An image stitching method and system for a UV inkjet printer

Abstract

This application relates to the field of image stitching technology for UV inkjet printers, and in particular to an image stitching method and system for UV inkjet printers. The method includes: capturing scene images using a camera to obtain multiple images to be stitched; detecting and identifying coded markers and their contours in each image to obtain a unique ID number for each coded marker; matching the same target coded marker in the multiple images based on the unique ID number; establishing a mapping relationship between feature points based on the pixel coordinates of at least four outer corner points and the geometric center point of the successfully matched target coded marker to obtain camera model parameters for each image to be stitched; transforming the multiple images to be stitched into the same camera model coordinate system based on the camera model parameters of each image, and performing image fusion processing on the overlapping areas to generate a seamless stitched image.

Description

Technical Field

[0001] This application relates to the field of image stitching technology for UV inkjet printers, and in particular to an image stitching method and system for UV inkjet printers. Background Technology

[0002] UV inkjet printers, as digital printing devices that integrate inkjet printing technology and ultraviolet curing process, have experienced rapid development globally in recent years. With advantages such as no need for plate making, instant drying, and compatibility with multiple materials, this technology is deeply penetrating from traditional printing fields into high-end manufacturing and emerging scenarios, becoming a core growth driver in the digital printing industry. Currently, the UV printer market is at a dual inflection point of scale expansion and quality upgrades. The traditional market is steadily expanding, while emerging fields are opening up new growth opportunities. At the technological level, intelligent, environmentally friendly, and high-precision upgrades have become core competitive advantages.

[0003] Currently, image stitching technology has important applications in fields such as visual positioning for UV printers. Image stitching technology mainly includes two key steps: image registration and image fusion, with image registration being the core component. Existing image registration algorithms are basically divided into two categories: frequency domain-based methods (phase correlation methods) and time domain-based methods. Phase correlation methods transform two images to be registered to the frequency domain and calculate the translation vector through cross-power spectrum to achieve registration; this is often used in aerial photographs and satellite remote sensing image registration. Time domain-based methods can be further divided into feature-based methods and region-based methods. Feature-based methods identify image feature points and determine their correspondences to find the transformation relationship, and are not sensitive to changes in lighting. Region-based methods use a patch of overlapping area from one image as a template to search for a matching patch in another image, offering higher accuracy.

[0004] However, existing image registration algorithms have certain drawbacks. Phase correlation methods typically require a large overlap ratio, and traditional phase correlation methods generally demand a high overlap ratio. Feature-based methods are highly dependent on the accuracy of feature point correspondences, making it difficult to find feature points in images with little or no texture. Region-based methods are computationally expensive. Furthermore, when stitching large-format, high-resolution images, traditional algorithms rely on feature point information, making it difficult to reduce systematic accumulated errors (multi-point probabilistic matching, where errors accumulate with the number of images).

[0005] Therefore, how to overcome the shortcomings of current image stitching technology in terms of overlap ratio requirements, feature point dependence, computational load, and cumulative error, and provide a more adaptable image stitching method that can reduce systematic cumulative error, is a technical problem that urgently needs to be solved. Summary of the Invention

[0006] The purpose of this application is to overcome the above-mentioned technical problems and provide an image stitching method and system for UV inkjet printers, which can improve the efficiency and accuracy of image stitching in UV inkjet printers.

[0007] Firstly, one embodiment of this application discloses an image stitching method for a UV inkjet printer, which employs the following scheme: An image stitching method for a UV inkjet printer includes: Image acquisition steps: Capture scene images with a camera to obtain multiple images to be stitched together, wherein adjacent images to be stitched together have overlapping areas, and the overlapping areas contain at least one preset coded mark. Feature extraction step: Detect and identify the coded markers and their contours in each image to be stitched together to obtain unique ID information corresponding to each coded marker, wherein the unique ID information includes the ID number, the pixel coordinates of at least four outer corner points and the geometric center point of each coded marker; Matching and mapping steps: Based on the unique ID number information, match the same target encoding mark in multiple images to be stitched; based on the pixel coordinates of at least four outer corner points and geometric center points of the successfully matched same target encoding mark, establish the mapping relationship between feature points to obtain the camera model parameters of each image to be stitched. Image stitching steps: Based on the camera model parameters of each image to be stitched, multiple images to be stitched are transformed to the same camera model coordinate system, and image fusion processing is performed on the overlapping areas to generate a seamless stitched image.

