Visual positioning method for welding of thermoplastic material and metal material

Abstract

The invention relates to the technical field of image processing, in particular to a visual positioning method for welding of a thermoplastic material and a metal material. According to the method, the moving track of the butt-joint material in the butt-joint process is monitored in real time and recorded, a moving track curve with obvious characteristics is finally formed, the positioning error of the butt-joint material in the moving process can be effectively analyzed based on the moving track curve, and in the welding process of the two materials, the material accumulation condition of the end face area is detected, so that the welding accuracy is improved. The positioning deformation in the welding process can be reflected through the material stacking condition, that is, the positioning adjustment function can be finally constructed through the end face stacking unevenness and the positioning error, the positioning defects generated in the butt joint process and the welding process can be adjusted and controlled through the positioning adjustment function, and then the welding quality is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and particularly to a welding vision positioning method for thermoplastic materials and metal materials. Background Art

[0002] As a new type of lightweight material, thermoplastic composites have been widely used in the aerospace field. The joining of dissimilar materials of metal and thermoplastic composites has received extensive attention because it can give full play to the respective advantages of the two materials. Friction stir welding, as a welding technology with low heat input, shows great application potential in the field of dissimilar welding of metal and thermoplastic composites. This process uses the frictional heat between a high-speed rotating stirring head and the workpiece as the heat input for welding, and has the advantages of short process cycle, simple operation and no environmental pollution. Because there are significant differences in the thermal expansion coefficient, thermal conductivity, melting point, etc. between thermoplastic materials and metal materials, problems such as thermal stress and deformation are likely to occur during the welding process; therefore, high-precision positioning technology is crucial for ensuring the accuracy of the welding position. Through a vision positioning system and image processing algorithms, a positioning accuracy of sub-millimeter level can be achieved. Through real-time monitoring and feedback, the errors during the welding process are compensated to improve the welding accuracy and consistency.

[0003] During the actual positioning process, because the material itself is not in a straight shape, the positioning error will occur due to the shape problem during the butt-friction process of the material, and the positioning error will cause uneven friction heating, which will further lead to welding defects during the end-face welding process. Summary of the Invention

[0004] In order to solve the technical problems in the prior art that the positioning error during the docking process cannot be recognized and the friction deformation during the welding process lead to poor welding effects, the purpose of the present invention is to provide a welding vision positioning method for thermoplastic materials and metal materials, and the specific technical solutions adopted are as follows:

[0005] The present invention proposes a welding vision positioning method for thermoplastic materials and metal materials, and the method includes:

[0006] Taking any one of the thermoplastic material and the metal material as the fixed material, and the other material as the docking material; during the process of the docking material moving to the fixed material for welding, obtaining the moving trajectory curve of the docking material and the welding end-face image;

[0007] Obtaining the positioning error according to the volatility of the moving trajectory curve; in the end-face area of the welding end-face image, constructing a search window with the end-face area as the center, and obtaining the end-face stacking non-uniformity according to the pixel value change between different preset areas within the search window;

[0008] Construct a positioning adjustment function based on the Gaussian function according to the unevenness of the end face stacking and the positioning error; use the positioning adjustment function to control the positioning of the two welding materials.

[0009] Further, the method for obtaining the movement trajectory curve includes:

[0010] Collect the movement images of the docking material during the docking process by the fixed camera installed on the fixed material; the end face of the docking material is included in the movement images. Taking the centroid point of the end face of the docking material as the reference point, use the horizontal distance during each movement of the reference point as the abscissa and the vertical distance as the ordinate to draw the movement trajectory curve.

[0011] Further, the method for obtaining the positioning error includes:

[0012] Obtain the positioning feature of the reference point in the movement image according to the coordinates of the reference point, and obtain the change fluctuation degree of the positioning feature during the movement of the docking material;

[0013] Decompose the movement trajectory curve to obtain the period, trend term and residual term of the movement trajectory curve; obtain the curve volatility of the movement trajectory curve according to the period, trend term and residual term;

[0014] Obtain the positioning error according to the change fluctuation degree and the curve volatility.

[0015] Further, the method for obtaining the change fluctuation degree includes:

[0016] Take the product of the range and variance of the positioning feature during the movement of the docking material as the change fluctuation degree.

