A method and system for detecting the uniformity of ink application on a printing stencil

Abstract

The present application relates to the technical field of printing ink detection, and particularly relates to a printing template ink uniformity detection method and system, which comprises the following steps: obtaining surface attribute information of a printing object before printing, and generating a detection control chart, the detection control chart carrying area attribute information registered with the position of printed graphics; when collecting the image of the printing object after printing, dynamically adjusting the imaging parameters according to the area attribute information corresponding to the current position, and obtaining an optimized image; performing spatial frequency separation on the optimized image to obtain a frequency component image; calling an analysis model associated with the area attribute information according to the area attribute information corresponding to the current position, and analyzing the frequency component image to obtain the uniformity detection result of the printing area. The present application realizes adaptive imaging and analysis through pre-printing collection and post-printing detection, avoids the problem that fixed parameter detection cannot take into account different surface attribute areas, and improves the accuracy and reliability of ink uniformity detection.

Description

Technical Field

[0001] This invention relates to the field of printing ink application testing technology, and in particular to a method and system for testing the uniformity of ink application on printing templates. Background Technology

[0002] In the field of aseptic packaging for liquid foods such as dairy products and beverages, the substrate is typically a multi-layered material composed of cardboard, aluminum foil, and polymer film. Ink on the printing stencil is transferred to the substrate surface through a transfer process to form images. The uniformity of ink distribution on the substrate surface is termed the ink uniformity. The substrate's ability to accept and spread ink is determined by its surface properties. When the surface energy of a localized area of ​​the substrate is lower than the critical wetting threshold required for ink spreading, ink transfer and spreading are hindered, resulting in a thinner ink layer in that area and uneven ink application.

[0003] Current technologies primarily rely on post-press machine vision inspection systems, where imaging parameters and image analysis algorithms remain constant throughout the inspection process, and uniformity is assessed for the entire image using a standardized approach. However, the surface of dairy packaging substrates typically contains both areas prone to ink wetting due to low surface energy and areas with high specular reflectivity, such as aluminum foil composite layers, varnish coatings, or hot stamping markings. Fixed imaging parameters cannot adequately account for these two distinctly different types of areas.

[0004] The information disclosed in this background section is intended only to enhance the understanding of the general background of this disclosure and should not be construed as an admission or in any way implying that the information constitutes prior art known to those skilled in the art. Summary of the Invention

[0005] In view of at least one of the above technical problems, the present invention provides a method and system for detecting the uniformity of ink application on printing templates. The method uses pre-press substrate surface attribute information to drive post-press adaptive imaging and differential analysis, so as to take into account the detection needs of different surface attribute areas and improve the accuracy and reliability of ink application uniformity detection.

[0006] This invention provides a method for detecting the uniformity of ink application on a printing template, comprising the following steps: The surface attribute information of the substrate before printing is obtained, and a detection control map is generated based on the surface attribute information. The detection control map carries regional attribute information that is registered with the position of the printed image. When acquiring images of the printed substrate, the imaging parameters are dynamically adjusted based on the region attribute information corresponding to the current position in the detection control diagram to obtain an optimized image. Spatial frequency separation is performed on the optimized image to obtain at least one frequency component image; Based on the regional attribute information corresponding to the current position in the detection control chart, the analysis model associated with the regional attribute information is invoked to analyze the frequency component image, thereby obtaining the uniformity detection result of the printing area.

[0007] Further, the step of obtaining surface property information of the substrate before printing and generating a detection control map based on the surface property information includes: Obtain the dyne value distribution on the surface of the substrate and the coordinate information of the printed image; The dyne value distribution is registered with the printed image based on the coordinate information to obtain a dyne value distribution map; Based on a preset wettability threshold, regions with dyne values ​​below the wettability threshold are marked as high wettability risk areas on the dyne value distribution map, and a first risk attribute is added to the high wettability risk areas as the region attribute information to generate the detection control map.

[0008] Further, the step of dynamically adjusting imaging parameters and obtaining an optimized image based on the region attribute information corresponding to the current position in the detection control map includes: When the detection control chart determines that the current location has the first risk attribute, the illuminance of the lighting source corresponding to the current location is increased, and / or the light source is switched to a preset spectral band sensitive to ink, to form the optimized image.

[0009] Further, the step of performing spatial frequency separation on the optimized image to obtain at least one frequency component image includes: Spatial frequency separation is performed on the optimized image to obtain low-frequency component image, mid-frequency component image and high-frequency component image.