[0008] By adopting the above technical solution, in the image stitching scenario of UV inkjet printers, the image acquisition step uses a camera to capture multiple images of the scene with overlapping areas containing preset coded markers, providing the necessary data foundation for subsequent stitching and ensuring that there are related parts between different images; the feature extraction step detects and identifies the coded markers and their contours to obtain unique ID number information, accurately locating the pixel coordinates of the outer corner points and geometric center points of the coded markers, providing key features for subsequent accurate matching and mapping; the matching and mapping step matches the same target coded markers based on the unique ID number information and establishes feature point mapping relationships to obtain camera model parameters, enabling different images to be associated under a unified coordinate system; the image stitching step transforms multiple images to be stitched to the same coordinate system based on the camera model parameters and performs fusion processing on the overlapping areas, generating seamless stitched images, effectively solving the problems of discontinuous and unnatural stitching that may occur during the image stitching process of UV inkjet printers, and improving the quality and efficiency of image stitching.

[0009] Optionally, before the image acquisition step, a camera calibration step is also included: controlling the printer platform to move at a preset interval to capture images containing the calibration board, obtaining multiple frames of calibration images; based on the calibration images, detecting feature points of the calibration board pattern, and obtaining the internal parameters and lens distortion coefficients of the camera based on the image coordinates and world coordinates of the feature points, thus obtaining camera calibration parameters.

[0010] By adopting the above technical solution, the printer platform is controlled to move at preset intervals to capture images containing the calibration board, resulting in multiple frames of calibration images. Calibration images at different positions can be obtained, providing rich data for subsequent accurate calculation of camera parameters. Based on the calibration images, feature points of the calibration board pattern are detected, and based on the image coordinates and world coordinates of the feature points, the camera's internal parameters and lens distortion coefficients are obtained, thus obtaining the camera calibration parameters. This allows for precise determination of the camera's internal parameters and lens distortion coefficients, providing an accurate camera model for subsequent image acquisition and processing, and improving the accuracy and quality of image stitching.

[0011] Optionally, before the feature extraction step, an image preprocessing step is also included: performing lens distortion correction on each image to be stitched according to the camera calibration parameters; and cropping the distortion-corrected image to be stitched according to a preset region of interest.

[0012] By adopting the above technical solution, image preprocessing is performed before the feature extraction step. Lens distortion correction is performed on the image to be stitched using camera calibration parameters, which can eliminate the influence of lens distortion on subsequent processing and make the image more consistent with the actual scene. Cropping the distortion-corrected image to be stitched according to the preset region of interest can remove unnecessary image parts, reduce the amount of data, and improve the efficiency and accuracy of subsequent feature extraction and image stitching. Combined with the previous camera calibration and image acquisition steps, as well as the subsequent feature extraction, matching and mapping, and image stitching steps, it helps to finally generate a high-quality seamless stitched image.

[0013] Optionally, the step of establishing a mapping relationship between feature points based on the pixel coordinates of at least four outer corner points and the geometric center point of the successfully matched target encoding marker to obtain the camera model parameters of each image to be stitched includes: The pixel coordinates of at least four outer corner points and the geometric center point of the target encoding marker of one of the two successfully matched images to be stitched are used as the source point coordinate set; Use the pixel coordinates of the corresponding feature points of the same target encoding tag of the other as the target point coordinate set; Based on the mapping relationship between the source point coordinate set and the target point coordinate set, the camera model parameters of each image to be stitched are obtained.

[0014] By adopting the above technical solution, the pixel coordinates of the target coded markers in the two successfully matched images to be stitched are respectively used to form a source point coordinate set and a target point coordinate set. Based on the mapping relationship between the two, the camera model parameters of each image to be stitched are obtained. This can accurately establish the mapping relationship between feature points, providing accurate parameter basis for transforming multiple images to be stitched to the same camera model coordinate system and performing image stitching, which helps to improve the precision and accuracy of image stitching.

[0015] Optionally, the image fusion processing of the overlapping regions includes fusion using a multi-band mixing algorithm.

[0016] By adopting the above technical solution and using a multi-band mixing algorithm to perform image fusion processing on the overlapping areas, multiple images to be stitched can be transformed into the same camera model coordinate system, which can better eliminate the stitching traces in the overlapping areas and generate seamless stitched images.

