A nozzle ink jet performance test platform and test method

Abstract

This invention relates to the field of inkjet printing technology, and provides a nozzle inkjet performance testing platform and method. The method includes: S1, detecting whether the bounding rectangle of the ink droplet profile completely falls within a preset grid area; S2, calculating the Euclidean distance between the ink droplet center and the theoretical landing point, and determining whether the positional accuracy is qualified; S3, using the ink droplet profile with the largest area as the main profile, identifying morphological abnormalities caused by nozzle wear or ink dripping, and simultaneously traversing non-main profiles to capture ink droplet splitting defects; S4, fusing the detection and identification results of S1 to S3 to obtain an inkjet performance diagnostic conclusion; wherein, steps S1 and S2 are independent and executed in parallel. This invention achieves complete quantitative analysis of ink droplet spatial attributes and morphological defects by independently and in parallel executing region determination and positional accuracy judgment, and pioneers a dual-branch feature extraction mechanism for main and non-main profiles, enabling the output of structured and operable diagnostic conclusions, breaking through the limitations of single-parameter detection.

Description

Technical Field

[0001] This invention relates to the field of inkjet printing technology, and more specifically, to a nozzle inkjet performance testing platform and testing method. Background Technology

[0002] In inkjet systems, the printhead plays a crucial role in accurately converting electrical signals into physical ink droplets. Its operating state directly determines the integrity of the droplet shape, the stability of its flight trajectory, and the accuracy of its landing position. These microscopic parameters collectively constitute the physical basis of the final image quality. Characteristics such as the regularity of the droplet contour, area distribution, center coordinate offset, and the presence of satellite points (tiny auxiliary droplets) have become core quantitative indicators for evaluating printhead performance and process stability in the industry.

[0003] In the field of inkjet quality inspection, existing technologies typically use high-speed image acquisition systems to obtain ink droplet ejection images and perform basic quality assessments based on image processing algorithms. These technologies often employ a single detection path (such as only region determination or only electrical signal analysis), making it difficult to simultaneously acquire multi-dimensional parameters such as the spatial assignment, positional offset, and morphological integrity of ink droplets in the same detection process, resulting in fragmented evaluation results. Summary of the Invention

[0004] Therefore, to address the problem that existing technologies, which use a single detection path, struggle to simultaneously acquire multi-dimensional parameters such as the spatial assignment, positional offset, and morphological integrity of ink droplets within the same detection process, leading to fragmented evaluation results, this invention provides a nozzle inkjet performance testing platform and method, the specific technical solution of which is as follows: A nozzle inkjet performance testing platform, comprising: The ink droplet region determination module is used to detect whether the bounding rectangle of the ink droplet outline completely falls into the preset grid area; The position accuracy determination module is used to calculate the Euclidean distance between the center of the ink droplet and the theoretical landing point. It determines whether the position accuracy is qualified based on the Euclidean distance and a preset threshold. If the Euclidean distance is not greater than the preset threshold, the position accuracy is qualified; otherwise, it is marked as an abnormal position offset. The ink droplet shape determination module is used to take the outline of the largest ink droplet as the main outline, simultaneously calculate its roundness and edge roughness, identify morphological abnormalities caused by nozzle wear or ink dripping, and at the same time traverse non-main outlines to count the number and area ratio of satellite points whose area exceeds the noise threshold, thereby capturing ink droplet splitting defects. The inkjet performance diagnostic module is used to integrate the detection and recognition results output by the droplet region determination module, the position accuracy determination module, and the droplet shape determination module to obtain inkjet performance diagnostic conclusions. Among them, the ink droplet region determination module and the position accuracy determination module are executed independently and in parallel.

[0005] In the nozzle inkjet performance testing platform described in this invention, the droplet region determination module and the position accuracy determination module are executed independently and in parallel. The determination conditions of the two are independent of each other and do not nest. It also pioneers a dual-branch feature extraction mechanism of main contour (roundness C, roughness R) and non-main contour (number of satellite points N, area ratio P), which can realize the complete quantitative analysis of the spatial attributes and morphological defects of the droplets. It can output structured and operable diagnostic conclusions, break through the limitation of single parameter detection, and solve the problem that the existing technology uses a single detection path, which makes it difficult to simultaneously obtain multi-dimensional parameters such as the spatial assignment, position offset and morphological integrity of the droplets in the same detection process, resulting in fragmented evaluation results.

[0006] Preferably, the nozzle inkjet performance testing platform includes: The ink droplet image acquisition and coordinate system calibration module is used to capture inkjet landing point images in real time, convert the images to grayscale and perform binarization segmentation, extract the contours of all connected regions, establish a pixel coordinate system with the theoretical landing point of the printhead as the origin, and calibrate the preset grid area (G). x1 G y1 ) to (G x2 G y2 ).

[0007] Preferably, the ink droplet region determination module includes: The outer rectangle acquisition unit is used to calculate the coordinates of the minimum bounding rectangle of each ink droplet profile (b). x1 ,b y1 ) to (b x2 ,b y2 ); The ink droplet boundary crossing detection unit is used to verify the four boundary conditions: b x1 ≥G x1 b x2 ≤G x2 b y1 ≥G y1 and b y2 ≤G y2 If any condition is not met, it is initially determined that the ink droplet has crossed the boundary; The boundary violation verification unit is used to calculate the area ratio between the actual ink droplet area and the theoretical grid area. When the area ratio is <0.95 or >1.05, the ink droplet is finally judged to have crossed the boundary.

[0008] Preferably, the ink droplet shape determination module includes: The main contour analysis unit is used to take the largest ink droplet contour as the main contour, and simultaneously calculate its roundness and edge roughness to identify morphological abnormalities caused by nozzle wear or ink dripping. If the roundness is <0.90, it is judged as roundness abnormality; if the edge roughness is >1.1, it is judged as edge roughness. The non-main contour analysis unit is used to traverse non-main contours and count the number N and area percentage of satellite points whose area exceeds the noise threshold. If N>2, the number of satellite points is determined to be excessive. If P>5%, the area percentage of satellite points is determined to be excessive. The area percentage P = (total area of ​​satellite points / area of ​​main contour) × 100%.

