A pre-printed card secondary printing self-adaptive positioning method and system based on visual recognition, a terminal and a storage medium

Abstract

This invention discloses a visual recognition-based adaptive positioning method, system, terminal, and storage medium for secondary printing of pre-printed cards. When a pre-printed card is delivered to the positioning station, a supplementary light source is activated to acquire a full-area image of the card surface and perform preprocessing. Standardized pre-printed image anchor points are extracted, invalid interference features are filtered out, local deformation areas and degrees of deformation of individual cards are identified, and deformation data is calculated. Standardized personalized printing layout reference parameters are pre-stored. Combined with the deformation data, global coordinate translation and angle calibration are performed on the overall card offset and rotation deviation. Local pixel coordinate point-to-point compensation correction is performed on local stretching and warping deformation caused by lamination to generate unique printing coordinate parameters. Based on the dynamically corrected unique printing coordinate parameters, the printer is driven to perform secondary printing, controlling the newly printed content to completely fit the pre-printed designated area. This invention achieves precise alignment between the personalized printed content of each pre-printed card and the original pre-printed content.

Description

Technical Field

[0001] This invention relates to the field of personalized card printing and thermal sublimation re-transfer printing precision positioning technology. Specifically, it relates to an adaptive visual positioning printing technology for pre-printed cards such as bank cards, social security cards, and government standard cards, which solves the alignment deviation of secondary personalized printing caused by card production deformation. It is applicable to batch personalized reprinting and graphic overprinting processing scenarios for ID cards, financial cards, and social security cards. Background Technology

[0002] Standard cards such as bank cards, social security cards, and resident service cards all adopt a production process of "mass pre-printing in the factory + personalized reprinting later". At the factory, the card base is made and the basic graphics and text are pre-printed through lamination, die-cutting, and coating processes. The pre-printed content includes fixed-format content such as card number area, name area, photo window, and logo. Subsequently, personalized variable content such as personal name, ID number, photo, and expiration date is printed and overprinted using dye-sublimation printers and re-transfer printers.

[0003] Existing secondary printing positioning technologies all employ a mechanical fixed positioning method combined with preset template coordinates for alignment. This means the card position is fixed via a printer track, and the system calls preset standard coordinate parameters to complete the personalized content printing. However, in actual mass production, cards exhibit inherent process deviations: Firstly, the card punching process has dimensional deviations ranging from micrometers to millimeters, resulting in inconsistent card outlines. Secondly, the lamination process of multiple layers of PVC and PET materials is prone to thermal expansion and contraction, localized warping, and planar offset, causing random shifts in the pre-printed image areas and window positions of each finished card, lacking a unified fixed coordinate reference.

[0004] Existing technologies have significant drawbacks, categorized into two types: defects in traditional mechanical positioning and defects in general visual positioning technology. Neither can adapt to the scenario of batch secondary overprinting of standardized cards: 1. The traditional mechanical fixed positioning + preset template coordinate scheme can only rely on the printer's card transmission track to fix the physical shape of the card. It cannot recognize the personalized offsets and non-standard deformations of pre-printed graphics on individual cards. The fixed printing coordinates do not match the actual layout of a single card, easily leading to problems such as printed content offset, edge overlap, graphics overlap, and window misalignment during mass production, resulting in extremely low yield rates. 2. Existing general printing visual positioning technologies are only suitable for non-deformable, uniformly formatted printing media such as paper and rigid boards. Their positioning logic is "collecting media boundaries + fixed feature point correction," which can only correct regular translations and angular deviations. They have a core technology blind spot: they cannot adapt to the irregular local warping, micro-stretching, and non-linear deformations caused by PVC / PET multi-layer card lamination processes. Furthermore, they cannot distinguish between pre-printed fixed graphics, blank areas, and reflective interference from lamination, easily leading to feature recognition failures and coordinate correction deviations. Currently, there is no dedicated adaptive positioning solution in the industry for the random deviations in punching standard cards and the nonlinear deformation of multi-layer materials. This technological gap has long existed, resulting in the inability to standardize the personalized overprinting quality of high-precision cards such as bank cards and social security cards, which has become a technical problem that the industry urgently needs to solve.

[0005] Therefore, existing technologies still need to be improved and developed. Summary of the Invention

[0006] The main objective of this invention is to provide a visual recognition-based adaptive positioning method, system, terminal, and computer-readable storage medium for secondary printing of pre-printed cards. This invention aims to solve the problems in the prior art regarding positioning of pre-printed images and text in secondary printing of standardized cards. Existing mechanical fixed positioning and fixed template printing methods cannot adapt to the punching deviation of a single card and the positional offset of pre-printed images and text caused by lamination deformation.

[0007] To achieve the above objectives, the present invention provides an adaptive positioning method for secondary printing of pre-printed cards based on visual recognition. The adaptive positioning method for secondary printing of pre-printed cards based on visual recognition includes the following steps: When the pre-printed card is conveyed to the positioning station by the feeding mechanism, the adjustable ring diffuse reflection supplementary light source is triggered to start, and the high-resolution industrial camera or high-speed flat panel scanner is controlled to vertically collect the full-area image of the card surface in diffuse reflection supplementary light mode. The full-area image of the card surface is preprocessed to obtain the full-area image of the target card surface. A library of feature anchor points specific to the standard card is preset. Standardized pre-printed image anchor points of the full-area image of the target card surface are extracted in a targeted manner. Invalid interference features are filtered out. Local deformation areas and deformation degree of a single card are identified. Deformation data is calculated. Standardized and personalized printing layout reference parameters are pre-stored. Combined with the deformation data, global coordinate translation and angle calibration are performed on the overall offset and rotation deviation of the card. Local pixel coordinate point-to-point compensation correction is performed on the local stretching and warping deformation caused by lamination. Exclusive printing coordinate parameters that are fully adapted to the actual pre-printing state of the current single card are generated. Based on the dynamically corrected dedicated print coordinate parameters, the printer is driven to complete the secondary printing of the preset content, so as to control the newly added print content to be completely aligned with the pre-printed designated area.

