ETFE Film Cutting Control Method and System Based on Image Recognition

Abstract

The present invention relates to the technical field of image analysis, specifically an ETFE film cutting control method and system based on image recognition, including the following steps: enhancing edges through multi-scale Gaussian filtering, extracting pixels with gray-scale change rate by Sobel operator to generate path segments and output coordinates, calculating gradient directions to construct a direction continuous sequence to record angle deviations, determining abnormal fluctuations by a dynamic window to trigger calibration, replacing the vector of the fluctuating segment to update the motion trajectory, and generating a trajectory by B-spline fitting to output a cutting instruction. In the present invention, continuous pixels are extracted by enhancing edges through combining multi-scale Gaussian filtering with Sobel operator, solving the fracture problem caused by a fixed threshold. The direction sequence is constructed by vector space projection to quantify angle deviations, and the dynamic window range difference is used to determine fluctuations to trigger calibration. The vector replacement strategy corrects the control point direction of the abnormal segment, and B-spline interpolation generates a smooth trajectory to reduce mechanical vibration, improving the cutting accuracy of flexible materials and the collaborative efficiency of equipment.

Description

Technical Field

[0001] The present invention relates to the technical field of image analysis, and particularly to an ETFE film cutting control method and system based on image recognition. Background Art

[0002] Image analysis technology performs structured processing on static or dynamic images through computer algorithms to extract information such as edges, shapes, and textures. It covers technologies such as image segmentation, object recognition, registration, and enhancement, and is widely used in fields such as industrial inspection, medical imaging, remote sensing mapping, security monitoring, and autonomous driving, emphasizing image space and semantic analysis to support high-level decision-making.

[0003] Among them, the ETFE film cutting control method based on image recognition uses technologies such as edge detection and image segmentation to achieve film material feature extraction and boundary recognition, and combines geometric matching and path planning algorithms to generate cutting paths. The image acquisition device captures the film material image, forms a path instruction after analysis, and drives the cutting execution mechanism to achieve automatic control. The existing technology uses fixed-threshold edge detection, which is difficult to adapt to the interference of film material reflection or wrinkles, and is prone to false detection and requires manual repair. The path planning ignores the pixel gradient continuity and global direction consistency, resulting in path conflicts and instability; lacking a real-time calibration mechanism, it cannot cope with deformation errors; linear interpolation does not adapt to the flexibility of the film material, and is prone to trajectory burrs, affecting the equipment life and accuracy. Summary of the Invention

[0004] The purpose of the present invention is to solve the deficiencies existing in the prior art, and to propose an ETFE film cutting control method and system based on image recognition.

[0005] To achieve the above purpose, the present invention adopts the following technical solutions: The ETFE film cutting control method based on image recognition includes the following steps:

[0006] S1: Perform edge enhancement on the ETFE film image through multi-scale Gaussian filtering, extract the pixel points whose gray change rate exceeds the dynamic threshold based on the Sobel gradient operator, generate candidate cutting path segments, and extract the starting and ending pixel coordinates of each path segment;

[0007] S2: For the adjacent pixel points of the candidate cutting path segments, call the vector space projection algorithm, calculate the gradient direction vectors in the horizontal and vertical directions of each point, construct a path segment direction continuity sequence, and store the angle deviation value between it and the preset cutting direction;

[0008] S3: Based on the path segment direction continuity sequence, use the dynamic window range determination method to perform window sliding calculation on the angle deviation values of 5 consecutive pixel points. If the range within the window exceeds the set threshold, mark this path segment as an abnormal fluctuation segment and trigger a calibration instruction;

[0009] S4: For the abnormal fluctuation segment, extract the average gradient vector of its adjacent path segments, replace the direction vectors of all control points within the fluctuation segment with the average vector through a vector replacement strategy to generate calibrated path data, and synchronously update the motion trajectory parameters of the cutting device.

[0010] As a further solution of the present invention, the candidate cutting path segment specifically includes the starting and ending point coordinates of the path segment and the gray change rate parameter. The path segment direction continuity sequence includes the horizontal gradient vector, the vertical gradient vector, and the angle deviation value. The abnormal fluctuation segment specifically refers to the abnormal path segment mark and the calibration trigger threshold. The calibrated path data includes the replaced control point vector and the motion trajectory parameter.

[0011] As a further solution of the present invention, the steps for obtaining the candidate cutting path segment are specifically as follows:

[0012] S101: Obtain the original data of the ETFE film image, perform convolution operations on the image using Gaussian kernels of multiple scales, linearly superimpose multiple sets of convolution results according to preset weight coefficients to generate a multi-scale edge-enhanced image;

[0013] S102: Based on the multi-scale edge-enhanced image, call the Sobel operator to calculate the pixel gradient components, synthesize the gradient amplitude matrix, set a dynamic threshold according to the statistical characteristics of the gradient amplitude distribution, screen the pixel points whose gradient amplitude exceeds the threshold, and generate a set of pixels with gradient exceeding the limit;

[0014] S103: For the set of pixels with gradient exceeding the limit, perform a morphological closing operation to connect adjacent pixel points, extract the circumscribed rectangle boundary of the closed region, eliminate small-area isolated regions, retain the remaining path segments, and generate a candidate cutting path segment.

[0015] As a further solution of the present invention, the steps for obtaining the path segment direction continuity sequence are specifically as follows:

[0016] S201: Call the horizontal and vertical convolution kernel matrices, perform point-by-point convolution operations on the adjacent pixel points of the candidate cutting path segment, record the horizontal gradient component and the vertical gradient component, and generate a gradient direction vector set;

[0017] S202: Calculate the direction angles of multiple pixel points based on the gradient direction vector set, define the difference in direction angles between adjacent pixel points and store them in a queue in sequence to construct a direction angle difference queue;

[0018] S203: Call the direction angle difference queue and use the formula:

[0019] ;

[0020] Calculate the path segment continuity score, compare it with a preset threshold to screen the path segments, and generate a path segment direction continuity sequence.

[0021] Among them, represents the path segment direction continuity score, is the difference in direction angles of adjacent pixel points, , are the gradient magnitudes of adjacent pixel points, and is the Euclidean distance between adjacent pixel points, is the total number of adjacent pixel point pairs within the path segment.