[0009] Preferably, the inkjet performance diagnostic module includes: The mapping rule base construction unit is used to establish an explicit mapping rule base between ink droplet image feature parameters, image category, and specific causes of printhead failure; The performance diagnosis conclusion unit is used to integrate the detection and recognition results output by the ink droplet region determination module, the position accuracy determination module, and the ink droplet shape determination module, and obtain inkjet performance diagnosis conclusions based on the established explicit mapping rule base. Among them, image categories include out-of-bounds, distortion, and excessive satellite dots, and specific causes of printhead failure include abnormal ink supply pressure, mechanical wear of the nozzles, excessively high drive voltage, and unsuitable ink viscosity.

[0010] A method for testing the inkjet performance of a nozzle includes the following steps: S1, Detect whether the bounding rectangle of the ink droplet contour completely falls into the preset grid area; S2, calculate the Euclidean distance between the center of the ink droplet and the theoretical landing point. Determine whether the position accuracy is qualified based on the Euclidean distance and the preset threshold. If the Euclidean distance is not greater than the preset threshold, the position accuracy is qualified; otherwise, it is marked as an abnormal position offset. S3 uses the largest ink droplet contour as the main contour, simultaneously calculates its roundness and edge roughness, identifies morphological abnormalities caused by nozzle wear or ink dripping, and traverses non-main contours to count the number and area ratio of satellite points whose area exceeds the noise threshold, thus capturing ink droplet splitting defects. S4, integrate the detection and identification results from steps S1 to S3 to obtain the inkjet performance diagnostic conclusion; Steps S1 and S2 are independent of each other and are executed in parallel.

[0011] Preferably, the nozzle inkjet performance testing method further includes the following steps: Real-time capture of inkjet landing point images is performed, the images are converted to grayscale and binarized, all connected region contours are extracted, a pixel coordinate system is established with the theoretical printhead landing point as the origin, and a preset grid area (G) is calibrated. x1 G y1 ) to (G x2 G y2 ).

[0012] Preferably, step S1 specifically includes: Calculate the coordinates of the minimum bounding rectangle of each ink droplet profile (b x1,b y1 ) to (b x2 ,b y2 ); Verify the four boundary conditions: b x1 ≥G x1 b x2 ≤G x2 b y1 ≥G y1 and b y2 ≤G y2 If any condition is not met, it is initially determined that the ink droplet has crossed the boundary; Calculate the area ratio between the actual ink droplet area and the theoretical grid area. When the area ratio is <0.95 or >1.05, the ink droplet is finally judged to be out of bounds.

[0013] Preferably, step S3 specifically includes: Using the largest ink droplet outline as the main outline, its roundness and edge roughness are calculated simultaneously to identify morphological abnormalities caused by nozzle wear or ink dripping. If the roundness is <0.90, it is judged as roundness abnormality; if the edge roughness is >1.1, it is judged as edge roughness. Simultaneously, non-main contours are traversed, and the number N and area percentage of satellite points with areas exceeding the noise threshold are counted. If N>2, the number of satellite points is determined to be excessive. If P>5%, the area percentage of satellite points is determined to be excessive. The area percentage P = (total area of ​​satellite points / area of ​​main contour) × 100%.

[0014] Preferably, step S4 specifically includes: Establish an explicit mapping rule base between ink droplet image feature parameters, image category, and specific causes of printhead malfunctions; By integrating the detection and identification results from steps S1 to S3, and based on the established explicit mapping rule base, an inkjet performance diagnostic conclusion is obtained. Among them, image categories include out-of-bounds, distortion, and excessive satellite dots, and specific causes of printhead failure include abnormal ink supply pressure, mechanical wear of the nozzles, excessively high drive voltage, and unsuitable ink viscosity. Attached Figure Description

[0015] The invention will be further understood from the following description taken in conjunction with the accompanying drawings. The components in the drawings are not necessarily drawn to scale, but rather the emphasis is on illustrating the principles of the embodiments. In different views, the same reference numerals designate corresponding parts.

[0016] Figure 1 This is a schematic diagram of the overall process of a nozzle inkjet performance testing method according to an embodiment of the present invention; Figure 2 This is a flowchart illustrating step S3 in one embodiment of the present invention; Figure 3This is a flowchart illustrating step S4 in one embodiment of the present invention. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to its embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and do not limit the scope of protection of the invention.

[0018] It should be noted that when an element is referred to as being "fixed to" another element, it can be directly attached to the other element or there may be an intervening element. When an element is referred to as being "connected to" another element, it can be directly connected to the other element or there may be an intervening element. The terms "vertical," "horizontal," "left," "right," and similar expressions used herein are for illustrative purposes only and do not represent the only possible implementation.

[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0020] In this invention, "first" and "second" do not represent a specific quantity or order, but are merely used to distinguish names.

[0021] Before describing the specific embodiments of the present invention, a brief description of the prior art will be given first.

[0022] As the core process of modern digital printing systems, inkjet printing technology has been widely used in diverse scenarios such as office document output, industrial marking and coding, textile digital printing, ceramic glaze decoration, biochip preparation and additive manufacturing (3D printing) due to its advantages such as non-contact operation, high-resolution imaging, strong color performance and wide substrate adaptability.