[0008] Optionally, the aforementioned visual recognition-based adaptive positioning method for secondary printing of pre-printed cards, wherein when the pre-printed card is conveyed to the positioning station by the feeding mechanism, the adjustable ring diffuse reflection supplementary light source is triggered to start, controlling a high-resolution industrial camera or high-speed planar scanner to vertically acquire a full-area image of the card surface in diffuse reflection supplementary light mode, further includes: An image reading mechanism is independently added to the card inlet of the printer. The image reading mechanism is a high-resolution industrial camera or a high-speed flatbed scanner, which serves as the core of image acquisition. An adjustable ring diffuse reflection supplementary light source is also provided to eliminate the reflection of the card coating and the interference of ambient light.

[0009] Optionally, in the visual recognition-based adaptive positioning method for secondary printing of pre-printed cards, the preprocessing includes grayscale conversion, filtering and noise reduction, and edge sub-pixel thinning processing. The standardized pre-printed image anchor points include the name box, card number border, four corners of the photo window, and official fixed logos; The invalid interference features include blank areas on the card, coating scratches, and dust noise. The deformation data includes lateral linear offset, longitudinal linear offset, rotation angle deviation, and local nonlinear deformation deviation data.

[0010] Optionally, the visual recognition-based adaptive positioning method for secondary printing of pre-printed cards, wherein the preprocessing of the full-area image of the card surface to obtain the full-area image of the target card surface, the pre-setting of a standard card-specific feature anchor point library, the directional extraction of standardized pre-printed image anchor points of the full-area image of the target card surface, the filtering of invalid interference features, the identification of local deformation areas and deformation degrees of a single card, and the calculation of deformation data specifically include: The entire surface image of the card is subjected to grayscale conversion, filtering and noise reduction, and edge subpixel thinning to obtain the entire surface image of the target card. A pre-defined feature anchor point library for standard cards is used to extract the name frame, card number border, four corners of the photo window, and official fixed logo from the full-area image of the target card surface based on the pre-defined feature anchor point library. Filter out blank areas, coating scratches, and dust noise in the full-area image of the target card surface; To address the local warping and nonlinear stretching deformation caused by card lamination, a local pixel comparison algorithm is used to identify the local deformation area and degree of deformation of a single card. Establish a dedicated three-dimensional reference coordinate system adapted to the non-standard deformation of a single card, and calculate the horizontal linear offset, vertical linear offset, rotation angle deviation, and local nonlinear deformation deviation data.

[0011] Optionally, the visual recognition-based adaptive positioning method for secondary printing of pre-printed cards, wherein the pre-stored standardized personalized printing layout reference parameters, combined with the deformation data, perform global coordinate translation and angle calibration on the overall card offset and rotation deviation, and perform local pixel coordinate point-to-point compensation correction on local stretching and warping deformation caused by lamination, generating exclusive printing coordinate parameters that are fully adapted to the actual pre-printing state of the current single card, specifically including: Retrieve the pre-stored standardized and personalized print layout reference parameters, and combine them with the horizontal linear offset, the vertical linear offset, the rotation angle deviation and the local nonlinear deformation deviation data. Use layered correction logic to uniformly translate and rotate the coordinates of all print points to eliminate overall card misalignment and tilt. Based on the local nonlinear deformation deviation in each region, the printing points of the portrait, name and certificate are independently fine-tuned at the pixel level in each block to correct the local point deviation caused by lamination warping and local stretching. By superimposing the two levels of correction results, a unique printing coordinate parameter is generated that is perfectly adapted to the actual pre-printing status of the current single card.

[0012] Optionally, in the visual recognition-based pre-printed card secondary printing adaptive positioning method, the preset content includes personalized text, avatar, and code.

[0013] Optionally, the visual recognition-based adaptive positioning method for secondary printing of pre-printed cards further includes: During the batch printing process, each card independently completes image acquisition, coordinate calculation, dynamic correction, and printing output, and each card is independently adapted.

[0014] Furthermore, to achieve the above objectives, the present invention also provides a visual recognition-based adaptive positioning system for secondary printing of pre-printed cards, wherein the visual recognition-based adaptive positioning system for secondary printing of pre-printed cards includes: The card image acquisition module is used to trigger the adjustable ring diffuse reflection supplementary light source when the pre-printed card is conveyed to the positioning station by the feeding mechanism, and control the high-resolution industrial camera or high-speed flatbed scanner to vertically acquire the full-area image of the card surface in diffuse reflection supplementary light mode. Anchor point extraction and deformation recognition module is used to preprocess the full-area image of the card surface to obtain the full-area image of the target card surface, preset the standard card-specific feature anchor point library, extract the standardized pre-printed image anchor points of the full-area image of the target card surface, filter invalid interference features, identify the local deformation area and deformation degree of a single card, and calculate deformation data. The layered dynamic coordinate compensation and correction module is used to pre-store standardized personalized printing layout reference parameters. Combined with the deformation data, it performs global coordinate translation and angle calibration on the overall offset and rotation deviation of the card, performs local pixel coordinate point-to-point compensation and correction on the local stretching and warping deformation caused by lamination, and generates exclusive printing coordinate parameters that are fully adapted to the actual pre-printing state of the current single card. The precision secondary printing module is used to drive the printer to complete the secondary printing of preset content based on the dynamically corrected exclusive printing coordinate parameters, so as to control the newly added printing content to completely fit the pre-printed specified area.

[0015] Furthermore, to achieve the above objectives, the present invention also provides a terminal, wherein the terminal includes: a memory, a processor, and a vision recognition-based pre-printed card secondary printing adaptive positioning program stored in the memory and executable on the processor, wherein when the vision recognition-based pre-printed card secondary printing adaptive positioning program is executed by the processor, it implements the steps of the vision recognition-based pre-printed card secondary printing adaptive positioning method as described above.

[0016] Furthermore, to achieve the above objectives, the present invention also provides a computer-readable storage medium, wherein the computer-readable storage medium stores a visual recognition-based adaptive positioning program for secondary printing of pre-printed cards, and when the visual recognition-based adaptive positioning program for secondary printing of pre-printed cards is executed by a processor, it implements the steps of the visual recognition-based adaptive positioning method for secondary printing of pre-printed cards as described above.

