Laser galvanometer distortion calibration method, device, equipment and medium

Abstract

This disclosure presents embodiments of a laser galvanometer distortion calibration method, apparatus, device, and medium. One specific implementation of the method includes: performing a dual-pose scan on a target plate to generate a first image set and a second image set; generating a first set of center pixel coordinates and a second set of center pixel coordinates based on the first and second image sets; performing a discreteness analysis on the first and second set of center pixel coordinates; selectively fusing the first and second set of center pixel coordinates based on axial label information to construct a synthetic coordinate matrix; generating a pixel deviation matrix based on the synthetic coordinate matrix and a reference coordinate matrix; generating a continuous distortion calibration model based on the pixel deviation matrix; and performing a calibration operation on the laser galvanometer. This implementation can utilize a common scanner to calibrate laser galvanometer distortion, thereby significantly reducing equipment investment costs.

Description

Technical Field

[0001] The embodiments disclosed herein relate to the field of computer technology, and more specifically to laser galvanometer distortion calibration methods, apparatus, devices, and media. Background Technology

[0002] Currently, existing techniques for calibrating laser galvanometer distortion mainly rely on specialized measuring equipment (such as 2D image measuring instruments). These devices establish coordinate mapping relationships by identifying marked points on a target plate to achieve calibration. However, when using this method to calibrate laser galvanometer distortion, the following technical problems often arise: these specialized measuring devices are extremely expensive, bulky, and used infrequently, resulting in high calibration costs and inconvenient operation, making it difficult to widely adopt and flexibly apply them in the field for a broad range of laser equipment users.

[0003] The information disclosed in this background section is only intended to enhance the understanding of the background of the inventive concept, and therefore may contain information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0004] The summary portion of this disclosure is intended to provide a brief overview of the concepts, which will be described in detail in the detailed description portion. This summary portion is not intended to identify key or essential features of the claimed technical solutions, nor is it intended to limit the scope of the claimed technical solutions.

[0005] Some embodiments of this disclosure provide laser galvanometer distortion calibration methods, apparatuses, electronic devices, and computer-readable media to address one or more of the technical problems mentioned in the background section above.

[0006] In a first aspect, some embodiments of this disclosure provide a laser galvanometer distortion calibration method, comprising: performing a dual-pose scan on a target plate to generate a first image set and a second image set, wherein the dual-pose scan includes a first orientation pose and a second orientation pose; generating a first center pixel coordinate set and a second center pixel coordinate set based on the first image set and the second image set; performing a discreteness analysis on the first center pixel coordinate set and the second center pixel coordinate set to generate axial label information; selectively fusing the first center pixel coordinate set and the second center pixel coordinate set based on the axial label information to construct a synthetic coordinate matrix; generating a pixel deviation matrix based on the synthetic coordinate matrix and a reference coordinate matrix; generating a continuous distortion calibration model based on the pixel deviation matrix; and performing a calibration operation on the laser galvanometer using the continuous distortion calibration model.

[0007] Secondly, some embodiments of this disclosure provide a laser galvanometer distortion calibration device, comprising: an execution unit configured to perform dual-pose scanning on a target plate to generate a first image set and a second image set, wherein the dual-pose scanning includes a first orientation pose and a second orientation pose; a first generation unit configured to generate a first center pixel coordinate set and a second center pixel coordinate set based on the first image set and the second image set; a discreteness analysis unit configured to perform discreteness analysis on the first center pixel coordinate set and the second center pixel coordinate set to generate axial label information; a fusion unit configured to selectively fuse the first center pixel coordinate set and the second center pixel coordinate set based on the axial label information to construct a synthetic coordinate matrix; a second generation unit configured to generate a pixel deviation matrix based on the synthetic coordinate matrix and a reference coordinate matrix; a third generation unit configured to generate a continuous distortion calibration model based on the pixel deviation matrix; and a calibration unit configured to perform calibration operations on a laser galvanometer using the continuous distortion calibration model.

[0008] Thirdly, some embodiments of this disclosure provide an electronic device, including: one or more processors; and a storage device having one or more programs stored thereon, such that when the one or more programs are executed by the one or more processors, the one or more processors implement the method as described in any implementation of the first aspect.

[0009] Fourthly, some embodiments of this disclosure provide a computer-readable medium having a computer program stored thereon, wherein the program, when executed by a processor, implements the method as described in any implementation of the first aspect.

[0010] The various embodiments disclosed above have the following beneficial effects: The laser galvanometer distortion calibration method of some embodiments of this disclosure can utilize a common scanner to calibrate laser galvanometer distortion, thereby significantly reducing equipment investment costs. Specifically, the high equipment investment cost is due to the fact that traditional calibration methods must rely on expensive dedicated measuring equipment (e.g., a two-dimensional image measuring instrument). Based on this, the image segmentation method of some embodiments of this disclosure first performs a dual-pose scan on the target plate to generate a first image set and a second image set. The dual-pose scan includes a first orientation pose and a second orientation pose. By scanning in two orthogonal directions, the differences in the accuracy characteristics of the scanner in different motion directions are actively captured, providing "raw material" containing directional error features. This is a fundamental data acquisition strategy to overcome the insufficient accuracy of a single scan and achieve low-cost substitution. Then, based on the first image set and the second image set, a first set of circle center pixel coordinates and a second set of circle center pixel coordinates are generated. The visual features (circles) in the first and second image sets are accurately quantized into computer-processable coordinate data, completing the key transformation from analog images to digital coordinates, providing structured input for subsequent quantitative analysis, comparison, and calculation. Next, a discreteness analysis is performed on the first and second circle-center pixel coordinate sets to generate axial label information. This discreteness analysis and axial label generation identifies coordinate stability and abnormal distributions, distinguishing reliable data from outliers and providing a basis for subsequent screening and fusion. Then, based on the axial label information, the first and second circle-center pixel coordinate sets are selectively fused to construct a synthetic coordinate matrix. This selective fusion of coordinates based on the axial label information preserves high-reliability data, eliminates outliers, and improves overall coordinate accuracy and consistency. Next, a pixel deviation matrix is ​​generated based on the synthetic coordinate matrix and the reference coordinate matrix. Finally, a continuous distortion calibration model is generated based on the pixel deviation matrix. Discrete, measurement-point-limited deviation data is transformed into a continuous function model applicable to the entire working area through mathematical fitting (e.g., polynomial surface fitting). This model can predict distortions at any unsampled point within the working area, providing a direct tool for implementing smooth, seamless spatial correction. Finally, the aforementioned continuous distortion calibration model is used to perform calibration operations on the laser galvanometer. This allows for real-time, automatic distortion compensation for any target coordinates sent to the galvanometer, ensuring the laser focus accurately reaches the theoretical position. This achieves laser galvanometer distortion calibration and significantly reduces equipment investment costs. Attached Figure Description

[0011] The above and other features, advantages, and aspects of the embodiments of this disclosure will become more apparent from the accompanying drawings and the following detailed description. Throughout the drawings, the same or similar reference numerals denote the same or similar elements. It should be understood that the drawings are schematic, and elements are not necessarily drawn to scale.

