Method for automatically identifying surface defects after copper wire drawing

Abstract

The invention is applicable to the technical field of surface defect detection, and particularly relates to a method for automatically identifying surface defects after copper wire drawing, which comprises the following steps: monitoring the surface temperature of the copper wire after drawing in real time; establishing a temperature attenuation model according to the surface temperature; dividing defect identification stages according to the temperature attenuation model; determining a first defect condition of the copper wire according to the defect identification stage; obtaining an annealing optimization parameter according to the first defect condition; and obtaining a second defect condition of the copper wire according to the first defect condition and the annealing optimization parameter. Therefore, the method for automatically identifying the surface defects after copper wire drawing provided by the embodiment of the invention is beneficial to accurately identifying the surface defects after copper wire drawing.

Description

Technical Field

[0001] This application belongs to the technical field of surface defect detection, and particularly relates to a method for automatically identifying surface defects after copper wire drawing. Background Art

[0002] Surface defect detection uses advanced machine vision detection technology to detect defects such as spots, pits, scratches, color differences, and defects on the surface of workpieces.

[0003] In the prior art, the detection of surface defects of copper wires often relies on the analysis of two-dimensional image data. However, two-dimensional image data has limitations in reflecting the three-dimensional morphology and minute defects on the surface of copper wires, especially when dealing with copper wires with complex shapes or variable surface textures after drawing and then annealing. Therefore, there is a problem that it is difficult to accurately identify the surface defects of copper wires after drawing with the current surface defect detection methods for copper wires. Summary of the Invention

[0004] The embodiments of this application provide a method for automatically identifying surface defects after copper wire drawing, which can solve the problem of difficultly and accurately identifying the surface defects of copper wires after drawing.

[0005] In a first aspect, the embodiments of this application provide a method for automatically identifying surface defects after copper wire drawing, including:

[0006] Real-time monitoring of the surface temperature of the copper wire after drawing;

[0007] Establishing a temperature decay model based on the surface temperature;

[0008] Dividing the defect identification stage according to the temperature decay model; wherein, the defect identification stage includes a high-temperature stage and a medium-low temperature stage;

[0009] Determining the first defect condition of the copper wire according to the defect identification stage; wherein, the first defect condition is used to reflect the surface defects of the copper wire after drawing and before annealing, and the surface defects include defect regions and defect types;

[0010] Obtaining annealing optimization parameters according to the first defect condition; wherein, the annealing optimization parameters include temperature and time;

[0011] Obtaining the second defect condition of the copper wire according to the first defect condition and the annealing optimization parameters; wherein, the second defect condition is used to reflect the surface defects of the copper wire after drawing and annealing.

[0012] The above technical solutions in the embodiments of this application have at least the following technical effects:

[0013] The automatic surface defect recognition method for copper wires provided by the embodiments of the present application monitors the surface temperature of copper wires in real time after wire drawing; establishes a temperature decay model based on the surface temperature; divides the defect recognition stage according to the temperature decay model; determines the first defect situation of the copper wire according to the defect recognition stage; obtains annealing optimization parameters according to the first defect situation; and obtains the second defect situation of the copper wire according to the first defect situation and the annealing optimization parameters. Therefore, the automatic surface defect recognition method for copper wires provided by the embodiments of the present application monitors the surface temperature of copper wires in real time after wire drawing and obtains annealing optimization parameters, thereby obtaining the second defect situation of the copper wire, establishing a closed-loop control chain of temperature-defect-process parameters, realizing the optimization of the annealing process, and being conducive to accurately identifying the surface defects of copper wires after wire drawing.

[0014] In a possible implementation manner of the first aspect, the determining the first defect situation of the copper wire according to the defect recognition stage includes:

[0015] For the process annealing path, thermal imaging data is obtained in the high-temperature stage, and stress wave signals are obtained in the medium-low temperature stage; wherein, the process annealing path refers to the process of annealing after large-drawing, medium-drawing, and small-drawing of the copper wire.

[0016] When the surface temperature is less than a preset threshold, a first image is obtained.

[0017] Abnormal hot spots and abnormal cold spots are obtained according to the thermal imaging data.

[0018] Signal mutation points are obtained according to the stress wave signals.

[0019] The defect area is obtained according to the signal mutation points, the abnormal hot spots, and the abnormal cold spots.

[0020] A defect image is obtained according to the first image and the defect area.