[0017] In this invention, when a pre-printed card is conveyed to the positioning station via the feeding mechanism, an adjustable ring diffuse reflection supplementary light source is activated, controlling a high-resolution industrial camera or high-speed planar scanner to vertically acquire a full-area image of the card surface in diffuse reflection supplementary light mode. The full-area image of the card surface is pre-processed to obtain a full-area image of the target card surface. A pre-defined feature anchor point library for standard cards is used to extract standardized pre-printed image anchor points from the full-area image of the target card surface, filtering out invalid interference features, identifying the local deformation areas and degree of deformation of a single card, and calculating deformation data. Standardized personalized printing format reference parameters are pre-stored. Combined with the deformation data, global coordinate translation and angle calibration are performed on the overall card offset and rotation deviation. Local pixel coordinate point-to-point compensation correction is performed on local stretching and warping deformation caused by lamination, generating exclusive printing coordinate parameters that are fully adapted to the actual pre-printing state of the current single card. Based on the dynamically corrected exclusive printing coordinate parameters, the printer is driven to complete the secondary printing of the preset content, ensuring that the newly added printing content completely fits the pre-printed designated area. This invention enables precise alignment of the personalized printed content of each pre-printed card with the original pre-printed content, significantly improving the alignment accuracy of secondary printing of standard cards, batch production consistency, and product yield, while reducing manual rework costs. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the overall structural assembly of the adaptive positioning printing device of the present invention; Figure 2 This is a flowchart of a preferred embodiment of the adaptive positioning method for secondary printing of pre-printed cards based on visual recognition according to the present invention; Figure 3 This is a flowchart of an adaptive positioning printing method based on nonlinear deformation correction in a preferred embodiment of the visual recognition-based pre-printed card secondary printing adaptive positioning method of the present invention. Figure 4 This is a structural diagram of a preferred embodiment of the adaptive positioning system for secondary printing of pre-printed cards based on visual recognition according to the present invention; Figure 5 This is a structural diagram of a preferred embodiment of the terminal of the present invention. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of this invention clearer and more explicit, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0020] This invention breaks through the conventional techniques of existing general vision printing positioning. Addressing two core industry-specific challenges—random deviations in standard card punching and non-linear, non-standard deformation in multi-layer lamination—it uniquely employs a proprietary technical solution: "pre-printed image anchor point benchmark positioning + non-linear deformation adaptive correction + dynamic adaptation of individual card coordinates." This solution abandons the traditional physical boundary positioning and fixed template correction methods, overcoming the technical bottlenecks of general vision technology's inability to adapt to non-standard card deformation and insufficient alignment accuracy. This invention does not rely on the physical shape of the card for positioning; it uses the pre-printed image features of each card as the sole benchmark, achieving precise overprinting of every deformed or offset card.

[0021] Figure 1 This is a schematic diagram of the core station structure of the modified card re-transfer printer of this invention, including the printer body (i.e., dye-sublimation printer / re-transfer printer), card feeding and conveying mechanism (i.e., card insertion mechanism), printing station, vision acquisition component, and main control correction module (not shown in the figure). The vision acquisition component consists of a high-resolution industrial camera (or high-speed planar scanner) and an adjustable ring diffuse reflection supplementary light source, fixedly mounted inside the sealed cavity between the feeding station and the printing station, facing the card conveying plane. The main control correction module is electrically connected to both the vision acquisition component and the printer's printing drive mechanism. This structure differs from the purely mechanical positioning structure of traditional printers. It achieves independent image acquisition of a single card through a front-mounted vision acquisition component, providing a hardware foundation for subsequent nonlinear deformation correction and avoiding recognition errors caused by lamination reflection and shape deviation. This is the core hardware architecture for achieving the adaptive positioning of this invention.

[0022] The core hardware modification of this invention lies in the independent addition of an image reading mechanism at the card inlet of the printer. This image reading mechanism can use a high-resolution industrial camera or a high-speed planar scanner as the image acquisition core, and is equipped with an adjustable ring diffuse reflection supplementary light source to eliminate reflections from the card lamination and interference from ambient light. The entire image reading mechanism is mounted above the card inlet of the printer, with the lens / scanning surface perpendicular to the card conveying plane. It can complete the full-area image acquisition of the card surface before the card enters the printing station inside the printer, fully covering the card printing area with no blind spots, and is suitable for high-speed batch printing of cards.

[0023] Based on the core station structure of the modified card re-transfer printer described above, the preferred embodiment of the present invention provides an adaptive positioning method for pre-printed cards re-printing based on visual recognition, such as... Figure 2 As shown, the visual recognition-based adaptive positioning method for secondary printing of pre-printed cards includes the following steps: Step S10: When the pre-printed card is conveyed to the positioning station by the feeding mechanism, the adjustable ring diffuse reflection light source is triggered to start, and the high-resolution industrial camera or high-speed flatbed scanner is controlled to vertically acquire the full-area image of the card surface in diffuse reflection light mode.

[0024] Specifically, before step S10, the method further includes: independently adding an image reading mechanism at the card inlet of the printer, wherein the image reading mechanism is a high-resolution industrial camera or a high-speed planar scanner, serving as the core of image acquisition; and providing an adjustable ring diffuse reflection supplementary light source to eliminate the reflection of the card coating and the interference of ambient light brightness.

[0025] Anti-interference image acquisition: After the pre-printed card is accurately conveyed to the positioning station by the feeding mechanism, the main control module triggers the adjustable ring diffuse reflection supplementary light source to start. The diffuse reflection supplementary light mode eliminates the interference of mirror reflection on the card coating and ambient light difference. In particular, the ring diffuse reflection supplementary light mode uses a 360° ring-shaped arrangement of soft light sources. The light is evenly scattered by the diffuse reflection diffusion layer before illuminating the card surface, which is different from the single-point strong light illumination of the direct light source. Its characteristics are as follows: (1) The light is concentrated in a non-directional direction, which can weaken the reflection of the PVC film mirror and prevent small anchor points such as card frame and photo window from being blocked by reflection; (2) The brightness, illumination height and tilt angle of the light source can be adjusted independently to adapt to different gloss levels and thicknesses of cards; (3) Achieve uniform illumination across the entire card, ensuring consistent brightness of the pre-printed images and text across the entire card, avoiding local overexposure and noise in dark areas, and guaranteeing stable and reliable anchor point recognition in the future.