[0017] Further, the method for obtaining the curve volatility includes:

[0018] Multiply the variance of the data in the residual term and the mean value in the trend term, and take the ratio of the product to the period size as the curve volatility.

[0019] Further, the method for obtaining the unevenness of the end face stacking includes:

[0020] In the search window, construct analysis regions centered on the end face region in sequence according to a preset step size, and use the normal material region as the comparison region; analyze the pixel value difference between the analysis regions of each scale and the comparison region to obtain the end face stacking degree; analyze the pixel value distribution stability within the analysis regions of each scale to obtain the end face stacking roughness; obtain the unevenness of the end face stacking according to the end face stacking degree and the end face stacking roughness.

[0021] Furthermore, the method for obtaining the stacking degree of the end face includes:

[0022] The difference between the average gray value between each analysis region and the corresponding comparison region is used as the stacking material appearance feature of each analysis region; the accumulated value of all analysis regions is used as the stacking degree of the end face.

[0023] Furthermore, the method for obtaining the roughness of the end face stacking includes:

[0024] For each analysis region at each scale, the difference between the gray value of each pixel point in the analysis region and the average gray value of the analysis region is used as the gray deviation of each gray value; the average gray deviation of all pixel points in the analysis region is used as the roughness of the analysis region;

[0025] The average roughness of all analysis regions is used as the roughness of the end face stacking.

[0026] Furthermore, the expression of the positioning adjustment function is:

[0027] where f represents the value of the positioning adjustment function, K is the unevenness of the end face stacking, q represents the positioning error, X represents the position deviation between the docking material at the current position and the docking material at the previous moment, and e represents the natural constant; when the welding has not started, X is the positive integer 1.

[0028] Furthermore, the method for obtaining the positioning feature includes:

[0029] Obtain the coordinates of the reference point in the moving image, and use the Euclidean distance between the coordinates and the origin of the moving image as the positioning feature.

[0030] The present invention has the following beneficial effects:

[0031] The present invention monitors and records the moving trajectory of the docking material during the docking process in real time, and finally forms a moving trajectory curve with obvious features. Based on the moving trajectory curve, the positioning error of the docking material during movement can be effectively analyzed, and then the positioning control can be performed according to the positioning error. Further, during the welding process of the two materials, the material stacking situation in the end face area is detected, and the positioning deformation during the welding process can be reflected by the material stacking situation. That is, finally, a positioning adjustment function can be constructed through the unevenness of the end face stacking and the positioning error, and the positioning defects generated during the docking process and the welding process can be regulated by using the positioning adjustment function, thereby improving the welding quality. Description of the Drawings

[0032] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the following will briefly introduce the accompanying drawings required for the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0033] Figure 1 A welding schematic diagram of the welding process between a thermoplastic material and a metal material provided by an embodiment of the present invention;

[0034] Figure 2 A flowchart of a welding vision positioning method for a thermoplastic material and a metal material provided by an embodiment of the present invention;

[0035] Figure 3 A schematic diagram of a moving trajectory curve provided by an embodiment of the present invention. Detailed implementation manners

[0036] In order to further elaborate on the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following will, in conjunction with the accompanying drawings and preferred embodiments, describe in detail the specific implementation manners, structures, features and effects of a welding vision positioning method for a thermoplastic material and a metal material proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0037] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present invention belongs.

[0038] The application scenario of the embodiment of the present invention is the friction stir welding process of thermoplastic materials and metal materials. Please refer to Figure 1 , which shows a welding schematic diagram of the welding process between a thermoplastic material and a metal material provided by an embodiment of the present invention. In the welding process, one material will be selected as the fixed material. In the embodiment of the present invention, welding material B is used as the fixed material and a camera is deployed above the fixed material. Welding material A is the butt joint material. The two welding materials are fixed by a fixing jig, and the butt joint between welding material A and welding material B is controlled. After the butt joint is completed, high-speed rotation is carried out. A large amount of heat is generated by rotational friction, melting the contact surfaces of the two materials, and thus welding is achieved.

[0039] The following will specifically describe the specific solution of a welding vision positioning method for a thermoplastic material and a metal material provided by the present invention in conjunction with the accompanying drawings.

