A method and system for controlling a cutting path of a panel pattern

Abstract

The application discloses a kind of plate pattern cutting path control method and system, the method includes obtaining plate pattern image;Plate pattern image is preprocessed, and first plate pattern image is obtained;Select the pixel point of color gradient change in first plate pattern image, connect each pixel point with each other, form first contour;Curve fitting and smoothing processing are carried out to first contour, and second contour is obtained;Excess line in second contour is rejected, and third contour is obtained;Whether third contour exceeds the width of cutting machine worktable is judged, if exceed, then remove the part of excess, obtain fourth contour, convert each pixel point coordinate of fourth contour into the world coordinate of cutting machine, generate cutting path;According to the control instruction of cutting path generation to cutting machine.The application solves the problem that the contour extracted is not enough in the traditional cutting path planning, which leads to the deviation and inaccuracy of the generated cutting path.

Description

Technical Field

[0001] This invention relates to the field of sheet metal pattern cutting technology, and in particular to a method and system for controlling the cutting path of sheet metal patterns. Background Technology

[0002] Cutting patterns onto sheet metal using a cutting machine is a crucial step in modern manufacturing. Typically, this requires planning the cutting path, which involves acquiring an image of the sheet metal pattern and preprocessing it to enhance contrast and facilitate contour extraction. However, traditional cutting path planning often fails to adequately process the extracted contours. For example, unnecessary lines or portions of the contour extending beyond the cutting machine's worktable are not removed, leading to inaccurate and deviated cutting paths. Summary of the Invention

[0003] To address the aforementioned shortcomings, this invention proposes a method and system for controlling the cutting path of sheet metal patterns. The purpose is to solve the problem that insufficient processing of the extracted contours in traditional cutting path planning leads to deviations and inaccuracies in the generated cutting paths.

[0004] To achieve this objective, the present invention adopts the following technical solution:

[0005] A method for controlling the cutting path of a pattern on a sheet metal board includes the following steps:

[0006] Step S1: Obtain the pattern image of the board material;

[0007] Step S2: Preprocess the board pattern image to obtain the first board pattern image;

[0008] Step S3: Select pixels in the first board pattern image where the color has a gradient change, and connect each pixel to form a first outline;

[0009] Step S4: Perform curve fitting and smoothing on the first contour to obtain the second contour;

[0010] Step S5: Remove the redundant lines from the second contour to obtain the third contour;

[0011] Step S6: Determine whether the third contour exceeds the working surface of the cutting machine. If it does, remove the excess part to obtain the fourth contour, and convert the coordinates of each pixel of the fourth contour into the world coordinates of the cutting machine to generate the cutting path; if it does not exceed the working surface, convert the coordinates of each pixel of the third contour into the world coordinates of the cutting machine to generate the cutting path.

[0012] Step S7: Generate control instructions for the cutting machine based on the cutting path.

[0013] Preferably, step S2 specifically includes the following sub-steps:

[0014] Step S21: Perform single-channel processing on the three channels of the board pattern image to obtain a grayscale image;

[0015] Step S22: Set the grayscale threshold of the grayscale image, adjust the grayscale values ​​of pixels with grayscale values ​​greater than the grayscale threshold to the maximum grayscale value, and adjust the grayscale values ​​of pixels with grayscale values ​​less than the grayscale threshold to the minimum grayscale value to obtain a binary image.

[0016] Preferably, step S3 specifically includes the following sub-steps:

[0017] Step S31: Scan each pixel in the first board pattern image, and replace the value of the center pixel of the Gaussian template with the weighted average gray value of the pixels in the neighborhood determined by the convolution kernel to obtain the first board pattern image after Gaussian filtering.

[0018] Step S32: Calculate the gradient of the first plate pattern image after Gaussian filtering;

[0019] Step S33: Interpolate by calculating the quadratic function of the gradient magnitude at three adjacent points along the gradient direction, that is, fit the quadratic equation using three coordinate points and obtain the compensation value.

[0020] Step S34: Add a compensation value to the midpoint of three adjacent points along the gradient direction to obtain the sub-pixel edge point;

[0021] Step S35: Connect the sub-pixel edge points to form the first contour.