[0010] Further, the analysis of the frequency component image includes: If the current location is marked as the high wetting risk area, then the preset ink layer gradient analysis model is invoked to analyze the low-frequency component image; The ink layer gradient analysis model is used to determine whether there is uneven ink volume gradient by comparing the deviation between the actual grayscale trend change of the low-frequency component image and the preset expected smooth trend.

[0011] Furthermore, the detection control chart also carries a second high reflectivity attribute that marks the high reflectivity areas of the substrate, and serves as the area attribute information.

[0012] Furthermore, the process of acquiring the optimized image includes: When the current position is determined to be the high reflectivity zone according to the detection control chart, the illuminance of the illumination source corresponding to the high reflectivity zone is reduced, and a polarized light source that intersects with the polarizer direction of the image acquisition device is enabled to suppress specular reflection.

[0013] Further, the analysis of the frequency component image includes: If the current location is marked as the high reflectivity area, the grayscale information of the current location is ignored, and a preset texture regularity analysis model is invoked to analyze the mid-frequency component image and the high-frequency component image. The texture regularity analysis model is used to determine whether there are micro-texture defects by analyzing the periodicity characteristics of the texture in the mid-frequency component image and / or the local variance statistical characteristics of the texture in the high-frequency component image.

[0014] The present invention also provides a printing template ink uniformity detection system, comprising: The surface property acquisition module is used to acquire surface property information of the substrate before printing. The control chart generation module is used to generate a detection control chart based on the surface attribute information, wherein the detection control chart carries regional attribute information that is registered with the position of the printed image; An adaptive imaging module is used to dynamically adjust imaging parameters to obtain an optimized image based on the region attribute information corresponding to the current position in the detection control map when acquiring images after printing. The image analysis module is used to perform frequency separation on the optimized image and call the associated analysis model for analysis based on the region attribute information; The comprehensive judgment module is used to generate printing uniformity test results.

[0015] Furthermore, the control chart generation module includes: The receiving unit acquires the dyne value distribution and coordinate information of the printed image on the surface of the substrate. The registration unit registers the dyne value distribution with the printed image based on the coordinate information to obtain a dyne value distribution map. The generation unit marks areas with dyne values ​​lower than the wettability threshold as high wettability risk areas on the dyne value distribution map according to a preset wettability threshold, adds a first risk attribute to the high wettability risk areas, and generates the detection control map.

[0016] The technical solution of this invention can achieve the following technical effects: By acquiring surface attribute information of the substrate before printing and generating a detection control chart that is registered with the printed image, the post-printing imaging parameters and analysis model are adaptively adjusted. This achieves the goal of taking into account the detection needs of different surface attribute areas in the same detection process, effectively avoiding false alarms caused by overexposure or signal attenuation in high reflective areas, effectively identifying weak defects such as uneven ink volume in low surface energy areas, and significantly improving the accuracy and reliability of ink uniformity detection.

[0017] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a flowchart illustrating the method for detecting the uniformity of ink application on a printing template in an embodiment of the present invention. Figure 2 This is a schematic diagram of the process for generating a detection control chart carrying a first risk attribute based on performance attribute information in an embodiment of the present invention; Figure 3 This is a schematic diagram illustrating the process of generating an optimized image based on regional attribute information in an embodiment of the present invention; Figure 4 This is a schematic diagram illustrating the process of spatial frequency separation of an optimized image and analysis using a corresponding analysis model in an embodiment of the present invention. Detailed Implementation

[0020] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0021] 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 in this specification 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.

[0022] This invention provides a method such as Figures 1 to 4 The method for detecting the uniformity of ink application on a printing template, as shown, includes the following steps: Obtain the surface attribute information of the substrate before printing, and generate a detection control chart based on the surface attribute information. The detection control chart carries the regional attribute information that is registered with the position of the printed image. When acquiring images of the printed substrate, the imaging parameters are dynamically adjusted based on the regional attribute information corresponding to the current position in the detection control chart to obtain an optimized image; Spatial frequency separation is performed on the optimized image to obtain at least one frequency component image; Based on the regional attribute information corresponding to the current position in the detection control chart, the analysis model associated with the regional attribute information is invoked to analyze the frequency component image and obtain the uniformity detection result of the printing area.

