print a different respective set of test marks. For example, Fig. 2 shows a first set 104a of test marks in a first color, represented in the figure by a first hatch pattern, and a second set 104b of test marks in a second color, represented by a second hatch pattern.

[0098] In some embodiments, the processor prints a predefined shape 106, such as a right triangle, on substrate 100, e.g., at the top of the substrate. Based on the location of shape 106 in the image of the substrate, the processor may more easily identify other features in the image.

[0099] Typically, each print bar prints a warmup ribbon 108 prior to printing any test marks. Advantageously, by printing the warmup ribbons, the system may expel, from the nozzles, any debris that might affect the quality of the test marks.

[0100] Typically, each of the warmup ribbons follows shape 106 and precedes first set 104a. Optionally, two or more of the warmup ribbons may be printed on the same portion of the substrate. For example, as shown in Fig. 2, all of the warmup ribbons may overlap with each other. Alternatively, for example, for embodiments with four print bars, two of the print bars may print a first set of overlapping warmup ribbons, and the other two print bars may print a second set of overlapping warmup ribbons.

[0101] Test marks 102 are typically arranged in multiple lines 114 arranged along the print axis, i.e., arranged such that successive lines 114 are spaced apart from one another along the print axis. In some embodiments, each print nozzle in the print bar prints a respective test mark 102. Thus, for example, assuming the print bar comprises 24 print heads arranged along the crossprint axis, each print head comprising 32 lines of 64 print nozzles each, the print bar may print 32 lines of 64x24 = 1536 test marks each. In other embodiments, some print nozzles do not print any test marks, e.g., due to the width of the substrate being less than the width of the print bar.

[0102] For convenience, the present specification assumes a sequential numbering of the test marks, whereby the first test marks in lines 114 are numbered 1 . . .N, the second test marks are numbered (N+l). . ,2N, etc., where N is the number of lines 114. In other words, any jthone of the test marks is a (floor((j-l) / N)+l)thone of the test marks in a ((j- 1)%N+I)t^1one of lines 114, where “%” indicates the modulo function. For example, assuming N = 32, the 3172ndtest mark is the 100thtest mark in the fourth line 114. ((3172- 1)%32+1 = 4; floor((3172-l) / 32)+l = 100.)

[0103] It is noted that the ordering of lines 114 does not necessarily correspond to the ordering of the nozzle lines, i.e., the kthline 114 is not necessarily printed by the kthnozzle line in each print head. Rather, the processor uses a predefined mapping between the nozzle lines and lines 114 to identify, for any given test mark, the nozzle that printed the test mark.

[0104] Typically, each of the test marks includes a (continuous) line or a (discontinuous) series of spots that is approximately parallel to the print axis, as shown in Fig. 2. Typically, the length of each test mark is between 2 mm and 4 cm.

[0105] Typically, the processor also prints respective duplicates of at least some of lines 114 so as to facilitate evaluating the print nozzles by processing an image of the substrate, as further described below with reference to Fig. 4. In particular, the duplicates include (i) a duplicate 116_N of the Nthline 114_N preceding the first line 114_1, and (ii) a duplicate 116_1 of first line 114_1 following Nthline 114_N. The duplicates may further include (i) a duplicate 116_(N- 1) of the (N-l)thline preceding duplicate 116_N, and (ii) a duplicate 116_2 of the second line following duplicate 116_1.

[0106] Typically, the processor causes each print bar to print a landmark ribbon 110 at a predefined distance from the test marks, e.g., preceding the test marks. Thus, for example, Figs. 2 shows a first landmark ribbon 110a, in the first color, at a predefined distance from first set 104a, and a second landmark ribbon 110b, in the second color, at a predefined distance from second set 104b.

[0107] Advantageously, the spacing between each landmark ribbon 110 and the nearest line 114, and the spacing between each pair of successive lines 114, are predefined. Hence, the processor may identify each line 114 in the image based on its distance from the preceding (or following) landmark ribbon.

[0108] In addition, landmark ribbons 110 may facilitate identifying the margins of the substrate, and thus, the beginning and end of each line 114. Alternatively or additionally, the processor may check the shape of each landmark ribbon, so as to identify any geometric distortion in the image, e.g., due to distortion in the ITM. Alternatively or additionally, based on the distance between successive landmark ribbons, the processor may compute the scale of the image.

[0109] Typically, using a subset of the print nozzles, the processor prints, on the substrate, respective orientation marks 112 arranged in one or more orientation lines 118. The distance between successive orientation marks 112 varies across each orientation line 118, such that the orientation marks facilitate identifying the print nozzle corresponding to (i.e., the print nozzle that printed) any given one of the test marks, as further described below. Typically, the orientation lines are parallel to lines 114.

[0110] For example, as shown in Fig. 2, the orientation lines may include a first orientation line preceding the test marks (e.g., preceding duplicate 116_(N-1)) and a second orientation line following the test marks (e.g., following duplicate 116_2).