[0022] Preferably, in step S31, the Gaussian template is calculated using a Gaussian function, the formula of which is as follows:

[0023]

[0024] Where G(x, y) represents x and y as the pixel coordinates of the Gaussian template at various positions in the image; x represents the x-coordinate in the coordinate system; y represents the y-coordinate in the coordinate system; and σ represents the standard deviation in the xy direction in the coordinate system.

[0025] Preferably, in step S6, the coordinate transformation specifically includes the following sub-steps:

[0026] Step S61: Generate camera parameters through camera calibration;

[0027] Step S62: Obtain the transformation matrix for the transformation based on the camera parameters;

[0028] Step S63: Convert pixel coordinates to world coordinates according to the transformation matrix.

[0029] Another aspect of this application provides a control system for a pattern cutting path of a sheet metal, the system comprising:

[0030] The acquisition module is used to acquire images of the board pattern.

[0031] The first processing module is used to preprocess the board pattern image to obtain the first board pattern image;

[0032] The contour forming module is used to select pixels with gradient color changes in the first board pattern image and connect each pixel to form a first contour.

[0033] The second processing module is used to perform curve fitting and smoothing on the first contour to obtain the second contour.

[0034] The removal module is used to remove redundant lines from the second contour to obtain the third contour.

[0035] The judgment module is used to determine whether the third contour exceeds the working table of the cutting machine. If it does, the excess part is removed to obtain the fourth contour, and the coordinates of each pixel point of the fourth contour are converted into the world coordinates of the cutting machine to generate the cutting path. If it does not exceed the working table, the coordinates of each pixel point of the third contour are converted into the world coordinates of the cutting machine to generate the cutting path.

[0036] The control command generation module is used to generate control commands for the cutting machine based on the cutting path.

[0037] Preferably, the first processing module includes:

[0038] The grayscale processing submodule is used to perform single-channel processing on the three channels of the board pattern image to obtain a grayscale image.

[0039] The binarization processing submodule is used to set the grayscale threshold of the grayscale image, adjust the grayscale values ​​of pixels with grayscale values ​​greater than the grayscale threshold to the maximum grayscale value, and adjust the grayscale values ​​of pixels with grayscale values ​​less than the grayscale threshold to the minimum grayscale value, thus obtaining a binary image.

[0040] Preferably, the contour forming module includes:

[0041] The scanning submodule is used to scan each pixel in the first board pattern image;

[0042] The substitution submodule is used to replace the value of the center pixel of the Gaussian template with the weighted average gray value of the pixels in the neighborhood determined by the convolution kernel, so as to obtain the first plate pattern image after Gaussian filtering.

[0043] The Gaussian template is calculated using the Gaussian function, the formula of which is as follows:

[0044]

[0045] Where G(x, y) represents x and y as the pixel coordinates of the Gaussian template at various positions in the image; x represents the x-coordinate in the coordinate system; y represents the y-coordinate in the coordinate system; and σ represents the standard deviation in the xy direction in the coordinate system.

[0046] The first calculation submodule is used to calculate the gradient of the first plate pattern image after Gaussian filtering.

[0047] The second calculation submodule interpolates the gradient magnitude at three adjacent points along the gradient direction using a quadratic function, that is, it fits a quadratic equation using three coordinate points and calculates the compensation value.

[0048] The third calculation submodule adds a compensation value to the midpoint of three adjacent points along the gradient direction to obtain the sub-pixel edge point.

[0049] The connection submodule is used to connect subpixel edge points to form a first contour.

[0050] Preferably, the judgment module includes a coordinate transformation submodule, which includes:

[0051] The generation subunit is used to generate camera parameters through camera calibration;

[0052] The sub-unit is used to obtain the transformation matrix during the transformation based on the camera parameters.

[0053] The transformation subunit is used to convert pixel coordinates into world coordinates based on the transformation matrix.

