A method for extracting a center line of a laser stripe based on Chebyshev moments

Abstract

The application discloses a laser stripe center line extraction method based on Chebyshev moments, and belongs to the technical field of line structured light vision detection. The Scharr gradient operator is used to detect the laser stripe edge, and then searching is performed along the normal direction of the edge point, so that the laser stripe cross-section gray distribution information is obtained. On the basis of analyzing the laser stripe cross-section gray distribution characteristics, a laser stripe cross-section gray distribution model is constructed, and the laser stripe cross-section center sub-pixel coordinates are solved by using Chebyshev moments, and the laser stripe center line is obtained by connecting the laser stripe cross-section centers. Compared with the Steger method, the gray gravity center method, the spatial moment method and the Legendre moment method, when the gray saturation laser stripe is processed, the center line detection precision can be improved while the algorithm real-time is ensured, the balance between the algorithm precision and speed is realized, and the measurement requirements of the line structured light measurement system can be met in the actual use.

Description

Technical Field

[0001] This invention belongs to the field of line structured light visual inspection technology, specifically relating to a method for extracting the center line of laser stripes based on Chebyshev moments. Background Technology

[0002] Structured light vision measurement technology is an active vision measurement technology that uses an additional light source (such as point, line, and surface structured light) to illuminate the surface of the object being measured, and measures the contour features of the object under the guidance of the additional light source. It has advantages such as high real-time performance and high accuracy, and is widely used in fields such as robotics, aerospace, medical, autonomous driving, and weld inspection.

[0003] Laser fringe centerline extraction algorithms are indispensable in line structured light measurement systems. They are primarily used for extracting laser plane feature points during line structured light system calibration and for extracting surface contour features of the measured object during line structured light image processing. The accuracy and real-time performance of the laser fringe centerline extraction method directly affect the measurement performance of the line structured light measurement system. Therefore, line structured light systems place stringent requirements on laser fringe centerline extraction algorithms: high accuracy and strong real-time performance. Currently, laser fringe centerline extraction mainly employs image morphology-based and grayscale distribution-based algorithms. While image morphology-based algorithms are faster, they only consider the morphological information of the laser fringe and do not utilize the grayscale information of the laser fringe cross-section in the image. Therefore, these methods have low accuracy under complex conditions and cannot meet the accuracy requirements of line structured light vision measurement systems. Algorithms based on grayscale distribution include the Steger method and the grayscale centroid method. While the Steger method offers high extraction accuracy, center extraction requires solving the Hessian matrix, necessitating numerous convolution operations and resulting in long detection times, which cannot meet the accuracy requirements of line structured light vision measurement systems. The grayscale centroid method and its improved algorithms, while computationally simple and fast, can meet the requirements of line structured light measurement systems to some extent. However, this method directly uses the centroid of the laser stripe cross-section as the center, leading to lower accuracy. Furthermore, traditional centerline extraction algorithms based on grayscale distribution assume that the grayscale distribution of the laser stripe cross-section follows a Gaussian distribution, but actual light stripes do not have an ideal Gaussian distribution. This results in grayscale saturation ("flat top") at certain pixels.

[0004] Traditional laser stripe centerline extraction algorithms sacrifice speed for high accuracy, and vice versa. Therefore, designing a line structured light measurement system that offers both high real-time performance and high accuracy requires an algorithm that balances speed and accuracy. Summary of the Invention

[0005] To address the problem that current laser stripe centerline extraction algorithms cannot simultaneously meet the requirements of high accuracy and strong real-time performance, this invention provides a laser stripe centerline extraction method based on Chebyshev moments.

[0006] To achieve the above objectives, the present invention employs the following technical solutions:

[0007] A method for extracting the centerline of laser stripes based on Chebyshev moments is proposed. This method utilizes the Scharr gradient operator to detect the edges of the laser stripes, and then searches along the normal direction of the edge points to obtain the grayscale distribution information of the laser stripe cross-section. Based on the analysis of the grayscale distribution characteristics of the laser stripe cross-section, a grayscale distribution model of the laser stripe cross-section is constructed, and the sub-pixel coordinates of the center of the laser stripe cross-section are solved using Chebyshev moments. Connecting the centers of each laser stripe cross-section yields the centerline of the laser stripe. The specific steps include:

[0008] Step 1: Convert the color image of the laser stripes to grayscale, reduce image noise using a smoothing filter operator, obtain the laser stripe region, and crop the laser stripe region as the region of interest.