[0111] Typically, the print nozzles used to print the orientation marks include (i) a first series of print nozzles corresponding to the test marks having first indices kl*dl for dl > 2N and kl = 1 . . ,floor(M / dl), and (ii) a second series of print nozzles corresponding to the test marks having second indices k2*d2 for k2 = 1 . . . floor(M / d2) and d2 > dl, where M is the total number of test marks. Typically, dl and d2 are prime; for example, for N = 32 and M = 49,152, dl may be 997 and d2 may be 1259. (Typically, the prime numbers are selected such that no first index is within a predefined distance of a second index.) By virtue of dl and d2 being prime, the distance between successive orientation marks 112 varies across each orientation line 118. Typically, each orientation mark includes a line that is approximately parallel to the print axis, as shown in Fig. 2. (This line may be shorter than each of the test marks.)

[0112] As noted above, orientation marks 112 facilitate identifying the nozzle corresponding to any given test mark. For example, for a test mark 102a, the processor may first ascertain that the coordinate of the test mark on the cross-print axis lies between one orientation mark 112a and another orientation mark 112b. The processor may then compute the distance between these two orientation marks. Based on the distance, the processor may ascertain the respective nozzles that printed these two orientation marks. Subsequently, based on the distance, along the cross-print axis, between test mark 102a and each of these orientation marks, the processor may ascertain the nozzle corresponding to test mark 102a.

[0113] Advantageously, the printing of multiple orientation lines may help compensate for any rotation of the substrate in the image.

[0114] For example, if, for test mark 102a, the processor were to refer only to orientation marks 112a and 112b as described above, the processor might ascertain an incorrect nozzle if the substrate were rotated. Hence, the processor may also refer to orientation marks 112c and 112d in the second orientation line 118. In other words, the processor may also ascertain that the coordinate of the test mark on the cross-print axis lies between orientation marks 112c and 112d, compute the distance (along the cross-print axis) between test mark 102a and each of these orientation marks, and then ascertain the nozzle corresponding to test mark 102a based on these distances and the distances with respect to orientation marks 112a and 112b.

[0115] NOZZLE INCONSISTENCY

[0116] Reference is now made to Fig. 3, which is a schematic illustration of a portion 102i of an image of substrate 100 (Fig. 2) corresponding to a test mark, in accordance with some embodiments of the present invention. (For convenience, the remainder of the description will refer to a portion of the image corresponding to a test mark simply as a “test mark,” but will use the reference number 102i, rather than 102, to indicate that reference is being made to the pixelated representation of a test mark in the image.)

[0117] Typically, by processing the image of the substrate, the processor computes respective inconsistency scores for the print nozzles. Each inconsistency score quantifies the inconsistency - i.e., the lack of uniformity in the printing intensity - of the nozzle for which the score was computed. The processor compares the inconsistency scores to a predefined threshold. Based on this comparison, the processor may output an output indicating those of the print nozzles whose inconsistency scores pass the threshold. For example, assuming a convention in which a greater inconsistency score indicates greater inconsistency, the output may indicate the print nozzles whose inconsistency scores exceed the threshold.

[0118] In particular, for each test mark, the processor computes respective differences between pairs of adjacent pixels 120, in the image, corresponding to the test mark. Based on the differences, the processor computes the inconsistency score for the print nozzle that was used to print the test mark.

[0119] Typically, for line-shaped test marks, the differences are computed along a line (e.g., a central line) of pixels parallel to the length of the test mark. By way of example, Fig. 3 shows one pair of pixels, including a pixel 120a and another pixel 120b, belonging to a line of pixels in test mark 102i.

[0120] Typically, the image includes multiple color channels. In such cases, typically, the processor computes the difference, for each pair of adjacent pixels, by first computing respective per-channel differences, for the color channels, between the pair of adjacent pixels. Subsequently, the processor computes the difference as a maximum of the per-channel differences.

[0121] For example, Fig. 3 shows respective components 102i_R, 102i_G, and 102i_B of test mark 102i for red (R), green (G), and blue (B) color channels. Pixel 120a includes a pixelcomponent 120a_R in component 102i_R, a pixel-component 120a_G in component 102i_G, and a pixel-component 120a_B in component 102i_B. Similarly, pixel 120b includes a pixelcomponent 120b_R in component 102i_R, a pixel-component 120b_G in component 102i_G, and a pixel-component 120b_B in component 102i_B. To compute the difference between pixels 120a and 120b, the processor may compute the maximum of (i) the difference between pixelcomponents 120a_R and 120b_R, (ii) the difference between pixel-components 120a_G and 120b_G, and (iii) the difference between pixel-components 120a_B and 120b_B.

[0122] For example, assuming the components of pixel 120a have the values 47, 82, and 143, respectively, and the components of pixel 120b have the values 49, 85, and 139, respectively, the processor may compute the difference between pixels 120a and 120b as max(49-47, 85-82, 143-139) = 4.

[0123] In some embodiments, to compute the inconsistency score, the processor computes a representative difference based on the differences between the pairs of adjacent pixels. In some embodiments, the representative difference is the average of (i) the largest one of the differences, and (ii) the (k-1 )th- largest one of the differences, where k = 3, for example. Subsequently, the processor may compute the score based on the representative difference, e.g., by setting the score equal to the representative difference or to the representative difference divided by a normalizing value. To compute the normalizing value, the processor may first compute the average value of the pixels that include the adjacent pairs in each of the color channels. Subsequently, the processor may compute a statistic, such as the maximum, minimum, median, or average, of these average values.