[0040] Please refer toFigure 2 , which shows a flowchart of a visual positioning method for welding a thermoplastic material and a metal material provided by an embodiment of the present invention. The method includes:

[0041] Step S1: Use either one of the thermoplastic material and the metal material as the fixed material, and the other as the docking material; during the process of moving the docking material to the fixed material for welding, obtain the moving trajectory curve of the docking material and the welding end face image.

[0042] In the embodiment of the present invention, the metal material is selected as the fixed material and the thermoplastic material is selected as the docking material. During the docking process of the docking material, the image of the docking material can be collected by computer vision method, and then the moving trajectory curve of the docking material can be drawn. According to the moving trajectory curve, information such as the positioning error during the movement of the docking material can be determined. Further, during the butt welding process, the welding end face image of the contact surface of the two materials is collected. In the subsequent steps, by determining the material accumulation situation of the welding end face, the positioning defects during the friction welding process can be further judged.

[0043] Preferably, in an embodiment of the present invention, the method for obtaining the moving trajectory curve includes:

[0044] Collect the moving image of the docking material during docking according to the fixed camera installed on the fixed material; the moving image includes the end face of the docking material. In the embodiment of the present invention, considering that the end face of the material processed by friction welding has a specific shape, such as circular, rectangular, etc., the moving image can be subjected to edge detection, and the end face area can be obtained by identifying the shape of the edge contour. It should be noted that methods such as edge detection and contour recognition are well-known technical means to those skilled in the art and will not be elaborated here.

[0045] Taking the centroid point of the end face of the docking material as the reference point, taking the horizontal distance during each movement of the reference point as the abscissa and the vertical distance as the ordinate, draw the moving trajectory curve. Please refer to Figure 3 , which shows a schematic diagram of a moving trajectory curve provided by an embodiment of the present invention.

[0046] In the embodiment of the present invention, the position data of the docking material is collected once per second, and then the moving trajectory curve is constructed.

[0047] Step S2: Obtain the positioning error according to the volatility of the moving trajectory curve; in the end face area in the welding end face image, construct a search window with the end face area as the center, and obtain the unevenness of end face accumulation according to the pixel value change between different preset areas within the search window.

[0048] During the docking process of the docking material and the fixing material, due to mechanical vibration and the shape of the material, there will be a certain bending deformation during the docking process, which will lead to the generation of positioning errors. The movement trajectory curve can characterize the position characteristics of the docking material during the movement process. Therefore, the positioning error during the docking process can be obtained based on the volatility of the movement trajectory curve, that is, the greater the volatility, the more positioning errors are generated.

[0049] Preferably, in the embodiments of the present invention, the method for obtaining the positioning error includes:

[0050] Obtain the positioning characteristics of the reference point in the moving image according to the coordinates of the reference point. The positioning characteristics are used to convert two-dimensional coordinate information into one-dimensional numerical feature data, which can facilitate the quantification of the position of the reference point. Furthermore, the degree of change and fluctuation of the positioning characteristics during the movement of the docking material can be obtained, that is, the degree of change and fluctuation can be obtained by statistically analyzing the changes in the positioning characteristics throughout the docking process.

[0051] In order to perform volatility analysis on the movement trajectory curve, the movement trajectory curve is decomposed to obtain the period, trend term, and residual term of the movement trajectory curve. The curve volatility of the movement trajectory curve is obtained based on the period, trend term, and residual term. That is, the smaller the period, the greater the number of changes on the curve, and the greater the volatility; the greater the trend term, the more complex the trend change, and the greater the curve volatility; the greater the residual term, the more obvious the fluctuation of the curve, and there is a certain deviation between the data, so the volatility is greater.

[0052] Obtain the positioning error based on the degree of change and fluctuation and the curve volatility. In the embodiments of the present invention, after quantifying both the degree of change and fluctuation and the curve volatility, the product of the two is used as the positioning error.

[0053] In the embodiments of the present invention, the method for obtaining the positioning characteristics includes:

[0054] Obtain the coordinates of the reference point in the moving image, and use the Euclidean distance between the coordinates and the origin of the moving image as the positioning characteristic. That is, the positioning characteristic of the reference point is effectively quantified through the Euclidean distance.