[0017] Optionally, the encoding tag is an ArUco code or an AprilTag.

[0018] By adopting the above technical solution and selecting ArUco code or AprilTag as the encoding mark, the encoding mark is easy to detect and identify, and can accurately obtain its corresponding unique ID number information. This helps to quickly and accurately match the same target encoding mark in multiple images to be stitched, thereby more accurately establishing the mapping relationship between feature points, obtaining camera model parameters, and ultimately achieving more accurate image stitching.

[0019] Secondly, another embodiment of this application discloses an image stitching system for a UV inkjet printer, which adopts the following solution: An image stitching system for a UV inkjet printer, for performing the method described in any of the preceding claims, comprising: The image acquisition module performs the following steps: capturing scene images using a camera to obtain multiple images to be stitched together, wherein adjacent images to be stitched together have overlapping areas, and the overlapping areas contain at least one preset coded marker; the feature extraction module performs the following steps: detecting and identifying the coded marker and its contour in each image to be stitched together to obtain a unique ID number for each coded marker, wherein the unique ID number includes an ID number, and the pixel coordinates of at least four outer corner points and the geometric center point of each coded marker; the matching module performs the following steps: matching the same target coded marker in the multiple images to be stitched together based on the unique ID number; establishing a mapping relationship between feature points based on the pixel coordinates of at least four outer corner points and the geometric center point of the successfully matched same target coded marker to obtain the camera model parameters of each image to be stitched together; the image stitching module performs the following steps: transforming the multiple images to be stitched together to the same camera model coordinate system based on the camera model parameters of each image to be stitched together, and performing image fusion processing on the overlapping areas to generate a seamless stitched image.

[0020] By adopting the above technical solution, the acquisition module can acquire multiple images with overlapping areas and coded markers to be stitched together, providing a foundation for subsequent stitching; the feature extraction module can obtain the unique ID number information of the coded markers, including the ID number, the pixel coordinates of the outer corner point and the geometric center point, which facilitates accurate identification and positioning; the matching module can match the target coded markers according to the ID number information and establish a mapping relationship, and obtain the camera model parameters, providing a basis for image transformation; the image stitching module transforms the images to the same coordinate system based on the camera model parameters and performs fusion processing to generate a seamless stitched image.

[0021] Optionally, a calibration module is also included to perform camera calibration steps: controlling the printer platform to move at a preset interval to capture images containing the calibration board, obtaining multiple frames of calibration images; based on the calibration images, detecting feature points of the calibration board pattern, and obtaining the internal parameters of the camera and the lens distortion coefficient based on the image coordinates and world coordinates of the feature points, thus obtaining camera calibration parameters.

[0022] By adopting the above technical solution, the printer platform is controlled to move at preset intervals to capture images containing the calibration board, resulting in multiple frames of calibration images. Feature points of the calibration board pattern are detected, and the camera's internal parameters and lens distortion coefficients are solved based on the images of the feature points and world coordinates to obtain the camera calibration parameters. This can improve the accuracy and precision of camera imaging, provide a more reliable foundation for subsequent image stitching processing, reduce image errors caused by the camera's own characteristics, and make the final seamless stitched image more in line with the actual scene.

[0023] Thirdly, another embodiment of this application discloses an electronic device that adopts the following solution: An electronic device, comprising: A processor, a memory, and a computer program stored in the memory and capable of running on the processor, the processor being configured to load and execute the computer program stored in the memory to cause the electronic device to perform the method as described in any of the foregoing.

[0024] Fourthly, another embodiment of this application discloses an electronic device that adopts the following solution: A computer-readable storage medium storing a computer program that, when loaded and executed by a processor, implements the method as described in any of the preceding claims.

