Composite metal wire coating thickness measurement method, evaluation method, terminal and medium

Abstract

The present invention discloses a composite metal wire coating thickness measurement method, evaluation method, terminal and medium, wherein the measurement method comprises: obtaining a cross-sectional image of the composite metal wire to be measured from which the coating contour and outer contour have been extracted; marking a first number of pixel points along the coating contour, and marking a second number of pixel points along the outer contour, wherein the second number is greater than the first number; applying a KD tree algorithm, finding the pixel points marked on the outer contour closest to the pixel points marked on the coating contour, and obtaining a first number of pixel point pairs; taking the distance between two pixel points in the pixel point pair as the coating thickness, calculating and saving the distance corresponding to the first number of pixel point pairs. The coating thickness is calculated by finding the distance between the coating contour and the nearest pixel point pair on the outer contour, thereby avoiding the coating thickness measurement error caused by the center position error and improving the coating thickness measurement accuracy; the closest pixel point pair is found based on the KD tree algorithm, which greatly improves the operating efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of composite metal coating evaluation, and in particular to a composite metal wire coating thickness measurement method, evaluation method, terminal and medium. Background Art

[0002] For composite wires with stainless steel as coating, such as coated steel bars, coated cables and coated pumping rods, the coating uniformity directly affects the mechanical properties, corrosion resistance and material cost of the products. As composite wires are in the stage of research and development and initial application, there is no mature method system for measuring and evaluating coating uniformity. Currently, only the following two methods can be retrieved.

[0003] The patent application with publication number CN112598723A discloses a method, device and storage medium for identifying the thickness of stainless steel coated steel bars. The pre-processing of the binary image includes: grayscale, median filtering noise reduction, sector partitioning, adaptive thresholding, etc. The process is complicated, and some steps need to be tested by multiple parameters according to the image quality to obtain better processing results. The center of the composite steel bar is identified by the Hough transform circle detection algorithm. The recognition quality of the algorithm is closely related to the standard of the circle and the image quality. The contour is an approximate circle, and the recognition result has a certain error. By making 15 equal-angle straight lines through the center of the circle to intersect with the inner and outer contours, the length of the two intersections on each line is determined as the coating thickness. This method has the following problems: the accuracy of the center recognition is crucial. If the center recognition position has a certain error, the result obtained by this method will amplify the error; the contour is an approximate circle, and the coating thickness obtained by this method is usually greater than the actual coating thickness; only 15 straight lines are too few, which has a certain degree of randomness. Moreover, this scheme also lacks an evaluation method for the uniformity of the coating.

[0004] The patent application with publication number CN114688981A discloses a composite steel bar coating thickness identification and thickness uniformity evaluation method, equipment and medium, wherein the center of the composite steel bar is calculated by a circumscribed rectangle. Because the contour is an approximate circle, the calculated center must have a certain error. By drawing 40 straight lines to measure the coating thickness at 40 locations, the number of measurements is still small, there is a certain degree of randomness, and the process of finding the nearest point by the minimum radius method is cumbersome and inefficient (40ms). The indicators involved in the coating uniformity evaluation method are only three statistical quantities: minimum thickness, range and standard deviation, and there is a lack of description of the contour thickness distribution morphology; the coating uniformity evaluation method involves the calculation of three weight coefficients, which are calculated by hierarchical analysis method (subjective) + entropy method (objective). The key step of the hierarchical analysis method is the construction of the judgment matrix, and the judgment matrix is ​​purely established by expert scoring, voting, etc., and the subjective factor is too high; the entropy method has two disadvantages in the application of this problem: ① If there is a large correlation between indicators, there will be a certain error in the allocation of weight coefficients by the entropy method. In this method, the minimum value and the range value have a high correlation; ② This method requires the provision of actual sample data. The more sample data, the better the result. In this scheme, only 5 sample data are used for calculation, and the number of samples is too small.