[0054] The technical solutions provided in this application embodiment may include the following beneficial effects:

[0055] This solution extracts a first contour from a pre-processed image of the board pattern. A second contour is obtained by curve fitting and smoothing the first contour. Excess lines in the second contour are removed to obtain a third contour, at which point the board pattern recognition is complete. Finally, it is determined whether the recognized board pattern exceeds the cutting machine's worktable. If it does, the excess portion is removed to obtain a fourth contour. The cutting path is then derived based on the fourth contour; otherwise, it is derived based on the third contour. This solution effectively avoids cutting path deviations by removing excess lines and excess portions from the extracted contours, resulting in a more accurate cutting path. Attached Figure Description

[0056] Figure 1This is a flowchart illustrating the steps involved in controlling the cutting path of a pattern on a sheet metal board. Detailed Implementation

[0057] Embodiments of the present invention are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0058] A method for controlling the cutting path of a pattern on a sheet metal board includes the following steps:

[0059] Step S1: Obtain the pattern image of the board material;

[0060] Step S2: Preprocess the board pattern image to obtain the first board pattern image;

[0061] Step S3: Select pixels in the first board pattern image where the color has a gradient change, and connect each pixel to form a first outline;

[0062] Step S4: Perform curve fitting and smoothing on the first contour to obtain the second contour;

[0063] Step S5: Remove the redundant lines from the second contour to obtain the third contour;

[0064] Step S6: Determine whether the third contour exceeds the working surface of the cutting machine. If it does, remove the excess part to obtain the fourth contour, and convert the coordinates of each pixel of the fourth contour into the world coordinates of the cutting machine to generate the cutting path; if it does not exceed the working surface, convert the coordinates of each pixel of the third contour into the world coordinates of the cutting machine to generate the cutting path.

[0065] Step S7: Generate control instructions for the cutting machine based on the cutting path.

[0066] This solution provides a method for controlling the cutting path of a pattern on a sheet metal substrate, such as... Figure 1As shown, the first step is to acquire the board pattern image. Specifically, a camera is set up within the working area of ​​the cutting machine, directly above the cutting machine's worktable. The board to be cut is placed on the cutting machine's worktable, and the camera acquires the board pattern image. The second step is to preprocess the board pattern image to obtain a first board pattern image. Specifically, the preprocessing includes grayscale processing, color processing, and binarization processing. Preprocessing the board pattern image enhances the image contrast. The third step is to select pixels in the first board pattern image with gradient color changes and connect each pixel to form a first contour. Specifically, extracting the first contour facilitates the generation of the subsequent cutting path. The fourth step is to perform curve fitting and smoothing processing on the first contour to obtain a second contour. Specifically, curve fitting processing on the first contour helps to make the extraction of the board pattern contour more accurate and reduce errors; smoothing processing on the first contour helps to make the contour lines smoother. The fifth step is to remove redundant lines from the second contour to obtain a third contour. Specifically, since gradient changes cause edge pixels, edge pixels are removed using a range-area segmentation method to obtain the third contour, completing the recognition of the board pattern. The sixth step is to determine whether the third contour exceeds the cutting machine's worktable. If it does, the excess portion is removed to obtain the fourth contour, and the coordinates of each pixel of the fourth contour are converted to the cutting machine's world coordinates to generate a cutting path. If it does not exceed the worktable's size, the coordinates of each pixel of the third contour are converted to the cutting machine's world coordinates to generate a cutting path. Specifically, since the cutting machine's worktable has a known size, but the size of the material to be cut is not uniform, when the size of the material to be cut is larger than the cutting machine's worktable, some parts of the identified material pattern's contour may not be within the cutting machine's cutting range. Therefore, the excess portion needs to be removed. The fourth contour obtained after removal is the actual cutting path that the cutting machine needs to cut, and this path will be more accurate. The seventh step is to generate control commands for the cutting machine based on the cutting path. Specifically, the cutting machine performs cutting processing on the material to be cut according to the control commands. In this embodiment, the cutting processing is laser cutting.

[0067] This solution extracts a first contour from a pre-processed image of the board pattern. A second contour is obtained by curve fitting and smoothing the first contour. Excess lines in the second contour are removed to obtain a third contour, at which point the board pattern recognition is complete. Finally, it is determined whether the recognized board pattern exceeds the cutting machine's worktable. If it does, the excess portion is removed to obtain a fourth contour. The cutting path is then derived based on the fourth contour; otherwise, it is derived based on the third contour. This solution effectively avoids cutting path deviations by removing excess lines and excess portions from the extracted contours, resulting in a more accurate cutting path.