[0023] The working principle of this invention is as follows: First, before the substrate enters the printing unit, the system acquires surface attribute information of the substrate through the pre-press inspection station. This refers to physical quantitative data that characterizes the substrate's ability to accept and spread ink, obtained through methods such as contact angle measurement, surface energy scanning, or spectral reflectance measurement. The surface attribute information of the substrate is stored in a spatial distribution form corresponding to the physical size of the substrate. Simultaneously, the system acquires the coordinate information of the printed graphics in this batch. This coordinate information originates from the digital pre-press file in the printing process, defining the precise position of each graphic element on the substrate plane. The system registers and aligns the spatial distribution of surface attribute information with the coordinates of the printed graphics, establishing a correspondence between each point on the surface attribute distribution map and the graphic content that will ultimately be printed at that location. Based on this, the system divides the surface attribute distribution map into regions according to preset classification rules. Different region attribute labels are assigned to different regions based on the magnitude or range of surface attribute values, ultimately generating an inspection control chart aligned with the coordinate system of the printed graphics. Each coordinate position on the inspection control chart carries the corresponding region attribute information.

[0024] After ink transfer is completed in the printing unit, the substrate enters the post-printing inspection station. The post-printing inspection station is equipped with zone-adjustable lighting sources and image acquisition devices. As the printed substrate passes through the inspection station at production line speed, the system reads the regional attribute information corresponding to the current imaging area in the inspection control chart in real-time or near real-time. Based on the read regional attribute information, the system dynamically adjusts the imaging parameters. For example, if the regional attribute information indicates that the current location has a first attribute, the system uses the first set of imaging parameters; if the regional attribute information indicates that the current location has a second attribute, the system uses the second set of imaging parameters. Adjustable imaging parameters include, but are not limited to, the illuminance of the lighting source, the spectral band distribution of the lighting source, and the polarization configuration between the light source and the image acquisition device. Through the above-mentioned zone-based differentiated imaging control based on regional attribute information, the system can obtain a zone-optimized digital image in a single scan or capture. The digital image exhibits relatively balanced signal quality in different surface attribute areas, forming an optimized image.

[0025] After acquiring the optimized image, the system enters the image analysis stage. First, spatial frequency separation is performed on the optimized image. This separation can be achieved using frequency domain transformation or spatial domain filtering. By decomposing information at different spatial scales in the image, the optimized image is divided into component images with several different spatial frequency ranges. After separation, at least one frequency component image is obtained. For example, low-frequency components reflecting large-area, slowly changing grayscale trends, mid-frequency components reflecting medium-scale, regular grayscale fluctuations, and high-frequency components reflecting subtle textures and local abrupt changes are extracted into their respective component images. After frequency separation, the system reads the regional attribute information corresponding to each position in the detection control chart. The system has pre-established and stored a database of relationships between regional attribute information and analysis models. This database defines which analysis model should be invoked for which frequency component image to diagnose based on different regional attribute labels. Based on the current region attribute information, the system selects the corresponding analysis model from the association database and applies the analysis model to the corresponding frequency component image of the region for analysis and calculation. Because different surface attribute regions may cause ink application anomalies that manifest differently and with varying degrees of significance in the image, this embodiment performs differentiated model matching and analysis based on region attributes. Compared to a unified analysis of the entire image, this approach can more accurately identify and quantify the ink application uniformity of each region. Finally, the analysis results of each region are aggregated to form a detection result that includes a conclusion on the ink application uniformity.

[0026] Building upon the above embodiments, a rotary encoder linked to the substrate conveying roller can be installed at the post-printing inspection station. The encoder pulse signals are used to calculate the substrate's displacement in real time, and the inspection control chart is pre-stored as a pixel-level mask image aligned with the printing coordinate system, with resolution matching the spatial sampling rate of the imaging device. During image acquisition, corresponding image blocks are dynamically extracted from the inspection control chart based on the current cumulative encoder displacement, and the regional attribute information carried by each pixel or predefined region within the image block is analyzed. If an offset is detected between the current substrate position and the pre-printing registration reference, the system performs real-time position correction using edge alignment or template matching algorithms to ensure accurate correspondence between regional attribute information and physical position. Based on the analyzed regional attribute information, the system dynamically adjusts imaging parameters according to a preset decision mapping table. This mapping table includes at least the correspondence between regional attribute labels, illuminance adjustment coefficients, spectral band selection identifiers, and polarization activation flags. This enables real-time and accurate acquisition of regional attribute information and rapid adjustment of imaging parameters on high-speed printing production lines, resulting in optimized images with balanced signal quality.

[0027] In some embodiments of the present invention, such as Figure 2 As shown, the surface property information of the substrate before printing is obtained, and a detection control chart is generated based on the surface property information, including: Obtain the dyne value distribution and coordinate information of the printed image on the substrate surface; The dyne value distribution is registered with the printed text based on the coordinate information to obtain the dyne value distribution map; Based on a preset wettability threshold, areas with dyne values ​​below the wettability threshold are marked as high wettability risk areas on the dyne value distribution map. A first risk attribute is added to the high wettability risk areas as regional attribute information, and a detection control map is generated.