[0009] Step 2: Use the Scharr edge operator to detect the edges of the laser stripe region and obtain the coordinates of the laser stripe edges. The gray-level gradient at the edge of the laser stripe is calculated, and then the normal direction of the edge of the laser stripe is calculated.

[0010] Step 3: Starting from the laser stripe edge point obtained in Step 2, search along the normal direction of the laser stripe edge. If the current point is another edge point corresponding to the normal direction of the current edge point, the search ends, thus obtaining the current grayscale cross-section distribution information, i.e., the set of grayscale values ​​of cross-section pixels. , , Take natural numbers;

[0011] Step 4: Based on the grayscale distribution of the laser stripe cross section, construct a grayscale distribution model of the laser stripe cross section, and use Chebyshev moments to obtain the center position of the laser stripe;

[0012] Step 5: Transform the laser stripe cross-section using the size transformation formula to obtain the true coordinates of the center.

[0013] Step 6: Repeat steps 3 and 5 to obtain grayscale distribution information of different laser stripe cross sections in turn, and calculate the sub-pixel coordinates of the center point of each laser stripe cross section. Connect these center points to obtain the center line of the laser stripe.

[0014] Furthermore, in step 2, the edges of the laser stripes in the laser stripe region are detected using the Scharr edge operator to obtain the coordinates of the laser stripe edges. The gray-level gradient at the edge points of the laser stripes is calculated, and then the normal direction of the laser stripe edge points is calculated. The specific method is as follows:

[0015] First, the gray-level gradient of the laser stripe edge points is calculated using the Scharr edge operator; the gray-level gradient of the edge points is expressed as... ,

[0016]

[0017]

[0018] Then calculate the normal direction of the edge points;

[0019] like The normal direction of the edge point is ;

[0020] like The normal direction of the edge point is .

[0021] Furthermore, step 3 starts from the laser stripe edge point obtained in step 2 and searches along the normal direction of the laser stripe edge. If the current point is another edge point corresponding to the normal direction of the edge point, the search ends. The specific method for obtaining the current grayscale cross-sectional distribution information is as follows:

[0022] like The normal direction of the edge point is The search method is as follows: , Its sign is the same as the normal direction. If the current point is the starting edge point and the corresponding edge point along the normal direction is another edge point, the search stops.

[0023] like The normal direction of the edge point is The search method is as follows: , Its sign is the same as the normal direction. If the current point is the starting edge point and the corresponding edge point along the normal direction, the search stops.

[0024] Furthermore, the grayscale distribution model of the laser stripe cross section in step 4 is expressed as follows:

[0025]

[0026] in: The length of the lower base of the trapezoid. The length of the upper base of the trapezoid. The x-coordinate of the trapezoid's centerline is... The background grayscale value, Let denoted as , where is the difference between the grayscale value at the flat top of the laser stripe and the background grayscale value, and 'x' represents the coordinate value of the pixel along the cross-section. Analysis of the laser stripe reveals a "flat top" phenomenon at grayscale saturation. The grayscale value change between the background pixel and the laser stripe saturation point is not a direct jump from the background grayscale to the maximum grayscale value, but rather a gradual transition from the background grayscale value to the grayscale saturation value. Therefore, this invention proposes a model to consider the "flat top" phenomenon and the gradual grayscale change process of the actual grayscale distribution across the laser stripe cross-section.

[0027] Furthermore, in the laser stripe cross-section grayscale distribution model, when A and B take specific values, the laser stripe grayscale distribution model is a rectangular pulse distribution or a triangular (roof) distribution; when A=B≠0, the laser stripe cross-section grayscale distribution is a rectangular pulse distribution; when A≠0 and B=0, the laser stripe cross-section grayscale distribution is a triangular (roof) distribution.

[0028] Furthermore, in step 4, the center position of the laser stripe is obtained using the Chebyshev moment. The specific method is as follows:

[0029] The formula for calculating Chebyshev moments is:

[0030]

[0031] in , The polynomials are Chebyshev polynomials, and their first four orders are:

[0032]

[0033] Based on the Chebyshev moment calculation formula and laser grayscale distribution information Calculate the first four Chebyshev moments corresponding to the actual laser stripe cross-section grayscale distribution model:

[0034]

[0035]

[0036]

[0037]

[0038] Based on the Chebyshev moment calculation formula and the laser fringe grayscale distribution model, calculate the first four Chebyshev moments corresponding to the ideal laser fringe cross-section grayscale distribution model:

[0039]

[0040]

[0041]

[0042]

[0043] From the first four Chebyshev moments corresponding to the actual laser fringe cross-section gray-level distribution model and the first four Chebyshev moments corresponding to the ideal laser fringe cross-section gray-level distribution model, we obtain:

[0044]

[0045] Simplifying the above equation, we get:

[0046] By the principle of moment invariance of Chebyshev's moments:

[0047]

[0048] get:

[0049] make , , , have to:

[0050]

[0051] like hour, This represents the coordinate value of the pixel in the cross-sectional direction. Compared to spatial moments and Legendre moments, Chebyshev moments offer higher extraction accuracy; therefore, this invention uses Chebyshev moments to solve for the coordinates of the centerline point.

[0052] Furthermore, the specific method for obtaining the true coordinates of the center of the laser stripe cross-section through size transformation formula is as follows:

[0053] The formula for size transformation is:

[0054]

[0055] In the formula, These are the actual coordinates of the center line point of the current laser stripe cross section; These are the coordinates of the starting point of the current laser stripe cross section. These are the coordinates of the current laser stripe cross-section termination point. Both coordinates can be obtained through... get.

[0056] Furthermore, before converting the laser stripe color image to grayscale, the laser stripe image is preprocessed, including grayscale conversion, noise reduction, scaling, and cropping.

[0057] The centerline data obtained by the above-mentioned Chebyshev-based laser strip centerline extraction method is used for line structured light machine vision inspection.

[0058] Compared with the prior art, the present invention has the following advantages:

[0059] This invention proposes a laser stripe center extraction method based on Chebyshev moments. It utilizes the Scharr gradient operator to detect the laser stripe edges and then searches along the normal direction of the edge points to obtain the grayscale distribution information of the laser stripe cross-section. Based on the inductive analysis of the grayscale distribution characteristics of the laser stripe cross-section, a double-sloping (trapezoidal) grayscale distribution model of the laser stripe cross-section is designed. The sub-pixel coordinates of the center of the laser stripe cross-section are solved using Chebyshev moments. Connecting the centers of each laser stripe cross-section yields the laser stripe centerline. Compared to the Steger method, grayscale centroid method, spatial moment method, and Legendre moment method, this invention significantly improves the extraction accuracy of the center extraction algorithm while maintaining real-time performance when the laser stripe exhibits a saturated "flat-top" phenomenon. It achieves a good balance between algorithm accuracy and speed and can also meet the measurement requirements of line structured light measurement systems in practical applications. Attached Figure Description

[0060] To more effectively illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the embodiments will be described in detail below.

[0061] Figure 1 This is a flowchart of the laser stripe center extraction method based on Chebyshev moments according to the present invention.

[0062] Figure 2 This is a schematic diagram of the Scharr operator template of the present invention;

[0063] Figure 3 This is a flowchart illustrating the process of obtaining grayscale distribution information on the cross-section of a laser stripe according to the present invention.

[0064] Figure 4 This is a schematic diagram of the laser stripe cross-section pixel points searched by the present invention;

[0065] Figure 5 This is a statistical analysis diagram of the grayscale distribution of laser stripe cross sections according to the present invention;

[0066] Figure 6 This is a grayscale distribution model diagram of a laser stripe cross section with a double-sloping (trapezoidal) shape according to the present invention;

[0067] Figure 7 This is a flowchart of the Chebyshev moment solution process of the present invention;

[0068] Figure 8 These are images showing the laser stripe center extraction results of various methods in the embodiments of the present invention. Detailed Implementation

[0069] To make the objectives, techniques, and advantages of this invention clearer, the specific embodiments of this invention will be described in detail below with reference to the accompanying drawings. Obviously, the accompanying drawings described below are merely some embodiments described in this invention, and not all embodiments. Other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort should fall within the scope of protection of this invention.

[0070] Example 1

[0071] The specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. The process of the laser stripe centerline extraction method based on Chebyshev moment is as follows: Figure 1 The diagram illustrates the process of extracting the center using the method of the present invention.