[0023] In inkjet systems, the printhead plays a crucial role in accurately converting electrical signals into physical ink droplets. Its operational status directly determines the droplet's morphological integrity, trajectory stability, and landing accuracy. These microscopic parameters collectively constitute the physical basis of the final image quality. Characteristics such as droplet contour regularity, area distribution, center coordinate offset, and the presence of satellite points (tiny auxiliary droplets) have become core quantitative indicators for evaluating printhead performance and process stability in the industry. With the continuous improvement in image precision and color consistency requirements in high-end printing, and the stringent standards for continuous equipment reliability and maintenance efficiency in industrial automated production lines, implementing scientific and systematic quality monitoring and status assessment of the inkjet process has become a vital step in ensuring production yield and optimizing process parameters.

[0024] Currently, the industry is actively promoting the evolution of inkjet testing technology towards multi-parameter fusion, high-precision quantification, and intelligent analysis. It is committed to acquiring comprehensive information such as the spatial distribution, positional accuracy, and morphological characteristics of ink droplets simultaneously through integrated testing methods. This provides comprehensive and objective data support for printhead R&D, production quality inspection, and equipment maintenance, helping inkjet printing technology to continue to deepen its application on the track of high-quality development.

[0025] In the field of inkjet quality inspection, existing technologies typically employ high-speed image acquisition systems to obtain images of ink droplet ejection and then use image processing algorithms to perform basic quality assessments. The position determination stage generally uses the centroid localization method: after extracting the ink droplet contour through threshold segmentation, its geometric centroid coordinates are calculated and compared with the Euclidean distance of a preset grid center point. The position is then judged to be acceptable based on a preset distance threshold (e.g., ±5μm). For shape inspection, mainstream solutions focus on calculating the contour roundness parameter (formula: C= Where A is the area of ​​the outline. For example, the perimeter of the outline is used to set a fixed threshold (such as C≥0.85) as a criterion for shape qualification; some systems are supplemented by the ratio analysis of the actual area of ​​the ink droplet to the theoretical spray area to identify area anomalies.

[0026] In terms of system workflow design, existing detection methods typically treat position determination and shape analysis as sequential processing modules. This means that all droplet positions are first screened, and then shape parameter calculations are performed on all samples (including droplets whose positions have been determined to be abnormal). Some industrial systems attempt to establish simple rule bases, such as triggering a "nozzle status attention" prompt when the roundness value is below a threshold for multiple consecutive frames, or performing basic classification based on the number of satellite points. However, the correlation logic between relevant parameters and equipment faults largely relies on manual experience, and a systematic multi-parameter fusion diagnostic framework has not yet been formed. Current technology can achieve preliminary quantitative assessment of droplet position and morphology, providing basic quality feedback for inkjet processes.

[0027] In summary, the existing technology has the following shortcomings: I. Fragmented Detection Dimensions: Existing technologies often employ a single detection path (such as only area determination or only electrical signal analysis), making it difficult to simultaneously acquire multi-dimensional parameters such as the spatial assignment, positional offset, and morphological integrity of ink droplets in the same detection process, resulting in fragmented evaluation results.

[0028] 2. Confusion in judgment logic: The region judgment (whether the ink droplet falls into the preset grid) and the position accuracy judgment (the degree of center coordinate offset) are often designed as serial or coupled logic, failing to clarify their independence, and are prone to misjudging situations such as "outside the region but the position is accurate" or "in the region but the offset is large".

[0029] Third, the shape analysis is superficial: existing image processing solutions mostly focus on basic parameters such as the area and centroid of the main ink droplet, lack quantitative calculation of the roundness of the main contour and the roughness of the edge, and generally ignore the identification and statistics of non-main contour features such as satellite points, making it difficult to capture subtle defects such as ink droplet splitting and deformation.

[0030] IV. Hollow Diagnostic Conclusions: The test outputs are mostly "qualified / unqualified" or isolated parameter values. No explicit mapping relationship is established between ink droplet image features (such as abnormal roundness, excessive number of satellite points) and specific printhead faults (such as nozzle wear, abnormal drive voltage). The value of operation and maintenance guidance is limited.

[0031] 5. Insufficient industrial adaptability: Some high-precision algorithms rely on complex models or a large amount of training data, resulting in high computational latency. This makes it difficult to meet the real-time detection requirements of high-speed inkjet printing (≥30 frames / second) on production lines, and the generalization ability to different ink characteristics and printhead models is weak.

[0032] One of the objectives of this invention is to provide a highly integrated, logically clear, and diagnostically accurate nozzle inkjet performance testing platform and method, aiming to systematically solve core problems in inkjet printhead status assessment such as the lack of multi-parameter coordination, fuzzy judgment logic, and difficulty in fault attribution.

[0033] Specifically, this invention aims to construct a unified detection framework: First, it explicitly designs the region determination of ink droplets (verifying whether the circumscribed rectangle of the ink droplet contour completely falls within the preset grid area) and the positional accuracy determination (calculating the Euclidean distance deviation between the ink droplet center and the theoretical landing point) as independent and parallel determination modules, ensuring that the two types of spatial attribute evaluations do not interfere with each other and that the conclusions are reliable; Second, it innovatively proposes a dual-branch analysis strategy for ink droplet shape determination—using the contour with the largest area as the main contour, and simultaneously calculating its roundness (C= ) and edge roughness (R= It accurately identifies morphological abnormalities caused by nozzle wear or ink buildup; simultaneously, it traverses non-primary contours and performs statistical analysis on the number (N) and area ratio (P) of satellite points whose area exceeds the noise threshold, effectively capturing ink droplet splitting defects; finally, it deeply integrates the above multi-dimensional detection results to establish an explicit mapping rule base of "ink droplet image feature parameters → image category (such as out of bounds, deformation, satellite point over-standard) → specific cause of printhead failure (such as abnormal ink supply pressure, nozzle mechanical wear, excessive drive voltage, unsuitable ink viscosity)", directly outputting structured and operable diagnostic conclusions.

[0034] This solution ensures a real-time detection efficiency of over 30 frames per second while significantly improving the completeness of detection dimensions, the rigor of judgment logic, and the directionality of fault tracing. It provides scientific, efficient, and easy-to-use technical tools for the research and development testing, production line quality inspection, and preventive maintenance of inkjet equipment, effectively promoting the upgrade of inkjet printing technology towards intelligent and refined operation and maintenance.