[0028] Before the substrate enters the printing unit, the surface of the substrate is first scanned and measured at the pre-press inspection station to obtain the dyne values ​​corresponding to each coordinate position on the substrate surface. This forms a dyne value distribution corresponding to the physical size of the substrate. The dyne value is a quantitative indicator characterizing the free energy of the substrate surface, directly reflecting the wetting and spreading ability of the ink at that position. Simultaneously, the system obtains the electronic file of the printed graphics for this batch from the digital front end of the printing process and extracts the two-dimensional coordinate information of the printed graphics on the substrate plane. The coordinate information defines the precise arrangement position of each graphic element on the substrate surface. Subsequently, the system spatially registers the dyne value distribution with the coordinate information of the printed graphics. By scaling, translating, and rotating the dyne value distribution data, it is precisely aligned with the coordinate system of the printed graphics, resulting in a dyne value distribution map that corresponds one-to-one with the position of the printed graphics. After registration, the system performs region classification processing on the dyne value distribution map according to a preset wettability threshold, determining whether the dyne value at each coordinate position is lower than the wettability threshold. For coordinate locations with dyne values ​​below the wettability threshold, the system marks them as high-wetting-risk areas and adds a first-risk attribute label to these areas, indicating a potential risk of ink spreading obstruction due to insufficient surface energy. The dyne value distribution map after marking becomes a detection control map carrying regional attribute information registered with the printed image location. This allows subsequent post-press inspection processes to adopt differentiated imaging and analysis strategies based on the regional attribute information corresponding to each printing location. By proactively identifying risk areas and generating guidance information before printing, this lays the foundation for improving the overall targeting and accuracy of the inspection.

[0029] Based on the above embodiments, the preset wettability threshold can be achieved by conducting a standard ink spreading test under laboratory conditions using sample materials of the same batch or specification as the substrate before the formal printing operation; or by adopting a statistical preset based on historical production data.

[0030] In some embodiments of the present invention, such as Figure 3 As shown, based on the region attribute information corresponding to the current position in the detection control chart, the imaging parameters are dynamically adjusted to obtain an optimized image, including: When the detection control chart determines that the current location has a first-risk attribute, the illuminance of the lighting source corresponding to the current location is increased, and / or the light source is switched to a preset spectral band that is sensitive to ink, to form an optimized image.

[0031] As the printed substrate passes through the post-press inspection station at production line speed, the regional attribute information corresponding to the current imaging window's coverage area in the inspection control chart is read line by line or frame by frame. If the region attribute information for the current location carries a first-risk attribute, it indicates that the current location was identified in the pre-press stage as an area at risk of ink spreading obstruction due to insufficient surface energy. For such areas, the signal contrast between ink and substrate is typically low. If conventional lighting parameters are used for imaging, subtle changes in ink layer thickness are difficult to create discernible grayscale differences in the image, easily leading to missed defects such as uneven ink distribution. To address this issue, after determining that the current location possesses a first-risk attribute, the system performs imaging parameter adjustments. These adjustments include: illuminance enhancement, which adjusts the driving current or pulse width of the illumination source zone corresponding to the current location, increasing the luminous flux illuminating that area. This enhances the intensity difference of reflected light between the ink layer and the substrate, making the grayscale representation of ink layer thickness in the image more distinct. Alternatively, spectral switching can be used, changing the emission band of the illumination source from conventional white light or broadband illumination to a preset narrowband spectral band with high absorption peaks for specific pigment components in water-based inks, such as the near-infrared band. In the near-infrared band, the absorption rate of the ink layer is significantly different from the reflectivity of the substrate, further enhancing the signal contrast between the ink and the substrate. Illuminance enhancement and spectral switching can be used individually or in combination. After adjustment, in the image acquired in the current imaging window, the ink details in high-risk areas with a first-risk attribute are effectively highlighted, and previously imperceptible gradual fluctuations in ink volume are clearly presented, thereby improving the accuracy and reliability of subsequent analysis in determining the uniformity of ink application in low surface energy areas.

[0032] In some embodiments of the present invention, please refer to Figure 4 Spatial frequency separation is performed on the optimized image to obtain at least one frequency component image, including: Spatial frequency separation is performed on the optimized image to obtain low-frequency component image, mid-frequency component image and high-frequency component image.