[0124] POSITIONAL DEVIATION

[0125] Reference is now made to Fig. 4, which is a schematic illustration of a portion 122 of an image of the substrate, in accordance with some embodiments of the present invention.

[0126] Typically, by processing the image, the processor computes respective locations of the test marks, along the cross-print axis, in the image. In other words, the processor computes the location, along the cross-print axis, of each test mark 102i. Based on the locations, the processor computes respective positional-deviation scores, which indicate respective positional deviations 124 along the cross-print axis, for the test marks. The processor compares the positional- deviation scores to a predefined threshold. Based on this comparison, the processor may output an output indicating those of the print nozzles that were used to print those of the test marks whose positional-deviation scores pass the threshold. For example, assuming a convention in which a greater positional-deviation score indicates a greater deviation, the output may indicate the print nozzles that were used to print those of the test marks whose positional-deviation scores exceed the threshold.

[0127] Typically, for any given test mark, the processor computes the positional-deviation score based on the location of the given test mark and the respective locations of others of the test marks having the same relative position as the given test mark in at least one of the lines 114 immediately preceding the line of the given test mark, and the same number of lines immediately following the line of the given test mark. In some embodiments, this number, which is represented herein by the variable “s,” equals 2. Thus, for example, the positional-deviation score for a test mark 102i_l, which is the k^1test mark in its line and has a location yl, may be based on the respective locations y2, y3, y4, and y5 of test marks 102i_2, 102i_3, 102i_4, and 102i_5, each of which is the k^1test mark in its respective line. (Thus, the printing of duplicate lines, as described above with reference to Fig. 2, may facilitate computing the positional-deviation score for the first and last lines.)

[0128] For example, the processor may compute multiple expected locations of the given test mark based on the respective locations of the others of the test marks, compute respective absolute differences between the expected locations and the location of the given test mark, and compute the positional-deviation score based on the absolute differences, e.g., based on the minimum of the absolute differences.

[0129] The expected locations may include, for example, a median location for the 2s+l test marks after correcting for any offset, along the cross-print axis, between the test mark in one line 114 and the test mark in the next line 114. For example, the offset may be 1 / n pixels, where n is a function of the positioning of the nozzles and the resolution of the camera. The median location for test mark 102i_l may therefore be calculated as median(y 1, y2+l / n, y3+2 / n, y4-l / n, y5-2 / n).

[0130] Alternatively or additionally, the expected locations may include the average of the locations in the surrounding lines. For example, for test mark 102i_l, the expected locations may include (y2+y4) / 2, (y3+y5) / 2, and / or (y2+y3+y4+y5) / 4.

[0131] When calculating the positional-deviation score, the processor may compensate for any shift of the print heads along the cross-print axis based on a “dY calibration” (in advance or on the fly) as described in “Unified calibrations in a digital printing system” (attorney ref. no. 1373- 2027 | LIP 6 / 017), whose disclosure is incorporated herein by reference.

[0132] Reference is now made to Fig. 5, which is a flow diagram for a method 126 for computing the locations of the test marks (such as yl . . ,y5 of Fig. 4), in accordance with some embodiments of the present invention.

[0133] Typically, the processor first converts the image to a grayscale image, at a converting step 128. For example, given an RGB image, the processor may convert the image to grayscale by setting each pixel equal to the minimum value from the three color channels.

[0134] Next, the processor smooths the grayscale image along the print axis, at a smoothing step 130. Optionally, at another smoothing step 131, the processor then smooths the grayscale image along the cross-print axis, e.g., using a filter such as [1 1.2 l] / 3.2.

[0135] Following the smoothing, the processor selects each line of test marks, typically by identifying the center of the line along the print axis, at a line-selecting step 132. For example, to select the line, the processor may add the requisite offset, along the print axis, from the landmark ribbon 110 (Fig. 2) that precedes or follows the test marks. As described above with reference to Fig. 2, the processor may also use the landmark ribbons to identify the margins of the substrate, and hence, the start and end of each line.

[0136] For each selected line, the processor, at a pixel-identifying step 134, identifies pixels in the line whose values are local minima along the cross-print axis and, typically, are less than a predefined threshold. Typically, each test mark is darkest at its center, such that the local minima correspond, approximately, with the centers of the test marks in the line.

[0137] Next, the processor selects each identified pixel at a pixel- selecting step 136. For each selected pixel, the processor, at a parabola-fitting step 138, fits a parabola to multiple pixel values in the neighborhood of the selected pixel, i.e., to respective values of the identified pixel, at least one other pixel immediately preceding the identified pixel, and at least one other pixel immediately following the identified pixel. Next, the processor, at a computing step 140, computes the location of the test mark as the location of the vertex of the parabola.

[0138] Following computing step 140, the processor checks, at a checking step 142, whether any identified local-minima pixels remain. If yes, the processor returns to pixel- selecting step 136. Otherwise, the processor checks, at another checking step 144, whether any more lines of test marks remain. If yes, the processor returns to line- selecting step 132. Otherwise, method 126 ends.