[0022] As a further solution of the present invention, the steps for obtaining the abnormal fluctuation segment are specifically as follows:

[0023] S301: Obtain the angle deviation values of 5 consecutive pixel points in the path segment direction continuity sequence, traverse the path segment in a window sliding manner, calculate the range of the angle deviation values within each window, and generate a dynamic window range value;

[0024] S302: Invoke the dynamic window range value, and based on the difference in the range change trend between adjacent windows and the path segment curvature change rate, use the formula:

[0025] ;

[0026] Perform an operation to obtain a range fluctuation determination coefficient, and generate a range fluctuation determination coefficient through weighted superposition;

[0027] Among them, represents the range fluctuation determination coefficient, represents the maximum angle deviation within the window, represents the minimum angle deviation within the window, represents the average value of the range change trend difference between adjacent windows, represents the path segment curvature change rate, represents the environmental interference factor;

[0028] S303: Invoke the range fluctuation determination coefficient, compare it with a preset path fluctuation determination threshold, and if the coefficient exceeds the threshold, mark the path segment covered by the corresponding window as an abnormal fluctuation segment.

[0029] As a further solution of the present invention, the steps for obtaining the calibrated path data are specifically as follows:

[0030] S401: Obtain the data of adjacent path segments on both sides of the abnormal fluctuation segment, extract the gradient vectors of the control points within the path segment, calculate the arithmetic mean of the adjacent segment gradient vectors, and generate an adjacent gradient mean vector;

[0031] S402: Based on the adjacent gradient mean vector, traverse the direction vectors of the control points within the abnormal fluctuation segment, compare the multi-component values of the direction vectors and the mean vector, replace the original components with the mean components, update the control point coordinates and direction parameters, and generate a set of replacement vector control points;

[0032] S403: Call the coordinates of the unaffected path segments in the original path, integrate the set of replacement vector control points, reconstruct the connection relationship and curvature parameters of adjacent control points, and generate calibrated path data.

[0033] As a further solution of the present invention, the method further includes:

[0034] S5: According to the calibrated path data, call the B-spline curve interpolation algorithm to perform smooth fitting on the path segment joints, generate the ETFE film cutting trajectory, drive the multi-axis linkage parameters of the cutting head based on the coordinate differences of adjacent path points in the trajectory, and output cutting instructions;

[0035] The ETFE film cutting trajectory is specifically the B-spline curve parameters and multi-axis linkage parameters.

[0036] As a further solution of the present invention, the steps for obtaining the ETFE film cutting trajectory are specifically as follows:

[0037] S501: Call the calibrated path data, extract the control point coordinates and curvature parameters at the path segment joints, calculate the distances and angle differences between adjacent control points, set the knot vector and weight coefficients, and generate a set of path node parameters;

[0038] S502: Based on the set of path node parameters, divide the interpolation interval and solve the cubic polynomial coordinate components of each interval, adjust the high-order derivative terms of the polynomial coefficient matrix to meet the curvature continuity condition, and generate a set of fitting trajectory coordinates;

[0039] S503: Traverse the set of fitting trajectory coordinates, calculate the curvature change rate of adjacent interpolation points, screen the jump points that exceed the continuity threshold, re-interpolate and compensate, and output a path sequence to generate the ETFE film cutting trajectory.

[0040] An ETFE film cutting control system based on image recognition, the ETFE film cutting control system based on image recognition is used to execute the above-mentioned ETFE film cutting control method based on image recognition, and the system includes:

[0041] An edge enhancement module, which is used to perform filtering and noise reduction on the ETFE film image through a multi-scale Gaussian filtering algorithm, call the Sobel gradient operator to calculate the horizontal and vertical gray-scale change rates, screen the set of continuous pixel points whose gray-scale change rates exceed the dynamic threshold, generate candidate cutting path segments and output their starting and ending point coordinates, and transfer the candidate cutting path segments to the direction continuity analysis module;

[0042] A direction continuity analysis module, which is used for adjacent pixel points of the candidate cutting path segment, constructs gradient direction vectors of each pixel point by using a vector space projection algorithm, calculates the direction angle between adjacent vectors, generates a path segment direction continuity sequence, stores the angle deviation value between the angle and a preset direction, and transmits the path segment direction continuity sequence to an abnormal fluctuation determination module;

[0043] An abnormal fluctuation determination module, which is used for based on the path segment direction continuity sequence, calls a dynamic window range difference determination method to perform a sliding window calculation on the angle deviation values of 5 consecutive pixel points, determines whether the range difference within the window exceeds a set threshold, marks the path segment with an excessive range difference as an abnormal fluctuation segment and generates a calibration instruction, and transmits the abnormal fluctuation segment to a path calibration module;

[0044] A path calibration module, which is used for extracting the mean value of the gradient direction vectors of adjacent path segments of the abnormal fluctuation segment, replacing the direction vectors of all control points within the fluctuation segment with the mean value vectors through a vector replacement strategy, generating calibrated path data and updating the motion trajectory parameters of the cutting device, and transmitting the calibrated path data to a trajectory generation module;

[0045] A trajectory generation module, which is used for based on the calibrated path data, calls a B-spline curve interpolation algorithm to perform smooth fitting on the connection of path segments, calculates the coordinate difference between adjacent path points to generate an ETFE film cutting trajectory, and drives the multi-axis linkage parameters of the cutting head to output a cutting instruction.

[0046] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0047] In the present invention, continuous pixels are enhanced by multi-scale Gaussian filtering combined with the Sobel operator to solve the fracture problem caused by a fixed threshold. The vector space projection constructs a direction sequence to quantify the angle deviation. The dynamic window range difference determines the fluctuation to trigger calibration. The vector replacement strategy corrects the direction of control points in the abnormal segment. The B-spline interpolation generates a smooth trajectory to reduce mechanical vibration, improving the cutting accuracy of flexible materials and the collaborative efficiency of the device. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 It is a schematic diagram of the working process of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0049] In order to make the objectives, technical solutions and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention. Embodiment

[0050] Please refer to Figure 1, the present invention provides a technical solution: an ETFE film cutting control method based on image recognition, comprising the following steps:

[0051] S1: Perform edge enhancement on the ETFE film image through multi-scale Gaussian filtering, extract pixel points with a gray-scale change rate exceeding the dynamic threshold based on the Sobel gradient operator, generate candidate cutting path segments, and extract the starting and ending pixel coordinates of each path segment;

[0052] S2: For adjacent pixel points of the candidate cutting path segments, call the vector space projection algorithm, calculate the gradient direction vectors in the horizontal and vertical directions for each point, construct a path segment direction continuity sequence, and store the angle deviation value from the preset cutting direction (that is, calculate the angle deviation between the direction vector of each pixel point in the sequence and the preset cutting direction and store it for subsequent determination);

[0053] S3: Based on the path segment direction continuity sequence, use the dynamic window range determination method to perform window sliding calculations on the angle deviation values of 5 consecutive pixel points. If the range within the window exceeds the set threshold (that is, the difference between the maximum angle deviation and the minimum angle deviation within the window exceeds the threshold, and this difference is the "range fluctuation value", defined as: Δθ = max(θᵢ) - min(θᵢ), and if Δθ > T, it is considered abnormal), mark this path segment as an abnormal fluctuation segment and trigger a calibration instruction;

[0054] S4: For the abnormal fluctuation segment, extract the average value of the gradient vectors of its adjacent path segments, replace the direction vectors of all control points within the fluctuation segment with the average value vector through the vector replacement strategy, generate calibrated path data, and synchronously update the motion trajectory parameters of the cutting device;

[0055] S5: According to the calibrated path data, call the B-spline curve interpolation algorithm to perform smooth fitting on the connections of the path segments, generate the ETFE film cutting trajectory, drive the multi-axis linkage parameters of the cutting head based on the coordinate differences between adjacent path points in the trajectory, and output a cutting instruction.