[0012] Figure 1 This is a flowchart of some embodiments of the laser galvanometer distortion calibration method according to the present disclosure; Figure 2 These are schematic diagrams of some embodiments of the laser galvanometer distortion calibration apparatus according to this disclosure; Figure 3 This is a schematic diagram of the structure of an electronic device suitable for implementing some embodiments of the present disclosure.

[0013] Figure 4 This can be a schematic diagram showing the error detection results of laser galvanometer distortion that has not been calibrated according to this disclosure.

[0014] Figure 5 This can be illustrated as a schematic diagram showing the error detection results of laser galvanometer distortion calibrated by the method of the present invention. Detailed Implementation

[0015] Embodiments of this disclosure will now be described in more detail with reference to the accompanying drawings. While some embodiments of this disclosure are shown in the drawings, it should be understood that this disclosure can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of this disclosure. It should be understood that the accompanying drawings and embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of protection of this disclosure.

[0016] It should also be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings. Unless otherwise specified, the embodiments and features described in this disclosure can be combined with each other.

[0017] It should be noted that the concepts of "first" and "second" mentioned in this disclosure are used only to distinguish different devices, modules or units, and are not used to limit the order of functions performed by these devices, modules or units or their interdependencies.

[0018] It should be noted that the terms "a" and "a plurality of" used in this disclosure are illustrative rather than restrictive, and those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0019] The names of messages or information exchanged between multiple devices in the embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of such messages or information.

[0020] This disclosure will now be described in detail with reference to the accompanying drawings and embodiments.

[0021] refer to Figure 1 The diagram illustrates a flow 100 of some embodiments of a laser galvanometer distortion calibration method according to the present disclosure. This laser galvanometer distortion calibration method includes the following steps: Step 101: Perform dual-pose scanning on the target plate to generate a first image set and a second image set.

[0022] In some embodiments, the execution entity (e.g., an electronic device) of the above-described laser galvanometer distortion calibration method can be hardware or software. When the computing device is hardware, it can be implemented as a distributed cluster of multiple servers or terminal devices, or as a single server or a single terminal device. When the computing device is software, it can be installed in the hardware devices listed above. It can be implemented as multiple software programs or software modules to provide distributed services, or as a single software program or software module. No specific limitations are made here.

[0023] In other embodiments, the execution entity can perform dual-pose scanning on the target plate to generate a first image set and a second image set. The dual-pose scanning includes a first orientation and a second orientation. The target plate can be a plate-like object printed with a specific pattern (e.g., an M×N matrix of hollow rings). Dual-pose scanning refers to performing two scanning operations on the same target plate in two different orientations. For example, first scanning laterally, then rotating the target plate 90 degrees, and then performing a longitudinal scan. The first and second orientations refer to the placement angles of the target plate relative to the scanner's scanning direction. The first and second orientations are typically two orthogonal directions, such as 0 degrees (lateral) and 90 degrees (longitudinal). The first image set can be a set of image data obtained under the first orientation, representing a single-view image set. The second image set can be a set of image data obtained under the second orientation, with a different viewpoint than the first image set.

[0024] In some optional implementations of certain embodiments, the execution entity may perform dual-pose scanning on the target plate to generate a first image set and a second image set, which may include the following steps: The first step involves using a target scanner to scan the target plate in the aforementioned first orientation to generate a first image set. The target scanner can be a general-purpose flatbed scanner used for image acquisition. In practice, the target plate is first fixed in the first orientation. Then, the target scanner is activated to image the target plate. Finally, all images from this viewpoint are acquired and saved to form the first image set.

[0025] The second step involves using a target scanner to scan the target plate in the aforementioned second orientation to generate a second image set. In practice, first, the target plate is adjusted to the second orientation and fixed. Then, the target plate is rescanned using the target scanner. Finally, images from this perspective are acquired and compiled into the second image set.

[0026] Step 102: Based on the first image set and the second image set, generate the first circle center pixel coordinate set and the second circle center pixel coordinate set.

[0027] In some embodiments, the execution entity may generate a first set of center pixel coordinates and a second set of center pixel coordinates based on the first image set and the second image set. The first set of center pixel coordinates may be a collection of two-dimensional coordinates of all center points in the image pixel coordinate system extracted from the first image set. The second set of center pixel coordinates may be a collection of two-dimensional coordinates of all center points in the image pixel coordinate system extracted from the second image set.

[0028] In some optional implementations of certain embodiments, the execution entity may generate a first set of center pixel coordinates and a second set of center pixel coordinates based on the first image set and the second image set, which may include the following steps: The first step involves identifying reference marker points in each image of the first and second image sets to generate a set of reference point coordinates. These reference marker points can be geometric patterns of a specific shape printed outside the circular array area of ​​the target plate, used for image alignment (e.g., four additional cross-shaped or solid dot markers printed outside the four corners of a 10×10 dot array). The reference point coordinates can be the position coordinates of the reference marker points located in each image within the image pixel coordinate system. For example, the reference point coordinates could be the center point coordinates of four cross-shaped markers: (50, 50), (1950, 50), (50, 2950), and (1950, 2950). In practice, firstly, each image in the first and second image sets is loaded using an image processing algorithm (e.g., an OpenCV-based image processing algorithm). Then, the image processing algorithm searches for a pattern of preset reference markers (e.g., four "L"-shaped corner markers) in each image and uses template matching or feature detection methods to accurately locate their positions in the image.

[0029] Finally, the pixel coordinates of each marker point are recorded, and this set of coordinates is the reference point coordinate set of the image.

[0030] The second step involves performing pose unification on each image in both the first and second image sets, based on the aforementioned reference point coordinate set, to generate a pose-unified image set. This pose unification can be achieved by using rotation and translation transformations based on the reference point coordinates to correct all images to a unified standard position and angle. The pose-unified image set can be a collection of images aligned to the same coordinate system after pose unification. In practice, the reference point coordinate set is used as the calibration basis first. Then, rotation and translation corrections are performed on the images in the first and second image sets. Finally, a pose-unified image set with a unified viewpoint is obtained.

[0031] The third step involves extracting the center coordinates and normalizing the coordinate system of each pose-unified image in the aforementioned image set to generate a first set of center pixel coordinates and a second set of center pixel coordinates. In practice, firstly, each image in the pose-unified image set (e.g., Aligned_H.jpg) is processed. Within the known annular array region, image processing techniques (e.g., Hough circle detection or contour analysis) are used to identify each annulus and calculate the pixel coordinates of its center. Then, to eliminate any overall translation that may occur during image correction, the offset of all center coordinates relative to the coordinates of the central annulus of the annular array is calculated. Finally, this offset is subtracted from all center coordinates to achieve coordinate system normalization. The coordinates from the first image set constitute the first set of center pixel coordinates, and the coordinates from the second image set constitute the second set of center pixel coordinates; both are in a comparable, unified coordinate system with the array center as the origin.