[0139] By way of illustration, Fig. 6, to which reference is now made, shows the values, in the smoothed grayscale image, of a number of pixels along a line. As described above with reference to Fig. 5, the processor identifies the local minima 146, fits parabolas 148 to the local minima, and sets the locations of the test marks equal to the locations of the vertices of the parabolas. For example, Fig. 6 shows one such location at approximately 7697.5 pixels.

[0140] RELATIVE INTENSITIES

[0141] Reference is again made to Fig. 4.

[0142] Typically, by processing the image of the substrate, the processor computes the respective intensities of the test marks. The processor further computes respective relativeintensity scores for the test marks, each of the relative-intensity scores quantifying the intensity of the test mark relative to one or more others of the intensities. The processor compares the relative-intensity scores to a predefined threshold. Based on this comparison, the processor may output an output indicating those of the print nozzles that were used to print those of the test marks whose relative-intensity scores pass the threshold. Depending on the convention used, the relative-intensity score may be either an increasing or a decreasing function of the darkness of the test mark.

[0143] For example, assuming a convention in which the relative-intensity score is an increasing function of the darkness, the relative-intensity score of a test mark 102i_6, which is relatively faint, may be less than the threshold. The output may therefore indicate that the print nozzle that printed test mark 102i_6 is problematic, e.g., by virtue of lacking sufficient ink.

[0144] In some embodiments, the processor compares the relative-intensity scores to two thresholds, so as to identify problematic nozzles whose printing is either darker or fainter than the norm.

[0145] Typically, to compute the intensities, the processor first converts the image to a grayscale image and smooths the grayscale image along the print axis, as described above with reference to Fig. 5. For each of the lines of test marks in the smoothed grayscale image, the processor identifies pixels whose values are local minima along the cross-print axis, as described above with reference to Figs. 5-6. Subsequently, the processor computes the intensities of those of the test marks in the line based on the values.

[0146] In general, the processor may compute the intensities using any suitable formula. For example, with reference to Fig. 6, for each local minimum 146 in the line, the processor may compute the average gray level of a set 150 of gray levels including the local minimum, the respective values of one or more (e.g., two) pixels immediately preceding the local minimum in the line, and the respective values of one or more (e.g., two) pixels immediately following the local minimum in the line. (This average accounts for the spread of the test mark along the crossprint axis.) Subsequently, the processor may compute the intensity based on this average. For example, the processor may subtract this average from a suitable baseline value, such as a statistic (e.g., the 75thpercentile) of the pixel values in the image.

[0147] Typically, the processor computes respective intensity ratios between the intensity of the given test mark and the intensities of others of the test marks, and computes the relative-intensity score for the given test mark based on the intensity ratios. For example, the processor may set the relative-intensity score (prior to any normalization, as described below) equal to a statistic, such as the minimum, maximum, median, or average, of the intensity ratios. For example, it will be assumed that the intensity of test mark 102i_6 is 140, and that the intensities of the other test marks are 148, 150, 152, and 154. In such a case, the processor may compute four intensity ratios: 140 / 148 = 0.95, 140 / 150 = 0.93, 140 / 152 = 0.92, and 140 / 154 = 0.91. The processor may then set the (unnormalized) relative-intensity score for test mark 102i_6 to 0.91 (the minimum) or 0.95 (the maximum), for example.

[0148] In some embodiments, the other test marks whose intensities are used for calculating the intensity ratios include the test mark immediately preceding the given test mark in the line of the given test mark, and the test mark immediately following the given test mark in the line of the given test mark. For example, with reference to Fig. 4, for test mark 102i_6, the other test marks may include the immediately-preceding test mark 102i_7 and the immediately -following test mark 102i_8.

[0149] Alternatively or additionally, the other test marks may include the test mark having the same relative position as the given test mark in one of the lines preceding the line of the given test mark, and the test mark having the same relative position as the given test mark in one of the lines following the line of the given test mark.

[0150] For example, the processor may use the immediately -preceding and immediately- following lines. For example, assuming test mark 102i_6 to be the test mark in its line 114, the other test marks may include a test mark 102i_9, which is the test mark in the immediately -preceding line, and a test mark 102i_10, which is the test mark in the immediately-following line.

[0151] (Thus, the printing of duplicate lines, as described above with reference to Fig. 2, may facilitate computing the relative-intensity scores for the first and last lines.)

[0152] Alternatively, or additionally, the other test marks may include those test marks that are most correlated, with the given test mark, with respect to a cross -print- axis spread.

[0153] In particular, due to the relatively low resolution of the camera, each test mark may appear to be spread unevenly along the cross-print axis, the nature of this spread (i.e., the “spread function” that defines this spread) varying between the test marks due to the different positions of the test marks with respect to the imaging sensor of the camera. The other test marks may include those test marks whose spread is most similar to that of the given test mark.

[0154] More generally, any given test mark with index jO may be compared to a group of test marks having indices {jO + k}, another group of test marks having indices {jO + k*C}, and / or another group of test marks having indices { jO + k*N}, where:

[0155] N is the number of lines 114 (e.g., 32),

[0156] C is an integer (e.g., 4) corresponding to an offset providing maximal correlation in cross-print-axis spread, and k is a set of integers such as { -kO, . . . - 1 } , { 1 , . . . ,k0 } , or the union of these two sets, where kO = 1, 2, C, or any other suitable integer.