[0005] In summary, the existing coating thickness measurement and evaluation methods have the following shortcomings:

[0006] (1) The contour extraction process is done by manually testing the parameters multiple times to obtain the results, and automated extraction cannot be achieved (testing the method in CN114688981A shows that it is only effective for simulated images, and is basically ineffective for real images);

[0007] (2) Both solutions adopt the method of finding the center of the circle first and then drawing the line, but the contour is an approximate circle, and there must be errors in the position of the center of the circle;

[0008] (3) The number of straight lines drawn in the two schemes is the number of thickness measurements. The number of measurements is small and there is a certain degree of randomness;

[0009] (4) The indicators involved in the coating uniformity evaluation method are only three statistical quantities: minimum thickness, range and standard deviation, and lack the description of the coating thickness distribution morphology;

[0010] (5) The solution to the calculation of the three weight coefficients involved in the coating uniformity evaluation method is a combination of subjective method and objective method. However, the determination of the weight coefficients is still highly subjective. At the same time, the objective method uses the entropy method, which is not suitable for solving this type of problem (indicators have strong correlation). Summary of the invention

[0011] In view of the deficiencies in the prior art, an object of the present invention is to provide a composite metal wire coating thickness measurement method, evaluation method, terminal and medium to at least solve at least one of the above-mentioned technical problems.

[0012] In a first aspect, a method for measuring the coating thickness of a composite metal wire is provided, comprising the following steps:

[0013] S1: Acquire a cross-sectional image of the composite metal wire to be tested with the coating contour and outer contour extracted;

[0014] S2: marking a first number of pixel points along the contour of the coating, and marking a second number of pixel points along the outer contour, wherein the second number is greater than the first number;

[0015] S3: Applying a KD tree algorithm to find the pixel points marked on the outer contour that are closest to the pixel points marked on the cover contour, and obtaining a first number of pixel point pairs;

[0016] S4: Taking the distance between two pixels in a pixel pair as the coating thickness, calculate and save the distances corresponding to the first number of pixel pairs.

[0017] Furthermore, in step S1, contour detection is performed on the cross-sectional image of the composite metal wire to be tested based on deep learning technology to obtain the cross-sectional image of the composite metal wire to be tested with the coating contour and outer contour extracted.

[0018] Furthermore, in step S1, the cross-sectional image of the composite metal wire to be tested, from which the coating contour and the outer contour have been extracted, is obtained by the following method:

[0019] Acquire a cross-sectional image of the composite metal wire to be tested;

[0020] The cross-sectional image of the composite metal wire to be tested is input into the trained coating detection model based on the U-Net neural network, the coating is detected and segmented, and a black-and-white binary image is generated;

[0021] The coating contour and outer contour in the black-and-white binary image are extracted by using an edge detection algorithm, and a cross-sectional image of the composite metal wire to be tested with the coating contour and outer contour extracted is obtained.

[0022] Furthermore, in step S2, equidistant marking is performed when marking pixel points on the coating contour and the outer contour; and the second number is more than twice the first number.

[0023] Furthermore, step S3 specifically includes:

[0024] Construct a KD tree based on the pixels marked on the outer contour;

[0025] Based on the KD tree, pixel points marked on the outer contour that are closest to each pixel point marked on the cover contour are searched one by one to obtain a first number of pixel point pairs.

[0026] In a second aspect, a method for evaluating the coating thickness of a composite metal wire is provided, comprising:

[0027] Using the composite metal wire coating thickness measurement method as described above to obtain a first number of coating thickness values;

[0028] The coating uniformity of the composite metal wire is comprehensively evaluated based on two basic coating thickness statistics indicators, namely coating thickness standard deviation and range, and two basic coating profile morphology indicators, namely coating profile curvature and coating thickness skewness.

[0029] Furthermore, the coating thickness consistency index is constructed based on the coating thickness standard deviation. , and expressed as follows:

[0030]

[0031] In the formula, = Defined as the coefficient of variation, is the standard deviation of coating thickness, is the mean coating thickness, α c is a positive adjustment coefficient used to control VC sensitivity to change;

[0032] Construct coating thickness range index based on coating thickness range , and expressed as follows:

[0033]

[0034] In the formula, is the range of coating thickness, α r is a positive adjustment coefficient used to control right sensitivity to change;

[0035] Constructing the coating profile smoothness index based on the coating profile curvature , and expressed as follows:

[0036]

[0037] In the formula, is the standard deviation of the curvature of the first number of pixels on the coating contour, α sm is a positive adjustment coefficient used to control right sensitivity to change;

[0038] Constructing the symmetry index of coating profile based on coating thickness skewness , and expressed as follows:

[0039]

[0040] In the formula, is the coating thickness skewness, α sk is a positive adjustment coefficient used to control |Yes sensitivity to change;

[0041] Comprehensive index of coating uniformity of composite metal wire It is expressed as follows:

[0042]

[0043] In the formula, The value range is 0 to 1. The larger the value, the better the coating uniformity, and the smaller the value, the worse the coating uniformity.