[0035] In one embodiment, the present invention provides a nozzle inkjet performance testing platform, which includes an ink droplet image acquisition and coordinate system calibration module, an ink droplet region determination module, a position accuracy determination module, an ink droplet shape determination module, and an inkjet performance diagnosis module.

[0036] The ink droplet image acquisition and coordinate system calibration module is used to capture inkjet landing point images in real time, convert the images to grayscale and perform binarization segmentation, extract the contours of all connected regions, establish a pixel coordinate system with the theoretical landing point of the printhead as the origin, and calibrate the preset grid area (G). x1 G y1 ) to (G x2 G y2 Specifically, high-resolution industrial cameras can be used to capture images of inkjet landing points in real time.

[0037] The droplet region determination module is used to detect whether the bounding rectangle of the droplet contour completely falls within the preset grid area. The position accuracy determination module is used to calculate the Euclidean distance between the center of the droplet contour and the theoretical landing point. Based on the Euclidean distance and a preset threshold (such as 0.5 pixels), it determines whether the position accuracy is qualified. If the Euclidean distance is not greater than the preset threshold, the position accuracy is qualified; otherwise, it is marked as an abnormal position offset.

[0038] In the spatial attribute analysis layer, the system executes two independent decisions in strict parallel: the ink droplet region determination module and the position accuracy determination module are executed independently and in parallel.

[0039] Specifically, the ink droplet region determination module includes an external rectangle acquisition unit, an ink droplet boundary crossing determination unit, and a boundary crossing reinforcement verification unit.

[0040] The outer rectangle acquisition unit is used to calculate the coordinates of the minimum bounding rectangle of each droplet profile (b). x1,b y1 ) to (b x2 ,b y2 The droplet boundary crossing detection unit is used to verify the four boundary conditions: b x1 ≥G x1 b x2 ≤G x2 b y1 ≥G y1 and b y2 ≤G y2 If any condition is not met, it is initially determined that the ink droplet has crossed the boundary. The boundary crossing enhancement verification unit is used to calculate the area ratio between the actual ink droplet area and the theoretical grid area. When the area ratio is <0.95 or >1.05, the ink droplet is finally determined to have crossed the boundary.

[0041] The droplet shape determination module uses the largest droplet outline as the main outline, simultaneously calculates its roundness and edge roughness, identifies morphological abnormalities caused by nozzle wear or ink dripping, and traverses non-main outlines to count the number and area ratio of satellite points whose area exceeds the noise threshold, thereby capturing droplet splitting defects.

[0042] In the shape analysis layer, an innovative dual-branch feature extraction method is implemented: the main contour branch locks the contour with the largest area, and the roundness C is calculated simultaneously. ) and roughness R ( ); Non-main contour branches traverse the remaining contours, filter satellite points whose area exceeds the noise threshold (0.05mm²), and count their number N and total area percentage P.

[0043] As a preferred technical solution, the ink droplet shape determination module includes a main contour analysis unit and a non-main contour analysis unit.

[0044] The main contour analysis unit uses the largest ink droplet contour as the main contour, and simultaneously calculates its roundness and edge roughness to identify morphological abnormalities caused by nozzle wear or ink dripping. If the roundness is <0.90, it is judged as roundness abnormality; if the edge roughness is >1.1, it is judged as edge roughness.

[0045] The non-main contour analysis unit is used to traverse non-main contours and count the number N and area percentage of satellite points whose area exceeds the noise threshold. If N>2, the number of satellite points is determined to be excessive. If P>5%, the area percentage of satellite points is determined to be excessive. The area percentage P = (total area of ​​satellite points / area of ​​main contour) × 100%.

[0046] The inkjet performance diagnostic module is used to integrate the detection and recognition results output by the droplet region determination module, the position accuracy determination module, and the droplet shape determination module to obtain inkjet performance diagnostic conclusions.

[0047] All parameters are matched with a preset mapping library by the rule engine, and finally a structured diagnostic report is output. The report clearly marks the "droplet image category" (such as out of bounds, abnormal roundness, and excessive satellite points) and the corresponding "printhead failure cause" (such as abnormal ink supply pressure, nozzle wear, excessive drive voltage, and unsuitable ink viscosity). This achieves accurate, traceable, and operable end-to-end diagnosis from image features to the root cause of the fault. The entire processing latency is less than 33ms, which meets the requirements of industrial-grade real-time detection.

[0048] As a preferred technical solution, the inkjet performance diagnostic module includes a mapping rule base construction unit and a performance diagnostic conclusion unit.

[0049] The mapping rule base construction unit is used to establish an explicit mapping rule base between ink droplet image feature parameters, image categories, and specific causes of printhead malfunctions; the performance diagnosis conclusion unit is used to integrate the detection and recognition results output by the ink droplet region determination module, position accuracy determination module, and ink droplet shape determination module, and obtain inkjet performance diagnosis conclusions based on the established explicit mapping rule base.

[0050] In summary, in the nozzle inkjet performance testing platform of this invention, the droplet region determination module and the position accuracy determination module are executed independently and in parallel. The determination conditions of the two are independent of each other and do not nest. Furthermore, it pioneers a dual-branch feature extraction mechanism based on the main contour (roundness C, roughness R) and non-main contour (number of satellite points N, area ratio P), which enables comprehensive quantitative analysis of the spatial attributes and morphological defects of droplets. It can output structured and operable diagnostic conclusions, breaking through the limitations of single-parameter detection. It also solves the problem that existing technologies use a single detection path, making it difficult to simultaneously obtain multi-dimensional parameters such as the spatial attribution, position offset, and morphological integrity of droplets in the same detection process, resulting in fragmented evaluation results.