[0157] MISSHAPEN TEST MARKS

[0158] Typically, by processing the image, the processor also identifies any test marks having a deviated shape, such as - in lieu of a straight line - a curved or slanted line, and outputs an indication of the corresponding nozzles.

[0159] For example, the processor may first convert the image to grayscale and then smooth the image along the print axis (optionally, with a different filter used in smoothing step 130 (Fig. 5)). Subsequently, for each test mark, the processor may identify the location, along the crossprint axis, of the minimum pixel value in the test mark, for several points along the print axis. The processor may then calculate the standard deviation of the locations. (A higher standard deviation indicates less straightness of the test mark.)

[0160] NORMALIZED NOZZLE PERFORMANCE

[0161] As described above with reference to Fig. 2, typically, the processor prints a separate set of test marks 102 for each color, using a different respective set of print nozzles. The sets of test marks may span multiple substrates, such as multiple sheets 50 (Fig. 1). Subsequently to receiving at least one image of the substrate(s), the processor, based on the sets of test marks, computes respective performance scores for the nozzles, such as the inconsistency scores, positional-deviation scores, relative-intensity scores, or scores relating to misshapen test marks described above.

[0162] Typically, prior to comparing each type of performance score to the relevant threshold(s), the processor normalizes the performance scores for the nozzles across the colors. The processor then compares the normalized performance scores to the threshold(s).

[0163] For example, for any given one of the test marks in any given one of the sets of test marks, the processor may identify the performance scores of those of the test marks having the same index as the given test mark in the other sets of test marks. In other words, for the Ithtest mark in the line of test marks in one of the sets, the processor may identify the performance scores of the i^1test marks in the j^1lines of test marks in the other sets. The processor may then normalize the performance score of the given test mark based on the identified performance scores.

[0164] For example, the processor may identify a percentile of a set of the performance scores that includes (i) the performance score of the given print nozzle and (ii) the identified performance scores. For example, assuming a convention in which a greater score indicates greater deviation from the norm, the processor may identify the second-lowest performance score.

[0165] Subsequently to identifying the percentile, the processor may set the normalized performance score of the given print nozzle equal to ±d, d being the difference between the percentile and the performance score of the given print nozzle. (Whether the normalized performance score is equal to d or -d depends on the convention.)

[0166] For example, assuming that the performance scores across four colors, for any given testmark position, are 4, 5, 6, and 7, the processor may identify 5 as the second-lowest of the scores, and in response thereto, compute the normalized scores as -1, 0, 1, and 2.

[0167] It will be appreciated by persons skilled in the art that the present invention is not limited to what has been particularly shown and described hereinabove. Rather, the scope of embodiments of the present invention includes both combinations and subcombinations of the various features described hereinabove, as well as variations and modifications thereof that are not in the prior art, which would occur to persons skilled in the art upon reading the foregoing description. Documents incorporated by reference in the present patent application are to be considered an integral part of the application except that to the extent any terms are defined in these incorporated documents in a manner that conflicts with the definitions made explicitly or implicitly in the present specification, only the definitions in the present specification should be considered.

Claims

CLAIMS1. A system, comprising: multiple print nozzles; and a processor, configured to: using the print nozzles, print, on a substrate, respective test marks for use in evaluating the print nozzles by processing an image of the substrate, and using a subset of the print nozzles, print, on the substrate, respective orientation marks arranged in one or more orientation lines, a distance between successive ones of the orientation marks varying across each of the orientation lines, such that the orientation marks facilitate identifying the print nozzle corresponding to any given one of the test marks.

2. The system according to claim 1, wherein the orientation lines include a first orientation line preceding the test marks and a second orientation line following the test marks.

3. The system according to any one of claims 1-2, wherein the test marks are arranged along N lines, each of which is parallel to an axis of the orientation lines, and wherein the subset includes: a first series of the print nozzles corresponding to those of the test marks having first indices kl*dl for dl > 2N and kl = 1 . . ,floor(M / dl), M being a total number of the test marks, and, a second series of the print nozzles corresponding to those of the test marks having second indices k2*d2 for d2 > dl and k2 = 1 . . ,floor(M / d2), anyone of the test marks being a (floor((j-l) / N)+l)t^1one of the test marks in a ((j- 1)%N+I)t^1one of the lines.

4. The system according to claim 3, wherein dl and d2 are prime.

5. A system, comprising: multiple print nozzles; and a processor, configured to: using the print nozzles, print, on a substrate, respective test marks for use in evaluating the print nozzles by processing an image of the substrate, the test marks being arranged in N lines, andusing the print nozzles, print respective duplicates of at least some of the lines so as to facilitate the evaluation, the duplicates including: a duplicate of an N^1one of the lines preceding a first one of the lines, and a duplicate of the first one of the lines following the N^1one of the lines.

6. The system according to claim 5, wherein the duplicates further include: a duplicate of an (N-l)^ one of the lines preceding the duplicate of the N^1one of the lines, and a duplicate of a second one of the lines following the duplicate of the first one of the lines.