[0044] Furthermore, the curvature of each pixel on the coating contour is obtained by the following method:

[0045] Find the current pixel and two pixels before and after it with a preset number of intervals, determine the center and radius of a unique circle based on these three pixels, and then calculate the curvature of the unique circle as the curvature of the current pixel;

[0046] Coating thickness deviation Calculated by the following formula:

[0047]

[0048] In the formula, is the standard deviation of coating thickness, is the mean coating thickness, n is the first number, is the thickness value of the i-th coating.

[0049] In a third aspect, an electronic terminal is provided, comprising:

[0050] a memory having a computer program stored thereon;

[0051] A processor is used to load and execute the computer program to implement the composite metal wire coating thickness measurement method or composite metal wire coating thickness evaluation method as described above.

[0052] In a fourth aspect, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the composite metal wire coating thickness measurement method or composite metal wire coating thickness evaluation method as described above is implemented.

[0053] The present invention proposes a composite metal wire coating thickness measurement method, evaluation method, terminal and medium, which have the following beneficial effects compared with the prior art:

[0054] (1) In the prior art, the contour extraction process obtains the result by manually testing the parameters multiple times, and cannot realize automatic extraction; however, the present invention adopts deep learning technology to realize automatic detection of the coating and automatic contour extraction, thereby improving the efficiency of contour extraction;

[0055] (2) The prior art adopts the method of first finding the center of the circle and then drawing the line, but the contour is an approximate circle and there must be an error in the position of the center of the circle. The present invention proposes to calculate the coating thickness by finding the distance between the coating contour and the nearest pixel point pair on the outer contour. The coating thickness calculated by this method is closer to the actual coating thickness, avoiding the coating thickness measurement error caused by the center position error, and improving the coating thickness measurement accuracy;

[0056] (3) The number of straight lines drawn in the prior art is the number of thickness measurements, which is relatively small, has a certain degree of randomness, and is inefficient. The present invention can avoid randomness by marking a large number of pixel points, and realize the search for the closest pixel point pair based on the KD tree algorithm, which greatly improves the operating efficiency.

[0057] (4) The indicators involved in the existing coating uniformity evaluation method are only three statistical quantities, namely, the minimum thickness, the range and the standard deviation, and lack the coating thickness distribution morphology quantity. The present invention proposes to evaluate the coating uniformity based on two basic coating thickness statistical quantity indicators, namely, the coating thickness standard deviation and the range, and two basic coating profile morphology quantity indicators, namely, the coating profile curvature and the coating thickness skewness, so that the evaluation is more accurate and comprehensive.

[0058] (5) The solution to the calculation of three weight coefficients involved in the existing coating uniformity evaluation method is a combination of subjective method and objective method, but the determination of the weight coefficient is still highly subjective. At the same time, the objective method adopts the entropy method, which is not suitable for solving this type of problem. The present invention proposes to standardize, normalize and multiply the indicator data based on the basic indicators of coating profile morphology and the basic indicators of coating thickness statistics, thereby avoiding the influence of subjective factors. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0060] Figure 1 is a flow chart of a composite metal wire coating thickness measurement method provided by an embodiment of the present invention;

[0061] Figure 2 is a flow chart of a method for training a coating detection model provided by an embodiment of the present invention;

[0062] Figure 3 It is a flow chart of obtaining a cross-sectional image of a composite metal wire to be tested with the coating contour and outer contour extracted provided by an embodiment of the present invention;

[0063] Figure 4 1 is a schematic diagram of a coating contour and outer contour extraction process provided by an embodiment of the present invention, wherein (a) is a cross-sectional image of a composite metal wire to be tested, (b) is a black-and-white binary image of the coating, and (c) is a cross-sectional image of the composite metal wire to be tested from which the coating contour and outer contour are extracted;

[0064] Figure 5 is a schematic diagram of a coating thickness measurement result of a first example provided by an embodiment of the present invention, wherein (a) is a black-and-white binary image of a coating identified in the first example, and (b) is a schematic diagram of a coating thickness measurement result measured in the first example;

[0065] Figure 6 is a schematic diagram of the coating thickness measurement result of the second example provided by an embodiment of the present invention, wherein (a) is a black and white binary image of the coating identified in the second example, and (b) is a schematic diagram of the coating thickness measurement result in the second example;

[0066] Figure 7 It is a flow chart of a method for evaluating the coating thickness of a composite metal wire provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0067] To make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be described in detail below. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other implementation methods obtained by ordinary technicians in this field without creative work belong to the scope of protection of the present invention.