[0055] Furthermore, the method for obtaining the degree of change and fluctuation includes:

[0056] Since the position data of the docking material is obtained at a certain sampling frequency, the positioning characteristics also have a certain sampling frequency, and a set of positioning characteristics will be obtained during the docking process. The product of the range and variance of the positioning characteristics during the movement of the docking material is used as the degree of change and fluctuation. The greater the range, the greater the change amplitude of the positioning characteristics, and the greater the variance, the greater the degree of change of the positioning characteristics. Therefore, the product of the two can effectively characterize the change volatility.

[0057] Further, the method for obtaining the curve volatility includes:

[0058] Multiply the variance of the data in the residual term by the mean in the trend term, and use the ratio of the product to the period size as the curve volatility. In the embodiments of the present invention, the STL decomposition algorithm is used to decompose the moving trajectory curve. The specific method is a well-known technical means for those skilled in the art and will not be elaborated here. The embodiments of the present invention construct positive and negative correlation relationships through the product and ratio methods to effectively quantify the curve volatility.

[0059] During friction welding, by rotating two materials at high speed to generate friction, heat is generated to melt the end faces of the two materials for welding. In this process, due to the relatively soft nature of thermoplastic materials, they will produce greater deformation compared to metal materials. This deformation will cause uneven stress on the materials during friction welding, leading to a new positioning error and welding defects. During friction welding, in order to weld the welding port firmly, a certain horizontal pressure is applied to the welding material, resulting in a certain degree of surplus material deformation of the welding material, that is, the situation of material accumulation at the friction port, and the welding material will slowly move towards the direction of the welding port. Therefore, the welding quality can be determined by monitoring the surplus material accumulation situation in the end face area during friction welding, and then it can be judged whether an error occurs. That is, the greater the degree of surplus material accumulation, the more the positioning deviation during welding causes an increase in surplus material. Therefore, in the embodiments of the present invention, image processing technology is used to determine the end face area in the welding end face image, and then a search window is constructed with the end face area as the center. By analyzing the pixel value changes between different preset areas within the search window, the unevenness of the end face accumulation can be obtained. That is, surplus material accumulation will occur in the search window, and there will inevitably be an obvious pixel value change between the pixel values of the accumulation area and the non-accumulation area. Therefore, by analyzing between different areas in the search window, the unevenness of the end face accumulation of the surplus material on the end face can be effectively obtained.

[0060] In the embodiments of the present invention, since the welding end face image is an image of the welding process that has already started, the end faces of the two materials have already come into contact. Therefore, the end face area should form a line segment in the image, and the line segment formed by the end face area can be directly obtained through edge detection.

[0061] Preferably, in an embodiment of the present invention, the method for obtaining the unevenness of the end face accumulation includes:

[0062] In the search window, analysis regions are sequentially constructed centered on the end face region according to a preset step size. In an embodiment of the present invention, the search window is set as a region that expands 20 times to both sides centered on the end face region, and the step size of each expansion is one pixel point. That is, the final search window is a rectangular region with a length of 41 and a width equal to the length of the line segment formed by the end face region. Set the construction compensation of the analysis region to 2, that is, each time it expands outward by 2 pixel point units to form a region. Taking the analysis region constructed for the first time as an example, this analysis region should be a rectangular region with a length of 5 and a width equal to the length of the line segment formed by the end face region; then the analysis region constructed for the second time is a rectangular region with a length of 9 and a width equal to the length of the line segment formed by the end face region. Keep expanding until the analysis region coincides with the search window, and end the analysis.

[0063] Since the analysis region is expanded centered on the end face region, the accumulation of leftover materials will be preferentially characterized in the analysis region, while the comparison region is a normal material region where no leftover material accumulation occurs. The leftover material accumulation region will show bright characteristics due to the influence of high temperature, while the normal material region is relatively dark. Therefore, analyze the pixel value difference between the analysis region at each scale and other regions in the search window to obtain the end face accumulation degree. That is, the greater the pixel value difference between the two regions, the more likely the analysis region is a leftover material accumulation region, and the greater the end face accumulation degree.

[0064] The accumulation of leftover materials in the image will not only show bright characteristics but also present the texture characteristics of the surface of the leftover materials. Therefore, further analyze the pixel value distribution stability within the analysis region at each scale to obtain the end face accumulation roughness degree. That is, the worse the pixel value distribution stability within the analysis region, the richer the texture characteristics, indicating that the analysis region is more likely to be a leftover material accumulation region.