[0025] In summary, this application includes at least one of the following beneficial technical effects: 1. In the image stitching scenario of UV inkjet printers, the image acquisition step uses a camera to capture multiple images of the scene, each with overlapping areas containing preset coded markers. This provides the necessary data foundation for subsequent stitching, ensuring that there are related parts between different images. The feature extraction step detects and identifies the coded markers and their contours to obtain unique ID information, accurately locating the pixel coordinates of the outer corner points and geometric center points of the coded markers, providing key features for accurate matching and mapping. The matching and mapping step matches the same target coded markers based on the unique ID information and establishes feature point mapping relationships to obtain camera model parameters, enabling different images to be associated under a unified coordinate system. The image stitching step transforms multiple images to be stitched to the same coordinate system based on the camera model parameters and performs fusion processing on the overlapping areas, generating seamless stitched images. This effectively solves the problems of discontinuous and unnatural stitching that may occur during UV inkjet printer image stitching, improving the quality and efficiency of image stitching. 2. Control the printer platform to move at preset intervals to capture images including the calibration board, obtaining multiple frames of calibration images. This allows for the acquisition of calibration images from different positions, providing rich data for subsequent accurate calculation of camera parameters. Based on the calibration images, feature points of the calibration board pattern are detected, and the camera's internal parameters and lens distortion coefficients are obtained based on the image coordinates and world coordinates of the feature points. This yields the camera calibration parameters, enabling precise determination of the camera's internal parameters and lens distortion coefficients. This provides an accurate camera model for subsequent image acquisition and processing, improving the accuracy and quality of image stitching. 3. Image preprocessing is performed before the feature extraction step. Lens distortion correction is performed on the image to be stitched using camera calibration parameters, which can eliminate the impact of lens distortion on subsequent processing and make the image more consistent with the actual scene. Cropping the distortion-corrected image to be stitched according to the preset region of interest can remove unnecessary image parts, reduce the amount of data, and improve the efficiency and accuracy of subsequent feature extraction and image stitching. Combined with the previous camera calibration and image acquisition steps, as well as the subsequent feature extraction, matching and mapping, and image stitching steps, it helps to generate a high-quality seamless stitched image in the end. Attached Figure Description

[0026] Figure 1 This is a schematic flowchart of an image stitching method for a UV inkjet printer disclosed in an embodiment of this application; Figure 2 A schematic diagram illustrating the process of a camera capturing images of a scene; Figure 3 A schematic diagram illustrating the identification of the four outer corner points and the geometric center point in the ArUco code; Figure 4 This is a schematic diagram illustrating the markings of the printing area and the origin of the print head when printing seamlessly stitched images. Figure 5 This is a schematic diagram of the structure of an image stitching system for a UV inkjet printer disclosed in another embodiment of this application. Detailed Implementation

[0027] The present application will be further described in detail below with reference to the accompanying drawings.

[0028] Embodiments of this application will now be described in more detail with reference to the accompanying drawings. While embodiments of this application are shown in the drawings, it should be understood that this application may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to make this application more thorough and complete, and to fully convey the scope of this application to those skilled in the art.

[0029] The terminology used in this application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. The singular forms “a” and “the” as used in this application and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any or all possible combinations of one or more of the associated listed items.

[0030] It should be understood that although the terms "first," "second," etc., may be used in this application to describe various information, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, without departing from the scope of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0031] The technical solutions of the embodiments of this application are described in detail below with reference to the accompanying drawings.

[0032] [First Embodiment] See Figure 1 The first embodiment of this application discloses an image stitching method for a UV inkjet printer, which includes an image acquisition step, a feature extraction step, a matching and mapping step, and an image stitching step.

[0033] The image acquisition step provides basic image data for subsequent feature extraction, the feature extraction step provides key coded label information for matching and mapping, the matching and mapping step establishes feature point mapping relationships to obtain camera model parameters, and the image stitching step achieves seamless image stitching based on camera model parameters. Through the orderly cooperation of these steps, efficient and accurate image stitching is achieved, reducing the problems of traditional algorithms in terms of overlap ratio requirements, feature point dependence, computational load, and cumulative error.

[0034] Specifically as follows: S10, Image Acquisition Steps: Multiple images to be stitched together are captured by taking pictures of the scene using a camera.

[0035] In this process, adjacent images to be stitched together have overlapping areas, and these overlapping areas contain at least one pre-defined coded marker, which is pre-printed regularly onto the scene images. For example, the shooting process is as follows: Figure 2 As shown, two adjacent patterns have overlapping ARUCO code regions. In this embodiment, the encoding marker can be an ArUco code or an AprilTag, etc. In practical applications, an appropriate encoding marker can be selected according to different scenarios and requirements.