[0032] Step 103: Perform a discreteness analysis on the first and second circle center pixel coordinate sets to generate axial label information.

[0033] In some embodiments, the execution entity may perform discrete analysis on the first set of center pixel coordinates and the second set of center pixel coordinates to generate axial label information. The axial label information may be a decision label used to identify which coordinate axis (X-axis, Y-axis) data in each set of center pixel coordinates is more reliable, integrating encoded information of the valid axial determination result to identify the valid axial axis of the coordinate set.

[0034] In some optional implementations of certain embodiments, the execution entity may perform discrete analysis on the first set of center pixel coordinates and the second set of center pixel coordinates to generate axial label information, which may include the following steps: The first step is to determine the row dispersion of each coordinate in the first and second center pixel coordinate sets to generate the first dispersion data. The row dispersion can be the degree of fluctuation of the X-axis coordinate values ​​of the center of the same row in the image, usually expressed as the standard deviation. The first dispersion data can refer to the set of all row dispersion values ​​obtained after calculating the row dispersion of the first and second center pixel coordinate sets. In practice, firstly, the first and second center pixel coordinate sets are read separately. For each coordinate set, grouping by row: extract the X-coordinates of all center points in the same row, calculate their standard deviation or range, and obtain the row dispersion of that row; repeat this operation for all rows to generate the first dispersion data.

[0035] The second step involves determining the column dispersion of each coordinate in both the first and second center pixel coordinate sets to generate second dispersion data. The column dispersion refers to the fluctuation of the Y-axis coordinate values ​​of the same row center in both sets, typically expressed as standard deviation. The second dispersion data can be the set of all column dispersion values ​​obtained after calculating the column dispersion of both sets. In practice, firstly, the first and second center pixel coordinate sets are read. For each coordinate set, it is grouped by column: all center Y-coordinates in the same column are extracted, and their standard deviation or range is calculated to obtain the column dispersion for that column; this operation is repeated for all columns to generate the second dispersion data.

[0036] The third step involves comparing the first and second discrete data sets to generate a discrete comparison result. This result can be a conclusion drawn from comparing the first and second discrete data sets within the same coordinate set, used to determine which direction's coordinate data is more stable and accurate. In practice, firstly, the corresponding first and second discrete data sets are taken. Then, the magnitude relationships are compared group by group. Finally, a discrete comparison result showing "more discrete rows / columns" is generated.

[0037] The fourth step involves determining the effective axis of the first circle center pixel coordinate set based on the aforementioned dispersion comparison results, thereby generating a first axis determination identifier. The effective axis can be an axis with smaller dispersion and a more regular distribution, serving as a valid reference. The first axis determination identifier can be an identifier after determining the effective axis of the first circle center coordinate set. For example, the first axis determination identifier could be the identifier "X," indicating that the effective axis of the first coordinate set is the X-axis. In practice, the dispersion comparison results are first used. Then, the axis with smaller dispersion in the first coordinate set is determined to be the effective axis. Finally, the corresponding first axis determination identifier is generated. For example, if the column dispersion is smaller, the effective axis is determined to be Y, and the identifier "Y" is output.

[0038] Fifth, based on the above dispersion comparison results, the effective axis of the second circle center pixel coordinate set is determined to generate second axis determination identifier information. This second axis determination identifier information can be an identifier after determining the effective axis of the second circle center coordinate set. For example, the second axis determination identifier information can be the identifier "Y" indicating that the effective axis of the second coordinate set is the Y-axis. In practice, first, the dispersion comparison results are used. Then, the effective axis of the second coordinate set is determined. Finally, the corresponding second axis determination identifier information is generated.

[0039] The sixth step involves integrating and encoding the first and second axial direction determination information to generate the axial label information. In practice, the first and second axial direction determination information are first obtained. Then, they are concatenated and encoded according to rules. Finally, axial label information is generated to identify the overall axial direction. For example, the integrated and encoded axial label information means: the X-coordinate is taken from the first coordinate set, and the Y-coordinate is taken from the second coordinate set.

[0040] Step 104: Based on the axial label information, selectively fuse the first circle center pixel coordinate set and the second circle center pixel coordinate set to construct a synthetic coordinate matrix.

[0041] In some embodiments, the execution entity may selectively fuse the first set of center pixel coordinates and the second set of center pixel coordinates based on the axial label information to construct a composite coordinate matrix. The composite coordinate matrix may be a high-precision pixel coordinate matrix containing the complete two-dimensional coordinates (X, Y) of all center pixels, generated after selective fusion. The selective fusion may be a process of selecting reliable axial data from different coordinate sets based on the axial label information and combining them into a complete coordinate matrix.

[0042] In some optional implementations of certain embodiments, the execution entity may selectively fuse the first set of center pixel coordinates and the second set of center pixel coordinates based on the axial label information to construct a composite coordinate matrix, which may include the following steps: The first step is to generate first valid axial information for the first set of center pixel coordinates based on the aforementioned axial label information. This first valid axial information can be an instruction parsed from the axial label information indicating which coordinate axis (X or Y) should be adopted for the first set of center pixel coordinates. The valid axis can refer to the coordinate axis direction determined by dispersion comparison to be more reliable and with less coordinate fluctuation. In practice, firstly, the valid axis corresponding to the first set of center pixel coordinates is extracted from the axial label information. Finally, the first valid axial information is generated.

[0043] The second step involves extracting the corresponding axis coordinate data from the first circle center pixel coordinate set based on the aforementioned first effective axis information, thereby generating a first effective coordinate subset. This first effective coordinate subset can be a set of coordinates extracted from the first circle center pixel coordinate set according to the effective axis. In practice, the first effective coordinate subset can be obtained by extracting the axis coordinates from the first circle center pixel coordinate set based on the first effective axis information.

[0044] The third step involves generating second valid axial information for the second set of center pixel coordinates, based on the aforementioned axial label information. This second valid axial information can be an instruction parsed from the axial label information, indicating which coordinate axis (X or Y) should be adopted for the second set of center pixel coordinates. In practice, first, the valid axial direction corresponding to the second set of center pixel coordinates is extracted from the axial label information. Finally, the second valid axial information is generated.

[0045] Fourth, based on the aforementioned second effective axial information, extract the coordinate data of the corresponding axial direction from the aforementioned second center pixel coordinate set to generate a second effective coordinate subset. This second effective coordinate subset can be a set of coordinates extracted from the second center pixel coordinate set according to the effective axial direction.

[0046] The fifth step involves pairing and combining the first and second effective axial information to generate a preliminary fused coordinate matrix. This preliminary fused coordinate matrix can be an initial two-dimensional coordinate matrix formed by pairing two effective coordinate subsets (e.g., an X subset and a Y subset) in point order. For example, the preliminary fused coordinate matrix can be formed by pairing 100 X coordinates (from the first effective coordinate subset) and 100 Y coordinates (from the second effective coordinate subset) in sequence to generate 100 (X, Y) points, constituting a preliminary matrix. In practice, firstly, it is ensured that the first and second effective coordinate subsets have the same number of points and correspond in order (i.e., both correspond to points 1 to 100 in the matrix). Then, elements at the same index position in the two one-dimensional sequences are paired to form a two-dimensional coordinate point. Finally, all paired two-dimensional coordinate points are arranged in matrix order to generate the preliminary fused coordinate matrix.