[0056] The candidate cutting path segments specifically include the starting and ending point coordinates of the path segment and the gray-scale change rate parameter. The path segment direction continuity sequence includes the horizontal gradient vector, the vertical gradient vector, and the angle deviation value. The abnormal fluctuation segment specifically refers to the abnormal path segment mark and the calibration trigger threshold. The calibrated path data includes the replaced control point vector and the motion trajectory parameter. The ETFE film cutting trajectory specifically includes the B-spline curve parameter and the multi-axis linkage parameter.

[0057] The specific steps for obtaining the candidate cutting path segments are as follows:

[0058] S101: Obtain the original image data of the ETFE film. This data is a grayscale image file with 1920x1080 pixels, named ETFE_raw.png. The average grayscale value in the main area of the film material in the image is stable at around 150, the average grayscale value in the background area is about 50, and the edge area shows a grayscale gradient due to the influence of light and the curved surface.

[0059] Perform convolution operations on the original image using Gaussian kernels of multiple scales. Three Gaussian kernels with different scales are selected, and their parameter settings are based on the response analysis of the edge features of the ETFE film image at different observation scales, aiming to capture both fine edge details and smooth overall contours simultaneously. The specific parameters are shown in Table 1.

[0060] Table 1 Gaussian kernel parameter table

[0061]

[0062] As shown in Table 1, the specific sizes and standard deviations of the three Gaussian kernels used in this embodiment are listed.

[0063] Perform the first convolution on the original image data matrix using the scale 1 Gaussian kernel (3x3, ) in Table 1. Align the center of the kernel with each pixel point in the image in turn, such as pixel point and its 3x3 neighborhood, calculate the weighted sum of the kernel coefficients and the corresponding neighborhood pixel values, and obtain the convolution result of this point at the first scale . Perform this operation on the entire image to generate the first set of convolution result matrices .

[0064] Next, perform the second convolution on the original image data matrix using the scale 2 Gaussian kernel (5x5, ) in Table 1. Similarly, traverse all pixel points, calculate the weighted sum, and obtain the convolution result of pixel point at the second scale , and generate the second set of convolution result matrices .

[0065] Then, perform the third convolution on the original image data matrix using the scale 3 Gaussian kernel (7x7, ) in Table 1. Calculate the convolution result of pixel point at the third scale , and generate the third set of convolution result matrices .

[0066] Combine these three sets of convolution results Perform linear superposition according to the preset weight coefficients. The setting of the weight coefficients aims to balance the contributions of information at different scales. According to empirical analysis, the medium scale (5x5 kernel) contributes the most to the extraction of the main contour of the ETFE film, while the small scale (3x3 kernel) helps to retain details, and the large scale (7x7 kernel) helps to suppress noise and connect discontinuous edges. Therefore, set the weight coefficients . Specifically, by evaluating the effects on a set of test images, it is found that when , the edge enhancement effect is the most balanced and satisfies . Calculate the final enhancement value of each pixel point. For the pixel point , its enhanced pixel value is calculated as . Perform this linear superposition calculation on all pixel points to generate the final multi-scale edge enhancement image .

[0067] S102: Based on the multi-scale edge enhancement image , call the Sobel operator to calculate the pixel gradient components. Use the standard Sobel horizontal direction gradient kernel and the vertical direction gradient kernel .

[0068] For each pixel point in, perform the convolution operation. First, extract the weighted sum of the pixel values in its 3x3 neighborhood and the kernel coefficients to obtain the horizontal gradient component of this point. Then, use the kernel to perform the convolution operation on the 3x3 neighborhood of the same point to obtain the vertical gradient component of this point. Repeat this process for all pixel points to generate the complete horizontal gradient component matrix and the vertical gradient component matrix .

[0069] Based on the calculated gradient components and , synthesize the gradient magnitude of each pixel point. The calculation formula is . Consider a pixel point , and its gradient components obtained by convolution calculation are and , then its gradient magnitude is . The unit of the gradient magnitude here is gray intensity unit per pixel. Perform this calculation on all pixel points to generate the gradient magnitude matrix .

[0070] According to the gradient magnitude matrix Set a dynamic threshold based on the statistical characteristics of the gradient magnitude distribution of all pixels . First, calculate the matrix the average value of all gradient magnitudes and the standard deviation . For the calculated in this embodiment gray levels / pixel, gray levels / pixel. The dynamic threshold is set using to adapt to the image contrast and noise level and identify pixels significantly higher than the average edge intensity. The value of the coefficient determines the strictness of the screening, and its setting is based on tests of multiple sample images containing clear edges and noise regions. When , too many noise points may be introduced; when , some weak edges may be lost. By testing, selecting can effectively suppress noise while retaining most of the real edges. Therefore, calculate the dynamic threshold gray levels / pixel.

[0071] Compare the gradient magnitude of each pixel in the gradient magnitude matrix with the calculated dynamic threshold . If , then determine that this pixel is a gradient-overrun pixel. For the point calculated above, its gradient magnitude , since , this point is screened out. Record the coordinates of all pixels that satisfy to generate a gradient-overrun pixel set .

[0072] S103: For the gradient-overrun pixel set , this set appears as a series of discrete points or short pixel chains on the image.

[0073] Perform a morphological closing operation to connect adjacent pixels. Define a 3x3 pixel square structuring element . First, perform a morphological dilation operation on the binary image (the overrun pixel value is 1 and the rest are 0) represented by : Use to slide over the image. If the center of or any pixel covered by it touches a pixel with a value of 1, then this The values of all pixels under the coverage are set to 1. This operation will expand the gradient overrun pixel area and connect adjacent pixels. Then, perform morphological erosion on the dilated result: use the same to slide over the image. Only when all the pixel values within the completely covered area are 1, the value of its central pixel is retained as 1; otherwise, it is set to 0. This operation removes small noise points and thin connecting bridges. The combination of dilation followed by erosion is the morphological closing operation, which can effectively fill small discontinuities and connect adjacent edge segments.