[0033] After acquiring the digital image optimized by adaptive imaging, the system enters the image frequency decomposition stage. This stage breaks down information at different spatial scales, assigning grayscale variation patterns reflecting different physical causes to their respective frequency channels, providing independent analytical objects for subsequent differential analysis. Frequency domain filtering can be used for this separation. First, a two-dimensional Fourier transform or wavelet transform is performed on the optimized image to convert it from the spatial domain to the frequency domain, obtaining a spectrum representing the energy distribution of each spatial frequency component. Then, the system applies a preset first set of bandpass filter parameters to the spectrum to extract the spectral components of the low-frequency region. After inverse transformation, a low-frequency component image is obtained, retaining only the large-area, slowly changing grayscale trend information while filtering out medium-scale stripe variations and subtle texture noise. Next, a second set of bandpass filter parameters is applied to extract the spectral components of the mid-frequency region. After inverse transformation, a mid-frequency component image is obtained, primarily containing regular bright and dark stripe information caused by anilox rollers, printing plate wear, etc. Finally, high-pass filter parameters are applied to extract the spectral components of the high-frequency region. After inverse transformation, the high-frequency component image is obtained. This high-frequency component image mainly contains subtle dot loss, white spots, dirt spots, or other random texture variations. Frequency separation can also be achieved directly in the spatial domain through convolutional filtering. By using spatial convolution kernels of different scales and weight distributions to perform pixel-by-pixel sliding calculations on the optimized image, the low-frequency component image after smoothing filtering, the mid-frequency component image after bandpass filtering, and the high-frequency component image after sharpening or residual extraction are obtained respectively.

[0034] Through the above spatial frequency separation operation, the gray-scale change patterns originally mixed together in the optimized image, caused by different process sources, are effectively decoupled into three independent component images. This significantly reduces the mutual interference between different defect signals in the subsequent analysis stage, allowing the most suitable analysis method to be used for diagnosis for each component image, thereby improving the resolution and accuracy of ink uniformity detection.

[0035] In some embodiments of the present invention, such as Figure 3 and Figure 4 As shown, the analysis of the frequency component image includes: If the current location is marked as a high wetting risk area, the preset ink layer gradient analysis model is invoked to analyze the low-frequency component image; The ink layer gradient analysis model is used to determine whether there is uneven ink volume gradient by comparing the deviation between the actual grayscale trend change of the low-frequency component image and the preset expected smooth trend.

[0036] First, the system reads the regional attribute information of the current analysis location in the detection control chart. When the current location is marked as a high wetting risk area, it retrieves the associated ink layer gradient analysis model from the analysis model library and inputs the corresponding low-frequency component image data into the ink layer gradient analysis model. The working logic is based on the fact that under normal printing conditions, the ink layer thickness in a certain area on the substrate should show a stable and consistent trend or a smooth gradient along the design direction. In the low-frequency component image, this trend is represented by the change of gray values ​​along the spatial direction following a continuous and smooth baseline. However, when the ink spread is hindered due to insufficient surface energy at the current location, the ink layer thickness will show irregular shifts or fluctuations in local areas, and the gray value change trend will deviate from the expected smooth shape. Therefore, when the model performs the analysis, it first extracts one or more gray-scale sampling lines along the design gradient direction of the printed image or the main direction of ink spread on the low-frequency component image to obtain the gray value sequence along the main direction as the actual gray-scale trend. At the same time, the model reads the standard gray-scale trend curve established in advance based on the digital file of the printing original and the standard proofing process, which describes the smooth change shape of the gray value along the sampling direction that the location should have under ideal printing conditions. Subsequently, the model compares the actual grayscale trend with the standard grayscale trend curve point by point, calculating the deviation between the actual grayscale value at each sampling point and the corresponding grayscale value on the standard curve. If the deviation values ​​of multiple consecutive sampling points along the sampling direction exceed the preset deviation tolerance range, and the length or area of ​​the consecutive abnormal segment exceeds the preset minimum defect scale threshold, then the current area is determined to have an ink volume gradient unevenness defect; otherwise, the current area is determined to have qualified ink uniformity. This effectively solves the problem of easy missed detection in low-contrast gradient areas by the fixed threshold method, and significantly improves the sensitivity and reliability of ink uniformity detection in high wetting risk areas.

[0037] In some embodiments of the present invention, such as Figure 3 As shown, the detection control chart also carries a second high reflectivity attribute that marks the high reflectivity areas of the substrate, and serves as regional attribute information.