[0068] Preferably, step S2 specifically includes the following sub-steps:

[0069] Step S21: Perform single-channel processing on the three channels of the board pattern image to obtain a grayscale image;

[0070] Step S22: Set the grayscale threshold of the grayscale image, adjust the grayscale values ​​of pixels with grayscale values ​​greater than the grayscale threshold to the maximum grayscale value, and adjust the grayscale values ​​of pixels with grayscale values ​​less than the grayscale threshold to the minimum grayscale value to obtain a binary image.

[0071] In this embodiment, the board pattern image is a color image with RGB three channels. Each pixel of the color image is composed of three components. To reduce the computational load and interference during image processing, the RGB three channels are converted into single-channel processing, transforming the color relationship into a brightness relationship, thereby obtaining a grayscale image. Each pixel in the grayscale image is determined by only one component. By calculating the grayscale value of each pixel in the grayscale image, the grayscale value is generally in the range of 0-255, where a grayscale value of 0 represents black and a grayscale value of 255 represents white. In one embodiment, the grayscale threshold of the grayscale image is set to 150. The grayscale values ​​of pixels with a grayscale value greater than 150 are adjusted to 255, and the grayscale values ​​of pixels with a grayscale value less than 150 are adjusted to 0, thereby obtaining a binary image. The binary image has a small amount of data, thus highlighting the outline of the target.

[0072] Preferably, step S3 specifically includes the following sub-steps:

[0073] Step S31: Scan each pixel in the first board pattern image, and replace the value of the center pixel of the Gaussian template with the weighted average gray value of the pixels in the neighborhood determined by the convolution kernel to obtain the first board pattern image after Gaussian filtering.

[0074] Step S32: Calculate the gradient of the first plate pattern image after Gaussian filtering;

[0075] Step S33: Interpolate by calculating the quadratic function of the gradient magnitude at three adjacent points along the gradient direction, that is, fit the quadratic equation using three coordinate points and obtain the compensation value.

[0076] Step S34: Add a compensation value to the midpoint of three adjacent points along the gradient direction to obtain the sub-pixel edge point;

[0077] Step S35: Connect the sub-pixel edge points to form the first contour.

[0078] In this embodiment, after preprocessing the board pattern image, the boundary color difference in the first board pattern image is already quite obvious, but there are still pixel jagged edges and noise. In order to make the outline edge of the board pattern that is identified smoother, Gaussian blur needs to be added. Gaussian blur can effectively reduce the noise of the image.

[0079] This embodiment uses the Canny edge detection algorithm to detect the edges of the pattern contour in the first plate pattern image after Gaussian filtering. However, the edges are at the pixel level, which has low accuracy and is prone to jagged edges. Instead of using the non-maximum suppression method in the Canny edge detection algorithm, the quadratic function interpolation of the gradient magnitude at three adjacent points in the gradient direction is used. Then, the interpolation intermediate values ​​are used to correct the jittery pattern contour edge lines from the horizontal and vertical directions, so that the edge lines of the pattern contour fit better.

[0080] Preferably, in step S31, the Gaussian template is calculated using a Gaussian function, the formula of which is as follows:

[0081]

[0082] Where G(x, y) represents x and y as the pixel coordinates of the Gaussian template at various positions in the image; x represents the x-coordinate in the coordinate system; y represents the y-coordinate in the coordinate system; and σ represents the standard deviation in the xy direction in the coordinate system.

[0083] In this embodiment, a 3×3 Gaussian template is used as an example. Sampling is performed with the center of the Gaussian template as the origin, and the value of the pixel at that position is replaced by the weighted average gray value of the pixels in the neighborhood determined by the convolution kernel. This achieves the effect of Gaussian filtering. Further explanation: if σ is small, the coefficients at the center of the generated template are larger, while the coefficients around it are smaller, resulting in a less noticeable smoothing effect on the image. Conversely, when σ is large, the coefficients of the generated template are not significantly different, similar to a mean template, leading to a more pronounced smoothing effect on the image.

[0084] Preferably, in step S6, the coordinate transformation specifically includes the following sub-steps:

[0085] Step S61: Generate camera parameters through camera calibration;

[0086] Step S62: Obtain the transformation matrix for the transformation based on the camera parameters;

[0087] Step S63: Convert pixel coordinates to world coordinates according to the transformation matrix.