[0074] Perform connected component analysis on the binary image after the closing operation to identify all independent regions composed of connected pixels. Calculate the bounding rectangle of each independent closed region, that is, determine the smallest rectangle range that can completely enclose the region and whose sides are parallel to the image coordinate axes, and record its upper left corner coordinates and the lower right corner coordinates .

[0075] Eliminate small isolated areas. Calculate the area of each bounding rectangle , with the unit of square pixels. Set an area threshold . The setting of this threshold is based on the statistical analysis of the sizes of typical noise spots and the minimum sizes of effective edge segments in the ETFE film image. It is observed that the area of isolated regions formed by sensor noise or tiny impurities is usually less than 50 square pixels, while the area of connected regions formed by meaningful edge segments is much larger than this value. Therefore, set square pixels. Compare the area calculated for each region with the threshold . If , then determine that the region is a small isolated area, and remove all the pixel points it contains from the set. A calculation example is that the upper left corner of the bounding rectangle of a connected region is (50, 50), the lower right corner is (54, 56), and its area square pixels. Because , this region is removed.

[0076] Retain the set of pixel points within all regions with an area . These retained sets of pixel points that have been morphologically connected and have a sufficient area constitute the path segments. Generate a set of candidate cutting path segments.

[0077] The steps to obtain the sequence of path segment direction continuity are as follows:

[0078] S201: Call each path segment in the set of candidate cutting path segments. Select one of the path segments , this path segment contains a series of ordered pixel points .

[0079] To calculate the local direction information of each pixel point, the gradient information needs to be utilized again. Here, the horizontal gradient component matrix calculated in S102 is called and the vertical gradient component matrix . For each pixel point in the path segment ( ), directly read the gradient component values corresponding to this coordinate from the matrices and , that is, and .

[0080] Combine the gradient components of each pixel point into a two-dimensional gradient direction vector . Taking the pixel point in S102 as an example, if it belongs to the point in the path segment , then its gradient direction vector is . Perform this operation on all pixel points in the path segment , record the gradient direction vectors of all points, and generate the gradient direction vector set of this path segment. Repeat this process for all path segments to obtain the gradient direction vector sets of all candidate path segments respectively.

[0081] S202: Based on the gradient direction vector set of a certain path segment , where .

[0082] Calculate the gradient direction angle of each pixel point in the path segment. Use the two-parameter arctangent function atan2 for calculation, and the formula is . This function can correctly judge the quadrant where the vector is located according to the signs of and , and return a radian value within the interval . For the vector in S201, its direction angle is radians (about 52.26 degrees). Calculate the direction angle ( ) of all pixel points in the path segment to obtain the direction angle sequence

[0083] Calculate the difference in orientation angles between adjacent pixel points. For the th and in the path segment (i.e., ), calculate the difference in their orientation angles. To handle the periodicity of angles (e.g., jumping from near to near ), the calculated difference needs to be normalized to the interval. The implementation is as follows: If , then ; if , then . For a specific calculation, if radians and radians, the direct subtraction gives radians, which is less than , so an adjustment is made: radians. If radians and radians, then radians, which is within and does not require adjustment.

[0084] Calculate the difference in orientation angles for all pairs of adjacent pixel points in the path segment in order. Store these calculated and normalized differences in orientation angles in a queue data structure in the order in which they appear in the path segment (from to ). Construct the queue of differences in orientation angles for this path segment .

[0085] S203: Call the queue of differences in orientation angles for the path segment , and prepare to calculate the continuity score for this path segment using a formula. The formula is: ,

[0086] Obtaining and explanation of each parameter in the formula: : The difference in orientation angles (in radians) for the th pair of adjacent pixel points ( and ), and its value is obtained in order from the queue . represents taking the absolute value of this difference. : Respectively, the gradient magnitudes (i.e., gradient amplitudes, in gray levels per pixel) of the adjacent pixel points and , and their values are from the gradient magnitude matrix calculated in S102 In it, it is obtained according to the points and coordinates and query. : Euclidean distance between adjacent pixel points and . On the pixel grid, if two points are horizontally or vertically adjacent, pixels; if diagonally adjacent, pixels. : Total number of adjacent pixel point pairs within the path segment, equal to the number of path segment pixels minus 1. : Find the queue in all direction angle differences square of the maximum value. Its square root is equal to , that is, the maximum absolute direction angle difference (unit: radian) in the entire path segment.

[0087] The calculation process of the formula is illustrated with an example path segment . This path segment contains 4 pixel points , so there are pairs of adjacent points. The relevant parameter values are shown in Table 2 below, where the gradient magnitude is obtained from S102, the direction angle difference is obtained from the queue constructed by S202, and the adjacent distance is calculated according to the pixel coordinates.

[0088] Table 2 Path segment Example parameter table

[0089]

[0090] As shown in Table 2, the specific parameter values for calculating the continuity score of the path segment are listed.

[0091] Substitute into the formula for calculation:

[0092] The first part (summation term):

[0093] : ;

[0094] : ;

[0095] : ;

[0096] Summation result = ;

[0097] Second part (maximum absolute difference term):

[0098] Calculate radians;

[0099] Or calculate according to the formula radians.

[0100] Final continuity score ;

[0101] This score is a comprehensive metric with mixed units (derived from a combination of radians, grayscale / pixel, pixels), mainly used for relative comparison. The smaller the value, the smoother and more continuous the direction change of the path segment.

[0102] Set a preset threshold for path segment continuity . The setting of this threshold is based on scoring calculations and manual evaluations of path segments extracted from a large number (such as 100) of ETFE film sample images. The path segments are divided into three categories according to visual effects: "continuously smooth" (scores generally lower than 25), "slightly fluctuating" (scores between 25 - 35), and "broken / abrupt turn" (scores higher than 35). To select high-quality initial path segments, set the threshold . This is an engineering setting value based on statistical analysis and application requirements, used to distinguish path segments with sufficiently good direction continuity.

[0103] Compare the calculated path segment continuity score with the preset threshold . Since , it indicates that the direction continuity of this path segment meets the requirements and is judged as qualified. Therefore, retain this path segment. Repeat the process of S201 - S203 for all candidate cropping path segments in , screen out all path segments that meet the conditions, and generate a path segment direction continuity sequence .

[0104] The specific steps for obtaining abnormally fluctuating segments are as follows:

[0105] S301: Obtain a path segment from the path segment direction continuity sequence , denoted as . This path segment contains a series of pixel points and its corresponding queue of direction angle differences , where (unit: radians) is the angle deviation value of the th pair of adjacent pixel points.