7. A system, comprising: a display; and a processor, configured to: receive an image of a substrate on which multiple test marks were printed using different respective print nozzles, compute respective inconsistency scores for the print nozzles by, for each of the test marks: computing respective differences between pairs of adjacent pixels, in the image, corresponding to the test mark, and based on the differences, computing the inconsistency score for the print nozzle that was used to print the test mark, compare the inconsistency scores to a predefined threshold, and based on the comparing, output, on the display, an output indicating those of the print nozzles whose inconsistency scores pass the predefined threshold.

8. The system according to claim 7, wherein the image includes multiple color channels, and wherein the processor is configured to compute the differences by, for each of the pairs of adjacent pixels: computing respective per-channel differences, for the color channels, between the pair of adjacent pixels, and computing the difference as a maximum of the per-channel differences.

9. The system according to any one of claims 7-8, wherein the processor is configured tocompute the inconsistency score by: based on the differences, computing a representative difference, and computing the score based on the representative difference.

10. The system according to claim 9, wherein the representative difference is an average of (i) a k^-largest one of the differences, and (ii) a (k-1 ^-largest one of the differences.

11. A system, comprising: a display; and a processor, configured to: receive an image of a substrate on which multiple test marks were printed using different respective print nozzles, compute respective locations of the test marks, along a cross-print axis, in the image, based on the locations, compute respective positional-deviation scores, which indicate respective positional deviations along the cross-print axis, for the test marks, compare the positional-deviation scores to a predefined threshold, and based on the comparing, output, on the display, an output indicating those of the print nozzles that were used to print those of the test marks whose positional-deviation scores pass the predefined threshold.

12. The system according to claim 11, wherein the test marks are arranged in multiple lines, each of which extends along the cross-print axis, and wherein, for any given one of the test marks, the processor is configured to compute the positional-deviation score for the given test mark based on: the location of the given test mark, and the respective locations of others of the test marks having the same relative position as the given test mark in s > 1 of the lines immediately preceding the line of the given test mark and s of the lines immediately following the line of the given test mark.

13. The system according to claim 12, wherein s = 2.

14. The system according to any one of claims 12-13, wherein the processor is configured to compute the positional-deviation score for the given test mark by: computing multiple expected locations of the given test mark, based on the respectivelocations of the others of the test marks, computing respective absolute differences between the expected locations and the location of the given test mark, and computing the positional-deviation score based on the absolute differences.

15. The system according to claim 14, wherein the processor is configured to compute the positional-deviation score based on a minimum of the absolute differences.

16. The system according to any one of claims 11-15, wherein the test marks are arranged in multiple lines, each of which extends along the cross-print axis, and wherein the processor is configured to compute the locations by: converting the image to a grayscale image, smoothing the grayscale image along a print axis, and for each of the lines in the smoothed grayscale image: identifying pixels whose values are local minima along the cross-print axis, and for each of the identified pixels: fitting a parabola to respective values of the identified pixel, at least one other pixel immediately preceding the identified pixel, and at least one other pixel immediately following the identified pixel, and computing one of the locations as a location of a vertex of the parabola.

17. The system according to claim 16, wherein the processor is configured to compute the locations by, prior to identifying the pixels whose values are the local minima, smoothing the grayscale image along the cross-print axis.

18. A system, comprising: a display; and a processor, configured to: receive an image of a substrate on which multiple test marks were printed using different respective print nozzles, compute respective intensities of the test marks, compute respective relative-intensity scores for the test marks, each of the relative-intensity scores quantifying the intensity of the test mark relative to one or more others of the intensities, compare the relative-intensity scores to a predefined threshold, and based on the comparing, output, on the display, an output indicating those of theprint nozzles that were used to print those of the test marks whose relative-intensity scores pass the predefined threshold.

19. The system according to claim 18, wherein, for any given one of the test marks, the processor is configured to compute the relative-intensity score for the given test mark by: computing respective intensity ratios between the intensity of the given test mark and the intensities of others of the test marks, and computing the relative-intensity score based on the intensity ratios.

20. The system according to claim 19, wherein the test marks are arranged in multiple lines, each of which extends along a cross-print axis, and wherein the others of the test marks include: the test mark immediately preceding the given test mark in the line of the given test mark, and the test mark immediately following the given test mark in the line of the given test mark.

21. The system according to claim 19, wherein the test marks are arranged in multiple lines, each of which extends along a cross-print axis, and wherein the others of the test marks include: the test mark having the same relative position as the given test mark in one of the lines preceding the line of the given test mark, and the test mark having the same relative position as the given test mark in one of the lines following the line of the given test mark.

22. The system according to claim 19, wherein the test marks are arranged in multiple lines, each of which extends along a cross-print axis, and wherein the others of the test marks include those of the test marks that are most correlated, with the given test mark, with respect to a spread along the cross-print axis, the spread being a function of a resolution of a camera that acquired the image.