[0088] In this embodiment, camera parameters include intrinsic and extrinsic parameters. Extrinsic parameters include a rotation matrix and a translation vector, which together describe the process of transforming coordinate points from the world coordinate system to the camera coordinate system. The rotation matrix describes the direction of the coordinate axes of the world coordinate system relative to the coordinate axes of the camera coordinate system, and the translation vector describes the position of the origin in the camera coordinate system. The transformation from pixel coordinates to world coordinates using the transformation matrix specifically includes the following steps: first, using intrinsic parameter transformation to convert the pixel coordinates to camera coordinates; finally, using extrinsic parameter transformation to convert the camera coordinates to world coordinates. This solution transforms the coordinates of each pixel point of the contour into the world coordinates of the cutting machine, enabling the cutting machine to accurately cut the board pattern according to the cutting path.

[0089] Another aspect of this application provides a control system for a pattern cutting path of a sheet metal, the system comprising:

[0090] The acquisition module is used to acquire images of the board pattern.

[0091] The first processing module is used to preprocess the board pattern image to obtain the first board pattern image;

[0092] The contour forming module is used to select pixels with gradient color changes in the first board pattern image and connect each pixel to form a first contour.

[0093] The second processing module is used to perform curve fitting and smoothing on the first contour to obtain the second contour.

[0094] The removal module is used to remove redundant lines from the second contour to obtain the third contour.

[0095] The judgment module is used to determine whether the third contour exceeds the working table of the cutting machine. If it does, the excess part is removed to obtain the fourth contour, and the coordinates of each pixel point of the fourth contour are converted into the world coordinates of the cutting machine to generate the cutting path. If it does not exceed the working table, the coordinates of each pixel point of the third contour are converted into the world coordinates of the cutting machine to generate the cutting path.

[0096] The control command generation module is used to generate control commands for the cutting machine based on the cutting path.

[0097] This solution provides a control system for cutting patterns on sheet metal. Through the cooperation of an acquisition module, a first processing module, a contour forming module, a second processing module, a rejection module, a judgment module, and a control command generation module, the system achieves the cutting of sheet metal patterns.

[0098] This solution extracts a first contour from a pre-processed image of the board pattern. A second contour is obtained by curve fitting and smoothing the first contour. Excess lines in the second contour are removed to obtain a third contour, at which point the board pattern recognition is complete. Finally, it is determined whether the recognized board pattern exceeds the cutting machine's worktable. If it does, the excess portion is removed to obtain a fourth contour. The cutting path is then derived based on the fourth contour; otherwise, it is derived based on the third contour. This solution effectively avoids cutting path deviations by removing excess lines and excess portions from the extracted contours, resulting in a more accurate cutting path.

[0099] Preferably, the first processing module includes:

[0100] The grayscale processing submodule is used to perform single-channel processing on the three channels of the board pattern image to obtain a grayscale image.

[0101] The binarization processing submodule is used to set the grayscale threshold of the grayscale image, adjust the grayscale values ​​of pixels with grayscale values ​​greater than the grayscale threshold to the maximum grayscale value, and adjust the grayscale values ​​of pixels with grayscale values ​​less than the grayscale threshold to the minimum grayscale value, thus obtaining a binary image.

[0102] In this embodiment, a grayscale image is obtained through a grayscale processing submodule. Acquiring a grayscale image reduces the computational load and interference in the image processing process. A binary image is obtained through a binarization processing submodule. The binary image contains less data, thus highlighting the outline of the target.

[0103] Preferably, the contour forming module includes:

[0104] The scanning submodule is used to scan each pixel in the first board pattern image;

[0105] The substitution submodule is used to replace the value of the center pixel of the Gaussian template with the weighted average gray value of the pixels in the neighborhood determined by the convolution kernel, so as to obtain the first plate pattern image after Gaussian filtering.

[0106] The Gaussian template is calculated using the Gaussian function, the formula of which is as follows:

[0107]

[0108] Where G(x, y) represents x and y as the pixel coordinates of the Gaussian template at various positions in the image; x represents the x-coordinate in the coordinate system; y represents the y-coordinate in the coordinate system; and σ represents the standard deviation in the xy direction in the coordinate system.