[0038] During the pre-press stage, in generating the inspection control chart, the system not only assesses the wettability risk of the dyne value distribution on the substrate surface but also simultaneously acquires information on the reflectivity distribution of the substrate surface. Due to differences in material composition, the substrate surface exhibits different optical reflectivity. Areas with aluminum foil composite layers, UV varnish coatings, and hot-stamped holographic anti-counterfeiting labels have significantly higher specular reflectivity than paper-plastic composite areas. In subsequent imaging, if conventional lighting parameters are used for image acquisition, the incident light will generate strong specular reflection components on the surface of these high-reflectivity areas. These specular reflection components, upon entering the camera lens, will cause localized overexposure or large areas of bright spots in the image, obscuring the true ink layer information in the current area and making it impossible to discern variations in ink thickness or texture details. To pre-identify areas requiring special imaging processing in the inspection control chart, one approach is to directly parse the digital files of the printed graphics. Based on the aluminum foil layer boundaries, varnish coating ranges, and hot stamping marking outlines defined in the files, the boundary coordinates of high-reflectivity areas can be extracted. Alternatively, a reflectivity measurement module can be added to the pre-press inspection station. By projecting standard illumination probe light onto the substrate surface and measuring the intensity of reflected light, reflectivity distribution data of the substrate surface can be obtained point-by-point or area-by-area. Areas with reflectivity exceeding a preset reflectivity threshold are identified as high-reflectivity areas. After identifying high-reflectivity areas, a second high-reflectivity attribute label is added to the inspection control chart, indicating that this location requires a differentiated imaging strategy to suppress specular reflection in subsequent post-press inspections. The second high-reflectivity attribute is integrated with the first risk attribute in the same inspection control chart, ensuring that the entire inspection control chart can fully cover the differentiated inspection needs of different surface attribute areas. This provides complete prior information for targeted adjustments to imaging parameters based on area attributes in subsequent post-press inspection stages.

[0039] In some embodiments of the present invention, such as Figure 3 As shown, obtaining the optimized image includes: When the current location is determined to be a high-reflectivity area according to the detection control chart, the illuminance of the illumination source corresponding to the high-reflectivity area is reduced, and a polarized light source that intersects with the polarizer direction of the image acquisition device is enabled to suppress specular reflection.

[0040] When the current location is determined to carry a mark with a second high reflectivity attribute, it indicates that the current location is a surface with significant specular reflection characteristics, such as a composite layer, varnish coating area, or hot stamping marking area. Corresponding imaging parameter adjustments are performed, including reducing illuminance by decreasing the drive current or luminous duty cycle of the illumination source zone corresponding to the current location, appropriately weakening the incident light flux illuminating the high-reflectivity area, controlling the intensity of specular reflection from the source, and preventing image sensor response saturation; and enabling cross-polarization by activating the polarization device in the illumination optical path to make the light emitted from the light source linearly polarized, while simultaneously activating the polarization analysis device in front of the camera lens and setting its polarization transmission direction to be orthogonal to the polarization direction of the light source. According to the principle of optical polarization, the specular reflected light generated on the substrate surface maintains the polarization state of the incident light and is effectively blocked by the polarization analysis device in front of the camera under the cross-polarization configuration, while the diffuse reflected light inside the ink layer is sufficiently randomized after multiple scatterings, and a considerable portion still passes through the polarization analysis device in front of the camera to enter the image sensor. By combining illumination reduction and cross-polarization adjustment, specular reflection interference in high reflectivity areas is significantly suppressed, and diffuse reflection information of the ink layer is preserved and clearly presented. This allows the ink layer texture and thickness variation that were originally hidden under strong reflection to be properly distinguished, effectively solving the contradiction between overexposure in high reflectivity areas and underexposure in low-risk areas under fixed parameter imaging, and improving the basic data quality of ink uniformity detection on the entire printing surface.

[0041] In some embodiments of the present invention, such as Figure 4 As shown, the analysis of the frequency component image includes: If the current location is marked as a highly reflective area, the grayscale information of the current location is ignored, and the preset texture regularity analysis model is called to analyze the mid-frequency component image and the high-frequency component image. Texture regularity analysis model is used to determine whether there are micro-texture defects by analyzing the periodicity characteristics of texture in mid-frequency component images and / or the local variance statistical characteristics of texture in high-frequency component images.