[0051] like Figure 1 As shown, an embodiment of the present invention also provides a method for testing the inkjet performance of a nozzle, which includes the following steps: S1, detect whether the bounding rectangle of the ink droplet contour falls completely into the preset grid area.

[0052] As a preferred technical solution, step S1 specifically includes: S11, calculate the coordinates of the minimum bounding rectangle of each ink droplet profile (b x1 ,b y1 ) to (b x2 ,b y2 ).

[0053] S12, Verify the four boundary conditions: b x1 ≥G x1 b x2 ≤G x2 b y1 ≥Gy1 and b y2 ≤G y2 If any condition is not met, it is initially determined that the ink droplet has crossed the boundary.

[0054] S13, Calculate the area ratio between the actual ink droplet area and the theoretical grid area. When the area ratio is <0.95 or >1.05, the ink droplet is ultimately judged to be out of bounds.

[0055] S2, calculate the Euclidean distance between the center of the ink droplet and the theoretical landing point. Determine whether the position accuracy is qualified based on the Euclidean distance and the preset threshold. If the Euclidean distance is not greater than the preset threshold, the position accuracy is qualified; otherwise, it is marked as an abnormal position offset.

[0056] Steps S1 and S2 are independent and executed in parallel. Specifically, the Euclidean distance between the centroid of the ink droplet profile and the center of the preset grid is calculated. = ;like If the value is ≤0.5 pixels (the threshold can be adjusted according to the device's precision), the position accuracy is considered acceptable; otherwise, the mark position is considered abnormally offset.

[0057] Specifically, the Euclidean distance deviation is based on the intuitive definition of the straight-line distance between two points in Euclidean geometry. It directly quantifies the absolute spatial offset between the centroid of the ink droplet profile and the theoretical center of the preset grid in the two-dimensional pixel coordinate system. Its characteristics are clear geometric meaning, simple calculation logic, and because the coordinate unit is uniformly pixel (dimensionless), it effectively avoids the potential defect of Euclidean distance being sensitive to scale. At the same time, it is strictly decoupled from the region determination module to ensure the independence and objectivity of the determination dimension of "whether the position is accurate", providing accurate quantitative basis for directional faults such as oblique spraying and piezoelectric crystal damage.

[0058] In terms of performance, Euclidean distance deviation calculation only requires basic arithmetic operations (squaring, summing, and square root), with a single judgment latency of less than microseconds, perfectly adapting to the real-time requirements of industrial inkjet inspection ≥30 frames / second; compared to complex metrics such as Mahalanobis distance that require covariance matrix inversion, it has no risk of "curse of dimensionality" in two-dimensional low-dimensional space, with high computational stability and low resource consumption; the threshold judgment mechanism (e.g., ≤0.5 pixels) has clear rules, is easy to optimize in engineering, and is highly compatible with the standardized pixel coordinate system of this solution, significantly improving the efficiency and reliability of position offset detection, and meeting the stringent requirements of high-speed quality inspection in production lines for lightweight and deterministic algorithms.

[0059] S3 uses the largest ink droplet contour as the main contour, simultaneously calculates its roundness and edge roughness, identifies morphological abnormalities caused by nozzle wear or ink dripping, and traverses non-main contours to count the number and area ratio of satellite points whose area exceeds the noise threshold, thereby capturing ink droplet splitting defects. As a preferred technical solution, such as Figure 2 As shown, step S3 specifically includes: S31, using the largest ink droplet outline as the main outline, calculate its area A and perimeter. Simultaneously calculate its roundness and edge roughness to identify morphological abnormalities caused by nozzle wear or ink buildup. If roundness C= If the value is less than 0.90, it is considered an abnormal roundness. If the edge roughness R = >1.1, is judged as rough edge. , These represent the mean and standard deviation of the distances from each point on the profile to the centroid, respectively.

[0060] As a preferred technical solution, roundness Where A is the actual projected area of ​​the main contour of the ink droplet, which is obtained through image binarization segmentation and contour analysis; This is the length of the main outline boundary of the ink droplet, i.e., the actual perimeter of the outline, which can be obtained through an outline tracking algorithm. These represent the ink viscosity correction factor and the average area of ​​normal ink droplets from the same printhead model, respectively.

[0061] Specifically, the ink viscosity correction factor is used to quantify the impact of ink viscosity on roundness calculation (high viscosity inks tend to cause droplet deformation). It can be directly calibrated according to the ink type, such as λ≈0.01 for low viscosity ink (water-based) and λ≈0.05 for high viscosity ink (UV-cured). Basic roundness. The term is used to quantify the regularity of the profile (value range [0,1], 1 represents a perfect circle) and identify defects such as nozzle wear; As a compensation term, it is used to correct ink property interference: when the ink viscosity is high, the ink droplet ejection speed is slow, which easily produces tailing or deformation. The enhancement term, through the ink viscosity correction factor, can make the roundness value more sensitive to the actual shape defects (rather than ink property interference).

[0062] The basic roundness is limited to no less than 0.1 to filter out invalid contours, such as small contours caused by ink droplet splatter.

[0063] In summary, this roundness function enhances the diagnostic accuracy of roundness parameters by introducing an ink property compensation mechanism, enabling explicit mapping between "ink droplet image feature parameters → specific causes of printhead malfunctions," such as associating abnormal roundness with "nozzle mechanical wear."

[0064] As a preferred technical solution, edge roughness = global roughness component + sensitivity coefficient × curvature compensation component. Wherein, the global roughness component = average distance from the profile point to the centroid / standard deviation of the distance from the profile point to the centroid. The average distance from the profile point to the centroid can be understood as the average distance from all boundary points of the main profile of the ink droplet to its geometric centroid, reflecting the average degree of expansion of the ink droplet profile; a larger value indicates a larger overall size of the ink droplet. The standard deviation of the distance from the profile point to the centroid can be understood as the dispersion of the distance from the profile point to the centroid, used to quantify the edge fluctuation intensity of the profile. A larger value indicates more significant local unevenness of the profile (such as burrs caused by ink dripping).