[0026] High-resolution industrial cameras or high-speed planar scanners vertically acquire full-area images of the card surface, preserving the original feature data of all pre-printed graphics, window borders, and marking lines on the card. This avoids the image distortion and feature loss problems of traditional acquisition methods, providing original data support for subsequent accurate calculations.

[0027] Step S20: Preprocess the full-area image of the card surface to obtain the full-area image of the target card surface, preset the standard card-specific feature anchor point library, extract the standardized pre-printed image anchor points of the full-area image of the target card surface, filter invalid interference features, identify the local deformation area and degree of deformation of a single card, and calculate the deformation data.

[0028] Specifically, the standardized pre-printed image anchor points include name boxes, card number borders, the four corners of the photo window, and official fixed logos; the invalid interference features include blank areas on the card, lamination scratches, and dust noise; the deformation data includes lateral linear offset, longitudinal linear offset, rotation angle deviation, and local nonlinear deformation deviation data.

[0029] Dedicated image and text anchor point extraction and non-standard deformation recognition: The entire image of the card surface is processed by grayscale conversion, filtering and noise reduction, and edge sub-pixel refinement to obtain the entire image of the target card surface. Unlike general full-area recognition methods, this invention pre-sets a dedicated feature anchor point library for standard cards. Based on the dedicated feature anchor point library, standardized pre-printed image and text anchor points such as name boxes, card number borders, photo window corners, and official fixed logos are extracted from the entire image of the target card surface. Invalid interference features such as blank areas of the card, lamination scratches, and dust noise in the entire image of the target card surface are filtered out. At the same time, for local warping and non-linear stretching deformation caused by card lamination, a local pixel comparison algorithm is used to identify the local deformation area and degree of deformation of a single card. The traditional overall offset correction mode is abandoned, and a dedicated three-dimensional reference coordinate system adapted to the non-standard deformation of a single card is established to accurately calculate the horizontal linear offset, vertical linear offset, rotation angle deviation, and local non-linear deformation deviation data.

[0030] Among them, the standard card-specific feature anchor point library is a standardized pre-printed image and text feature set specifically established for bank cards, social security cards, and government cards, pre-stored within the embedded main control software. Its characteristics are as follows: (1) Fixed identification features stored in the database: name frame border, card number area border, four right angles of the photo window, official fixed logo / identification lines of the card and other fixed graphics and text uniformly printed at the factory; (2) The library has built-in feature filtering rules that can automatically distinguish invalid interference pixels such as valid anchor points, blank areas, coating scratches, and dust noise. (3) It is only for custom PVC multi-layer pre-printed cards, which is different from the general printing image feature library. It extracts the card's exclusive positioning reference and does not recognize the card's outer boundary.

[0031] The complete image recognition algorithm process consists of the following steps: (1) Image preprocessing: grayscale conversion, filtering and noise reduction, and subpixel edge refinement are performed on the original card images captured by the camera / scanner to remove dust, scratches and noise, and enhance the edge lines of the images and text. (2) Anchor point orientation extraction: Call the standard card-specific feature anchor point library to match and recognize standard anchor points such as name box, four corners of photo window, and card number edge in the image; (3) Deformation area detection: The local pixel difference comparison algorithm is used to compare the anchor point pixel position of the standard template with the actual anchor point pixel coordinates of the current card; (4) Deviation data calculation: Synchronously output four types of deviation data of the whole card: horizontal offset, vertical offset, rotation angle, and local area stretching / warping nonlinear deformation, and establish a single card exclusive three-dimensional reference coordinate system.

[0032] Step S30: Pre-store standardized personalized printing layout reference parameters, combine the deformation data, perform global coordinate translation and angle calibration on the overall card offset and rotation deviation, perform local pixel coordinate point-to-point compensation correction on the local stretching and warping deformation caused by lamination, and generate exclusive printing coordinate parameters that are fully adapted to the actual pre-printing state of the current single card.

[0033] Specifically, the layered dynamic coordinate compensation and correction involves: the main control module pre-storing standardized and personalized printing layout reference parameters, retrieving these parameters, and combining them with the horizontal linear offset, the vertical linear offset, the rotation angle deviation, and the local nonlinear deformation deviation data. Layered correction logic is then used to perform global coordinate translation and angle calibration on the overall card offset and rotation deviation. For local stretching and warping deformation caused by lamination, point-to-point compensation and correction of local pixel coordinates is performed. This completely solves the technical defect of general correction algorithms that can only correct overall offset and cannot adapt to local non-standard deformations, ultimately generating exclusive printing coordinate parameters that are fully adapted to the actual pre-printing state of the current single card.

[0034] The complete layered coordinate compensation and correction process is as follows: (1) Retrieve the standard personalized printing template reference parameters stored in the device; (2) Global unified calibration: Based on the horizontal and vertical offset and rotation angle data, the coordinates of all printed points are uniformly translated and rotated to eliminate the overall misalignment and tilt of the card; (3) Local point-to-point compensation: Based on the local nonlinear deformation deviation of each area, the printing points such as portrait, name, and ID number are independently fine-tuned at the pixel level in blocks to correct the local point deviation caused by lamination warping and local stretching. (4) Overlay the two-level correction results to generate new printing coordinate parameters that are adapted to the current single card.

[0035] Step S40: Based on the dynamically corrected exclusive print coordinate parameters, drive the printer to complete the secondary printing of the preset content, so as to control the newly added print content to completely fit the pre-printed specified area.

[0036] Specifically, precise secondary printing: Based on the dynamically calibrated exclusive printing coordinate parameters, the printer is driven to complete the secondary printing of personalized text, portraits, codes, and other content, so that the newly printed content is completely aligned with the pre-printed designated area, achieving precise overprinting.

[0037] After global calibration and local point-to-point deformation compensation superposition calculation, the final set of coordinates is the exclusive printing coordinate parameters adapted to the current deformed card. Due to the different punching deviations and lamination deformations of each card, the generated coordinate parameters are independent of each other. The printer completes the personalized content accurate secondary overprinting based on this set of exclusive coordinates.