[0036] S20. Feature extraction steps: Detect and identify the coded markers and their outlines in each image to be stitched together to obtain the unique ID number information corresponding to each coded marker; The detection of coded markers can utilize image recognition algorithms, while contour recognition can employ edge detection algorithms. Through these operations, a unique ID number is obtained for each coded marker. See also... Figure 3 The unique ID information includes the ID number, at least c pixel coordinates for each coded tag, and the geometric center point is the intersection of the diagonals of at least four outer corner points.

[0037] For example, see Figure 2 The scene image contains 16 Aruco codes, arranged from left to right and top to bottom according to their ID numbers (0-15). Each Aruco code has a unique corner number. The pixel coordinates of the geometric center point of each Aruco code can be obtained from the pixel coordinates of its four outer corner points. The corresponding unique ID number includes the ID number and five sets of coordinate point information, which are used for subsequent matching of identical coded markers to improve matching accuracy.

[0038] S30. Matching and Mapping Steps: S31. Match the same target encoding mark in multiple images to be stitched together based on the unique ID number information; This step can be understood as finding images with the same pattern among numerous puzzle pieces. Unique ID information allows for quick and accurate identification of identical coded markers, fundamentally eliminating false matches caused by similar appearances. Furthermore, the matching stage only requires integer-to-string comparison or table lookup operations on a small number of decoded IDs, resulting in extremely low computational complexity and enabling real-time, high-frequency image matching and localization.

[0039] S32. Based on the pixel coordinates of at least four outer corner points and geometric center points of the same target encoding mark that have been successfully matched, establish the mapping relationship between feature points to obtain the camera model parameters of each image to be stitched. In this process, the pixel coordinates of at least four outer corner points and the geometric center point of the target coded marker in one of the two successfully matched images to be stitched are used as the source point coordinate set, and the pixel coordinates of the corresponding feature points of the same target coded marker in the other image are used as the target point coordinate set. Based on the mapping relationship between the source point coordinate set and the target point coordinate set, the camera model parameters for each image to be stitched are obtained. The camera model parameters include intrinsic parameters and extrinsic parameters.

[0040] Specifically, the mapping relationship between feature points can be described by perspective transformation. For a planar scene (where marker points are coplanar), this relationship can be represented by a 3x3 homography matrix H: Where (u,v) are the image pixel coordinates, (Xw,Yw) are the world plane coordinates (Z=0), and s is the scale factor.

[0041] H = K[R∣t], where K is the intrinsic parameter matrix and [R∣t] are the first two columns of the extrinsic parameter matrix (because the third column is determined by the plane normal vector, which is usually 0).

[0042] After obtaining the homography matrix H, the camera parameters need to be obtained through matrix decomposition. The mathematical method for establishing the mapping can use the Direct Linear Transform (DLT) algorithm. It should be noted that the above calculation method is an existing technology and will not be elaborated upon further.

[0043] S40. Image stitching steps: Based on the camera model parameters of each image to be stitched, multiple images to be stitched are transformed to the same camera model coordinate system, and image fusion processing is performed on the overlapping areas to generate a seamless stitched image.

[0044] Since each image to be stitched was taken from a different position and angle, the position, shape, and size of the same object are different in the two images (there is perspective distortion). Direct superposition will produce ghosting and misalignment. Therefore, it is necessary to remap the images and then fuse the overlapping areas.

[0045] Image remapping: This step can be understood as using camera parameters to project images originally captured from different viewpoints onto the same virtual "canvas," eliminating parallax differences. Specifically: Using the camera model parameters obtained in the previous step (mainly extrinsic parameters: rotation matrix R and translation vector t, and intrinsic parameter K), a projection transformation matrix (WarpingMatrix) is calculated. For each pixel in the image, its new position on the panoramic image is calculated using this matrix, resulting in multiple "aligned" images in the same spatial coordinate system (at this point, static objects in the scene should be roughly aligned in the overlapping area).

[0046] Overlapping region fusion: A multi-band mixing algorithm is used to decompose the image into different frequency bands (high-frequency details, low-frequency colors). A wide transition band is used to smoothly blend the low-frequency part to eliminate color differences, while a narrow transition band is used in the high-frequency part to preserve the clarity of details, thereby generating the most "seamless" image (i.e. seamless stitched image).