[0065] Curvature compensation component = Edge curvature variance / Calibrated maximum curvature variance. The edge curvature variance is the variance of curvature at each point on the main contour, used to capture abrupt changes in local curvature, such as sharp protrusions caused by crystallization. The calibrated maximum curvature variance serves as the denominator, normalizing the curvature variance to avoid underestimating the curvature fluctuations of large ink droplets. The sensitivity coefficient is a configurable parameter that adjusts the weight of the curvature component, adapting to different imaging conditions and ink types. For high-resolution cameras (>5μm / pixel), it is set to 0.08 to enhance sensitivity to micro-defects; for low-viscosity inks, it is set to 0.03 to suppress flow noise interference.

[0066] Here, to integrate and filter small outlines of ink droplet splatter, it is mandatory to constrain the number of outline points to be no less than a threshold such as 15, the average centroid distance to be no less than a preset minimum distance threshold to filter small outlines, and the global roughness component to be no greater than a roughness threshold such as 0.5. That is, the edge roughness R = global roughness component + sensitivity coefficient × curvature compensation component is calculated only when all of the following conditions are met simultaneously: the number of outline points is no less than 15, the average centroid distance is no less than the preset minimum threshold to filter small outlines, and the global roughness component is no greater than 0.5. If any condition is not met, it is directly judged as an invalid splatter outline.

[0067] The global roughness component directly reflects the overall irregularity of the profile; for example, a value >0.25 indicates edge burrs. The curvature compensation component is mainly used for: 1. Microscopic enhancement, which is sensitive to local sharp defects (such as crystal protrusions), where the curvature variance increases dramatically; 2. Curvature variance normalization, eliminating the influence of droplet size, making curvature fluctuations of small droplets easier to detect. Thus, by making the edge roughness R = global roughness component + sensitivity coefficient × curvature compensation component, a dual-component collaborative diagnostic mechanism can be achieved, thereby accurately identifying microscopic defects such as nozzle ink buildup and nozzle crystallization, supporting the output of the fault mapping rule base (e.g., R > 1.1 → "nozzle ink buildup / nozzle crystallization").

[0068] S32. Simultaneously traverse non-main contours and filter contours with an area > 0.05 mm² (noise threshold) as satellite points. Perform statistics on the number N and area percentage of satellite points. If N > 2, it is determined that the number of satellite points exceeds the standard. If P > 5%, it is determined that the area percentage of satellite points exceeds the standard. Wherein, the area percentage P = (total area of ​​satellite points / area of ​​main contour) × 100%.

[0069] The droplet shape dual-branch feature extraction method in step S2, based on the differences in the physical causes of inkjet defects, decouples the shape analysis into two logically parallel and clearly defined branches: the main contour branch (locking the contour with the largest area) simultaneously calculates roundness and roughness, i.e., the ratio of the standard deviation to the mean of the distance from the contour point to the centroid, quantifying edge fluctuations; the non-main contour branch systematically traverses the remaining contours, using an area threshold (0.05mm)... 2 After filtering out noise, satellite points are identified, and the number N and area ratio P are counted. The dual-branch design is not a simple parameter superposition, but rather a corresponding separation of "nozzle body damage defects" (such as wear leading to decreased roundness, crystallization causing edge burrs) and "jet dynamics instability defects" (such as voltage abnormalities causing droplet splitting) at the algorithm level, realizing targeted tracing of the root cause of defects.

[0070] This dual-branch feature extraction method for shapes overcomes the limitations of existing technologies that rely solely on shallow analysis using a single area / perimeter parameter. The dual branches construct a three-dimensional evaluation system that integrates "main body morphological integrity + quantification of auxiliary defects": roundness is sensitive to 0.05mm-level contour deformation (C<0.90 indicates early warning of nozzle wear), roughness can capture micron-level edge burrs (R>1.1 is associated with ink dripping / crystallization), and satellite point statistics accurately identify micro-split phenomena that are difficult to discern with the naked eye (N>2 or P>5% indicates drive / ink problems); three thresholds form complementary judgment logic, and the algorithm only contains basic geometric and statistical operations (single frame <10ms), with no model training dependency. The parameter thresholds can be flexibly configured according to the printhead model, combining high robustness and engineering applicability. S4. Combine the detection and identification results from steps S1 to S3 to obtain the inkjet performance diagnostic conclusion.

[0071] As a preferred technical solution, before step S1, the nozzle inkjet performance testing method further includes the following steps: real-time capture of inkjet landing point images (i.e., ink droplet images), conversion of the images to grayscale images and binarization segmentation, extraction of the contours of all connected regions, establishment of a pixel coordinate system with the theoretical landing point of the printhead as the origin, and calibration of the preset grid area (G x1 G y1 ) to (G x2 G y2 ).

[0072] As a preferred technical solution, such as Figure 3 As shown, step S4 specifically includes: S41. Establish an explicit mapping rule library between ink droplet image feature parameters, image category, and specific causes of printhead malfunctions.

[0073] S42, integrate the detection and identification results from steps S1 to S3, and obtain inkjet performance diagnostic conclusions based on the established explicit mapping rule base.

[0074] Among them, image categories include out-of-bounds, distortion, and excessive satellite dots, and specific causes of printhead failure include abnormal ink supply pressure, mechanical wear of the nozzles, excessively high drive voltage, and unsuitable ink viscosity.