[0068] It should be noted that, in the description of the present invention, the terms "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or order. In addition, in the description of the present invention, unless otherwise specified, the meaning of "plurality" refers to at least two.

[0069] like Figure 1 As shown, an embodiment of the present invention provides a method for measuring the coating thickness of a composite metal wire, comprising the following steps:

[0070] S1: Obtain a cross-sectional image of the composite metal wire to be tested with the coating contour and outer contour extracted.

[0071] In order to solve the problem that the contour extraction process in the prior art obtains results by manually testing parameters multiple times and cannot realize automatic extraction, the present invention proposes to realize automatic extraction of coating contour and outer contour based on deep learning technology.

[0072] Specifically, the detection and segmentation of the coating can be realized by first using the trained coating contour detection model based on the neural network model, and then the edge detection algorithm can be used to extract the coating contour and the outer contour. Among them, the neural network model can be selected from one of the neural networks such as FCN, U-Net, Seg-Net, deeplabv3, etc. The following is an example based on the implementation of the U-Net neural network. U-Net is a convolutional neural network for image segmentation. It can extract high-level features in the image through learning and training, and classify the original image at the pixel level, thereby realizing the extraction of the target object in the image.

[0073] First, if Figure 2 As shown, the coating detection model based on U-Net neural network can be obtained by the following method:

[0074] 1.1. Collect images of different steel bar sections and cut them to the same length and width;

[0075] 1.2. Use annotation software to annotate the steel bar cross-section image to obtain a sample data set;

[0076] 1.3. Divide the sample data set into training set and test set according to the ratio of 8:2;

[0077] 1.4. Input the training set and the test set into the U-Net neural network for training to obtain the optimal weight and the corresponding network model, that is, the coating detection model.

[0078] After the coating detection model is trained, it can be used for subsequent extraction of coating contours and outer contours, such as Figure 3 As shown, the specific steps include:

[0079] 2.1. Obtain the cross-sectional image of the composite metal wire to be tested (e.g. Figure 4 (a) and input;

[0080] 2.2. Determine whether the length and width of the input image are consistent. If not, the longer dimension of the image is cropped symmetrically on both sides to make the length and width consistent;

[0081] 2.3. Input the cropped cross-sectional image of the composite metal wire to be tested into the trained coating detection model based on the U-Net neural network, detect and segment the coating, and generate a black and white binary image, such as Figure 4 As shown in (b);

[0082] 2.4. Use the Canny edge detection algorithm to extract the coating contour and outer contour in the black and white binary image, obtain the cross-sectional image of the composite metal wire to be tested with the coating contour and outer contour extracted, and output it, such as Figure 4 As shown in (c).

[0083] S2: Mark a first number of pixel points along the contour of the coating, and mark a second number of pixel points along the outer contour, where the second number is greater than the first number.

[0084] When marking pixel points on the coating contour and the outer contour, it is preferred to mark them equidistantly; and the second number is more than twice the first number. The first number and the second number are set to different values ​​according to actual conditions. When the first number is smaller, the multiple of the second number relative to the first number is larger. When the first number is larger, the multiple of the second number relative to the first number can be relatively smaller. The value of the first number is not less than 10, and generally takes a value of 100 to 5000. In this embodiment, the first number is 3600 and the second number is 36000 as an example for explanation.

[0085] S3: Apply a KD tree algorithm to find the pixel points marked on the outer contour that are closest to the pixel points marked on the cover contour, and obtain a first number of pixel point pairs.

[0086] Step S3 specifically includes:

[0087] Construct a KD tree based on the pixels marked on the outer contour;

[0088] Based on the KD tree, pixel points marked on the outer contour that are closest to each pixel point marked on the cover contour are searched one by one to obtain a first number of pixel point pairs.