23. The system according to any one of claims 18-19, wherein the test marks are arranged in multiple lines, each of which extends along a cross-print axis, and wherein the processor is configured to compute the intensities by: converting the image to a grayscale image, smoothing the grayscale image along a print axis, and for each of the lines in the smoothed grayscale image: identifying pixels whose values are local minima along the cross-print axis, and computing the intensities of those of the test marks in the line based on the values.

24. The system according to claim 23, wherein the processor is configured to compute theintensity of any given one of the test marks in the line based on an average of: the local minimum corresponding to the given test mark, respective values of one or more pixels immediately preceding the local minimum in the line of the given test mark, and respective values of one or more pixels immediately following the local minimum in the line of the given test mark.

25. A system, comprising: a display; and a processor, configured to: receive at least one image of at least one substrate on which multiple sets of test marks were printed, in different respective colors, using different respective sets of print nozzles, compute respective performance scores for the test marks, normalize the performance scores across the colors, compare the normalized performance scores to a predefined threshold, and based on the comparing, output, on the display, an output indicating those of the print nozzles corresponding to those of the test marks whose performance scores pass the predefined threshold.

26. The system according to claim 25, wherein the processor is configured to normalize the performance scores by: for any given one of the test marks in any given one of the sets of test marks, identifying the performance scores of those of the test marks having the same index as the given test mark in others of the sets of test marks, and normalizing the performance score of the given test mark based on the identified performance scores.

27. The system according to claim 26, wherein the processor is configured to normalize the performance score of the given test mark by: identifying a percentile of a set of the performance scores that includes (i) the performance score of the given test mark and (ii) the identified performance scores, and setting the normalized performance score of the given test mark equal to ±d, d being a difference between the percentile and the performance score of the given test mark.

28. A method, comprising:using multiple print nozzles, printing, on a substrate, respective test marks for use in evaluating the print nozzles by processing an image of the substrate; and using a subset of the print nozzles, printing, on the substrate, respective orientation marks arranged in one or more orientation lines, a distance between successive ones of the orientation marks varying across each of the orientation lines, such that the orientation marks facilitate identifying the print nozzle corresponding to any given one of the test marks.

29. The method according to claim 28, wherein the orientation lines include a first orientation line preceding the test marks and a second orientation line following the test marks.

30. The method according to any one of claims 28-29, wherein the test marks are arranged along N lines, each of which is parallel to an axis of the orientation lines, and wherein the subset includes: a first series of the print nozzles corresponding to those of the test marks having first indices kl*dl for dl > 2N and kl = 1 . . ,floor(M / dl), M being a total number of the test marks, and, a second series of the print nozzles corresponding to those of the test marks having second indices k2*d2 for d2 > dl and k2 = 1 . . ,floor(M / d2), anyone of the test marks being a (floor((j-l) / N)+l)t^1one of the test marks in a ((j- 1)%N+I)t^1one of the lines.

31. The method according to claim 30, wherein dl and d2 are prime.

32. A method, comprising: using multiple print nozzles, printing, on a substrate, respective test marks for use in evaluating the print nozzles by processing an image of the substrate, the test marks being arranged in N lines; and using the print nozzles, printing respective duplicates of at least some of the lines so as to facilitate the evaluation, the duplicates including: a duplicate of an N^1one of the lines preceding a first one of the lines, and a duplicate of the first one of the lines following the N^1one of the lines.

33. The method according to claim 32, wherein the duplicates further include:a duplicate of an (N-l)^ one of the lines preceding the duplicate of the N^1one of the lines, and a duplicate of a second one of the lines following the duplicate of the first one of the lines.

34. A method, comprising: receiving an image of a substrate on which multiple test marks were printed using different respective print nozzles; computing respective inconsistency scores for the print nozzles by, for each of the test marks: computing respective differences between pairs of adjacent pixels, in the image, corresponding to the test mark, and based on the differences, computing the inconsistency score for the print nozzle that was used to print the test mark; comparing the inconsistency scores to a predefined threshold; and based on the comparing, outputting an output indicating those of the print nozzles whose inconsistency scores pass the predefined threshold.

35. The method according to claim 34, wherein the image includes multiple color channels, and wherein computing the differences comprises, for each of the pairs of adjacent pixels: computing respective per-channel differences, for the color channels, between the pair of adjacent pixels, and computing the difference as a maximum of the per-channel differences.

36. The method according to any one of claims 34-35, wherein computing the inconsistency score comprises: based on the differences, computing a representative difference; and computing the score based on the representative difference.

37. The method according to claim 36, wherein the representative difference is an average of (i) a k^-largest one of the differences, and (ii) a (k-1 ^-largest one of the differences.

38. A method, comprising: receiving an image of a substrate on which multiple test marks were printed using different respective print nozzles; computing respective locations of the test marks, along a cross-print axis, in the image;based on the locations, computing respective positional-deviation scores, which indicate respective positional deviations along the cross-print axis, for the test marks; comparing the positional-deviation scores to a predefined threshold; and based on the comparing, outputting an output indicating those of the print nozzles that were used to print those of the test marks whose positional-deviation scores pass the predefined threshold.

39. The method according to claim 38, wherein the test marks are arranged in multiple lines, each of which extends along the cross-print axis, and wherein, for any given one of the test marks, computing the positional-deviation score for the given test mark comprises computing the positional-deviation score based on: the location of the given test mark, and the respective locations of others of the test marks having the same relative position as the given test mark in s > 1 of the lines immediately preceding the line of the given test mark and s of the lines immediately following the line of the given test mark.