[0038] Furthermore, during the batch printing process, each card independently completes image acquisition, coordinate calculation, dynamic correction, and printing output, and each card is independently adapted, completely avoiding the uniform alignment error problem caused by batch card deformation and punching deviation.

[0039] like Figure 3 As shown, Figure 3 The core process of this invention's proprietary adaptive printing control flow is as follows: card loading and positioning → anti-interference full-domain image acquisition → extraction of pre-printed text and images from dedicated anchor points and filtering of invalid features → global offset / angle deviation detection + local nonlinear deformation recognition → layered dynamic coordinate compensation and correction (global calibration + local point-to-point compensation) → adaptive and precise secondary printing → batch cyclic operation. This process abandons the general logic of traditional fixed template printing and single offset correction. Through a dual-dimensional layered correction mechanism, it specifically solves the industry problems of non-standard deformation in multi-layer card lamination and random deviation in punching.

[0040] This invention integrates a visual recognition and coordinate correction module into an existing commercial card re-transfer printer. The main body of the equipment requires no modification; only visual acquisition hardware and a dedicated control program are added. Pre-printed cards are transported from the production line to the printer's feeding station for automatic alignment. A ring-shaped diffuse reflection supplementary light source is activated to eliminate glare interference from the PVC-coated mirror surface. An industrial camera vertically captures the entire surface image of each card, fully preserving key anchor point features such as the pre-printed name frame, card number borders, and the four corners of the photo window.

[0041] The device's main control program calls upon a pre-defined library of feature anchor points specific to each card type to perform grayscale conversion, noise reduction filtering, and sub-pixel edge refinement on the acquired images. It accurately extracts pre-defined standard image and text anchor points and automatically filters out invalid interference features such as scratches, dust, and blank areas on the card surface. Simultaneously, through a local pixel difference comparison algorithm, it identifies local warping, micro-stretching, and non-linear deformation areas on a single card caused by the lamination process. It also simultaneously detects the overall lateral offset, vertical offset, and rotation angle deviation of the card, establishing a dedicated three-dimensional reference coordinate system adapted to the current single card.

[0042] The main control module retrieves the pre-stored standard personalized printing layout parameters and executes a layered correction strategy: for the overall physical offset and rotation angle error of the card, it performs global coordinate translation and angle uniform calibration; for local nonlinear deformation areas, it adopts a point-to-point pixel coordinate compensation algorithm to precisely fine-tune the local printing position, thoroughly correcting non-standard deformation errors that cannot be handled by general correction algorithms, and generating dynamic printing coordinates that are uniquely adapted to the current single card.

[0043] The printer driver uses the corrected real-time coordinate parameters to perform secondary overprinting of personalized content such as name, photo, ID number, and expiration date, ensuring that the printed content perfectly matches the pre-printed area without misalignment, edge overlap, or other issues. During mass production, each card independently completes image acquisition, deformation recognition, coordinate correction, and printing, achieving fully automated adaptive production.

[0044] Verified in actual working conditions, this implementation method can effectively adapt to the non-standard deformation and punching deviation of various pre-printed standard cards. The registration accuracy is much higher than that of traditional fixed template printing and general visual positioning solutions. The consistency and yield of mass production are greatly improved, making it suitable for large-scale industrial production scenarios.

[0045] The key innovations of this invention include: (1) A unique pre-printed graphic anchor point benchmark positioning mechanism breaks through the boundaries of existing technologies. Abandoning the traditional physical boundary positioning of media and the global positioning method of general images, a feature anchor point library is established for the exclusive pre-printed versions of bank cards and social security cards. The pre-printed fixed graphics are used as the only positioning benchmark, which completely avoids the alignment interference caused by card punching size deviation and shape error, and adapts to the differentiated process error of individual cards.

[0046] (2) The invention is the first to propose a layered correction algorithm for card nonlinear deformation, which solves the blind spot of core technology in the industry. Unlike the single overall correction logic of general printing equipment, the invention adopts a layered processing mode of "global offset calibration + local deformation compensation", which specifically solves the non-standard deformation problems such as local warping and nonlinear stretching caused by the PVC / PET multi-layer card lamination process. It fills the technical gap that the existing visual positioning technology cannot adapt to the non-standard deformation of cards, and has substantial innovation.

[0047] (3) Optimized recognition logic specifically for card-coated media. Optimized supplementary lighting mode and image recognition algorithm to address the reflective, smooth, and single-color characteristics of standard card-coated media. This avoids the problems of reflective distortion and feature misjudgment in traditional visual recognition, and accurately adapts to the high-precision overprinting requirements of special printing equipment for thermal sublimation and re-transfer cards. It is highly scenario-specific and is not a simple transplantation of general technology.

[0048] The technical effects that this invention can bring are as follows: (1) Completely solve the long-standing technical pain points in the industry. Break through the technical limitations of traditional mechanical fixed positioning and fixed template printing, and specifically solve the problems of inconsistent pre-printed image positions and misalignment during secondary printing caused by die-cutting size deviation, thermal deformation of laminated materials, and warping deformation of standard cards such as bank cards and social security cards, thereby eliminating single-card alignment deviation from the root.

[0049] (2) Precisely solve the problem of non-standard deformation overprinting and significantly improve product consistency. In response to the problem of local nonlinear deformation of card lamination and random deviation of punching that cannot be handled by existing technology, high-precision bonding between personalized printing content and pre-printed area is achieved through layer coordinate correction and point-to-point deformation compensation, which completely eliminates defects such as misalignment, edge pressing and text overlap, and unifies the printing quality of batch cards from the technical root, solving the long-standing pain point of poor consistency in batch production in the industry.

[0050] (3) Adapt to batch automated production, reduce costs and increase efficiency. The entire process of image acquisition, analysis and correction and adaptive printing is completed automatically. There is no need for manual alignment or manual rework of defective products, which greatly reduces labor costs and material losses. It is suitable for large-scale standardized card personalized batch production scenarios and significantly improves production efficiency.