[0042] First, the regional attribute information of the current analysis position in the detection control chart is read. When the position is determined to be a high-reflectivity area, the associated texture regularity analysis model is retrieved from the analysis model library, and the corresponding mid-frequency component image and high-frequency component image data are input into the model. Furthermore, after cross-polarization imaging processing in the high-reflectivity area, the absolute grayscale value of the image can no longer accurately reflect the ink layer thickness. This is because after the specular reflection component is filtered out by the polarization device, the intensity of diffuse reflection light entering the camera mainly depends on the internal scattering characteristics of the ink layer rather than the ink thickness. Therefore, when processing high-reflectivity areas, the model actively ignores grayscale information and instead performs uniformity diagnosis from the texture dimension. The working logic of the texture regularity analysis model is based on the fact that under normal printing conditions, high-reflectivity areas should exhibit a uniform texture after polarization imaging, with no irregular bright and dark stripes in the mid-frequency range and only stable substrate noise corresponding to the microstructure of the substrate surface in the high-frequency range. However, when the printing plate experiences slight clogging or local anomalies in ink transfer, alternating bright and dark stripes with specific periodicity or quasi-periodicity will be generated in the mid-frequency component image, which will manifest as a significant deviation of local texture roughness from the normal substrate noise level in the high-frequency component image. The model employs a dual-channel parallel processing strategy during analysis. In the mid-frequency analysis channel, the model extracts several equally spaced analysis windows along the direction perpendicular to the paper feed in the mid-frequency component image. Autocorrelation analysis or spectral analysis is performed on the pixel grayscale fluctuations within each analysis window to detect the presence of periodic signal components exceeding a preset amplitude threshold. If significant periodic stripe features are detected within a certain analysis window and the period length falls within a preset range matching the anilox roller pitch or printing plate wear cycle, then a mid-frequency ink streak defect is determined to exist in that area. In the high-frequency analysis channel, the model divides the high-frequency component image into several equally sized analysis sub-blocks. For each sub-block, the local variance of the pixel grayscale value is calculated as a texture roughness index. The local variance of each sub-block is compared with a preset normal texture variance baseline. If the local variance of a sub-block is significantly higher than the baseline and exceeds a preset anomaly threshold, it is determined that a minor texture defect caused by plate clogging or ink particle abnormalities exists at the current sub-block location. Furthermore, the mid-frequency and high-frequency dual-channel analyses can provide independent judgment results or mutually corroborate each other for comprehensive evaluation. By replacing grayscale analysis with texture analysis, the detection of ink uniformity in highly reflective areas is no longer limited by the inherent limitations of grayscale distortion after polarization imaging, significantly reducing the false negative and false positive rates of defects in highly reflective areas.

[0043] Based on the same inventive concept as the printing template ink uniformity detection method in the foregoing embodiments, the present invention also provides a printing template ink uniformity detection system, comprising: The surface property acquisition module is used to acquire surface property information of the substrate before printing. The control chart generation module is used to generate inspection control charts based on surface attribute information. The inspection control charts carry regional attribute information that is registered with the position of the printed graphics. The adaptive imaging module is used to dynamically adjust imaging parameters to obtain an optimized image based on the region attribute information corresponding to the current position in the detection control chart when acquiring images after printing. The image analysis module is used to perform frequency separation on the optimized image and call the associated analysis model for analysis based on the region attribute information; The comprehensive judgment module is used to generate printing uniformity test results.

[0044] The detection system described above in this invention can effectively realize the method for detecting the uniformity of ink application on printing templates, and the technical effects it can achieve are as described in the above embodiments, which will not be repeated here.

[0045] In some embodiments of the present invention, the control chart generation module includes: The receiving unit acquires the dyne value distribution and coordinate information of the printed image on the surface of the substrate. The registration unit registers the dyne value distribution with the printed image based on the coordinate information to obtain the dyne value distribution map; The generation unit marks areas with dyne values ​​below the wettability threshold as high wettability risk areas on the dyne value distribution map according to the preset wettability threshold, adds a first risk attribute to the high wettability risk areas, and generates a detection control chart.

[0046] Similarly, the above-mentioned optimization schemes for the system can also respectively achieve the optimization effects corresponding to the methods in the embodiments, which will not be repeated here.

[0047] Although this application has been described in conjunction with specific features and embodiments, it is obvious that various modifications and combinations can be made thereto without departing from the spirit and scope of this application. Accordingly, this specification and drawings are merely exemplary illustrations of the application as defined herein, and are to be considered as covering any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Thus, if such modifications and modifications fall within the scope of this application and its equivalents, this application intends to include such modifications and modifications.