[0047] Furthermore, prior to step S10, the image acquisition step, the following steps are also included: S01. Camera calibration steps: S011. Control the printer platform to move at a preset interval to capture images containing the calibration board, and obtain multiple frames of calibration images; The preset spacing is set to a default value before execution. This ensures that the printer platform moves a fixed distance after each photo is taken, guaranteeing that the acquired image sequence is spatially evenly distributed and avoiding uneven data density. Each time the printer platform moves to a preset position, the system triggers the camera to take a picture, ensuring that the calibration board (such as a checkerboard or dot array) appears clearly and completely in the image. The final output is an image sequence, such as dozens or even hundreds of images from position 1 to position N.

[0048] S012. Based on the calibration image, detect the feature points of the calibration board pattern, and based on the image coordinates and world coordinates of the feature points, obtain the camera's internal parameters and lens distortion coefficients to obtain the camera calibration parameters.

[0049] This step is performed based on Zhang Zhengyou's camera calibration method, which uses images of a planar calibration board (such as a checkerboard) from multiple perspectives to solve for the camera physical model by minimizing the reprojection error.

[0050] By performing a camera calibration procedure, the camera's internal parameters and lens distortion coefficients can be accurately obtained, enabling distortion correction in subsequent image processing. This helps eliminate the impact of lens distortion on image stitching, improving the quality and accuracy of the stitched images. Compared to methods without camera calibration, this embodiment is better adaptable to different cameras and printing environments, reducing stitching errors caused by lens distortion and providing a more reliable foundation for image stitching.

[0051] Furthermore, prior to step S20, the feature extraction step, the following steps are also included: S02, Image preprocessing steps: S021. Based on the camera calibration parameters, perform lens distortion correction on each image to be stitched; The camera lens is not an ideal pinhole imaging model. Due to manufacturing limitations, light bends as it passes through the lens edge, causing straight lines at the image edges to curve (radial distortion) or the overall image to shift tangentially (tangential distortion). If this is not corrected, directly stitching the images together will result in the following problems: Feature point mismatch: The positions of the same physical point in adjacent images may not be accurately aligned due to different degrees of distortion.

[0052] Seam gaps: Non-linear deformation caused by distortion can prevent overlapping areas from perfectly aligning, resulting in ghosting or breaks.

[0053] Therefore, this step is a key step in image stitching preprocessing: lens distortion correction. Using the calibrated camera physical model, the "bent" lines in each image to be stitched (original image) caused by lens physical defects are restored to "straight" lines, providing accurate input for subsequent feature point matching and geometric transformation.

[0054] S022. Based on the preset region of interest, crop the image to be stitched after distortion correction.

[0055] The preset "Region of Interest (ROI)" is used to define the part that contains only the valid scene, so as to avoid these invalid pixels from interfering with subsequent feature point matching. In other words, it is to crop out unnecessary parts of the image and reduce the amount of computation in subsequent processing.

[0056] Image preprocessing steps can improve image quality and processing efficiency. Lens distortion correction can eliminate the effects of lens distortion on the image, making it more realistic. Cropping operations can reduce irrelevant information, focus on processing regions of interest, and improve the accuracy and speed of feature extraction and stitching. Therefore, through image preprocessing, image data can be better utilized, providing higher-quality input for subsequent image stitching, thereby improving the performance of the entire image stitching process.

[0057] See Figure 4 After obtaining a seamlessly stitched image, the image and the actual size to be printed can be aligned with the printer nozzle origin with a precision of 10:1. The white edge of the printing area is reserved with an offset of (10, 5). After printing, it can be cropped again to a print size of 310X420mm (image resolution 3100X4200).

[0058] The implementation principle of this embodiment is as follows: By introducing coded tags, the excessive reliance on feature points in traditional image stitching algorithms is avoided, reducing the difficulty in finding feature points in images with little or no texture. Simultaneously, based on the matching and mapping of coded tags, camera model parameters can be obtained more accurately, thereby reducing systemic issues. The use of a multi-band mixing algorithm improves the quality of image fusion, making the stitched image more seamless. This method improves the efficiency and accuracy of image stitching for UV inkjet printers, providing a more effective solution for large-format, high-resolution image stitching in UV inkjet printers. Compared to traditional methods, it has significant advantages and can better adapt to the needs of practical applications.

[0059] [Second Embodiment] See Figure 5 The second embodiment of this application discloses an image stitching system for a UV inkjet printer, including: an acquisition module 210, a feature extraction module 220, a matching module 230 and an image stitching module 240.