[0065] Obtain the end face accumulation non-uniformity based on the end face accumulation degree and the end face accumulation roughness degree. In an embodiment of the present invention, after quantifying the end face accumulation degree and the end face roughness degree, multiply them, and the product is used as the end face accumulation non-uniformity.

[0066] Furthermore, in an embodiment of the present invention, the method for obtaining the end face accumulation degree includes:

[0067] Take the difference between the average gray value of each analysis region and the corresponding comparison region as the stockpiling manifestation feature of each analysis region. That is, the greater the difference in the average gray value, the brighter the analysis region, indicating that the analysis region is more likely to be entirely formed by the accumulation of leftover materials. Take the cumulative value of all analysis regions as the end face accumulation degree.

[0068] Furthermore, in an embodiment of the present invention, the method for obtaining the end face accumulation roughness degree includes:

[0069] For each analysis region at each scale, the difference between the grayscale value of each pixel in the analysis region and the average grayscale value of the analysis region is used as the grayscale deviation of each grayscale value; the average grayscale deviation of all pixels in the analysis region is used as the roughness of the analysis region. The larger the average grayscale deviation, the more uneven the pixel value distribution within the analysis region, the stronger the texture feature, and the greater the roughness within the region.

[0070] Since there are multiple scales of analysis regions in the search window, the average roughness of all analysis regions is used as the rough degree of the end face stacking.

[0071] Step S3: Construct a positioning adjustment function based on the Gaussian function according to the unevenness of the end face stacking and the positioning error; use the positioning adjustment function to control the positioning of the two welding materials.

[0072] Based on the above steps, the analysis results of real-time visual monitoring based on image information during the butt joint process and the welding process can be obtained. Based on the unevenness of the end face stacking and the positioning error, a positioning adjustment function during the positioning process can be constructed. In the embodiment of the present invention, the high-speed function is used as the objective function. Through the special function shape of the Gaussian function, an effective control response to the positioning process can be achieved.

[0073] Preferably, in the embodiment of the present invention, the expression of the positioning adjustment function is:

[0074] Where f represents the value of the positioning adjustment function, K is the unevenness of the end face stacking, q represents the positioning error, X represents the position deviation between the butt joint materials at the current position and the butt joint materials at the previous moment, and e represents the natural constant; when the welding has not started, X is a positive integer 1. It should be noted that when the welding has not started, it means that this is the butt joint process at this time. There is no welding process and no analysis of the unevenness of the end face stacking. Therefore, X can be directly set to 1, only retaining the positioning error information.

[0075] It should be noted that since X represents the position deviation of the butt joint materials at adjacent moments, and in the embodiment of the present invention, the position of the butt joint materials determined during the butt joint process is obtained through the end face of the butt joint materials, so X can be set as the difference in the positioning features between two moments; while in the welding process, there is no complete material end face, only the line segment formed by the contact surface between the two materials. Therefore, taking the midpoint of the line segment as the positioning point, the difference in the positioning features of the positioning point can also be obtained to get X.

[0076] The control module of the welding device can adaptively control the adjustment amount and perform positioning adjustment with reference to the output result of the positioning adjustment function. The specific adjustment feedback method is a well-known technical means for those skilled in the art and will not be elaborated here.

[0077] In summary, the embodiments of the present invention monitor the movement trajectory of the butt-joint material during the butt-joint process in real time and record it, and finally form a movement trajectory curve with obvious features. Based on the movement trajectory curve, the positioning error of the butt-joint material during movement can be effectively analyzed. During the welding process of two materials, the material accumulation situation in the end face area is detected, and the positioning deformation during the welding process can be reflected through the material accumulation situation. That is, finally, a positioning adjustment function can be constructed through the unevenness of the end face accumulation and the positioning error. The present invention can use the positioning adjustment function to control the positioning defects generated during the butt-joint process and the welding process, thereby improving the welding quality.