[0075] Specifically, by integrating all the judgment results from steps S1 to S3, a precise diagnostic report is generated based on a preset mapping rule base: Area judgment exceeding boundaries (including...) Ratio anomaly verification) → "Ink supply pressure abnormal / partial nozzle blockage"; position accuracy abnormal ( >0.5 pixels) → "Slanted spray / damaged piezoelectric crystal / local contamination of nozzle causing spray direction deviation"; Abnormal roundness of main contour (C<0.90) → "Nozzle aging / mechanical wear of nozzle"; Rough edge of main contour (R>1.1) → "Ink residue on nozzle / crystallization at the nozzle"; Excessive number of satellite dots (N>2) or excessive area ratio (P>5%) → "Drive voltage too high / Ink surface tension failure / Improper waveform settings / Ink viscosity too low". The report clearly labels the "droplet image category" (such as out of bounds, position deviation, abnormal roundness, rough edge, abnormal satellite dots, etc.) and the corresponding "specific cause of printhead failure" in a structured format.

[0076] The core advantage of this invention lies in achieving a dual breakthrough in "logical decoupling" and "deep attribution." Addressing the limitations of existing solutions where region determination and position accuracy judgment are logically coupled, and shape analysis relies solely on a single parameter, this invention innovatively constructs an independent and parallel dual-space determination module (region boundary verification combined with area ratio verification, and independent calculation of position offset). It also pioneers a dual-branch feature extraction mechanism for the main contour (roundness C, roughness R) and non-main contour (number of satellite points N, area ratio P), achieving comprehensive quantitative analysis of ink droplet spatial attributes and morphological defects. More importantly, the system accurately maps out-of-specification errors to specific root causes such as "oblique spraying / damaged piezoelectric crystal / dirty nozzles," and associates out-of-specification satellite points with "driving voltage / ink properties," completely overcoming the diagnostic gaps inherent in existing technologies that "only output qualified / unqualified or isolated parameter values."

[0077] This invention does not rely on a large amount of training data or complex models. This solution not only significantly improves the accuracy of printhead fault identification, but also transforms the detection results into structured maintenance guidelines, enabling maintenance personnel to quickly locate specific problems such as "nozzle wear" and "improper waveform settings," greatly reducing downtime. It provides a solution for intelligent quality inspection and preventive maintenance of inkjet equipment that combines technical rigor with industrial applicability.

[0078] In summary, the nozzle inkjet performance testing platform and method described in this invention have the following technical advantages: 1. Independent Parallel Architecture for Region Determination and Position Accuracy Judgment: The "whether the ink droplet completely falls into the preset grid area" (region determination) and the "offset of the ink droplet center relative to the theoretical landing point" (position accuracy judgment) are explicitly designed as logically decoupled, synchronously executed dual-determination modules. The determination conditions for these two modules are independent and not nested. Region determination uses the four-boundary coordinate verification of the circumscribed rectangle (b... x1 ≥G x1 b x2 ≤G x2 b y1 ≥G y1 and b y2 ≤G y2 Combined with area ratio ( A dual verification mechanism (<0.95 or >1.05) is used to independently calculate the Euclidean distance for position accuracy determination. It compares the results with a threshold, thus fundamentally avoiding misjudgments caused by logical coupling in existing technologies, such as "accurate location outside the region" or "large offset within the region".

[0079] The dual-branch feature extraction method for determining the shape of ink droplets: Based on the contour with the largest area, two analysis paths are executed simultaneously - (1) Main contour branch: accurately calculate the roundness C= With roughness R= , and set C<0.90 to determine nozzle wear, R>1.1 to determine ink hanging / crystallization; (2) Non-main contour branches: traverse the remaining contours, with an area>0.05mm 2 Satellite points are selected based on a threshold, and the number N and area ratio P are statistically analyzed. Splitting defects are determined when N>2 or P>5%. This dual-branch structure achieves full coverage analysis of the main shape of the ink droplet and its associated defects, breaking through the limitations of single-parameter detection.

[0080] Explicit mapping rule base between ink droplet characteristic parameters and printhead malfunctions: Establishing a precise correspondence between detection results and root causes of malfunctions, including: Exceeding limits → “Slanted spraying / damaged piezoelectric crystal / localized nozzle contamination”; Area exceeding boundaries → “Abnormal ink supply pressure / partial nozzle blockage”; Abnormal roundness → “Nozzle aging / mechanical wear of nozzle”; Rough edges → “Ink residue on nozzle / crystallization at the nozzle”; Exceeding satellite point limits → “Drive voltage too high / ink surface tension failure / improper waveform settings / insect viscosity too low”; This mapping rule transforms abstract image features into operable maintenance guidelines, enabling structured and traceable output of diagnostic conclusions.

[0081] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0082] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these all fall within the protection scope of the present invention. Therefore, the protection scope of this invention patent should be determined by the appended claims.

Claims

1. A nozzle inkjet performance testing platform, characterized in that, The nozzle inkjet performance testing platform includes: The ink droplet region determination module is used to detect whether the bounding rectangle of the ink droplet outline completely falls into the preset grid area; The position accuracy determination module is used to calculate the Euclidean distance between the center of the ink droplet and the theoretical landing point. It determines whether the position accuracy is qualified based on the Euclidean distance and a preset threshold. If the Euclidean distance is not greater than the preset threshold, the position accuracy is qualified; otherwise, it is marked as an abnormal position offset. The ink droplet shape determination module is used to take the outline of the largest ink droplet as the main outline, simultaneously calculate its roundness and edge roughness, identify morphological abnormalities caused by nozzle wear or ink dripping, and at the same time traverse non-main outlines to count the number and area ratio of satellite points whose area exceeds the noise threshold, thereby capturing ink droplet splitting defects. The inkjet performance diagnostic module is used to integrate the detection and recognition results output by the droplet region determination module, the position accuracy determination module, and the droplet shape determination module to obtain inkjet performance diagnostic conclusions. Among them, the ink droplet region determination module and the position accuracy determination module are executed independently and in parallel.