[0047] The sixth step is to center-align the preliminary fused coordinate matrix to construct the composite coordinate matrix. In practice, first, the coordinates of the geometric center point of the preliminary fused coordinate matrix are determined (for example, the average of the X coordinates of all points is taken as Xc, and the average of the Y coordinates is taken as Yc). Then, the coordinates of each point in the matrix are subtracted from the coordinates of this center point. Finally, the resulting new coordinate matrix is ​​the composite coordinate matrix centered with the reference coordinate matrix.

[0048] Step 105: Generate a pixel deviation matrix based on the synthesized coordinate matrix and the reference coordinate matrix.

[0049] In some embodiments, the execution entity can generate a pixel deviation matrix based on the synthesized coordinate matrix and the reference coordinate matrix. The reference coordinate matrix can be a preset standard two-dimensional matrix of circle center coordinates, serving as a deviation comparison benchmark and possessing known high precision. The pixel deviation matrix can be a numerical matrix, where each element represents the physical deviation (e.g., millimeters) between corresponding points in the synthesized coordinate matrix and the reference coordinate matrix. For example, the pixel deviation matrix can be a 10×10 matrix, where each element is a vector (ΔX, ΔY) representing the positional deviation of the corresponding point before and after calibration, in millimeters.

[0050] In addressing the technical problems mentioned above, and considering the application scenario: many laser equipment users lack expensive dedicated measurement equipment, yet urgently need a low-cost, easy-to-operate method for galvanometer distortion calibration. This often presents the following technical challenges: while ordinary scanners are low-cost, they suffer from mechanical feed jitter errors, resulting in insufficient accuracy in direct measurement data and making it difficult to meet precision calibration requirements. Given the following requirements for this application scenario: low cost, high portability, and ease of operation, and without increasing hardware costs, to compensate for scanner errors and extract high-precision deviation data, we decided to adopt the following solution: In some optional implementations of certain embodiments, the execution entity may generate a pixel deviation matrix based on the synthesized coordinate matrix and the reference coordinate matrix, which may include the following steps: The first step is to pair the synthesized coordinate matrix with the reference coordinate matrix to generate a sequence of coordinate point pairs. This sequence can be a list of paired coordinates obtained by matching each coordinate point in the synthesized and reference coordinate matrices at the same array position. In practice, first, the synthesized coordinate matrix (from the laser under test) and the reference coordinate matrix (from a distortion-free reference laser or theoretical design values) are loaded. Then, based on the array indices (row and column numbers) of the annexes in the two matrices, the coordinate points at identical array positions are matched one-to-one. Finally, an ordered sequence of coordinate point pairs is generated.

[0051] The second step involves performing element-wise subtraction of pixel coordinate values ​​for each coordinate point pair in the above sequence to generate the original pixel deviation vector set. This original pixel deviation vector set can be a set of two-dimensional offset vectors in pixels, generated by performing subtraction on each coordinate point pair, for example (Δx = +3.2 pixels, Δy = -1.5 pixels). In practice, first, each coordinate pair in the sequence is traversed. Then, vector subtraction is performed on each coordinate pair: deviation vector = composite coordinate - reference coordinate, i.e., calculating the X-axis difference Δx and the Y-axis difference Δy respectively. Finally, all calculated two-dimensional deviation vectors are organized in the original array order to generate the original pixel deviation vector set.

[0052] The third step involves generating a first pixel deviation component set and a second pixel deviation component set based on the original pixel deviation vector set. The first pixel deviation component set can be the set of all X-axis deviation values ​​extracted from the original pixel deviation vector set. The second pixel deviation component set can be the set of all Y-axis deviation values ​​extracted from the original pixel deviation vector set. In practice, firstly, the X-axis component (Δx) of each vector is extracted from the original pixel deviation vector set. Finally, these X-axis components are assembled in array order to form the first pixel deviation component set (X-axis deviation set). Similarly, all Y-axis components (Δy) are extracted to form the second pixel deviation component set (Y-axis deviation set).

[0053] The fourth step is to determine the pixel-to-physical-size conversion factor based on the pre-acquired image resolution parameters. The pre-acquired image resolution parameters can be the optical resolution value set by the scanner during scanning, representing the number of pixels per inch. For example, 600 DPI (dots per inch). The pixel-to-physical-size conversion factor can be the conversion factor needed to convert pixel units to physical units such as millimeters. For example, 600 DPI corresponds to a conversion factor of 600 / 25.4 ≈ 23.62 pixels / mm. In practice, first, obtain the image resolution parameter, for example, 600 DPI. Then, according to the physical conversion formula 1 inch = 25.4 millimeters, calculate the pixel-to-physical-size conversion factor: K = DPI / 25.4 (unit: pixels / mm). Finally, output this conversion factor K as the pixel-to-physical-size conversion factor.

[0054] The fifth step involves using the aforementioned pixel-to-physical-size conversion coefficients to convert the first and second pixel deviation component sets into physical dimensions, thereby generating a pixel deviation matrix. In practice, firstly, using the pixel-to-physical-size conversion coefficient K obtained in the fourth step, each Δx value in the first pixel deviation component set is divided by K to convert it into a physical deviation Δx_mm in millimeters. Then, similarly, each Δy value in the second pixel deviation component set is divided by K to convert it into a physical deviation Δy_mm. Finally, the converted X-axis and Y-axis physical deviations are recombine into a two-dimensional deviation vector (Δx_mm, Δy_mm) and organized according to the original array structure to generate the final pixel deviation matrix (actually a physical deviation matrix).

[0055] The above-described operation steps, combined with step 107, constitute an inventive point of this disclosure, solving the technical problem mentioned in the background art: "Although ordinary scanners are low in cost, they inherently suffer from mechanical feed jitter errors, resulting in insufficient accuracy in directly measured data and making it difficult to meet the requirements of precision calibration." The reasons for this technical problem are as follows: the calibration path highly depends on the native accuracy of the hardware, lacking technical means to compensate for the inherent defects of low-cost equipment through software algorithms. This invention, by performing a structured comparison between the synthesized coordinate matrix and the reference coordinate matrix, and combining this with the scanner resolution parameters to accurately trace pixel deviations to the physical space, achieves the extraction of high-confidence full-field distortion information from noisy, low-cost equipment data. This saves on the expensive procurement and maintenance costs of dedicated measurement equipment and significantly lowers the entry barrier for high-precision calibration of galvanometer systems.

[0056] Step 106: Generate a continuous distortion calibration model based on the pixel deviation matrix.