[0109] The first calculation submodule is used to calculate the gradient of the first plate pattern image after Gaussian filtering.

[0110] The second calculation submodule interpolates the gradient magnitude at three adjacent points along the gradient direction using a quadratic function, that is, it fits a quadratic equation using three coordinate points and calculates the compensation value.

[0111] The third calculation submodule adds a compensation value to the midpoint of three adjacent points along the gradient direction to obtain the sub-pixel edge point.

[0112] The connection submodule is used to connect subpixel edge points to form a first contour.

[0113] In this embodiment, the extraction of the first contour from the first board pattern image is achieved through the cooperation of the scanning submodule, the substitution submodule, the first calculation submodule, the second calculation submodule, the third calculation submodule, and the connection submodule.

[0114] Specifically, after preprocessing the board pattern image, the boundary color difference in the first board pattern image is already quite obvious, but there are still pixel jagged edges and noise. In order to make the outline edge of the board pattern that is identified smoother, Gaussian blur needs to be added. Gaussian blur can effectively reduce the noise of the image.

[0115] This embodiment uses the Canny edge detection algorithm to detect the edges of the pattern contour in the first plate pattern image after Gaussian filtering. However, the edges are at the pixel level, which has low accuracy and is prone to jagged edges. Instead of using the non-maximum suppression method in the Canny edge detection algorithm, the quadratic function interpolation of the gradient magnitude at three adjacent points in the gradient direction is used. Then, the interpolation intermediate values ​​are used to correct the jittery pattern contour edge lines from the horizontal and vertical directions, so that the edge lines of the pattern contour fit better.

[0116] Preferably, the judgment module includes a coordinate transformation submodule, which includes:

[0117] The generation subunit is used to generate camera parameters through camera calibration;

[0118] The sub-unit is used to obtain the transformation matrix during the transformation based on the camera parameters.

[0119] The transformation subunit is used to convert pixel coordinates into world coordinates based on the transformation matrix.

[0120] In this embodiment, by the cooperation of the generation sub-unit, the acquisition sub-unit, and the transformation sub-unit, the coordinates of each pixel point of the outline are converted into the world coordinates of the cutting machine, so that the cutting machine can accurately cut the board pattern according to the cutting path.

[0121] Furthermore, the functional units in the various embodiments of the present invention can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0122] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A method for controlling the cutting path of a pattern on a sheet metal substrate, characterized in that: Includes the following steps: Step S1: Obtain the pattern image of the board material; Step S2: Preprocess the board pattern image to obtain the first board pattern image; Step S3: Select pixels in the first board pattern image where the color has a gradient change, and connect each pixel to form a first outline; Step S4: Perform curve fitting and smoothing on the first contour to obtain the second contour; Step S5: Remove the redundant lines from the second contour to obtain the third contour; Step S6: Determine whether the third contour exceeds the working surface of the cutting machine. If it does, remove the excess part to obtain the fourth contour, and convert the coordinates of each pixel of the fourth contour into the world coordinates of the cutting machine to generate the cutting path; if it does not exceed the working surface, convert the coordinates of each pixel of the third contour into the world coordinates of the cutting machine to generate the cutting path. Step S7: Generate control commands for the cutting machine based on the cutting path; Step S3 specifically includes the following sub-steps: Step S31: Scan each pixel in the first board pattern image, and replace the value of the center pixel of the Gaussian template with the weighted average gray value of the pixels in the neighborhood determined by the convolution kernel to obtain the first board pattern image after Gaussian filtering. Step S32: Calculate the gradient of the first plate pattern image after Gaussian filtering; Step S33: Interpolate by calculating the quadratic function of the gradient magnitude at three adjacent points along the gradient direction, that is, fit the quadratic equation using three coordinate points and obtain the compensation value. Step S34: Add a compensation value to the midpoint of three adjacent points along the gradient direction to obtain the sub-pixel edge point; Step S35: Connect the sub-pixel edge points to form the first contour.