[0021] The defect image is input into a classification model to obtain the defect type.

[0022] In a possible implementation manner of the first aspect, the obtaining annealing optimization parameters according to the first defect situation includes:

[0023] Historical parameters are obtained; wherein, the historical parameters include defect types, annealing parameters, and wire drawing parameters.

[0024] A defect correlation matrix is established according to the historical parameters; wherein, the defect correlation matrix is used to reflect the sensitivity of defect types to annealing parameters and wire drawing parameters.

[0025] According to the first defect condition and the defect correlation matrix, the annealing optimization parameters are obtained using a multi-objective optimization function.

[0026] In a possible implementation manner of the first aspect, obtaining the second defect condition of the copper wire according to the first defect condition and the annealing optimization parameters includes:

[0027] For the process annealing path, after annealing according to the annealing optimization parameters, the second defect condition is obtained according to the first defect condition;

[0028] In a possible implementation manner of the first aspect, after obtaining the second defect condition according to the first defect condition, the method further includes:

[0029] Adjust the wire drawing parameters during drawing and small drawing in the copper wire according to the second defect condition and the defect correlation matrix.

[0030] In a possible implementation manner of the first aspect, determining the first defect condition of the copper wire according to the defect identification stage further includes:

[0031] For the post-processing annealing path, three-dimensional point cloud data of the copper wire is obtained before final annealing; wherein, the three-dimensional point cloud data includes multiple segments of data of the copper wire at different angles, and the three-dimensional point cloud data includes three-dimensional coordinates and a second image;

[0032] Register the three-dimensional coordinates to obtain a three-dimensional mesh model;

[0033] Map the second image onto the three-dimensional mesh model to obtain a three-dimensional model;

[0034] Determine the defect position of the copper wire based on the three-dimensional model; wherein, the defect position is the coordinate position of the defect of the copper wire on the three-dimensional model;

[0035] Determine the first defect condition according to the defect position.

[0036] In a possible implementation manner of the first aspect, determining the defect position of the copper wire based on the three-dimensional model includes:

[0037] Determine the bending points of the copper wire according to the three-dimensional model;

[0038] Determine the copper wire segments of the copper wire according to the bending points; wherein, each copper wire segment contains at most one bending point;

[0039] Detect the eddy current signals of each copper wire segment one by one;

[0040] Determine the defective segments of the copper wire according to the eddy current signals.

[0041] Obtain the ultrasonic signal of the defective section;

[0042] Obtain the defect position based on the ultrasonic signal and the three-dimensional model.

[0043] In a possible implementation manner of the first aspect, the method further includes:

[0044] Judge whether the shape of the copper wire conforms to the standard range according to the three-dimensional point cloud data; wherein, the shape includes diameter and length;

[0045] If the shape conforms to the standard range, it is determined that the copper wire has no shape defect;

[0046] If the shape does not conform to the standard range, it is determined that the copper wire has the shape defect;

[0047] Mark the three-dimensional model according to the shape defect.

[0048] In a possible implementation manner of the first aspect, the determining the first defect situation according to the defect position includes:

[0049] Determine the defect texture according to the defect position; wherein, the defect texture includes gray scale, structure and direction;

[0050] Grow outward according to the defect position and the defect texture to obtain the first defect situation.

[0051] In a possible implementation manner of the first aspect, the growing outward according to the defect position and the defect texture to obtain the first defect situation includes:

[0052] Determine the seed region according to the defect position and the defect texture;

[0053] Obtain the defect type according to the seed region and the classification model;

[0054] Calculate the local gradient direction according to the seed region and the defect type;

[0055] Grow the seed region outward along the local gradient direction;

[0056] When the growth process reaches the preset maximum growth times or the number of newly added pixel points is less than the number threshold, stop growing to obtain the defect region.

[0057] In a possible implementation manner of the first aspect, the obtaining the second defect situation of the copper wire according to the first defect situation and the annealing optimization parameters further includes:

[0058] For the post-treatment annealing path, after annealing according to the annealing optimization parameters, the second defect condition is obtained based on the first defect condition.