[0051] (4) Strong equipment compatibility and low modification cost. This invention only requires existing commercial dye-sublimation and re-transfer printers with integrated vision acquisition and correction modules. It does not require replacement of the original core printing equipment or modification of the production line. It is compatible with mainstream card printing equipment on the market and has extremely strong practicality and versatility.

[0052] (5) Clear Innovation Barriers. This invention is not a simple superposition and scene transfer of existing visual positioning and printing correction technologies. Instead, it is designed for the unique production process defects and non-standard deformation conditions of standard cards. It has created a unique anchor point recognition mechanism and a hierarchical nonlinear correction algorithm, which overcomes the inherent technical biases and blind spots of existing technologies and produces a precise overprinting effect that cannot be achieved by existing technologies.

[0053] Furthermore, such as Figure 4 As shown, based on the above-mentioned adaptive positioning method for pre-printed cards secondary printing based on visual recognition, the present invention also provides a corresponding adaptive positioning system for pre-printed cards secondary printing based on visual recognition, wherein the adaptive positioning system for pre-printed cards secondary printing based on visual recognition includes: The card image acquisition module 51 is used to trigger the adjustable ring diffuse reflection supplementary light source when the pre-printed card is conveyed to the positioning station by the feeding mechanism, and control the high-resolution industrial camera or high-speed flatbed scanner to vertically acquire the full-area image of the card surface in diffuse reflection supplementary light mode. Anchor point extraction and deformation recognition module 52 is used to preprocess the full-area image of the card surface to obtain the full-area image of the target card surface, preset the standard card-specific feature anchor point library, extract the standardized pre-printed image anchor points of the full-area image of the target card surface, filter invalid interference features, identify the local deformation area and deformation degree of a single card, and calculate deformation data. The layered dynamic coordinate compensation and correction module 53 is used to pre-store standardized personalized printing layout reference parameters. Combined with the deformation data, it performs global coordinate translation and angle calibration on the overall offset and rotation deviation of the card, performs local pixel coordinate point-to-point compensation and correction on the local stretching and warping deformation caused by lamination, and generates exclusive printing coordinate parameters that are fully adapted to the actual pre-printing state of the current single card. The precision secondary printing module 54 is used to drive the printer to complete the secondary printing of preset content based on the dynamically corrected exclusive printing coordinate parameters, so as to control the newly added printing content to completely fit the pre-printed specified area.

[0054] Furthermore, such as Figure 5 As shown, based on the above-mentioned adaptive positioning method and system for secondary printing of pre-printed cards based on visual recognition, the present invention also provides a terminal, which includes a processor 10, a memory 20 and a display 30. Figure 5 Only some of the terminal components are shown; however, it should be understood that it is not required to implement all of the components shown, and more or fewer components may be implemented instead.

[0055] In some embodiments, the memory 20 may be an internal storage unit of the terminal, such as a hard disk or memory. In other embodiments, the memory 20 may be an external storage device of the terminal, such as a plug-in hard disk, smart media card (SMC), secure digital card (SD), flash card, etc. Further, the memory 20 may include both internal and external storage devices. The memory 20 is used to store application software and various types of data installed on the terminal, such as program code installed on the terminal. The memory 20 can also be used to temporarily store data that has been output or will be output. In one embodiment, the memory 20 stores a visual recognition-based pre-printed card secondary printing adaptive positioning program 40, which can be executed by the processor 10 to implement the visual recognition-based pre-printed card secondary printing adaptive positioning method of this application.

[0056] In some embodiments, the processor 10 may be a central processing unit (CPU), a microprocessor, or other data processing chip, used to run program code stored in the memory 20 or process data, such as executing the vision recognition-based pre-printed card secondary printing adaptive positioning method.

[0057] In some embodiments, the display 30 may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, or an OLED (Organic Light-Emitting Diode) touchscreen. The display 30 is used to display information on the terminal and to display a visual user interface. The terminal's processor 10, memory 20, and display 30 communicate with each other via a system bus.

[0058] In one embodiment, when the processor 10 executes the vision-based preprint card secondary printing adaptive positioning program 40 in the memory 20, the following steps are performed: When the pre-printed card is conveyed to the positioning station by the feeding mechanism, the adjustable ring diffuse reflection supplementary light source is triggered to start, and the high-resolution industrial camera or high-speed flat panel scanner is controlled to vertically collect the full-area image of the card surface in diffuse reflection supplementary light mode. The full-area image of the card surface is preprocessed to obtain the full-area image of the target card surface. A library of feature anchor points specific to the standard card is preset. Standardized pre-printed image anchor points of the full-area image of the target card surface are extracted in a targeted manner. Invalid interference features are filtered out. Local deformation areas and deformation degree of a single card are identified. Deformation data is calculated. Standardized and personalized printing layout reference parameters are pre-stored. Combined with the deformation data, global coordinate translation and angle calibration are performed on the overall offset and rotation deviation of the card. Local pixel coordinate point-to-point compensation correction is performed on the local stretching and warping deformation caused by lamination. Exclusive printing coordinate parameters that are fully adapted to the actual pre-printing state of the current single card are generated. Based on the dynamically corrected dedicated print coordinate parameters, the printer is driven to complete the secondary printing of the preset content, so as to control the newly added print content to be completely aligned with the pre-printed designated area.

[0059] The process of triggering the adjustable ring diffuse lighting source when the pre-printed card is conveyed to the positioning station via the feeding mechanism, controlling a high-resolution industrial camera or high-speed planar scanner to vertically acquire a full-area image of the card surface in diffuse lighting mode, and prior to this, also includes: An image reading mechanism is independently added to the card inlet of the printer. The image reading mechanism is a high-resolution industrial camera or a high-speed flatbed scanner, which serves as the core of image acquisition. An adjustable ring diffuse reflection supplementary light source is also provided to eliminate the reflection of the card coating and the interference of ambient light.

[0060] The preprocessing includes grayscale conversion, filtering and noise reduction, and edge subpixel thinning. The standardized pre-printed image anchor points include the name box, card number border, four corners of the photo window, and official fixed logos; The invalid interference features include blank areas on the card, coating scratches, and dust noise. The deformation data includes lateral linear offset, longitudinal linear offset, rotation angle deviation, and local nonlinear deformation deviation data.