[0060] The acquisition module 210 is used to perform the image acquisition step: to capture scene images through a camera and acquire multiple images to be stitched together, wherein adjacent images to be stitched together have overlapping areas, and the overlapping areas contain at least one preset coded mark. The feature extraction module 220 is used to perform the feature extraction steps: detect and identify the coded markers and the contours of the coded markers in each image to be stitched, so as to obtain the unique ID number information corresponding to each coded marker, wherein the unique ID number information includes the ID number, the pixel coordinates of at least four outer corner points and the geometric center point of each coded marker; The matching module 230 is used to perform the matching and mapping steps: matching the same target encoding mark in multiple images to be stitched according to the unique ID number information; establishing the mapping relationship between feature points based on the pixel coordinates of at least four outer corner points and geometric center points of the successfully matched same target encoding mark, so as to obtain the camera model parameters of each image to be stitched. The image stitching module 240 is used to perform the image stitching steps: based on the camera model parameters of each image to be stitched, the multiple images to be stitched are transformed to the same camera model coordinate system, and the overlapping areas are processed by image fusion to generate a seamless stitched image.

[0061] Furthermore, the system also includes a calibration module for performing camera calibration steps: The printer platform is controlled to move at preset intervals to capture images containing the calibration board, resulting in multiple frames of calibration images. Based on the calibration image, feature points of the calibration board pattern are detected, and based on the image coordinates and world coordinates of the feature points, the internal parameters of the camera and the lens distortion coefficient are obtained to obtain the camera calibration parameters.

[0062] It should be noted that the image stitching system for a UV inkjet printer disclosed in this embodiment implements the image stitching method for a UV inkjet printer as described in the above embodiment, and therefore will not be described in detail here. Optionally, the various modules and other operations or functions in this embodiment are respectively for implementing the methods in the foregoing embodiments.

[0063] [Third Embodiment] The third embodiment of this application discloses an electronic device, which includes a memory and a processor. The memory is used to store a computer program; the processor is used to execute the computer program to implement the steps of the image stitching method for a UV inkjet printer described in the first embodiment above. For details, please refer to the above description and will not be described in detail here.

[0064] The technical effect of the electronic device provided in this embodiment in practical application is the same as the technical effect of the image stitching method for a UV inkjet printer in the first embodiment.

[0065] [Fourth Embodiment] A computer-readable storage medium is disclosed in the fourth embodiment of this application. The computer-readable storage medium is, for example, a non-volatile memory, such as magnetic media (e.g., hard disks, floppy disks, and magnetic tapes), optical media (e.g., CD-ROMs and DVDs), magneto-optical media (e.g., optical discs), and hardware devices specifically configured to store and execute computer-executable instructions (e.g., read-only memory (ROM), random access memory (RAM), flash memory, etc.). A computer program is stored on the computer-readable storage medium. The computer-readable storage medium can be executed by one or more processors or processing devices to implement an image stitching method for a UV inkjet printer as described in the foregoing embodiments.

[0066] Furthermore, it is understood that the foregoing embodiments are merely illustrative examples of the present invention. Provided that the technical features do not conflict, the structure is not contradictory, and the purpose of the invention is not violated, the technical solutions of the various embodiments can be arbitrarily combined and used.

[0067] In the embodiments provided by this invention, it should be understood that the disclosed methods, systems, and measuring devices can be implemented in other ways. For example, the modules included in the systems described above are merely illustrative, and the division of modules is only a logical functional division. In actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0068] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0069] Furthermore, in the various embodiments of the present invention, the functional units / modules can be integrated into one processing unit / module, or each unit / module can exist physically separately, or two or more units / modules can be integrated into one unit / module. The integrated unit / module described above can be implemented in hardware or in the form of hardware plus software functional units / modules.