[0059] In a second aspect, an embodiment of the present application provides an automatic surface defect recognition device for copper wires after wire drawing, including:

[0060] A monitoring module for real-time monitoring of the surface temperature of the copper wire after wire drawing;

[0061] A temperature module for establishing a temperature decay model based on the surface temperature;

[0062] A defect recognition module for dividing a defect recognition stage according to the temperature decay model; wherein, the defect recognition stage includes a high-temperature stage and a medium-low temperature stage;

[0063] A first defect module for determining the first defect condition of the copper wire according to the defect recognition stage; wherein, the first defect condition is used to reflect the surface defects of the copper wire after wire drawing and before annealing, and the surface defects include defect regions and defect types;

[0064] An annealing optimization module for obtaining annealing optimization parameters according to the first defect condition; wherein, the annealing optimization parameters include temperature and time;

[0065] A second defect module for obtaining the second defect condition of the copper wire according to the first defect condition and the annealing optimization parameters; wherein, the second defect condition is used to reflect the surface defects of the copper wire after wire drawing and annealing.

[0066] In a third aspect, an embodiment of the present application provides an automatic surface defect recognition device for copper wires after wire drawing, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the method described in any one of the above first aspects is implemented.

[0067] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium storing a computer program, and when the computer program is executed by a processor, the method described in any one of the above first aspects is implemented.

[0068] In a fifth aspect, an embodiment of the present application provides a computer program product. When the computer program product runs on an automatic surface defect recognition device for copper wires after wire drawing, the automatic surface defect recognition device for copper wires after wire drawing is caused to execute the method described in any one of the above first aspects.

[0069] It can be understood that the beneficial effects of the above second aspect to fifth aspect can refer to the relevant descriptions in the above first aspect, and will not be elaborated here. Brief Description of the Drawings

[0070] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0071] Figure 1 is a schematic flowchart of an automatic surface defect recognition method for copper wires after wire drawing provided by an embodiment of the present application;

[0072] Figure 2 is a schematic flowchart of the implementation of step S400 in the automatic surface defect recognition method for copper wires after wire drawing provided by an embodiment of the present application;

[0073] Figure 3 is a schematic flowchart of the implementation of steps S400, S404, and S405 in the automatic surface defect recognition method for copper wires after wire drawing provided by an embodiment of the present application;

[0074] Figure 4 is a schematic flowchart of the implementation of steps S500 and S600 in the automatic surface defect recognition method for copper wires after wire drawing provided by an embodiment of the present application;

[0075] Figure 5 is a schematic structural diagram of an automatic surface defect recognition device for copper wires after wire drawing provided by an embodiment of the present application;

[0076] Figure 6 is a schematic structural diagram of an automatic surface defect recognition device for copper wires after wire drawing provided by an embodiment of the present application. Detailed Description of the Embodiments

[0077] In the following description, specific details such as specific system structures and technologies are proposed for the purpose of illustration rather than limitation, so as to thoroughly understand the embodiments of the present application. However, those skilled in the art should understand that the present application can also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid unnecessary details from interfering with the description of the present application.

[0078] It should be understood that when used in the specification of the present application and the appended claims, the term "comprising" indicates the presence of the described features, wholes, steps, operations, elements, and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their combinations.

[0079] It should also be understood that the term "and / or" as used in the specification and appended claims of this application refers to any combination and all possible combinations of one or more of the associated listed items, and includes such combinations.

[0080] As used in the specification and appended claims of this application, the term "if" can be interpreted as "when" or "once" or "in response to determining" or "in response to detecting" depending on the context. Similarly, the phrase "if determined" or "if [the described condition or event] is detected" can be interpreted as meaning "once determined" or "in response to determining" or "once [the described condition or event] is detected" or "in response to detecting [the described condition or event]" depending on the context.

[0081] In addition, in the description of the specification and appended claims of this application, the terms "first", "second", "third", etc. are only used for distinguishing descriptions and cannot be understood as indicating or implying relative importance.

[0082] Reference to "one embodiment" or "some embodiments" or the like described in the specification of this application means that a specific feature, structure, or characteristic described in connection with the embodiment is included in one or more embodiments of this application. Thus, statements such as "in one embodiment", "in some embodiments", "in other some embodiments", "in still other embodiments", etc. that appear in different places in this specification do not necessarily all refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "comprising", "including", "having", and their variants mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0083] In the related art, the detection of surface defects of copper wires often relies on two-dimensional image data for analysis. However, two-dimensional image data has limitations in reflecting the three-dimensional morphology and minute defects on the surface of copper wires, especially when dealing with copper wires with complex shapes or variable surface textures after wire drawing and then annealing. Therefore, there is a problem that it is difficult to accurately identify the surface defects of copper wires after wire drawing in the current surface defect detection methods for copper wires after wire drawing.