[0061] Specifically, the preprocessing of the full-area image of the card surface to obtain the full-area image of the target card surface, the pre-defined feature anchor point library specific to the card type, the targeted extraction of standardized pre-printed image anchor points from the full-area image of the target card surface, the filtering of invalid interference features, the identification of local deformation areas and deformation degrees of a single card, and the calculation of deformation data include: The entire surface image of the card is subjected to grayscale conversion, filtering and noise reduction, and edge subpixel thinning to obtain the entire surface image of the target card. A pre-defined feature anchor point library for standard cards is used to extract the name frame, card number border, four corners of the photo window, and official fixed logo from the full-area image of the target card surface based on the pre-defined feature anchor point library. Filter out blank areas, coating scratches, and dust noise in the full-area image of the target card surface; To address the local warping and nonlinear stretching deformation caused by card lamination, a local pixel comparison algorithm is used to identify the local deformation area and degree of deformation of a single card. Establish a dedicated three-dimensional reference coordinate system adapted to the non-standard deformation of a single card, and calculate the horizontal linear offset, vertical linear offset, rotation angle deviation, and local nonlinear deformation deviation data.

[0062] The pre-stored standardized personalized print layout reference parameters, combined with the deformation data, perform global coordinate translation and angle calibration on the overall card offset and rotation deviation, and perform local pixel coordinate point-to-point compensation correction on local stretching and warping deformation caused by lamination, generating exclusive print coordinate parameters that are fully adapted to the actual pre-printing state of the current single card, specifically including: Retrieve the pre-stored standardized and personalized print layout reference parameters, and combine them with the horizontal linear offset, the vertical linear offset, the rotation angle deviation and the local nonlinear deformation deviation data. Use layered correction logic to uniformly translate and rotate the coordinates of all print points to eliminate overall card misalignment and tilt. Based on the local nonlinear deformation deviation in each region, the printing points of the portrait, name and certificate are independently fine-tuned at the pixel level in each block to correct the local point deviation caused by lamination warping and local stretching. By superimposing the two levels of correction results, a unique printing coordinate parameter is generated that is perfectly adapted to the actual pre-printing status of the current single card.

[0063] The preset content includes personalized text, avatars, and codes.

[0064] The visual recognition-based adaptive positioning method for secondary printing of pre-printed cards further includes: During the batch printing process, each card independently completes image acquisition, coordinate calculation, dynamic correction, and printing output, and each card is independently adapted.

[0065] The present invention also provides a computer-readable storage medium, wherein the computer-readable storage medium stores a vision-based adaptive positioning program for secondary printing of pre-printed cards, and the vision-based adaptive positioning program for secondary printing of pre-printed cards, when executed by a processor, implements the steps of the vision-based adaptive positioning method for secondary printing of pre-printed cards as described above.

[0066] In summary, this invention provides a visual recognition-based adaptive positioning method, system, terminal, and computer-readable storage medium for secondary printing of pre-printed cards. The method includes: when a pre-printed card is conveyed to the positioning station by a feeding mechanism, triggering the activation of an adjustable ring diffuse reflection supplementary light source, controlling a high-resolution industrial camera or high-speed planar scanner to vertically acquire a full-area image of the card surface in diffuse reflection supplementary light mode; preprocessing the full-area image of the card surface to obtain a full-area image of the target card surface; pre-setting a standard card-specific feature anchor point library; and directionally extracting standardized pre-printed image and text anchor points from the full-area image of the target card surface. This invention filters out invalid interference features, identifies the local deformation areas and degree of deformation of a single card, and calculates deformation data. It pre-stores standardized personalized printing template reference parameters and, combined with the deformation data, performs global coordinate translation and angle calibration on the overall card offset and rotation deviation. It also performs point-to-point compensation correction on local pixel coordinates for local stretching and warping deformation caused by lamination, generating exclusive printing coordinate parameters that perfectly adapt to the actual pre-printing state of the current single card. Based on the dynamically corrected exclusive printing coordinate parameters, it drives the printer to complete the secondary printing of preset content, ensuring that the newly printed content completely fits the designated pre-printed area. This invention achieves precise alignment between the personalized printed content and the original pre-printed content on each pre-printed card, significantly improving the alignment accuracy, batch production consistency, and product yield of standard cards during secondary printing, while reducing manual rework costs.

[0067] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or terminal that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or terminal. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or terminal that includes that element.

[0068] Of course, those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware (such as a processor, controller, etc.). The program can be stored in a computer-readable storage medium, and when executed, it can include the processes described in the above method embodiments. The computer-readable storage medium can be a memory, magnetic disk, optical disk, etc.

[0069] It should be understood that the application of the present invention is not limited to the examples above. Those skilled in the art can make improvements or modifications based on the above description, and all such improvements and modifications should fall within the protection scope of the appended claims.

Claims

1. An adaptive positioning method for secondary printing of pre-printed cards based on visual recognition, characterized in that, The visual recognition-based adaptive positioning method for secondary printing of pre-printed cards includes: When the pre-printed card is conveyed to the positioning station by the feeding mechanism, the adjustable ring diffuse reflection supplementary light source is triggered to start, and the high-resolution industrial camera or high-speed flat panel scanner is controlled to vertically collect the full-area image of the card surface in diffuse reflection supplementary light mode. The full-area image of the card surface is preprocessed to obtain the full-area image of the target card surface. A library of feature anchor points specific to the standard card is preset. Standardized pre-printed image anchor points of the full-area image of the target card surface are extracted in a targeted manner. Invalid interference features are filtered out. Local deformation areas and deformation degree of a single card are identified. Deformation data is calculated. Standardized and personalized printing layout reference parameters are pre-stored. Combined with the deformation data, global coordinate translation and angle calibration are performed on the overall offset and rotation deviation of the card. Local pixel coordinate point-to-point compensation correction is performed on the local stretching and warping deformation caused by lamination. Exclusive printing coordinate parameters that are fully adapted to the actual pre-printing state of the current single card are generated. Based on the dynamically corrected dedicated print coordinate parameters, the printer is driven to complete the secondary printing of the preset content, so as to control the newly added print content to be completely aligned with the pre-printed designated area.