[0106] Set an analysis window of a fixed size, where the window size is defined as containing 5 consecutive pixel points. Such a window covers 4 consecutive angular deviation values (since the angular deviation is between adjacent points). Traverse the path segment in the way of window sliding .

[0107] Initially, the first window covers the to 5 pixel points at the start of the path segment. Therefore, the angular deviation values to be analyzed are . Calculate the range of these 4 angular deviation values within this window (denoted as ), that is, find the maximum value and the minimum value among these 4 values, and calculate the range . Let the sequence of deviation values within window be radians. Then radians, radians, and the range of this window radians.

[0108] Record the range value radians of the first window. Then, slide the window forward by one pixel position along the path segment. The second window covers the to 5 pixel points, and the involved angular deviation values are . Calculate the range of the angular deviation values within this new window. Repeat this sliding and calculating process, and the windows are successively . For each position on the path segment that can form a complete 5 - point window, calculate and record its corresponding angular deviation range value. Generate a dynamic window range value sequence .

[0109] S302: Call the dynamic window range value sequence . For each range value in the sequence (corresponding to the th window), use the following formula to calculate the range fluctuation determination coefficient of this window:

[0110] ;

[0111] Obtaining and explanation of each parameter in the formula (taking the calculation of of the th window as an example): : They are respectively the angular deviation values within the The maximum and minimum values (unit: radians). These two values are used to calculate the range in S301 and have been obtained. : The mean value of the difference in the change trend of the range between adjacent windows corresponding to the th window (unit: radians). This value reflects the local smoothness of the range change of the current window. The calculation method is as follows: Consider the current window and one window before and after it and for the range (if or , only consider one side). Calculate the absolute value of the change amount of adjacent ranges and . Take the average value of these absolute values of the change amounts. If and , then . : The absolute value of the average curvature change rate of the path segment covered by the th window. The curvature can be approximately calculated from the angular deviation and the pixel pitch , that is (unit: radians / pixel). The curvature change rate is the average value of the difference in curvature of adjacent points calculated within the window . Therefore . Note that the unit here is (radians / pixel) / pixel = radians / pixel². : The environmental interference factor, which is a dimensionless adjustment coefficient. Its setting is based on the evaluation of the overall quality of the input image, especially the noise level. It is quantified by calculating the signal-to-noise ratio (SNR) of the image. Setting rules: If SNR > 30 dB (determined as a low-noise image), ; if 20 dB < SNR 30 dB (determined as a medium-noise image), ; if SNR 20 dB (determined as a high-noise image), . This factor is used to adjust the sensitivity of the algorithm to noise fluctuations.

[0112] Continue with the example of S301 to calculate the coefficient of the first window . Given rad, rad, rad. It is necessary to calculate . This requires knowing . Assume the window The deviation values are [-0.10, 0.08, 0.02, 0.06] rad, then , the range rad. Since it is the first window, only consider the difference from : . To avoid a zero denominator, a small positive lower limit will be set in practical applications, or the average value over a wider range (such as two windows before and after) will be used. Here, the correction rule is adopted: if the calculated is less than a tiny value (such as ), then it is set to this tiny value. In this example, let the lower limit be , then rad. It is necessary to calculate . The points involved in window are , and the angular deviation is . The corresponding pixel distance is assumed to be pixels.

[0113] Calculate the approximate curvature:

[0114] , , , (unit: rad / pixel).

[0115] Calculate the absolute value of the curvature change:

[0116] ;

[0117] ;

[0118] .

[0119] Calculate the average curvature change rate rad / pixel²;

[0120] Set the environmental interference factor . By analyzing the input image ETFE_raw.png, its SNR is calculated to be about 25 dB, belonging to the medium noise level. According to the set rules, take .

[0121] Substitute the obtained values into the formula to calculate :

[0122] ;

[0123] ;

[0124] ;

[0125] ;

[0126] This value is very large, mainly because it is close to zero. Let's reconsider the calculation, using a more stable method, such as calculating the average of the absolute values of the range changes of two windows before and after the calculation window , and setting a reasonable minimum denominator value. Or, taking the calculation window as an example (assuming is not a boundary), its range . The range of its previous window is , and the range of the next window is .

[0127] Calculate rad;

[0128] Using the rad / pixel² calculated above (assuming this value also applies to the window ) and .

[0129] ;

[0130] ;

[0131] ;

[0132] ;

[0133] , this coefficient is a unitless comprehensive score used to judge whether there are abnormal fluctuations within the window. Calculate the corresponding value for each valid window position on the path segment.

[0134] S303: Call the range fluctuation determination coefficient of each window position, such as .

[0135] Set a preset path fluctuation determination threshold . The setting of this threshold is based on the calculation of the coefficient for a large number of path segment samples containing known normal bends (such as designed arcs) and known abnormal fluctuations (such as material wrinkles, sawteeth caused by image acquisition jitter). Statistical analysis shows that the The values are usually distributed between 5 and 10, while for the abnormal fluctuation segments, the values are significantly higher, and most of them exceed 15. To effectively identify abnormal fluctuations and avoid misjudging normal large curvature changes, a threshold is set. This is an engineering setting value based on empirical data and discrimination.

[0136] Compare the range fluctuation determination coefficient calculated for a specific window with the preset threshold . Perform the judgment: If , then it is determined that there is an abnormal fluctuation in the path segment (i.e., the part from the starting pixel point to the ending pixel point of the window) covered by this window . In our calculation example, . Because , the path segment covered by this window is not marked as an abnormal fluctuation segment. If there is another window , and its calculated coefficient . Because , the path segment covered by this window will be marked as an abnormal fluctuation segment.

[0137] Perform this comparison and judgment on all the calculated values on the path segment. Mark the path segments covered by all the windows whose coefficients are determined to exceed the threshold . Since the window is sliding, there may be cases where multiple adjacent or overlapping windows are marked. Merge the sets of pixel points covered by all the marked and possibly overlapping windows to form one or more continuous abnormal fluctuation segments. Finally, output the set of these abnormal fluctuation segments.

[0138] The specific steps for obtaining the calibrated path data are as follows:

[0139] S401: Obtain a marked abnormal fluctuation segment . At the same time, from the path segment direction continuity sequence , identify the normal path segment that is not marked as abnormal and is immediately adjacent in the path order before , and the normal path segment that is not marked as abnormal and is immediately adjacent in the path order after .

[0140] Extract the gradient vectors of all the control points (pixel points) inside these two adjacent normal path segments and . These vectors is obtained in S201. Let contain control points, and its gradient vector set is . Let contain control points, and its gradient vector set is .