[0057] In some embodiments, the execution entity can generate a continuous distortion calibration model based on the pixel deviation matrix. This continuous distortion calibration model can be a continuous mathematical function describing the distortion at any location within the working area of ​​the galvanometer. The continuous distortion calibration model can take any theoretical coordinate as input and output the deviation caused by the galvanometer distortion at that point, thereby achieving continuous calibration across the entire domain.

[0058] In addressing the technical problems mentioned above, and considering the application scenario: the laser galvanometer system has obtained distortion data from discrete sampling points during the calibration process, but actual processing requires continuous compensation capability at any location within the working area. This often presents the following technical challenges: discrete sampling points cannot cover the entire processing area, easily leading to abrupt compensation changes between sampling points, resulting in an uneven processing trajectory; simply increasing the number of sampling points significantly increases calibration time and cost. Given the following requirements for this application scenario: achieving high-precision compensation across the entire processing area to ensure the continuity of complex trajectories, we decided to adopt the following solution: In some optional implementations of certain embodiments, the execution entity may generate a continuous distortion calibration model based on the pixel deviation matrix, which may include the following steps: The first step is to decouple the pixel deviation matrix to generate a first deviation dataset and a second deviation dataset. This decoupling involves separating the two-dimensional deviation vector (ΔX, ΔY) in the pixel deviation matrix into independent X-direction and Y-direction deviation datasets. The first deviation dataset can be a set of all physical deviation values ​​in the X-axis direction obtained after decoupling. The second deviation dataset can be a set of all physical deviation values ​​in the Y-axis direction obtained after decoupling. In practice, firstly, all elements in the matrix are traversed, all Δx values ​​are extracted, and arranged in the original array order to generate the first deviation dataset. Then, all Δy values ​​are extracted and arranged in the original array order to generate the second deviation dataset.

[0059] The second step involves filtering the first and second deviation datasets to generate a calibration training sample set. This calibration training sample set can be high-quality deviation data and their corresponding theoretical coordinates selected for model fitting. For example, it could be a set of 200 sample points with a high signal-to-noise ratio. In practice, firstly, the first and second deviation datasets are read, and the theoretical coordinates (X, Y) for each deviation value are obtained. Then, statistical methods (six-day, box plot, 3σ criterion) are used to identify and remove outliers from the deviation values, while also removing samples from edge regions with low signal-to-noise ratios. Finally, the selected high-quality deviation data and their theoretical coordinates are combined to form the calibration training sample set.

[0060] The third step involves performing multi-order polynomial fitting on the aforementioned calibration training sample set to generate a first fitting curve and a second fitting curve. The first fitting curve can be a continuous function surface or curve fitted to the relationship between the X-axis deviation and the theoretical coordinates. The second fitting curve can be a continuous function surface or curve fitted to the relationship between the Y-axis deviation and the theoretical coordinates. In practice, firstly, the calibration training sample set is read and divided into two groups: one group is (Δx, X, Y), and the other group is (Δy, X, Y). Then, bivariate polynomial surface fitting is performed on both groups of data, starting from a low order (e.g., second order), and the polynomial coefficients are solved. Finally, a first fitting curve (actually a surface) describing the relationship between Δx and the coordinates and a second fitting curve describing the relationship between Δy and the coordinates are generated. For example, the first fitting curve could be: Where a0, a1, a2, a3, a4, and a5 are polynomial coefficients, and X and Y are coordinate values.

[0061] The fourth step involves determining the accuracy of the first and second fitted curves to generate fitting accuracy parameters. These fitting accuracy parameters can be indicators that quantify the degree of fit between the fitted curves and the original data. For example, the fitting accuracy parameter could be the root mean square error (RMSE). In practice, firstly, the theoretical coordinates from the calibration training sample set are substituted into the first and second fitted curves respectively to calculate the prediction deviation value for each sample point. Then, the prediction deviation value is compared point-by-point with the actual deviation value in the sample set to calculate the residual. Finally, various fitting accuracy parameters are statistically generated, such as the root mean square error (RMSE), the maximum residual, and the coefficient of determination (R²).

[0062] The fifth step involves iteratively optimizing the fitted curve based on the aforementioned fitting accuracy parameters to generate the optimal distortion fitting model. This optimal distortion fitting model can be the best combination of fitting function parameters that achieves the preset accuracy requirement after multiple iterations. For example, determining the optimal polynomial coefficient set [a0, a1, ..., an]. In practice, first, it is determined whether the fitting accuracy parameters generated in the fourth step reach the preset threshold (e.g., RMSE < 0.02 mm). Then, if the accuracy is not met, the polynomial order is increased (e.g., from second to third order), or the fitting algorithm is switched (e.g., adding a regularization term), and the third and fourth steps are re-executed. Finally, when the accuracy meets the requirement or the maximum number of iterations is reached, optimization is stopped, the current optimal polynomial order and coefficient combination are saved, and the optimal distortion fitting model is generated.

[0063] The sixth step involves encapsulating the optimal distortion fitting model using coordinate mapping to generate the continuous distortion calibration model. In practice, firstly, the mathematical expression and coefficient parameters of the optimal distortion fitting model are extracted. Then, according to the interface specifications of the target galvanometer control system, it is encapsulated into an executable software module. Common encapsulation forms include: C / C++ dynamic link libraries (.dll), Python function packages, JSON / XML configuration files, or embedded controller firmware. Finally, the continuous distortion calibration model is output. This model accepts any input theoretical coordinates (X, Y) and can output the required distortion compensation amount (Δx, Δy) for that point in real time. The continuous distortion calibration model can be a third-order polynomial formula (e.g., ,in, to Here are the polynomial coefficients, X and Y are the coordinates of the point, and ΔX(X,Y) is the predicted X-direction distortion compensation at the coordinate point (X,Y).

[0064] The above-described operation steps, combined with step 107, constitute an inventive point of this disclosure, solving the technical problem mentioned in the background art: "Discrete sampling points cannot cover the entire processing area, easily leading to abrupt compensation changes between sampling points, resulting in an uneven processing trajectory; simply increasing the number of sampling points will significantly increase calibration time and cost." The reasons for the above technical problems are as follows: the deviation data was not decoupled, fitted, iteratively optimized, and industrially encapsulated. This invention, by decoupling, fitting, iteratively optimizing, and industrially encapsulating the deviation data, achieves the construction of a high-precision, full-area, smooth, and continuous distortion compensation function using only a small number of sampling points. This makes the calibration results truly industrially applicable, saving the time and computational costs associated with dense sampling, while avoiding processing quality defects caused by discontinuous compensation.

[0065] Step 107: Perform calibration operation on the laser galvanometer using the continuous distortion calibration model.

[0066] In some embodiments, the aforementioned execution entity can utilize the aforementioned continuous distortion calibration model to perform calibration operations on the laser galvanometer. This calibration operation can be a process of using the continuous distortion calibration model to compensate the laser processing path in real time, ensuring that the actual processing position matches the theoretical target position.