[0089] KD tree (K-dimensional tree) is a data structure used to organize points in K-dimensional space, supporting efficient search operations such as nearest neighbor search, range search, etc. KD tree recursively divides K-dimensional space into smaller subspaces, with each node representing a K-dimensional point and a splitting hyperplane passing through the point, thereby organizing the point set into a binary tree.

[0090] The construction principle of the KD tree is: select a dimension from all points and find the median point of the dimension as the split point to split the data space into two subspaces. This splitting process is performed recursively, and the splitting dimension selected each time is rotated in order until all points are included in the nodes of the KD tree. Each node stores a split point and a split hyperplane, dividing the space into two left and right subspaces.

[0091] Search principle: When performing the nearest neighbor search, start from the root node and search recursively downward according to the positional relationship between the coordinates of the target point and the splitting hyperplane until a leaf node is reached. Then, the algorithm backtracks to the upper nodes of the tree, checks whether there are closer points in the subspace on the other side, and updates the closest point information. This process utilizes the properties of the splitting hyperplane, greatly reducing the number of points that need to be compared, thereby improving the search efficiency.

[0092] S4: Taking the distance between two pixels in a pixel pair as the coating thickness, calculate and save the distances corresponding to the first number of pixel pairs.

[0093] Taking the random generation of 3600 coating contour pixel points and 36000 outer contour points as an example, that is, calculating 3600 coating thickness values, it takes a total of 8.5 milliseconds, which is much better than the speed of 40 milliseconds required for calculating 40 coating thickness values ​​using the minimum radius method in the patent application with publication number CN114688981A.

[0094] Figure 5 is a schematic diagram of a coating thickness measurement result of a first example provided by an embodiment of the present invention, wherein (a) is a black-and-white binary image of a coating identified in the first example, and (b) is a schematic diagram of a coating thickness measurement result measured in the first example; Figure 6 1 is a schematic diagram of the coating thickness measurement result of the second example provided by an embodiment of the present invention, wherein (a) is a black and white binary image of the coating identified in the second example, and (b) is a schematic diagram of the coating thickness result measured in the second example.

[0095] The above-mentioned embodiment provides a composite metal wire coating thickness measurement method with the following advantages:

[0096] (1) In the prior art, the contour extraction process obtains the result by manually testing the parameters multiple times, and cannot realize automatic extraction; however, the present invention adopts deep learning technology to realize automatic detection of the coating and automatic contour extraction, thereby improving the efficiency of contour extraction;

[0097] (2) The prior art adopts the method of first finding the center of the circle and then drawing the line, but the contour is an approximate circle and there must be an error in the position of the center of the circle. The present invention proposes to calculate the coating thickness by finding the distance between the coating contour and the nearest pixel point pair on the outer contour. The coating thickness calculated by this method is closer to the actual coating thickness, avoiding the coating thickness measurement error caused by the center position error, and improving the coating thickness measurement accuracy;

[0098] (3) The number of straight lines drawn in the prior art is the number of thickness measurements, which is relatively small, has a certain degree of randomness, and is inefficient. The present invention can avoid randomness by marking a large number of pixel points, and realize the search for the closest pixel point pair based on the KD tree algorithm, which greatly improves the operating efficiency.

[0099] The embodiment of the present invention also provides a method for evaluating the coating thickness of a composite metal wire. Figure 7 As shown, the following steps are included:

[0100] S1: Acquire a cross-sectional image of the composite metal wire to be tested with the coating contour and outer contour extracted;

[0101] S2: marking a first number of pixel points along the contour of the coating, and marking a second number of pixel points along the outer contour, wherein the second number is greater than the first number;

[0102] S3: Applying a KD tree algorithm to find the pixel points marked on the outer contour that are closest to the pixel points marked on the cover contour, and obtaining a first number of pixel point pairs;

[0103] S4: taking the distance between two pixels in the pixel pair as the coating thickness, calculating and saving the distances corresponding to the first number of pixel pairs (i.e., the first number of coating thickness values);

[0104] S5: Comprehensively evaluate the coating uniformity of the composite metal wire based on the two basic coating thickness statistics indicators, namely, coating thickness standard deviation and range, and the two basic coating profile morphology indicators, namely, coating profile curvature and coating thickness skewness.