[0141] Calculate the arithmetic mean of all gradient vectors in these two adjacent normal segments and to obtain an adjacent gradient mean vector representing the stable gradient direction of this region. The calculation method is: ;

[0142] Taking a specific scenario as an example, contains points, and its gradient vector is . contains points, and its gradient vector is . The two segments have a total of points. Calculate the average x component:

[0143] ;

[0144] Calculate the average y component:

[0145] ;

[0146] The finally generated adjacent gradient mean vector is .

[0147] S402: Based on the adjacent gradient mean vector . Traverse all control points (pixel points) within the marked abnormal fluctuation segment . Let the th control point within the abnormal segment be , and its original gradient direction vector is (from S201).

[0148] For each control point within the abnormal fluctuation segment , perform the direction vector replacement operation: directly replace its original gradient direction vector with the adjacent gradient mean vector calculated in S401. That is, the updated vector . The basis for this operation is that it is considered that the true direction of the abnormal segment should smoothly transition to the directions of its adjacent stable segments, so the average gradient directions on both sides are used to correct the direction information of all points within the abnormal segment.

[0149] After replacing the direction vector, the direction parameters of the control point must be updated synchronously, mainly the direction angle . Use the replaced vector to calculate the new direction angle : radians (about 50.9 degrees). For all control points within the abnormal segment , their direction angles will be uniformly updated to this calculated radians. In this step, the coordinates of the control points remain unchanged, and only their associated direction parameters are updated.

[0150] Collect all the control points within all abnormal fluctuation segments after direction vector replacement and direction parameter update to generate a set of replacement vector control points .

[0151] S403: Call the data of the unaffected path segments in the original path, that is, the generated in S203, excluding all abnormal fluctuation segments marked in S303 The remaining part is denoted as . This part of the data contains the coordinates and original parameters of the path segments considered to be reliable. At the same time, call the set of replacement vector control points that have been directionally calibrated and generated in S402 (which represents the calibrated version of the original abnormal fluctuation segment ).

[0152] Integrate these two parts of data and . According to their order in the original complete path, insert into the corresponding position in , that is, replace the original with the calibrated part to form a preliminary integrated complete path point sequence containing all points.

[0153] On this basis, reconstruct the connection relationship and curvature parameters between adjacent control points on the path. Especially at the interfaces between the normal segment and the calibrated segment (that is, between the end point of and the start point of , and between the end point of and the start point of ), it is necessary to recalculate the distance and the direction angle difference . And according to the new, integrated point sequence and the updated direction parameters (from the original parameters of and from After calibration, recalculate the curvature parameters at each point along the entire path. At the connection points, to ensure smooth transition of the path, it is necessary to check and ensure the continuity of curvature. If there is an obvious jump, local smoothing processing is required, such as by fine-tuning the parameters of the control points near the connection points. Finally, generate the calibrated path data including all control point coordinates, updated direction parameters, and recalculated connection relationships and curvature parameters. .

[0154] The steps to obtain the cutting trajectory of the ETFE film are as follows:

[0155] S501: Call the calibrated path data. This data is a series of ordered control points. , each point has coordinates (unit: pixel) and updated direction or curvature parameters (unit: radian / pixel).

[0156] Extract the key information required for subsequent curve fitting. Select all control points on the path as the nodes for fitting. Extract the coordinates and curvature parameters of each node. Calculate the Euclidean distance and between adjacent control points (unit: pixel). Calculate the angle difference between adjacent tangent vectors , or the angle change (unit: radian) defined by three consecutive control points .

[0157] Set the knot vector for B-spline or NURBS curve fitting . The setting of the knot vector affects the shape and parameterization process of the curve. Use the chord length parameterization method to set the knot vector, which can better reflect the spatial distance between control points. Calculate the cumulative chord length: , for . Then normalize the entire node sequence to the interval [0,1] by dividing by the total chord length as follows: . For cubic B-spline (d = 3), the complete knot vector also needs to add repeated nodes at both ends, in the form of . Consider a path containing 4 control points: pixels.

[0158] Calculate the distance:

[0159] Pixel;

[0160] Pixel;

[0161] Pixel.

[0162] Calculate the cumulative chord length: Total length .

[0163] Normalize internal nodes:

[0164] ;

[0165] .

[0166] If using cubic B-spline, the knot vector .

[0167] Set the weight coefficients . For standard B-spline curve fitting, the weight coefficients of all control points are all set to 1.

[0168] Integrate the extracted control point coordinates , curvature , the calculated distance , the angle difference , and the generated knot vector and weight coefficients to generate a set of path node parameters for curve fitting .

[0169] S502: Based on the set of path node parameters , especially the control point coordinates , knot vector and weight , perform fitting using a cubic B-spline curve.

[0170] Divide the parameter domain (usually [0, 1]) into multiple parameter subintervals according to the knot vector . In each parameter subinterval , the points on the curve are given by the weighted sum of the control points and the corresponding cubic B-spline basis functions : , where the basis function is defined by the knot vector through the Cox-deBoor recurrence formula. Calculate the coordinate components in each subinterval and The specific cubic polynomial expression (i.e., determining the coefficients and ).

[0171] To ensure that the generated cutting trajectory is smooth enough to meet the cutting process requirements, it is necessary to adjust the curve parameters to meet continuity, i.e., curvature continuity. The standard cubic B-spline curve automatically satisfies continuity at non-repeated nodes. If the node vector design or control point arrangement results in non-satisfaction of continuity at some nodes (this is not common in the standard construction, unless there are special designs of the node vector or degenerate cases such as collinear control points), then it is necessary to adjust the positions of the control points or increase the number of control points to ensure that at all connection points, the first derivative (tangent) and the second derivative (related to curvature) of the curve are continuous.

[0172] Perform B-spline curve calculations for all parameter intervals. By densely sampling in the parameter range from 0 to 1, a series of coordinate points that define the entire smooth curve are generated. The selection of the sampling step determines the density of the trajectory points. According to the required cutting accuracy, set . For each sampled parameter , calculate the corresponding curve point . Generate the fitted trajectory coordinate set , where . The coordinate unit of these points is still pixels.

[0173] S503: Traverse the fitted trajectory coordinate set , where are the trajectory points arranged in order (unit: pixels).

[0174] Calculate the curvature change rate between adjacent interpolation points on the trajectory and perform unit conversion to match the physical cutting requirements. First, establish the conversion rule between pixel coordinates and physical coordinates (millimeters, mm). Through camera calibration or referring to the known object size, determine the physical scale of the image. Set the conversion rule: 1 millimeter corresponds to 10 pixels in the image. That is, the scale factor pixels_per_mm = 10 pixels / mm. Therefore, 1 pixel = 0.1 mm.