2. The nozzle inkjet performance testing platform as described in claim 1, characterized in that, The aforementioned nozzle inkjet performance testing platform includes: The ink droplet image acquisition and coordinate system calibration module is used to capture inkjet landing point images in real time, convert the images to grayscale and perform binarization segmentation, extract the contours of all connected regions, establish a pixel coordinate system with the theoretical landing point of the printhead as the origin, and calibrate the preset grid area (G). x1 G y1 ) to (G x2 G y2 ).

3. The nozzle inkjet performance testing platform as described in claim 2, characterized in that, The ink droplet region determination module includes: The outer rectangle acquisition unit is used to calculate the coordinates of the minimum bounding rectangle of each ink droplet profile (b). x1 ,b y1 ) to (b x2 ,b y2 ); The ink droplet boundary crossing detection unit is used to verify the four boundary conditions: b x1 ≥G x1 b x2 ≤G x2 b y1 ≥G y1 and b y2 ≤G y2 If any condition is not met, it is initially determined that the ink droplet has crossed the boundary; The boundary violation verification unit is used to calculate the area ratio between the actual ink droplet area and the theoretical grid area. When the area ratio is <0.95 or >1.05, the ink droplet is finally judged to have crossed the boundary.

4. The nozzle inkjet performance testing platform as described in claim 3, characterized in that, The ink droplet shape determination module includes: The main contour analysis unit is used to take the largest ink droplet contour as the main contour, and simultaneously calculate its roundness and edge roughness to identify morphological abnormalities caused by nozzle wear or ink dripping. If the roundness is <0.90, it is judged as roundness abnormality; if the edge roughness is >1.1, it is judged as edge roughness. The non-main contour analysis unit is used to traverse non-main contours and count the number N and area percentage of satellite points whose area exceeds the noise threshold. If N>2, the number of satellite points is determined to be excessive. If P>5%, the area percentage of satellite points is determined to be excessive. The area percentage P = (total area of ​​satellite points / area of ​​main contour) × 100%.

5. The nozzle inkjet performance testing platform as described in claim 4, characterized in that, The inkjet performance diagnostic module includes: The mapping rule base construction unit is used to establish an explicit mapping rule base between ink droplet image feature parameters, image category, and specific causes of printhead failure; The performance diagnosis conclusion unit is used to integrate the detection and recognition results output by the ink droplet region determination module, the position accuracy determination module, and the ink droplet shape determination module, and obtain inkjet performance diagnosis conclusions based on the established explicit mapping rule base. Among them, image categories include out-of-bounds, distortion, and excessive satellite dots, and specific causes of printhead failure include abnormal ink supply pressure, mechanical wear of the nozzles, excessively high drive voltage, and unsuitable ink viscosity.

6. A method for testing the inkjet performance of a nozzle, characterized in that, The nozzle inkjet performance testing method includes the following steps: S1, Detect whether the bounding rectangle of the ink droplet contour completely falls into the preset grid area; S2, calculate the Euclidean distance between the center of the ink droplet and the theoretical landing point. Determine whether the position accuracy is qualified based on the Euclidean distance and the preset threshold. If the Euclidean distance is not greater than the preset threshold, the position accuracy is qualified; otherwise, it is marked as an abnormal position offset. S3 uses the largest ink droplet contour as the main contour, simultaneously calculates its roundness and edge roughness, identifies morphological abnormalities caused by nozzle wear or ink dripping, and traverses non-main contours to count the number and area ratio of satellite points whose area exceeds the noise threshold, thus capturing ink droplet splitting defects. S4, integrate the detection and identification results from steps S1 to S3 to obtain the inkjet performance diagnostic conclusion; Steps S1 and S2 are independent of each other and are executed in parallel.

7. The method for testing the inkjet performance of a nozzle as described in claim 6, characterized in that, The nozzle inkjet performance testing method also includes the following steps: Real-time capture of inkjet landing point images is performed, the images are converted to grayscale and binarized, all connected region contours are extracted, a pixel coordinate system is established with the theoretical printhead landing point as the origin, and a preset grid area (G) is calibrated. x1 G y1 ) to (G x2 G y2 ).

8. The method for testing the inkjet performance of a nozzle as described in claim 7, characterized in that, Step S1 specifically includes: Calculate the coordinates of the minimum bounding rectangle of each ink droplet profile (b x1 ,b y1 ) to (b x2 ,b y2 ); Verify the four boundary conditions: b x1 ≥G x1 b x2 ≤G x2 b y1 ≥G y1 and b y2 ≤G y2 If any condition is not met, it is initially determined that the ink droplet has crossed the boundary; Calculate the area ratio between the actual ink droplet area and the theoretical grid area. When the area ratio is <0.95 or >1.05, the ink droplet is finally judged to be out of bounds.

9. The method for testing the inkjet performance of a nozzle as described in claim 8, characterized in that, Step S3 specifically includes: Using the largest ink droplet outline as the main outline, its roundness and edge roughness are calculated simultaneously to identify morphological abnormalities caused by nozzle wear or ink dripping. If the roundness is <0.90, it is judged as roundness abnormality; if the edge roughness is >1.1, it is judged as edge roughness. Simultaneously, non-main contours are traversed, and the number N and area percentage of satellite points with areas exceeding the noise threshold are counted. If N>2, the number of satellite points is determined to be excessive. If P>5%, the area percentage of satellite points is determined to be excessive. The area percentage P = (total area of ​​satellite points / area of ​​main contour) × 100%.

10. The method for testing the inkjet performance of a nozzle as described in claim 9, characterized in that, Step S4 specifically includes: Establish an explicit mapping rule base between ink droplet image feature parameters, image category, and specific causes of printhead malfunctions; By integrating the detection and identification results from steps S1 to S3, and based on the established explicit mapping rule base, an inkjet performance diagnostic conclusion is obtained. Among them, image categories include out-of-bounds, distortion, and excessive satellite dots, and specific causes of printhead failure include abnormal ink supply pressure, mechanical wear of the nozzles, excessively high drive voltage, and unsuitable ink viscosity.

Images