[0105] The evaluation of coating uniformity is defined as two-dimensional evaluation in the quantitative process: ① coating thickness statistical index; ② coating profile morphology index. Among them, the coating thickness statistical index includes standard deviation and range, which can reflect the consistency of coating thickness in the numerical statistics of coating thickness. However, this statistic cannot effectively reflect the changes in coating profile morphology, such as coating profile smoothness, local curvature, whether the coating thickness is evenly distributed, and other uniformities in profile morphology. Therefore, a second dimension of evaluation index is added in this embodiment, namely, coating profile morphology index.

[0106] The statistical indicators of coating thickness include:

[0107] (1) Standard deviation: It reflects the degree of dispersion of coating thickness data points relative to the average value and is a key indicator for measuring coating thickness consistency. A low standard deviation means a tight thickness distribution, indicating a higher consistency.

[0108] (2) Range: It can directly reflect the fluctuation range of thickness data and can reflect the abnormal areas that may exist in the coating.

[0109] The coating profile morphology indicators include:

[0110] (1) Curvature: It can reflect the local bending degree of the coating profile, and can reflect the smoothness of the coating profile and possible defects, such as unevenness, ripples, or instability. The curvature of each pixel on the coating profile is obtained by the following method: finding the current pixel and the two pixels before and after it with a preset number of intervals, determining the center and radius of a unique circle based on these three pixels, and then calculating the curvature of the unique circle as the curvature of the current pixel;

[0111] (2) Skewness: It can measure the symmetry of the coating profile distribution, that is, whether it is biased towards the maximum or minimum value. It can effectively evaluate whether the coating profile is evenly distributed, and further reflect whether there is a deviation in the hot rolling process of the coating profile. Calculated by the following formula:

[0112]

[0113] In the formula, is the standard deviation of coating thickness, is the mean coating thickness, n is the first number, is the thickness value of the i-th coating. This formula is used to calculate the skewness value of a set of data. When the skewness is equal to 0, it means that the data distribution is symmetrical; when the skewness is less than 0, it means that the data distribution is skewed in a certain direction; when the skewness is greater than 0, it means that the data distribution is skewed in another opposite direction. That is, when the skewness is less than or greater than 0, it means that during hot rolling, there may be some hole-type rolling processes, and the material may not be centered when passing through these hole-types, and tilt in a certain direction.

[0114] Specifically, the coating thickness consistency index is constructed based on the coating thickness standard deviation. , and expressed as follows:

[0115]

[0116] In the formula, = Defined as the coefficient of variation, is the standard deviation of coating thickness, is the mean coating thickness, αc is a positive adjustment coefficient used to control VC The sensitivity of the change, initialized to 1.

[0117] Coating thickness consistency index It is used to evaluate the consistency of coating thickness data. Specifically, it measures the degree of dispersion of coating thickness values ​​relative to the average value. Its value range is 0 to 1. When VC is small, it means that the degree of dispersion is low and the thickness values ​​are relatively consistent. close to 1; when VC increases, the exponential term decreases rapidly, resulting in a value of It decreases and approaches 0, reflecting a higher degree of dispersion, indicating that the coating thickness values ​​are relatively discrete and the coating thickness values ​​are uneven.

[0118] Construct coating thickness range index based on coating thickness range , and expressed as follows:

[0119]

[0120] In the formula, is the range of coating thickness, α r is a positive adjustment coefficient used to control right The sensitivity of the change, initialized to 1.

[0121] Coating thickness range index It is used to evaluate the difference between the maximum and minimum values ​​of the coating thickness. Its value range is 0 to 1. When it is smaller, it means the range is smaller and the thickness value is more consistent. The value is close to 1; when When it is larger, it means the range is larger and the thickness value difference is larger. The value is close to 0, indicating that the difference between the maximum and minimum values ​​of the coating thickness increases and the uniformity decreases.

[0122] Constructing the coating profile smoothness index based on the coating profile curvature , and expressed as follows:

[0123]

[0124] In the formula, is the standard deviation of the curvature of the first number of pixels on the coating contour, α sm is a positive adjustment coefficient used to control right The sensitivity of the change, initialized to 1.