[0072] A method for extracting the centerline of laser stripes based on Chebyshev moments is proposed. This method utilizes the Scharr gradient operator to detect the edges of the laser stripes, and then searches along the normal direction of the edge points to obtain the grayscale distribution information of the laser stripe cross-section. Based on the analysis of the grayscale distribution characteristics of the laser stripe cross-section, a grayscale distribution model of the laser stripe cross-section is constructed, and the sub-pixel coordinates of the center of the laser stripe cross-section are solved using Chebyshev moments. Connecting the centers of each laser stripe cross-section yields the centerline of the laser stripe. The specific steps include:

[0073] Step 1: Convert the color image of the laser stripes to grayscale, reduce image noise using a smoothing filter operator, obtain the laser stripe region, and crop the laser stripe region as the region of interest.

[0074] Step 2: Use the Scharr edge operator to detect the edges of the laser stripe region and obtain the coordinates of the laser stripe edges. The gray-level gradient at the edge of the laser stripe is calculated, and then the normal direction of the edge of the laser stripe is calculated.

[0075] First, the gray-level gradient of the laser stripe edge points is calculated using the Scharr edge operator; the gray-level gradient of the edge points is expressed as... ,

[0076]

[0077]

[0078] Then calculate the normal direction of the edge points;

[0079] like The direction of the normal to the edge point is ;

[0080] like The direction of the normal to the edge point is .

[0081] Step 3: Starting from the laser stripe edge point obtained in Step 2, search along the normal direction of the laser stripe edge. If the current point is another edge point corresponding to the normal direction of the current edge point, the search ends, thus obtaining the current grayscale cross-section distribution information, i.e., the set of grayscale values ​​of cross-section pixels. , , Take natural numbers;

[0082] like The normal direction of the edge point is The search method is as follows: , Its sign is the same as the normal direction. If the current point is the starting edge point and the corresponding edge point along the normal direction is another edge point, the search stops.

[0083] like The normal direction of the edge point is The search method is as follows: , Its sign is the same as the normal direction. If the current point is the starting edge point and the corresponding edge point along the normal direction, the search stops.

[0084] Step 4: Based on the grayscale distribution of the laser stripe cross section, construct a grayscale distribution model of the laser stripe cross section, and use Chebyshev moments to obtain the center position of the laser stripe;

[0085] The grayscale distribution model of the laser stripe cross section is expressed as:

[0086]

[0087] in: The length of the lower base of the trapezoid. The length of the upper base of the trapezoid. The x-coordinate of the trapezoid's centerline is... The background grayscale value, Let denoted as , where is the difference between the grayscale value at the flat top of the laser stripe and the background grayscale value, and 'x' represents the coordinate value of the pixel along the cross-section. Analysis of the laser stripe reveals a "flat top" phenomenon at grayscale saturation. The grayscale value change between the background pixel and the laser stripe saturation point is not a direct jump from the background grayscale to the maximum grayscale value, but rather a gradual transition from the background grayscale value to the grayscale saturation value. Therefore, this invention proposes a model to consider the "flat top" phenomenon and the gradual grayscale change process of the actual grayscale distribution across the laser stripe cross-section.

[0088] Furthermore, in the laser stripe cross-section grayscale distribution model, when A and B take specific values, the laser stripe grayscale distribution model is a rectangular pulse distribution or a triangular (roof) distribution; when A=B≠0, the laser stripe cross-section grayscale distribution is a rectangular pulse distribution; when A≠0 and B=0, the laser stripe cross-section grayscale distribution is a triangular (roof) distribution.

[0089] The formula for calculating Chebyshev moments is:

[0090]

[0091] in , The polynomials are Chebyshev polynomials, and their first four orders are:

[0092]

[0093] Based on the Chebyshev moment calculation formula and laser grayscale distribution information Calculate the first four Chebyshev moments corresponding to the actual laser stripe cross-section grayscale distribution model:

[0094]

[0095]

[0096]

[0097]

[0098] Based on the Chebyshev moment calculation formula and the laser fringe grayscale distribution model, calculate the first four Chebyshev moments corresponding to the ideal laser fringe cross-section grayscale distribution model:

[0099]

[0100]

[0101]

[0102]

[0103] From the first four Chebyshev moments corresponding to the actual laser fringe cross-section gray-level distribution model and the first four Chebyshev moments corresponding to the ideal laser fringe cross-section gray-level distribution model, we obtain:

[0104]

[0105] Simplifying the above equation, we get:

[0106] By the principle of moment invariance of Chebyshev's moments:

[0107]

[0108] get:

[0109] make , , , have to:

[0110]

[0111] like hour, The coordinates of the centerline point are given. Compared to spatial moments and Legendre moments, Chebyshev moments offer higher extraction accuracy; therefore, this invention uses Chebyshev moments to solve for the coordinates of the centerline point.