[0070] The integrated units / modules implemented as software functional units / modules described above can be stored in a computer-readable storage medium. These software functional units, stored in a storage medium, include several instructions to cause one or more processors of a computer measurement device (which may be a personal computer, server, or network measurement device, etc.) to execute some steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0071] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An image stitching method for a UV inkjet printer, characterized in that, include: Image acquisition steps: Capture scene images with a camera to obtain multiple images to be stitched together, wherein adjacent images to be stitched together have overlapping areas, and the overlapping areas contain at least one preset coded mark. Feature extraction step: Detect and identify the coded markers and their contours in each image to be stitched together to obtain unique ID information corresponding to each coded marker, wherein the unique ID information includes the ID number, the pixel coordinates of at least four outer corner points and the geometric center point of each coded marker; Matching and mapping steps: Based on the unique ID number information, match the same target encoding mark in multiple images to be stitched; based on the pixel coordinates of at least four outer corner points and geometric center points of the successfully matched same target encoding mark, establish the mapping relationship between feature points to obtain the camera model parameters of each image to be stitched. Image stitching steps: Based on the camera model parameters of each image to be stitched, multiple images to be stitched are transformed to the same camera model coordinate system, and image fusion processing is performed on the overlapping areas to generate a seamless stitched image.

2. The method according to claim 1, characterized in that, Prior to the image acquisition step, a camera calibration step is also included: The printer platform is controlled to move at preset intervals to capture images containing the calibration board, resulting in multiple frames of calibration images. Based on the calibration image, feature points of the calibration board pattern are detected, and based on the image coordinates and world coordinates of the feature points, the internal parameters of the camera and the lens distortion coefficient are obtained to obtain the camera calibration parameters.

3. The method according to claim 2, characterized in that, Prior to the feature extraction step, an image preprocessing step is also included: Based on the camera calibration parameters, lens distortion correction is performed on each image to be stitched. The image to be stitched is cropped based on the preset region of interest after distortion correction.

4. The method according to claim 1, characterized in that, The mapping relationship between feature points is established based on the pixel coordinates of at least four outer corner points and the geometric center point of the successfully matched target encoding marker, in order to obtain the camera model parameters of each image to be stitched, including: The pixel coordinates of at least four outer corner points and the geometric center point of the target encoding marker of one of the two successfully matched images to be stitched are used as the source point coordinate set; Use the pixel coordinates of the corresponding feature points of the same target encoding tag of the other as the target point coordinate set; Based on the mapping relationship between the source point coordinate set and the target point coordinate set, the camera model parameters of each image to be stitched are obtained.

5. The method according to claim 1, characterized in that, The image fusion processing of the overlapping regions includes using a multi-band mixing algorithm for fusion.

6. The method according to claim 1, characterized in that, The encoding tag is either ArUco code or AprilTag.

7. An image stitching system for a UV inkjet printer, characterized in that, For performing the method according to any one of claims 1 to 6, comprising: The acquisition module is used to perform the image acquisition step: to capture scene images through a camera and acquire multiple images to be stitched together, wherein adjacent images to be stitched together have overlapping areas, and the overlapping areas contain at least one preset coded mark. The feature extraction module is used to perform the feature extraction steps: detect and identify the coded markers and the contours of the coded markers in each image to be stitched, so as to obtain the unique ID number information corresponding to each coded marker, wherein the unique ID number information includes the ID number, the pixel coordinates of at least four outer corner points and the geometric center point of each coded marker; The matching module is used to perform the matching and mapping steps: matching the same target encoding mark in multiple images to be stitched according to the unique ID number information; establishing the mapping relationship between feature points based on the pixel coordinates of at least four outer corner points and geometric center points of the successfully matched same target encoding mark, so as to obtain the camera model parameters of each image to be stitched. The image stitching module is used to perform the image stitching steps: based on the camera model parameters of each image to be stitched, multiple images to be stitched are transformed to the same camera model coordinate system, and image fusion processing is performed on the overlapping areas to generate a seamless stitched image.

8. The system according to claim 7, characterized in that, It also includes a calibration module for performing camera calibration steps: The printer platform is controlled to move at preset intervals to capture images containing the calibration board, resulting in multiple frames of calibration images. Based on the calibration image, feature points of the calibration board pattern are detected, and based on the image coordinates and world coordinates of the feature points, the internal parameters of the camera and the lens distortion coefficient are obtained to obtain the camera calibration parameters.

9. An electronic device, characterized in that, include: A processor, a memory, and a computer program stored in the memory and capable of running on the processor, the processor being configured to load and execute the computer program stored in the memory to cause the electronic device to perform the method as described in claims 1 to 6.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is loaded and executed by the processor, it implements the method described in claims 1 to 6.

Images