[0125] Coating profile smoothness index It is used to evaluate the local curvature of the coating surface profile, which can reflect the smoothness of the coating profile and possible defects, such as unevenness, ripples, or instability. Its value range is 0 to 1. When it is small, it means that the curvature dispersion is low, and the transition of the coating contour curve is smoother and relatively uniform. Close to 1; when As it increases, the exponential term decreases rapidly, resulting in a value of It decreases and approaches 0, indicating that the coating profile curve has large fluctuations and mutations and poor smoothness, that is, it may be uneven, rippled, or unstable.

[0126] Constructing the symmetry index of coating profile based on coating thickness skewness , and expressed as follows:

[0127]

[0128] In the formula, is the coating thickness skewness, α sk is a positive adjustment coefficient used to control |Yes Sensitivity to change.

[0129] Cladding profile symmetry index It is used to evaluate the symmetry of the distribution of the coating profile thickness value, and can effectively reflect whether the coating profile is evenly distributed. Its value range is 0 to 1. When it is close to 0, it indicates that the distribution of the coating profile thickness value is almost completely symmetrical, which means that the coating surface profile distribution is uniform. Close to 1; when As it increases, the exponential term decreases rapidly, resulting in a value of It decreases, indicating that the asymmetry of the coating profile distribution increases, that is, the uniformity of the coating profile decreases.

[0130] For the four positive adjustment coefficients of the above four indicators, in this embodiment, the initial values ​​are all set to 1, which has a good effect. However, in other embodiments, such as for other types of image detection, if the initial value 1 cannot distinguish well, there is another method to adjust the adjustment coefficient. This method is to fine-tune the adjustment coefficient according to the difference between the maximum and minimum values ​​of the calculation results of the indicator of 100 random instances, so that the difference between its maximum and minimum values ​​has a certain size. The maximum value of the indicator in 100 random instances can be quantitatively set to 0.9, and then the value of the adjustment coefficient can be calculated by reverse calculation of parameters such as the extreme value corresponding to the maximum value (because the maximum value of each of these four indicators is 1 in theory, and the minimum value is infinitely close to 0, so the difference between the maximum value and the minimum value is infinitely close to 1 in theory, but if the actual maximum value of the difference in the calculation results of 100 random instances is 0.05, then the discrimination may not be enough, and it is necessary to reduce the value of the adjustment coefficient and increase the discrimination of the indicator).

[0131] Comprehensive index of coating uniformity of composite metal wire It is expressed as follows:

[0132]

[0133] In the formula, The value range is 0 to 1. According to the above index analysis, when The closer the value is to 1, the better the coating uniformity of the composite wire sample is. The closer the value is to 0, the worse the coating uniformity of the composite wire sample is.

[0134] The composite metal wire coating thickness evaluation method provided by the above embodiment has the following advantages: the indicators involved in the existing coating uniformity evaluation method are only three statistical quantities, namely, the minimum thickness, the range and the standard deviation, and lack the coating thickness distribution morphological quantity; the present invention proposes to evaluate the coating uniformity according to two basic coating thickness statistical quantity indicators, namely, the coating thickness standard deviation and the range, and two basic coating profile morphological quantity indicators, namely, the coating profile curvature and the coating thickness skewness, so that the evaluation is more accurate and comprehensive; the solution idea for calculating the three weight coefficients involved in the existing coating uniformity evaluation method is a combination of subjective method + objective method, but the determination of the weight coefficient is still highly subjective, and the objective method adopts the entropy method, which is not suitable for solving this type of problem; the present invention proposes to standardize, normalize and multiply the indicator data based on the basic indicators of the coating profile morphological quantity and the basic indicators of the coating thickness statistics, thereby avoiding the influence of subjective factors.

[0135] An embodiment of the present invention further provides an electronic terminal, including:

[0136] a memory having a computer program stored thereon;

[0137] A processor is used to load and execute the computer program to implement the composite metal wire coating thickness measurement method or composite metal wire coating thickness evaluation method as described above.

[0138] An embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the composite metal wire coating thickness measurement method or composite metal wire coating thickness evaluation method as described above is implemented.

[0139] It can be understood that the same or similar parts of the above embodiments can be referenced to each other, and the contents not described in detail in some embodiments can refer to the same or similar contents in other embodiments.

[0140] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0141] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0142] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0143] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0144] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present invention.