[0112] Step 5: Transform the laser stripe cross-section using the size transformation formula to obtain the true coordinates of the center; the size transformation formula is:

[0113]

[0114] In the formula, These are the actual coordinates of the center line point of the current laser stripe cross section; These are the coordinates of the starting point of the current laser stripe cross section. These are the coordinates of the current laser stripe cross-section termination point. Both coordinates can be obtained through... get.

[0115] Step 6: Repeat steps 3 and 5 to obtain grayscale distribution information of different laser stripe cross sections in turn, and calculate the sub-pixel coordinates of the center point of each laser stripe cross section. Connect these center points to obtain the center line of the laser stripe.

[0116] Example 2: Performance comparison test with existing technology:

[0117] 1. Centerline extraction accuracy comparison test

[0118] Five algorithms were used to extract the center lines of 10 sets of images (20 images per set) containing laser stripes. The extraction results of the laser stripe center lines by each method are as follows: Figure 8 As shown, since the laser stripes in the image are approximately straight lines, the extracted center line points are fitted with straight lines, and the root mean square error between the center point and the fitted straight line is taken as the test standard.

[0119] Table 1. Root Mean Square Error (RMSE) of each method

[0120]

[0121] As can be seen from the table above, in this embodiment, the method is significantly better than the gray-scale centroid method, spatial moment method and Legendre moment method in terms of extraction accuracy, and is on par with the Steger method with the highest accuracy.

[0122] The centerline extraction results of each method are as follows: Figure 8 As shown in the figure, the center lines extracted by the gray-scale centroid method, spatial moment method, and Legendre moment method exhibit significant jitter, indicating that these methods have low extraction accuracy. Although the center lines extracted by the Steger method show almost no jitter, they are broken. The center lines extracted by this method not only have very little jitter but also do not have any broken lines, resulting in excellent extraction performance.

[0123] 2. Runtime Comparison Test

[0124] Five different algorithms were used to extract the center of 200 images containing laser stripes, and the average running time was used as the test standard.

[0125] Table 2 Average running time of each method

[0126]

[0127] As can be seen from the table above, in this embodiment, the method is significantly better than the Steger method in terms of running time, which is only half that of the Steger method and almost on par with the gray-scale centroid method with the shortest running time, thus demonstrating good real-time performance.

[0128] Contents not described in detail in this specification are prior art known to those skilled in the art. Although illustrative specific embodiments of the invention have been described above to facilitate understanding by those skilled in the art, it should be understood that the invention is not limited to the scope of the specific embodiments. Various modifications are readily apparent to those skilled in the art as long as they fall within the spirit and scope of the invention as defined and determined by the appended claims, and all inventions utilizing the concept of this invention are protected.

Claims

1. A method for extracting the center line of laser stripes based on Chebyshev moments, characterized in that: The edges of the laser stripes are detected using the Scharr gradient operator, and then a search is performed along the normal direction of the edge points to obtain the gray-scale distribution information of the laser stripe cross section. Based on the analysis of the gray-scale distribution characteristics of the laser stripe cross section, a gray-scale distribution model of the laser stripe cross section is constructed, and the sub-pixel coordinates of the center of the laser stripe cross section are solved using Chebyshev moments. The center lines of the laser stripe cross sections are then connected to obtain the center line of the laser stripe. Specifically, the following steps are included: Step 1: Convert the color image of the laser stripes to grayscale, reduce image noise using a smoothing filter operator, obtain the laser stripe region, and crop the laser stripe region as the region of interest. Step 2: Use the Scharr edge operator to detect the edges of the laser stripe region and obtain the coordinates of the laser stripe edges. The gray-level gradient at the edge of the laser stripe is calculated, and then the normal direction of the edge of the laser stripe is calculated. Step 3: Starting from the laser stripe edge point obtained in Step 2, search along the normal direction of the laser stripe edge. If the current point is another edge point corresponding to the normal direction of the current edge point, the search ends, thus obtaining the current grayscale cross-section distribution information, i.e., the set of grayscale values ​​of cross-section pixels. , , Take natural numbers; Step 4: Based on the grayscale distribution of the laser stripe cross section, construct a grayscale distribution model of the laser stripe cross section, and use Chebyshev moments to obtain the center position of the laser stripe; Step 5: Transform the laser stripe cross-section using the size transformation formula to obtain the true coordinates of the center. Step 6: Repeat steps 3 and 5 to obtain grayscale distribution information of different laser stripe cross sections, and calculate the sub-pixel coordinates of the center point of each laser stripe cross section. Connect these center points to obtain the center line of the laser stripe. The grayscale distribution model of the laser stripe cross section in step 4 is expressed as follows: ； in: The length of the lower base of the trapezoid. The length of the upper base of the trapezoid. The x-coordinate of the trapezoid's centerline is... The background grayscale value, denoted as the difference between the grayscale value at the top of the laser stripe and the grayscale value of the background, where x represents the coordinate value of the pixel in the cross-sectional direction; In step 4, the center position of the laser stripe is obtained using Chebyshev moments. The specific method is as follows: The formula for calculating Chebyshev moments is: ； in , The polynomials are Chebyshev polynomials, and their first four orders are: ； Based on the Chebyshev moment calculation formula and laser grayscale distribution information Calculate the first four Chebyshev moments corresponding to the actual laser stripe cross-section grayscale distribution model: ； ； ； ； Based on the Chebyshev moment calculation formula and the laser fringe grayscale distribution model, calculate the first four Chebyshev moments corresponding to the ideal laser fringe cross-section grayscale distribution model: ； ； ； ； From the first four Chebyshev moments corresponding to the actual laser fringe cross-section gray-level distribution model and the first four Chebyshev moments corresponding to the ideal laser fringe cross-section gray-level distribution model, we obtain: ； Simplifying the above equation, we get: ; By the principle of moment invariance of Chebyshev's moments: ； get: ; make , , , have to: ; ； like hour, Represents the coordinate value of a pixel in the cross-section direction.