Claims

1. A method of detecting uniformity of inking of a printing stencil, characterized by, Includes the following steps: The surface attribute information of the substrate before printing is obtained, and a detection control map is generated based on the surface attribute information. The detection control map carries regional attribute information that is registered with the position of the printed image. When acquiring images of the printed substrate, the imaging parameters are dynamically adjusted based on the region attribute information corresponding to the current position in the detection control diagram to obtain an optimized image. Spatial frequency separation is performed on the optimized image to obtain at least one frequency component image; Based on the regional attribute information corresponding to the current position in the detection control chart, the analysis model associated with the regional attribute information is invoked to analyze the frequency component image, thereby obtaining the uniformity detection result of the printing area.

2. The method for detecting the uniformity of ink application on a printing template according to claim 1, characterized in that, The step of acquiring surface property information of the substrate before printing and generating a detection control chart based on the surface property information includes: Obtain the dyne value distribution on the surface of the substrate and the coordinate information of the printed image; The dyne value distribution is registered with the printed image based on the coordinate information to obtain a dyne value distribution map; Based on a preset wettability threshold, regions with dyne values ​​below the wettability threshold are marked as high wettability risk areas on the dyne value distribution map, and a first risk attribute is added to the high wettability risk areas as the region attribute information to generate the detection control map.

3. The method for detecting the uniformity of ink application on a printing template according to claim 2, characterized in that, The step of dynamically adjusting imaging parameters and obtaining an optimized image based on the region attribute information corresponding to the current position in the detection control map includes: When the detection control chart determines that the current location has the first risk attribute, the illuminance of the lighting source corresponding to the current location is increased, and / or the light source is switched to a preset spectral band sensitive to ink, to form the optimized image.

4. The method for detecting the uniformity of ink application on a printing template according to claim 2, characterized in that, The step of performing spatial frequency separation on the optimized image to obtain at least one frequency component image includes: Spatial frequency separation is performed on the optimized image to obtain low-frequency component image, mid-frequency component image and high-frequency component image.

5. The method for detecting the uniformity of ink application on a printing template according to claim 4, characterized in that, The analysis of the frequency component image includes: If the current location is marked as the high wetting risk area, then the preset ink layer gradient analysis model is invoked to analyze the low-frequency component image; The ink layer gradient analysis model is used to determine whether there is uneven ink volume gradient by comparing the deviation between the actual grayscale trend change of the low-frequency component image and the preset expected smooth trend.

6. The method for detecting the uniformity of ink application on a printing template according to claim 4, characterized in that, The detection control chart also carries a second high reflectivity attribute that marks the high reflectivity areas of the substrate, and serves as the area attribute information.

7. The method for detecting the uniformity of ink application on a printing template according to claim 6, characterized in that, The process of obtaining the optimized image includes: When the current position is determined to be the high reflectivity zone according to the detection control chart, the illuminance of the illumination source corresponding to the high reflectivity zone is reduced, and a polarized light source that intersects with the polarizer direction of the image acquisition device is enabled to suppress specular reflection.

8. The method for detecting the uniformity of ink application on a printing template according to claim 7, characterized in that, The analysis of the frequency component image includes: If the current location is marked as the high reflectivity area, the grayscale information of the current location is ignored, and a preset texture regularity analysis model is invoked to analyze the mid-frequency component image and the high-frequency component image. The texture regularity analysis model is used to determine whether there are micro-texture defects by analyzing the periodicity characteristics of the texture in the mid-frequency component image and / or the local variance statistical characteristics of the texture in the high-frequency component image.

9. A system for detecting the uniformity of ink application on printing templates, characterized in that, The detection method described in any one of claims 1 to 8 includes: The surface property acquisition module is used to acquire surface property information of the substrate before printing. The control chart generation module is used to generate a detection control chart based on the surface attribute information, wherein the detection control chart carries regional attribute information that is registered with the position of the printed image; An adaptive imaging module is used to dynamically adjust imaging parameters to obtain an optimized image based on the region attribute information corresponding to the current position in the detection control map when acquiring images after printing. The image analysis module is used to perform frequency separation on the optimized image and call the associated analysis model for analysis based on the region attribute information; The comprehensive judgment module is used to generate printing uniformity test results.

10. The printing template ink uniformity detection system according to claim 9, characterized in that, The control chart generation module includes: The receiving unit acquires the dyne value distribution and coordinate information of the printed image on the surface of the substrate. The registration unit registers the dyne value distribution with the printed image based on the coordinate information to obtain a dyne value distribution map. The generation unit marks areas with dyne values ​​lower than the wettability threshold as high wettability risk areas on the dyne value distribution map according to a preset wettability threshold, adds a first risk attribute to the high wettability risk areas, and generates the detection control map.

Images