[0175] Calculate the curvature of each sampling point on the trajectory. Since the analytical expression of the B-spline curve is known, the curvature can be accurately calculated through its first and second derivatives: ;

[0176] The calculated curvature is in the unit of 1 / pixel. According to the conversion rule, convert it to the physical unit 1 / mm:

[0177] Calculate the physical distance between adjacent trajectory points and :

[0178] ;

[0179] Calculate the physical curvature change rate between adjacent points (unit: (1 / mm) / mm = 1 / mm²):

[0180] ;

[0181] Set a curvature continuity threshold . The setting of this threshold is based on the specific characteristics of the ETFE film material and the dynamic performance of the cutting tool used (such as a laser cutting head). Excessive curvature changes may lead to unstable cutting speed, degraded cutting edge quality, or material tearing. Through experimental tests, for laser cutting of an ETFE film with a thickness of 0.2 mm, it is found that when the curvature change rate exceeds 0.8 mm⁻², the cutting quality begins to become unstable. Therefore, set mm⁻².

[0182] Compare the calculated curvature change rate of each point with the threshold . Screen out all points that satisfy , and these points are marked as "curvature jump points". .

[0183] For each screened curvature jump point , perform re - interpolation compensation on the local path segment near it. The method is to increase the control point density of the B - spline near this point, or adjust the positions of nearby control points, so that the curvature change of the re - fitted local curve at this point is smoother, ensuring that its value is lower than the threshold mm⁻².

[0184] After compensating all detected curvature jump points, traverse and check the curvature change rate of the entire trajectory again to ensure that all points have . The final output path point sequence, whose coordinates need to be converted to physical units (millimeters). For each point (pixels), the output coordinate is (millimeters). This coordinate sequence after smoothing processing and unit conversion is the final cutting trajectory of the ETFE film . Table 3 below shows a small segment of example data of the final trajectory.

[0185] Table 3 Example Table of ETFE Film Cutting Trajectory Segments

[0186]

[0187] As shown in Table 3, a small segment of data in the finally generated ETFE film cutting trajectory is listed, including physical coordinates, the curvature of each point, and the curvature change rate between adjacent points. All curvature change rates are lower than the set threshold of 0.8 mm⁻².

[0188] The ETFE film cutting control system based on image recognition is used to execute the above-mentioned ETFE film cutting control method based on image recognition. The system includes:

[0189] An edge enhancement module for performing filtering and noise reduction on the ETFE film image through a multi-scale Gaussian filtering algorithm, calling the Sobel gradient operator to calculate the horizontal and vertical gray-scale change rates, screening a set of continuous pixel points with gray-scale change rates exceeding the dynamic threshold, generating candidate cutting path segments and outputting their start and end point coordinates, and transmitting the candidate cutting path segments to the direction continuity analysis module;

[0190] A direction continuity analysis module for, for adjacent pixel points of the candidate cutting path segment, constructing gradient direction vectors of each pixel point using a vector space projection algorithm, calculating the direction angle between adjacent vectors, generating a path segment direction continuity sequence, storing the angle deviation value between the angle and the preset direction, and transmitting the path segment direction continuity sequence to the abnormal fluctuation determination module;

[0191] An abnormal fluctuation determination module for, based on the path segment direction continuity sequence, calling the dynamic window range difference determination method to perform a sliding window calculation on the angle deviation values of 5 consecutive pixel points, determining whether the range difference within the window exceeds the set threshold, marking the path segment with an excessive range difference as an abnormal fluctuation segment and generating a calibration instruction, and transmitting the abnormal fluctuation segment to the path calibration module;

[0192] A path calibration module for extracting the mean value of the gradient direction vectors of adjacent path segments of the abnormal fluctuation segment, replacing the direction vectors of all control points within the fluctuation segment with the mean vector through a vector replacement strategy, generating calibrated path data and updating the motion trajectory parameters of the cutting device, and transmitting the calibrated path data to the trajectory generation module;

[0193] A trajectory generation module for, based on the calibrated path data, calling the B-spline curve interpolation algorithm to perform smooth fitting at the connection of path segments, calculating the coordinate differences between adjacent path points to generate the ETFE film cutting trajectory, and driving the multi-axis linkage parameters of the cutting head to output a cutting instruction.

[0194] The above are only the preferred embodiments of the present invention, and do not limit the present invention in other forms. Any person skilled in the relevant art may use the technical content disclosed above to make changes or modifications into equivalent embodiments with equivalent changes and apply them to other fields. However, as long as it does not depart from the technical solution content of the present invention, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention still fall within the protection scope of the technical solution of the present invention.

Claims

1. An ETFE film cutting control method based on image recognition, characterized in that, It includes the following steps: S1: Perform edge enhancement on the ETFE film image through multi-scale Gaussian filtering, extract the pixel points whose gray-scale change rate exceeds the dynamic threshold based on the Sobel gradient operator, generate candidate cutting path segments, and extract the starting and ending pixel coordinates of each path segment; S2: For the adjacent pixel points of the candidate cutting path segments, call the vector space projection algorithm, calculate the gradient direction vectors of each point in the horizontal and vertical directions, construct a path segment direction continuity sequence, and store the angle deviation value from the preset cutting direction; S3: Based on the path segment direction continuity sequence, perform window sliding calculation on the angle deviation values of 5 consecutive pixel points using the dynamic window range determination method. If the range within the window exceeds the set threshold, mark this path segment as an abnormal fluctuation segment and trigger a calibration instruction; S4: For the abnormal fluctuation segment, extract the average gradient vector of its adjacent path segments, replace the direction vectors of all control points within the fluctuation segment with the average vector through the vector replacement strategy, generate calibrated path data, and synchronously update the motion trajectory parameters of the cutting device; The candidate cutting path segment specifically includes the starting and ending point coordinates of the path segment and the gray-scale change rate parameter. The path segment direction continuity sequence includes the horizontal gradient vector, the vertical gradient vector, and the angle deviation value. The abnormal fluctuation segment specifically refers to the abnormal path segment mark and the calibration trigger threshold. The calibrated path data includes the replaced control point vector and the motion trajectory parameter; The specific steps for obtaining the calibrated path data are as follows: S401: Obtain the data of the adjacent path segments on both sides of the abnormal fluctuation segment, extract the gradient vectors of the control points within the path segment, calculate the arithmetic average of the adjacent segment gradient vectors, and generate an adjacent gradient average vector; S402: Based on the adjacent gradient average vector, traverse the direction vectors of the control points within the abnormal fluctuation segment, compare the multi-component values of the direction vector and the average vector, replace the original component with the average component, update the control point coordinates and direction parameters, and generate a set of replaced vector control points; S403: Call the coordinates of the unaffected path segments in the original path, integrate the set of replaced vector control points, reconstruct the connection relationship and curvature parameters of the adjacent control points, and generate calibrated path data.