[0078] It should be noted that the above sequence of the embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0079] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

Claims

1. A visual positioning method for welding thermoplastic materials and metal materials, characterized in that: The method comprises: Any one of the thermoplastic material and the metal material is used as a fixed material, and the other material is used as a butt joint material; in the process of the butt joint material moving to the fixed material for welding, a moving trajectory curve of the butt joint material and a welding end face image are obtained; The positioning error is obtained according to the volatility of the moving trajectory curve; in the end face area in the welding end face image, a search window is constructed with the end face area as the center, and the end face stacking unevenness is obtained according to the change of pixel values ​​between different preset areas in the search window; A positioning adjustment function is constructed based on the Gaussian function according to the end face stacking unevenness and the positioning error; and the positioning of the two welding materials is controlled by using the positioning adjustment function.

2. A method for visually positioning a thermoplastic material and a metal material for welding according to claim 1, characterized in that: The method for obtaining the moving trajectory curve includes: The moving image of the docking material during the docking process is collected according to the fixed camera installed on the fixed material; the moving image includes the end face of the docking material, and the center of mass of the end face of the docking material is used as the reference point, the lateral distance of the reference point during each movement is used as the horizontal coordinate, and the longitudinal distance is used as the vertical coordinate to draw the moving trajectory curve.

3. A visual positioning method for welding thermoplastic material and metal material according to claim 2, characterized in that: The method for obtaining the positioning error includes: Obtaining a positioning feature of the reference point in the moving image according to the coordinates of the reference point, and obtaining a degree of fluctuation of the positioning feature during the movement of the butted material; Decomposing the moving trajectory curve to obtain a period, a trend term and a residual term of the moving trajectory curve; obtaining the curve volatility of the moving trajectory curve according to the period, the trend term and the residual term; The positioning error is obtained according to the variation fluctuation degree and the curve fluctuation.

4. A visual positioning method for welding thermoplastic material and metal material according to claim 3, characterized in that: The method for obtaining the degree of change fluctuation includes: The product of the range and the variance of the positioning feature during the movement of the butted materials is taken as the variation fluctuation degree.

5. The method for visually positioning a thermoplastic material and a metal material for welding according to claim 3, characterized in that: The method for obtaining the curve volatility includes: The variance of the data in the residual term and the mean value in the trend term are multiplied, and the ratio of the product to the cycle size is taken as the volatility of the curve.

6. The method for visually positioning a thermoplastic material and a metal material for welding according to claim 1, characterized in that: The method for obtaining the end face stacking unevenness comprises: In the search window, analysis areas are constructed in sequence according to a preset step size with the end face area as the center, and the normal material area is used as the comparison area; the pixel value difference between the analysis area and the comparison area at each scale is analyzed to obtain the end face stacking degree; the pixel value distribution stability within the analysis area at each scale is analyzed to obtain the end face stacking roughness; the end face stacking unevenness is obtained according to the end face stacking degree and the end face stacking roughness.

7. A visual positioning method for welding thermoplastic material and metal material according to claim 6, characterized in that: The method for obtaining the end face accumulation degree comprises: The difference between the mean grayscale values ​​of each analysis area and the corresponding comparison area is used as the pile material appearance feature of each analysis area; and the accumulated value of all analysis areas is used as the end face accumulation degree.

8. The method for visually positioning a thermoplastic material and a metal material for welding according to claim 6, characterized in that: The method for obtaining the end surface accumulation roughness comprises: For each scale analysis area, the difference between the grayscale value of each pixel in the analysis area and the grayscale mean of the analysis area is taken as the grayscale deviation of each grayscale value; the average grayscale deviation of all pixels in the analysis area is taken as the roughness of the analysis area; The average roughness of all analyzed areas is taken as the end surface stacking roughness.

9. A visual positioning method for welding thermoplastic material and metal material according to claim 1, characterized in that: The expression of the positioning adjustment function is: Wherein f represents the positioning adjustment function value, K is the end face stacking unevenness, q represents the positioning error, X represents the position deviation between the butt joint material at the current position and the butt joint material at the previous moment, and e represents the natural constant; when welding has not started, X is a positive integer 1.

10. The method for visually positioning a thermoplastic material and a metal material for welding according to claim 3, characterized in that: Methods for obtaining positioning features include: The coordinates of the reference point in the moving image are obtained, and the Euclidean distance between the coordinates and the origin of the moving image is used as the positioning feature.

Images