[0067] In addressing the technical challenges of the aforementioned background technologies, and considering the application scenario—the generated continuous distortion calibration model needs to be used for online compensation during high-speed, high-precision real-time machining—and the complex and variable machining paths containing numerous straight lines and frequent corners and arcs, the following technical issues arise: industrial machining demands extremely high real-time performance (microsecond-level response), but continuous distortion calibration models are typically high-order nonlinear functions. Performing complete model calculations for every interpolation point would result in a huge computational load, leading to control cycle timeouts or stuttering. Conversely, using simple interpolation for all points would produce significant compensation lag and contour errors in areas of abrupt curvature changes. Given the following requirements for this application scenario: to significantly reduce the computational overhead of model calls while ensuring compensation accuracy, enabling the high-precision calibration algorithm to operate in real-time in conjunction with the high-speed galvanometer control system, we decided to adopt the following solution: In some optional implementations of certain embodiments, the execution entity may utilize the aforementioned continuous distortion calibration model to perform calibration operations on the laser galvanometer, which may include the following steps: The first step is to generate an initial target coordinate sequence based on the received target machining coordinate command. This initial target machining coordinate command can be a theoretical position command describing the geometric contour of the workpiece to be machined, issued by the user or a host computer. The initial target coordinate sequence can be a set of theoretical coordinate points arranged in the machining sequence, generated according to the target machining coordinate command. For example: [(0, 0), (10, 0), (20, 0)]. In practice, firstly, in response to the received target machining coordinate command, the coordinate information and motion commands of the target machining coordinate command are parsed, each theoretical machining path point is extracted, and the initial target coordinate sequence is generated according to the machining sequence.

[0068] The second step involves generating a high-priority coordinate point queue and a standard query path based on the original target coordinate sequence. The high-priority coordinate point queue can be a set of coordinate points in the processing path that require precise model calculation at critical locations such as those with large curvature or abrupt changes in direction. For example, the high-priority coordinate point queue could be the starting and ending points of corners or arcs. The standard query path can be a sequence of coordinate points in the processing path that can be simplified for straight or smooth sections. For example, the standard query path could be intermediate interpolation points on straight lines. In practice, firstly, the original target coordinate sequence is subjected to geometric feature analysis, calculating the curvature, the angle between adjacent points, and the rate of change of velocity for each point. Then, coordinate points with curvature exceeding a threshold (e.g., >0.1), angles greater than a set value (e.g., >15°), or located in acceleration / deceleration segments are marked as high-priority and stored in the high-priority coordinate point queue. Finally, the coordinate points of the remaining straight and uniform speed segments constitute the standard query path. For example, 5 corner points enter the high-priority queue, and 95 straight interpolation points enter the standard query path.

[0069] The third step involves using the high-priority coordinate point queue and the standard query path, along with a pre-defined hybrid query strategy, to invoke the continuous distortion calibration model and generate precise compensation vectors and interpolated compensation vectors in parallel. The pre-defined hybrid query strategy can be a predefined combination of precise calculations for high-priority points and interpolated calculations for standard path points. The precise compensation vector can be a high-precision compensation amount calculated by fully invoking the continuous distortion calibration model. The interpolated compensation vector can be a compensation amount quickly generated based on adjacent precise compensation points using linear or spline interpolation; for example, compensation values ​​can be allocated between two points according to their distance ratio. In practice, firstly, two parallel computing threads are started. One thread traverses the high-priority coordinate point queue, fully invoking the continuous distortion calibration model for each coordinate point to solve for higher-order functions and generate precise compensation vectors. Then, the other thread traverses the standard query path, and for standard path points between two adjacent precise compensation points, performs linear or spline interpolation on the preceding and following precise compensation vectors based on their relative position ratios to quickly generate interpolated compensation vectors.

[0070] The fourth step involves recombining and temporally aligning the precise compensation vectors and interpolation compensation vectors according to the original target coordinate sequence to generate a distortion compensation sequence synchronized with each original coordinate point. This distortion compensation sequence can be a final compensation vector sequence arranged in the original target coordinate order, corresponding one-to-one with each theoretical point. In practice, firstly, the index position and timestamp information of the original coordinate point corresponding to each compensation vector in the original target coordinate sequence are obtained. Then, all precise compensation vectors and interpolation compensation vectors are rearranged according to their index order, filling in missing positions to ensure the sequence is continuous and complete. Finally, a distortion compensation sequence that strictly corresponds one-to-one with the original target coordinate sequence and is in the exact same order is output.

[0071] The fifth step involves generating a corrected coordinate sequence based on the original target coordinate sequence and the distortion compensation sequence. This corrected coordinate sequence can be the actual processing position sequence generated by point-by-point superposition of the original target coordinates and the distortion compensation sequence. In practice, first, the original target coordinate sequence and the distortion compensation sequence are loaded, both having equal lengths and corresponding indices. Then, vector addition is performed on each position in the sequence: corrected coordinates = original target coordinates + distortion compensation amount. Finally, all corrected coordinates are organized in their original order to generate the corrected coordinate sequence.

[0072] The sixth step is to convert the corrected coordinate sequence into galvanometer control signals to generate a galvanometer drive instruction set. This instruction set can be a sequence of analog voltages or digital signals that the galvanometer controller can recognize, converted from the corrected coordinate sequence. In practice, first, each coordinate point in the corrected coordinate sequence is read, and the conversion rule is determined according to the galvanometer controller's interface protocol (e.g., ±10V analog voltage, XY2-100 digital protocol). Then, the coordinate values ​​are mapped to the corresponding voltage values, digital codes, or pulse widths. Finally, a control instruction stream, i.e., the galvanometer drive instruction set, is continuously generated according to the machining sequence. For example, an X-coordinate of 100mm corresponds to a +5V analog voltage output.

[0073] The seventh step involves sending the aforementioned galvanometer drive command set to the galvanometer control system to drive the laser focus to the corrected position and to perform material processing while the laser focus is at the corrected position, thus obtaining the processing result. This processing result can be the actual processing trajectory or morphological features formed by the laser on the workpiece after distortion compensation. In practice, firstly, the galvanometer drive command set is sent to the galvanometer servo driver in real-time at microsecond intervals via a data acquisition card, motion control card, or fieldbus. Then, the galvanometer motor responds to the command, causing the lens to deflect and driving the laser focus to move precisely to the corrected position. Finally, at the precise moment the laser focus reaches each corrected position, the laser is synchronously triggered to emit light, performing ablation, welding, or marking on the workpiece surface, ultimately forming the processing result.

[0074] Step 8: Verify the above processing results to complete the calibration operation. In practice, firstly, after processing, use the camera (CCD) integrated into the device to capture images of the processing results. Finally, if the deviation is within the tolerance range, the calibration is considered successful and a report is generated; otherwise, a prompt is made indicating that recalibration or model inspection is required.