2. The method for extracting the center line of laser stripes based on Chebyshev moments according to claim 1, characterized in that: In step 2, the Scharr edge operator is used to detect the edges of the laser stripes in the laser stripe region, and the coordinates of the laser stripe edges are obtained. The gray-level gradient at the edge points of the laser stripes is calculated, and then the normal direction of the laser stripe edge points is calculated. The specific method is as follows: First, the gray-level gradient of the laser stripe edge points is calculated using the Scharr edge operator; the gray-level gradient of the edge points is expressed as... , ； Then calculate the normal direction of the edge points; like The normal direction of the edge point is ; like The normal direction of the edge point is .

3. The method for extracting the center line of laser stripes based on Chebyshev moments according to claim 1, characterized in that: Step 3 starts from the laser stripe edge point obtained in step 2 and searches along the normal direction of the laser stripe edge. If the current point is another edge point corresponding to the normal direction of the edge point, the search ends. The specific method for obtaining the current grayscale cross-sectional distribution information is as follows: like The normal direction of the edge point is The search method is as follows: , Its sign is the same as the normal direction. If the current point is the starting edge point and the corresponding edge point along the normal direction is another edge point, the search stops. like The direction of the normal to the edge point is The search method is as follows: , Its sign is the same as the normal direction. If the current point is the starting edge point and the corresponding edge point along the normal direction, the search stops.

4. The method for extracting the center line of laser stripes based on Chebyshev moments according to claim 1, characterized in that: In the laser stripe cross-section grayscale distribution model, when A and B take specific values, the laser stripe grayscale distribution model is a rectangular pulse distribution or a triangular distribution; when A=B≠0, the laser stripe cross-section grayscale distribution is a rectangular pulse distribution; when A≠0 and B=0, the laser stripe cross-section grayscale distribution is a triangular distribution.

5. The method for extracting the center line of laser stripes based on Chebyshev moments according to claim 1, characterized in that: The specific method for obtaining the true coordinates of the center of the laser stripe cross section by transforming it using the size transformation formula is as follows: The formula for size transformation is: In the formula, These are the actual coordinates of the center line point of the current laser stripe cross section; These are the coordinates of the starting point of the current laser stripe cross section. These are the coordinates of the current laser stripe cross-section termination point. Both coordinates can be obtained through... get.

6. The method for extracting the center line of laser stripes based on Chebyshev moments according to claim 1, characterized in that: Before converting the laser stripe color image to grayscale, the laser stripe image is preprocessed, including grayscale conversion, noise reduction, scaling, and cropping.

7. A method for extracting the center line of laser stripes based on Chebyshev moments according to any one of claims 1 to 6, characterized in that, The centerline data obtained by the laser strip centerline extraction method based on Chebyshev moments is used for line structured light machine vision inspection.

Images