2. The adaptive positioning method for secondary printing of pre-printed cards based on visual recognition according to claim 1, characterized in that, When the pre-printed card is conveyed to the positioning station by the feeding mechanism, the adjustable ring diffuse reflection supplementary light source is triggered to start, controlling a high-resolution industrial camera or high-speed planar scanner to vertically acquire a full-area image of the card surface in diffuse reflection supplementary light mode. Prior to this, the following steps were also included: An image reading mechanism is independently added to the card inlet of the printer. The image reading mechanism is a high-resolution industrial camera or a high-speed flatbed scanner, which serves as the core of image acquisition. An adjustable ring diffuse reflection supplementary light source is also provided to eliminate the reflection of the card coating and the interference of ambient light.

3. The adaptive positioning method for secondary printing of pre-printed cards based on visual recognition according to claim 1, characterized in that, The preprocessing includes grayscale conversion, filtering and noise reduction, and edge subpixel thinning. The standardized pre-printed image anchor points include the name box, card number border, four corners of the photo window, and official fixed logos; The invalid interference features include blank areas on the card, coating scratches, and dust noise. The deformation data includes lateral linear offset, longitudinal linear offset, rotation angle deviation, and local nonlinear deformation deviation data.

4. The adaptive positioning method for secondary printing of pre-printed cards based on visual recognition according to claim 3, characterized in that, The process involves preprocessing the full-area image of the card surface to obtain a target card surface full-area image, pre-setting a feature anchor point library specific to a particular card type, extracting standardized pre-printed image anchor points from the target card surface full-area image, filtering out invalid interference features, identifying local deformation areas and degrees of deformation on a single card, and calculating deformation data. Specifically, this includes: The entire surface image of the card is subjected to grayscale conversion, filtering and noise reduction, and edge subpixel thinning to obtain the entire surface image of the target card. A pre-defined feature anchor point library for standard cards is used to extract the name frame, card number border, four corners of the photo window, and official fixed logo from the full-area image of the target card surface based on the pre-defined feature anchor point library. Filter out blank areas, coating scratches, and dust noise in the full-area image of the target card surface; To address the local warping and nonlinear stretching deformation caused by card lamination, a local pixel comparison algorithm is used to identify the local deformation area and degree of deformation of a single card. Establish a dedicated three-dimensional reference coordinate system adapted to the non-standard deformation of a single card, and calculate the horizontal linear offset, vertical linear offset, rotation angle deviation, and local nonlinear deformation deviation data.

5. The adaptive positioning method for secondary printing of pre-printed cards based on visual recognition according to claim 4, characterized in that, The pre-stored standardized personalized print layout reference parameters, combined with the deformation data, perform global coordinate translation and angle calibration on the overall card offset and rotation deviation, and perform local pixel coordinate point-to-point compensation correction on local stretching and warping deformation caused by lamination, generating exclusive print coordinate parameters that are fully adapted to the actual pre-printing state of the current single card, specifically including: Retrieve the pre-stored standardized and personalized print layout reference parameters, and combine them with the horizontal linear offset, the vertical linear offset, the rotation angle deviation and the local nonlinear deformation deviation data. Use layered correction logic to uniformly translate and rotate the coordinates of all print points to eliminate overall card misalignment and tilt. Based on the local nonlinear deformation deviation in each region, the printing points of the portrait, name and certificate are independently fine-tuned at the pixel level in each block to correct the local point deviation caused by lamination warping and local stretching. By superimposing the two levels of correction results, a unique printing coordinate parameter is generated that is perfectly adapted to the actual pre-printing status of the current single card.

6. The adaptive positioning method for secondary printing of pre-printed cards based on visual recognition according to claim 1, characterized in that, The preset content includes personalized text, avatars, and codes.

7. The adaptive positioning method for secondary printing of pre-printed cards based on visual recognition according to claim 1, characterized in that, The visual recognition-based pre-printed card secondary printing adaptive positioning method also includes: During the batch printing process, each card independently completes image acquisition, coordinate calculation, dynamic correction, and printing output, and each card is independently adapted.

8. A visual recognition-based adaptive positioning system for secondary printing of pre-printed cards, characterized in that, The vision recognition-based pre-printed card secondary printing adaptive positioning system includes: The card image acquisition module is used to trigger the adjustable ring diffuse reflection supplementary light source when the pre-printed card is conveyed to the positioning station by the feeding mechanism, and control the high-resolution industrial camera or high-speed flatbed scanner to vertically acquire the full-area image of the card surface in diffuse reflection supplementary light mode. Anchor point extraction and deformation recognition module is used to preprocess the full-area image of the card surface to obtain the full-area image of the target card surface, preset the standard card-specific feature anchor point library, extract the standardized pre-printed image anchor points of the full-area image of the target card surface, filter invalid interference features, identify the local deformation area and deformation degree of a single card, and calculate deformation data. The layered dynamic coordinate compensation and correction module is used to pre-store standardized personalized printing layout reference parameters. Combined with the deformation data, it performs global coordinate translation and angle calibration on the overall offset and rotation deviation of the card, performs local pixel coordinate point-to-point compensation and correction on the local stretching and warping deformation caused by lamination, and generates exclusive printing coordinate parameters that are fully adapted to the actual pre-printing state of the current single card. The precision secondary printing module is used to drive the printer to complete the secondary printing of preset content based on the dynamically corrected exclusive printing coordinate parameters, so as to control the newly added printing content to completely fit the pre-printed specified area.

9. A terminal, characterized in that, The terminal includes: a memory, a processor, and a vision-based adaptive positioning program for pre-printed cards secondary printing, stored in the memory and executable on the processor. When the vision-based adaptive positioning program for pre-printed cards secondary printing is executed by the processor, it implements the steps of the vision-based adaptive positioning method for pre-printed cards secondary printing as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a vision-based adaptive positioning program for pre-printed card reprinting, which, when executed by a processor, implements the steps of the vision-based adaptive positioning method for pre-printed card reprinting as described in any one of claims 1-7.

Images