[0075] The above-described operation steps, as an inventive point of this disclosure, solve the technical problem mentioned in the background art: "Industrial processing sites have extremely high real-time requirements (microsecond-level response), but continuous distortion calibration models are usually high-order nonlinear functions. If a complete model calculation is performed for each interpolation point, it will cause a huge computational load, leading to control cycle timeouts or stuttering." The reason for the above technical problem is as follows: Industrial processing sites have extremely high real-time requirements (microsecond-level response), but continuous distortion calibration models are usually high-order nonlinear functions. If a complete model calculation is performed for each interpolation point, it will cause a huge computational load, leading to control cycle timeouts or stuttering. The inventive point of this invention, by adopting a hybrid parallel query strategy, combines the pre-calculation and real-time calculation of calibration compensation, significantly reducing query latency, saving high hardware upgrade costs and development cycles, and completely eliminating processing contour errors, corner overshoot, and efficiency losses caused by calculation latency.

[0076] The various embodiments disclosed above have the following beneficial effects: The laser galvanometer distortion calibration method of some embodiments of this disclosure can utilize a common scanner to calibrate the laser galvanometer distortion, thereby significantly reducing equipment investment costs. Specifically, the high equipment investment cost is due to the fact that traditional calibration methods must rely on expensive dedicated measuring equipment (e.g., a two-dimensional image measuring instrument). Based on this, the image segmentation method of some embodiments of this disclosure first performs a dual-pose scan on the target plate to generate a first image set and a second image set. The dual-pose scan includes a first orientation pose and a second orientation pose. By scanning in two orthogonal directions, the differences in the accuracy characteristics of the scanner in different motion directions are actively captured, providing "raw material" containing directional error features. This is a fundamental data acquisition strategy to overcome the insufficient accuracy of a single scan and achieve low-cost substitution. Then, based on the first image set and the second image set, a first set of circle center pixel coordinates and a second set of circle center pixel coordinates are generated. The visual features (circles) in the first and second image sets are accurately quantized into computer-processable coordinate data, completing the key transformation from analog images to digital coordinates, providing structured input for subsequent quantitative analysis, comparison, and calculation. Next, a discreteness analysis is performed on the first and second circle-center pixel coordinate sets to generate axial label information. This discreteness analysis and axial label generation identifies coordinate stability and abnormal distributions, distinguishing reliable data from outliers and providing a basis for subsequent screening and fusion. Then, based on the axial label information, the first and second circle-center pixel coordinate sets are selectively fused to construct a synthetic coordinate matrix. This selective fusion of coordinates based on the axial label information preserves high-reliability data, eliminates outliers, and improves overall coordinate accuracy and consistency. Next, a pixel deviation matrix is ​​generated based on the synthetic coordinate matrix and the reference coordinate matrix. Finally, a continuous distortion calibration model is generated based on the pixel deviation matrix. Discrete, measurement-point-limited deviation data is transformed into a continuous function model applicable to the entire working area through mathematical fitting (e.g., polynomial surface fitting). This model can predict distortions at any unsampled point within the working area, providing a direct tool for implementing smooth, seamless spatial correction. Finally, the aforementioned continuous distortion calibration model is used to perform calibration operations on the laser galvanometer. This allows for real-time, automatic distortion compensation for any target coordinates sent to the galvanometer, ensuring the laser focus accurately reaches the theoretical position. This achieves laser galvanometer distortion calibration and significantly reduces equipment investment costs.

[0077] Further reference Figure 2 As an implementation of the methods shown in the above figures, this disclosure provides some embodiments of a laser galvanometer distortion calibration device, which are similar to... Figure 1 Corresponding to the method embodiments shown, this laser galvanometer distortion calibration device can be specifically applied to various electronic devices.

[0078] like Figure 2 As shown, a laser galvanometer distortion calibration device 200 includes: an execution unit 201, a first generation unit 202, a discreteness analysis unit 203, a fusion unit 204, a second generation unit 205, a third generation unit 206, and a calibration unit 207. The execution unit 201 is configured to perform a dual-pose scan on a target plate to generate a first image set and a second image set, wherein the dual-pose scan includes a first orientation pose and a second orientation pose. The first generation unit 202 is configured to generate a first center pixel coordinate set and a second center pixel coordinate set based on the first image set and the second image set. The discreteness analysis unit 203 is configured to perform discreteness analysis on the first center pixel coordinate set and the second center pixel coordinate set to generate axial label information. The fusion unit 204 is configured to selectively fuse the first center pixel coordinate set and the second center pixel coordinate set based on the axial label information to construct a synthetic coordinate matrix. The second generation unit 205 is configured to generate a pixel deviation matrix based on the aforementioned synthesized coordinate matrix and reference coordinate matrix. The third generation unit 206 is configured to generate a continuous distortion calibration model based on the aforementioned pixel deviation matrix. The calibration unit 207 is configured to perform a calibration operation on the laser galvanometer using the aforementioned continuous distortion calibration model.

[0079] It is understandable that the units described in the laser galvanometer distortion calibration device 200 are related to the reference. Figure 1 The steps in the described method correspond accordingly. Therefore, the operations, features, and beneficial effects described above for the method also apply to the laser galvanometer distortion calibration device 200 and the units contained therein, and will not be repeated here.

[0080] The following is for reference. Figure 3 It shows a schematic diagram of the structure of an electronic device (e.g., an electronic device) 300 suitable for implementing some embodiments of the present disclosure. Figure 3 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments of this disclosure.

[0081] like Figure 3As shown, the electronic device 300 may include a processing unit (e.g., a central processing unit, a graphics processing unit, etc.) 301, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 302 or a program loaded from a storage device 308 into a random access memory (RAM) 303. The RAM 303 also stores various programs and data required for the operation of the electronic device 300. The processing unit 301, ROM 302, and RAM 303 are interconnected via a bus 304. An input / output (I / O) interface 305 is also connected to the bus 304.

[0082] Typically, the following devices can be connected to I / O interface 305: input devices 306 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 307 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 308 including, for example, magnetic tapes, hard disks, etc.; and communication devices 309. Communication device 309 allows electronic device 300 to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 3 An electronic device 300 with various devices is shown; however, it should be understood that it is not required to implement or possess all of the devices shown. More or fewer devices may be implemented or possessed alternatively. Figure 3 Each box shown can represent a device or multiple devices as needed.

[0083] In particular, according to some embodiments of this disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, some embodiments of this disclosure include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication device 309, or installed from storage device 308, or installed from ROM 302. When the computer program is executed by processing device 301, it performs the functions defined in the methods of some embodiments of this disclosure.

[0084] Figure 4 This indicates the error between the measured value of the laser galvanometer before calibration and the theoretical value, obtained using a two-dimensional measuring device. The upper left corner shows an error of 2.24 mm before calibration.

[0085] Figure 5 This indicates the error between the measured value of the laser galvanometer after calibration, obtained using a two-dimensional measuring device, and the theoretical value. The upper left corner shows a calibration error of 0.069 mm, verifying that the present invention can achieve calibration accuracy comparable to that of dedicated measuring equipment using a common scanner.

[0086] It should be noted that, in some embodiments of this disclosure, the computer-readable medium described above may be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In some embodiments of this disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In some embodiments of this disclosure, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wires, optical fibers, RF (radio frequency), etc., or any suitable combination thereof.