2. The method for controlling the cutting path of a plate pattern according to claim 1, characterized in that: Step S2 specifically includes the following sub-steps: Step S21: Perform single-channel processing on the three channels of the board pattern image to obtain a grayscale image; Step S22: Set the grayscale threshold of the grayscale image, adjust the grayscale values ​​of pixels with grayscale values ​​greater than the grayscale threshold to the maximum grayscale value, and adjust the grayscale values ​​of pixels with grayscale values ​​less than the grayscale threshold to the minimum grayscale value to obtain a binary image.

3. The method for controlling the cutting path of a plate pattern according to claim 1, characterized in that: In step S31, the Gaussian template is calculated using a Gaussian function, the formula of which is as follows: in, Let x and y represent the pixel coordinates of the Gaussian template at various locations in the image; x represents the x-coordinate in the coordinate system; and y represents the y-coordinate in the coordinate system. This represents the standard deviation in the xy direction of the coordinate system.

4. The method for controlling the cutting path of a plate pattern according to claim 1, characterized in that: In step S6, the coordinate transformation specifically includes the following sub-steps: Step S61: Generate camera parameters through camera calibration; Step S62: Obtain the transformation matrix for the transformation based on the camera parameters; Step S63: Convert pixel coordinates to world coordinates according to the transformation matrix.

5. A control system for cutting patterns on sheet metal, characterized in that: Using the control method for the cutting path of a sheet metal pattern as described in any one of claims 1-4, the system comprises: The acquisition module is used to acquire images of the board pattern. The first processing module is used to preprocess the board pattern image to obtain the first board pattern image; The contour forming module is used to select pixels with gradient color changes in the first board pattern image and connect each pixel to form a first contour. The second processing module is used to perform curve fitting and smoothing on the first contour to obtain the second contour. The removal module is used to remove redundant lines from the second contour to obtain the third contour. The judgment module is used to determine whether the third contour exceeds the working table of the cutting machine. If it does, the excess part is removed to obtain the fourth contour, and the coordinates of each pixel point of the fourth contour are converted into the world coordinates of the cutting machine to generate the cutting path. If it does not exceed the working table, the coordinates of each pixel point of the third contour are converted into the world coordinates of the cutting machine to generate the cutting path. The control command generation module is used to generate control commands for the cutting machine based on the cutting path. The contour forming module includes: The scanning submodule is used to scan each pixel in the first board pattern image; The substitution submodule is used to replace the value of the center pixel of the Gaussian template with the weighted average gray value of the pixels in the neighborhood determined by the convolution kernel, so as to obtain the first plate pattern image after Gaussian filtering. The Gaussian template is calculated using the Gaussian function, the formula of which is as follows: in, Let x and y represent the pixel coordinates of the Gaussian template at various locations in the image; x represents the x-coordinate in the coordinate system; and y represents the y-coordinate in the coordinate system. This represents the standard deviation in the xy direction of the coordinate system; The first calculation submodule is used to calculate the gradient of the first plate pattern image after Gaussian filtering. The second calculation submodule interpolates the gradient magnitude at three adjacent points along the gradient direction using a quadratic function, that is, it fits a quadratic equation using three coordinate points and calculates the compensation value. The third calculation submodule adds a compensation value to the midpoint of three adjacent points along the gradient direction to obtain the sub-pixel edge point. The connection submodule is used to connect subpixel edge points to form a first contour.

6. The control system for the cutting path of a plate pattern according to claim 5, characterized in that: The first processing module includes: The grayscale processing submodule is used to perform single-channel processing on the three channels of the board pattern image to obtain a grayscale image. The binarization processing submodule is used to set the grayscale threshold of the grayscale image, adjust the grayscale values ​​of pixels with grayscale values ​​greater than the grayscale threshold to the maximum grayscale value, and adjust the grayscale values ​​of pixels with grayscale values ​​less than the grayscale threshold to the minimum grayscale value, thus obtaining a binary image.

7. The control system for the cutting path of a plate pattern according to claim 5, characterized in that: The judgment module includes a coordinate transformation submodule, which includes: The generation subunit is used to generate camera parameters through camera calibration. The sub-unit is used to obtain the transformation matrix during the transformation based on the camera parameters. The transformation subunit is used to convert pixel coordinates into world coordinates based on the transformation matrix.

Images