Claims

1. A method for evaluating the coating thickness of a composite metal wire, characterized in that: The steps include: S1: Acquire a cross-sectional image of the composite metal wire to be tested with the coating contour and outer contour extracted; S2: marking a first number of pixel points along the contour of the coating, and marking a second number of pixel points along the outer contour, wherein the second number is greater than the first number; S3: Applying a KD tree algorithm to find the pixel points marked on the outer contour that are closest to the pixel points marked on the cover contour, and obtaining a first number of pixel point pairs; S4: taking the distance between two pixels in the pixel pair as the coating thickness, calculating and saving the distances corresponding to the first number of pixel pairs; Based on the two basic indicators of coating thickness statistics, namely, coating thickness standard deviation and range, and the two basic indicators of coating profile morphology, namely, coating profile curvature and coating thickness skewness, the coating uniformity of the composite metal wire is comprehensively evaluated. Constructing coating thickness consistency index based on coating thickness standard deviation , and expressed as follows: ； In the formula, = Defined as the coefficient of variation, is the standard deviation of coating thickness, is the mean coating thickness, α c is a positive adjustment coefficient used to control VC sensitivity to change; Construct coating thickness range index based on coating thickness range , and expressed as follows: ； In the formula, is the range of coating thickness, α r is a positive adjustment coefficient used to control right sensitivity to change; Constructing the coating profile smoothness index based on the coating profile curvature , and expressed as follows: ； In the formula, is the standard deviation of the curvature of the first number of pixels on the coating contour, α sm is a positive adjustment coefficient used to control right sensitivity to change; Constructing the symmetry index of coating profile based on coating thickness skewness , and expressed as follows: ； In the formula, is the coating thickness skewness, α sk is a positive adjustment coefficient used to control |Yes sensitivity to change; Comprehensive index of coating uniformity of composite metal wire It is expressed as follows: ； In the formula, The value range is 0 to 1. The larger the value, the better the coating uniformity, and the smaller the value, the worse the coating uniformity.

2. The method for evaluating the coating thickness of a composite metal wire according to claim 1, characterized in that: In step S1, contour detection is performed on the cross-sectional image of the composite metal wire to be tested based on deep learning technology to obtain the cross-sectional image of the composite metal wire to be tested with the coating contour and outer contour extracted.

3. The method for evaluating the coating thickness of a composite metal wire according to claim 1, characterized in that: In step S1, the cross-sectional image of the composite metal wire to be tested, from which the coating contour and the outer contour have been extracted, is obtained by the following method: Acquire a cross-sectional image of the composite metal wire to be tested; The cross-sectional image of the composite metal wire to be tested is input into the trained coating detection model based on the U-Net neural network, the coating is detected and segmented, and a black-and-white binary image is generated; The coating contour and outer contour in the black-and-white binary image are extracted by using an edge detection algorithm, and a cross-sectional image of the composite metal wire to be tested with the coating contour and outer contour extracted is obtained.

4. The method for evaluating the coating thickness of a composite metal wire according to claim 1, characterized in that: In step S2, pixel points are marked on the coating contour and the outer contour at equal intervals; and the second number is more than twice the first number.

5. The method for evaluating the coating thickness of a composite metal wire according to claim 1, characterized in that: Step S3 specifically includes: Construct a KD tree based on the pixels marked on the outer contour; Based on the KD tree, pixel points marked on the outer contour that are closest to each pixel point marked on the cover contour are searched one by one to obtain a first number of pixel point pairs.

6. The method for evaluating the coating thickness of a composite metal wire according to claim 1, characterized in that: The curvature of each pixel on the cover contour is obtained as follows: Find the current pixel and two pixels before and after it with a preset number of intervals, determine the center and radius of a unique circle based on these three pixels, and then calculate the curvature of the unique circle as the curvature of the current pixel; Coating thickness deviation Calculated by the following formula: ； In the formula, is the standard deviation of coating thickness, is the mean coating thickness, n is the first number, is the thickness value of the i-th coating.

7. An electronic terminal, characterized in that: include: a memory having a computer program stored thereon; A processor, configured to load and execute the computer program to implement the method for evaluating the coating thickness of a composite metal wire as claimed in any one of claims 1 to 6.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for evaluating the coating thickness of a composite metal wire according to any one of claims 1 to 6 is implemented.

Images