[0087] In some implementations, clients and servers can communicate using any currently known or future-developed network protocol such as HTTP (Hypertext Transfer Protocol) and can interconnect with digital data communication (e.g., communication networks) of any form or medium. Examples of communication networks include local area networks (“LANs”), wide area networks (“WANs”), the Internet (e.g., the Internet of Things), and peer-to-peer networks (e.g., ad hoc peer-to-peer networks), as well as any currently known or future-developed networks.

[0088] The aforementioned computer-readable medium may be included in the aforementioned electronic device; or it may exist independently and not assembled into the electronic device. The aforementioned computer-readable medium carries one or more programs that, when executed by the electronic device, cause the electronic device to: perform a dual-pose scan on the target plate to generate a first image set and a second image set, the dual-pose scan including a first orientation pose and a second orientation pose; generate a first center pixel coordinate set and a second center pixel coordinate set based on the first image set and the second image set; perform discrete analysis on the first center pixel coordinate set and the second center pixel coordinate set to generate axial label information; selectively fuse the first center pixel coordinate set and the second center pixel coordinate set based on the axial label information to construct a synthetic coordinate matrix; generate a pixel deviation matrix based on the synthetic coordinate matrix and the reference coordinate matrix; generate a continuous distortion calibration model based on the pixel deviation matrix; and perform a calibration operation on the laser galvanometer using the continuous distortion calibration model.

[0089] Computer program code for performing operations of some embodiments of this disclosure can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0090] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0091] The units described in some embodiments of this disclosure can be implemented in software or hardware. The described units can also be housed in a processor; for example, a processor may be described as including an execution unit, a first generation unit, a discrete analysis unit, a fusion unit, a second generation unit, a third generation unit, and a calibration unit. The names of these units do not necessarily limit the specific unit; for example, the execution unit may also be described as "a unit that performs dual-pose scanning on a target plate to generate a first image set and a second image set."

[0092] The functions described above in this document can be performed at least in part by one or more hardware logic components. For example, exemplary types of hardware logic components that can be used, without limitation, include: field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), system-on-a-chip (SoCs), complex programmable logic devices (CPLDs), and so on.

[0093] The above description is merely a selection of preferred embodiments of this disclosure and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of the invention involved in the embodiments of this disclosure is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-described inventive concept. For example, technical solutions formed by substituting the above-described features with (but not limited to) technical features with similar functions disclosed in the embodiments of this disclosure.

Claims

1. A method for calibrating laser galvanometer distortion, comprising: A dual-attitude scan is performed on the target plate to generate a first image set and a second image set, wherein the dual-attitude scan includes a first orientation attitude and a second orientation attitude. Based on the first image set and the second image set, a first set of center pixel coordinates and a second set of center pixel coordinates are generated; Discreteness analysis is performed on the first set of center pixel coordinates and the second set of center pixel coordinates to generate axial label information; Based on the axial label information, the first set of center pixel coordinates and the second set of center pixel coordinates are selectively fused to construct a composite coordinate matrix; Based on the synthesized coordinate matrix and the reference coordinate matrix, a pixel deviation matrix is ​​generated; Based on the pixel deviation matrix, a continuous distortion calibration model is generated; The laser galvanometer is calibrated using the continuous distortion calibration model.

2. The method according to claim 1, wherein, The process of performing dual-pose scanning on the target plate to generate a first image set and a second image set includes: Using a target scanner, the target plate in the first orientation position is scanned to generate a first image set; Using a target scanner, the target plate in the second orientation is scanned to generate a second image set.

3. The method according to claim 1, wherein, The step of generating a first set of center pixel coordinates and a second set of center pixel coordinates based on the first image set and the second image set includes: Reference point coordinates are generated by identifying reference marker points in each image of the first image set and the second image set. Based on the reference point coordinate set, pose unification is performed on each image in the first image set and the second image set to generate a pose-unified image set. The center coordinates of each pose-unified image in the pose-unified image set are extracted and the coordinate system is normalized to generate a first set of center pixel coordinates and a second set of center pixel coordinates.

4. The method according to claim 1, wherein, The step of performing discrete analysis on the first set of center pixel coordinates and the second set of center pixel coordinates to generate axial label information includes: The row dispersion of each coordinate in the first set of center pixel coordinates and the second set of center pixel coordinates is determined to generate the first dispersion data; The column-oriented discreteness of each coordinate in the first set of center pixel coordinates and the second set of center pixel coordinates is determined to generate the second discreteness data. The first discrete data and the second discrete data are compared to generate a discrete comparison result; Based on the discreteness comparison results, the effective axis determination is performed on the first circle center pixel coordinate set to generate the first axis determination identification information; Based on the discreteness comparison results, the effective axis determination is performed on the second circle center pixel coordinate set to generate second axis determination identification information; The first axial determination identifier information and the second axial determination identifier information are integrated and encoded to generate the axial label information.

5. The method according to claim 1, wherein, The selective fusion of the first set of center pixel coordinates and the second set of center pixel coordinates based on the axial label information to construct a synthetic coordinate matrix includes: Based on the axial label information, generate first effective axial information for the first set of center pixel coordinates; Based on the first valid axial information, coordinate data of the corresponding axial direction are extracted from the first set of center pixel coordinates to generate a first valid coordinate subset; Based on the axial label information, generate second effective axial information for the second set of center pixel coordinates; Based on the second effective axial information, coordinate data of the corresponding axial direction are extracted from the second set of center pixel coordinates to generate a second effective coordinate subset; The first effective axial information and the second effective axial information are paired and combined to generate a preliminary fused coordinate matrix; The preliminary fused coordinate matrix is ​​centered to construct the composite coordinate matrix.

6. A laser galvanometer distortion calibration device, comprising: The execution unit is configured to perform a dual-pose scan on the target plate to generate a first image set and a second image set, the dual-pose scan including: a first orientation pose and a second orientation pose; The first generation unit is configured to generate a first set of center pixel coordinates and a second set of center pixel coordinates based on the first image set and the second image set. The discreteness analysis unit is configured to perform discreteness analysis on the first set of center pixel coordinates and the second set of center pixel coordinates to generate axial label information; The fusion unit is configured to selectively fuse the first set of center pixel coordinates and the second set of center pixel coordinates based on the axial label information to construct a synthetic coordinate matrix; The second generation unit is configured to generate a pixel deviation matrix based on the synthesized coordinate matrix and the reference coordinate matrix; The third generation unit is configured to generate a continuous distortion calibration model based on the pixel deviation matrix; The calibration unit is configured to perform calibration operations on the laser galvanometer using the continuous distortion calibration model.

7. An electronic device, comprising: One or more processors; Storage device, on which one or more programs are stored, When the one or more programs are executed by the one or more processors, the one or more processors implement the method as described in any one of claims 1-5.

8. A computer-readable medium having a computer program stored thereon, wherein, When the program is executed by the processor, it implements the method as described in any one of claims 1-5.

Images