2. The ETFE film cutting control method based on image recognition according to claim 1, characterized in that, The specific steps for obtaining the candidate cutting path segment are as follows: S101: Obtain the original data of the ETFE film image, perform convolution operations on the image using Gaussian kernels of multiple scales, linearly superimpose multiple sets of convolution results according to the preset weight coefficients, and generate a multi-scale edge-enhanced image; S102: Based on the multi-scale edge-enhanced image, call the Sobel operator to calculate the pixel gradient components, synthesize the gradient amplitude matrix, set a dynamic threshold according to the statistical characteristics of the gradient amplitude distribution, screen the pixel points whose gradient amplitude exceeds the threshold, and generate a set of gradient-exceeding pixels; S103: For the set of gradient-exceeding pixels, perform a morphological closing operation to connect adjacent pixel points, extract the circumscribed rectangle boundary of the closed region, eliminate small-area isolated regions, retain the remaining path segments, and generate candidate cutting path segments.

3. The method for controlling the cutting of ETFE film based on image recognition according to claim 2, wherein, The specific steps for obtaining the path segment direction continuity sequence are as follows: S201: Invoke the horizontal and vertical convolution kernel matrices to perform point-by-point convolution operations on adjacent pixel points of the candidate cropping path segment, record the horizontal gradient component and the vertical gradient component, and generate a gradient direction vector set; S202: Calculate the direction angles of multiple pixel points based on the gradient direction vector set, define the difference in direction angles between adjacent pixel points and store them in a queue in sequence to construct a direction angle difference queue; S203: Invoke the direction angle difference queue and use the formula: ； Calculate the path segment continuity score, compare with a preset threshold to filter the path segments, and generate a path segment direction continuity sequence; Among them, represents the path segment direction continuity score, is the difference in the direction angles of adjacent pixel points, , is the gradient magnitude of adjacent pixel points, is the Euclidean distance between adjacent pixel points, is the total number of adjacent pixel point pairs within the path segment.

4. The method for controlling the cutting of ETFE film based on image recognition according to claim 3, wherein, The specific steps for obtaining the abnormal fluctuation segment are as follows: S301: Obtain the angle deviation values of 5 consecutive pixel points in the path segment direction continuity sequence, traverse the path segment in a window sliding manner, calculate the range of the angle deviation values within each window, and generate a dynamic window range value; S302: Invoke the dynamic window range value, based on the difference in the range change trend between adjacent windows and the path segment curvature change rate, use the formula: ； Perform operations to obtain a range fluctuation determination coefficient, and generate a range fluctuation determination coefficient through weighted superposition; Among them, represents the range fluctuation determination coefficient, represents the maximum angle deviation within the window, represents the minimum angle deviation within the window, represents the mean value of the difference in the range change trend between adjacent windows, represents the curvature change rate of the path segment, represents the environmental interference factor; S303: Invoke the range fluctuation determination coefficient, compare it with a preset path fluctuation determination threshold. If the coefficient exceeds the threshold, mark the path segment covered by the corresponding window as an abnormal fluctuation segment.

5. The method for controlling the cutting of ETFE film based on image recognition according to claim 4, wherein The method further includes: S5: According to the calibrated path data, invoke the B-spline curve interpolation algorithm to perform smooth fitting on the connection of path segments, generate an ETFE film cutting trajectory, drive the multi-axis linkage parameters of the cutting head based on the coordinate differences between adjacent path points in the trajectory, and output a cutting instruction; The ETFE film cutting trajectory is specifically the B-spline curve parameters and multi-axis linkage parameters.

6. The method for controlling the cutting of ETFE film based on image recognition according to claim 5, characterized in that, The specific steps for obtaining the ETFE film cutting trajectory are as follows: S501: Invoke the calibrated path data, extract the control point coordinates and curvature parameters at the connection of path segments, calculate the distance and angle difference between adjacent control points, set the knot vector and weight coefficients, and generate a path node parameter set; S502: Based on the path node parameter set, divide the interpolation interval and solve the cubic polynomial coordinate components of each interval, adjust the high-order derivative terms of the polynomial coefficient matrix to meet the curvature continuity condition, and generate a fitting trajectory coordinate set; S503: Traverse the fitting trajectory coordinate set, calculate the curvature change rate between adjacent interpolation points, filter out the jump points that exceed the continuity threshold, re-interpolate and compensate, and output a path sequence to generate an ETFE film cutting trajectory.

7. An ETFE film cutting control system based on image recognition, characterized in that, The system is used to implement the image recognition-based ETFE film cutting control method according to any one of claims 1-6. The system includes: An edge enhancement module, which is used to perform filtering and noise reduction on the ETFE film image through a multi-scale Gaussian filtering algorithm, invoke the Sobel gradient operator to calculate the horizontal and vertical gray scale change rates, filter out a set of continuous pixel points whose gray scale change rates exceed the dynamic threshold, generate a candidate cropping path segment and output its start and end point coordinates, and transfer the candidate cropping path segment to the direction continuity analysis module; The direction continuity analysis module is used to construct the gradient direction vectors of each pixel point for the adjacent pixel points of the candidate cutting path segment by using the vector space projection algorithm, calculate the direction angle between adjacent vectors, generate the path segment direction continuity sequence, store the angle deviation value between the angle and the preset direction, and transfer the path segment direction continuity sequence to the abnormal fluctuation determination module; The abnormal fluctuation determination module is used to perform a sliding window calculation on the angle deviation values of 5 consecutive pixel points by calling the dynamic window range difference determination method based on the path segment direction continuity sequence, determine whether the range difference within the window exceeds the set threshold, mark the path segment with an excessive range difference as an abnormal fluctuation segment and generate a calibration instruction, and transfer the abnormal fluctuation segment to the path calibration module; The path calibration module is used to extract the mean value of the gradient direction vectors of the adjacent path segments of the abnormal fluctuation segment, replace the direction vectors of all control points within the fluctuation segment with the mean value vector through the vector replacement strategy, generate the calibrated path data and update the motion trajectory parameters of the cutting device, and transfer the calibrated path data to the trajectory generation module; The trajectory generation module is used to perform smooth fitting on the connection of the path segments by calling the B-spline curve interpolation algorithm based on the calibrated path data, calculate the coordinate difference between adjacent path points to generate the ETFE film cutting trajectory, and drive the multi-axis linkage parameters of the cutting head to output a cutting instruction.

Images