40. The method according to claim 39, wherein s = 2.

41. The method according to any one of claims 39-40, wherein computing the positional- deviation score for the given test mark comprises: computing multiple expected locations of the given test mark, based on the respective locations of the others of the test marks; computing respective absolute differences between the expected locations and the location of the given test mark; and computing the positional-deviation score based on the absolute differences.

42. The method according to claim 41, wherein computing the positional-deviation score comprises computing the positional-deviation score based on a minimum of the absolute differences.

43. The method according to any one of claims 38-42, wherein the test marks are arranged in multiple lines, each of which extends along the cross-print axis, and wherein computing the locations comprises: converting the image to a grayscale image; smoothing the grayscale image along a print axis; and for each of the lines in the smoothed grayscale image:identifying pixels whose values are local minima along the cross-print axis, and for each of the identified pixels: fitting a parabola to respective values of the identified pixel, at least one other pixel immediately preceding the identified pixel, and at least one other pixel immediately following the identified pixel, and computing one of the locations as a location of a vertex of the parabola.

44. The method according to claim 43, wherein computing the locations further comprises, prior to identifying the pixels whose values are the local minima, smoothing the grayscale image along the cross-print axis.

45. A method, comprising: receiving an image of a substrate on which multiple test marks were printed using different respective print nozzles; computing respective intensities of the test marks; computing respective relative-intensity scores for the test marks, each of the relativeintensity scores quantifying the intensity of the test mark relative to one or more others of the intensities; comparing the relative-intensity scores to a predefined threshold; and based on the comparing, outputting an output indicating those of the print nozzles that were used to print those of the test marks whose relative-intensity scores pass the predefined threshold.

46. The method according to claim 45, wherein, for any given one of the test marks, computing the relative-intensity score for the given test mark comprises: computing respective intensity ratios between the intensity of the given test mark and the intensities of others of the test marks; and computing the relative-intensity score based on the intensity ratios.

47. The method according to claim 46, wherein the test marks are arranged in multiple lines, each of which extends along a cross-print axis, and wherein the others of the test marks include: the test mark immediately preceding the given test mark in the line of the given test mark, and the test mark immediately following the given test mark in the line of the given test mark.

48. The method according to claim 46, wherein the test marks are arranged in multiple lines, each of which extends along a cross-print axis, and wherein the others of the test marks include:the test mark having the same relative position as the given test mark in one of the lines preceding the line of the given test mark, and the test mark having the same relative position as the given test mark in one of the lines following the line of the given test mark.

49. The method according to claim 46, wherein the test marks are arranged in multiple lines, each of which extends along a cross-print axis, and wherein the others of the test marks include those of the test marks that are most correlated, with the given test mark, with respect to a spread along the cross-print axis, the spread being a function of a resolution of a camera that acquired the image.

50. The method according to any one of claims 45-46, wherein the test marks are arranged in multiple lines, each of which extends along a cross-print axis, and wherein computing the intensities comprises: converting the image to a grayscale image; smoothing the grayscale image along a print axis; and for each of the lines in the smoothed grayscale image: identifying pixels whose values are local minima along the cross-print axis, and computing the intensities of those of the test marks in the line based on the values.

51. The method according to claim 50, wherein computing the intensity of any given one of the test marks in the line comprises computing the intensity based on an average of: the local minimum corresponding to the given test mark, respective values of one or more pixels immediately preceding the local minimum in the line of the given test mark, and respective values of one or more pixels immediately following the local minimum in the line of the given test mark.

52. A method, comprising: receiving at least one image of at least one substrate on which multiple sets of test marks were printed, in different respective colors, using different respective sets of print nozzles; computing respective performance scores for the test marks; normalizing the performance scores across the colors; comparing the normalized performance scores to a predefined threshold; and based on the comparing, outputting an output indicating those of the print nozzles corresponding to those of the test marks whose performance scores pass the predefinedthreshold.

53. The method according to claim 52, wherein normalizing the performance scores comprises: for any given one of the test marks in any given one of the sets of test marks, identifying the performance scores of those of the test marks having the same index as the given test mark in others of the sets of test marks; and normalizing the performance score of the given test mark based on the identified performance scores.

54. The method according to claim 53, wherein normalizing the performance score of the given test mark comprises: identifying a percentile of a set of the performance scores that includes (i) the performance score of the given test mark and (ii) the identified performance scores; and setting the normalized performance score of the given test mark equal to ±d, d being a difference between the percentile and the performance score of the given test mark.

Citation Information

Patent Citations

  • Determination of thresholds to detect missing printing nozzles

    US20190105895A1

  • Method for detecting a defective print nozzle using a variable print nozzle test pattern

    US20190152219A1

  • Method and device for determining faulty print nozzles of a printing device

    US20210178754A1

  • Enhanced print defect detection

    US20210183036A1

  • Ejection control using substrate alignment features and print region alignment features

    US20210379888A1

Cited By

  • Digital printing process

    US12384170B2