[0084] To solve the above problems, an embodiment of the present application provides an automatic identification method for surface defects after copper wire drawing. In this method, the surface temperature of the copper wire after drawing is monitored in real time; a temperature decay model is established based on the surface temperature; a defect identification stage is divided according to the temperature decay model; a first defect condition of the copper wire is determined according to the defect identification stage; annealing optimization parameters are obtained according to the first defect condition; and a second defect condition of the copper wire is obtained according to the first defect condition and the annealing optimization parameters. Therefore, the automatic identification method for surface defects after copper wire drawing provided by the embodiment of the present application monitors the surface temperature of the copper wire after drawing in real time and obtains the annealing optimization parameters, thereby obtaining the second defect condition of the copper wire, establishing a closed-loop control chain of temperature-defect-process parameters, and realizing the optimization of the annealing process, which is beneficial to accurately identifying the surface defects after copper wire drawing.

[0085] The automatic identification method for surface defects after copper wire drawing provided by the embodiment of the present application can be applied to an automatic identification device for surface defects after copper wire drawing. At this time, the automatic identification device for surface defects after copper wire drawing is the execution subject of the automatic identification method for surface defects after copper wire drawing provided by the embodiment of the present application. The embodiment of the present application does not impose any restrictions on the specific type of the automatic identification device for surface defects after copper wire drawing.

[0086] For example, the automatic identification device for surface defects after copper wire drawing can be a cellular phone, a mobile phone, a tablet computer, an augmented reality (AR) / virtual reality (VR) device, a notebook computer, an ultra-mobile personal computer (UMPC), a netbook, a personal digital assistant (PDA), a desktop computer, a smart large screen, a smart TV, a computing device or other processing devices connected to a wireless modem, a computer, a laptop computer, a handheld computing device, etc., but not limited thereto.

[0087] To better understand the automatic identification method for surface defects after copper wire drawing provided by the embodiment of the present application, the following provides an exemplary introduction to the specific implementation process of the automatic identification method for surface defects after copper wire drawing provided by the embodiment of the present application.

[0088] Figure 1 The schematic flowchart of the automatic identification method for surface defects after copper wire drawing provided by the embodiment of the present application is shown. The automatic identification method for surface defects after copper wire drawing includes:

[0089] S100, monitoring the surface temperature of the copper wire after drawing in real time.

[0090] Exemplarily, the surface temperature of the copper wire after wire drawing can be monitored in real time by a non-contact infrared thermometer.

[0091] S200, establish a temperature decay model based on the surface temperature.

[0092] Exemplarily, a temperature decay model (such as an exponential decay model, etc.) can be established based on the initial temperature of the copper wire and environmental conditions (such as air flow, heat dissipation coefficient, etc.). For example, the initial temperature is T0 = 500 °C, and after time t, the temperature decays to Tt. According to the exponential decay model Tt = T0e -kt , where k is the decay coefficient. By collecting multiple sets of temperature data and fitting them by the least squares method, k = 0.02 / s is obtained. After 10 seconds, the temperature will decay to Tt = 500e -0.2 ≈367.9 °C.

[0093] S300, divide the defect identification stage according to the temperature decay model. Among them, the defect identification stage includes a high-temperature stage and a medium-low temperature stage.

[0094] Exemplarily, it can be divided into a high-temperature stage and a medium-low temperature stage according to the temperature decay model. For example, the threshold for the high-temperature stage is set at 400 °C, and the threshold for the medium-low temperature stage is 200 °C. When the temperature of the copper wire is higher than 400 °C, it enters the high-temperature stage; when the temperature drops below 200 °C, it enters the medium-low temperature stage.

[0095] S400, determine the first defect condition of the copper wire according to the defect identification stage. Among them, the first defect condition is used to reflect the surface defects of the copper wire after wire drawing and before annealing. The surface defects include defect regions and defect types.

[0096] Exemplarily, image processing techniques (including image preprocessing, image segmentation, image feature extraction, etc.) can be used to process and analyze the image of the copper wire surface to identify the defect regions and defect types.

[0097] In a possible implementation, please refer to Figure 2 , S400, determine the first defect condition of the copper wire according to the defect identification stage, including:

[0098] S410, for the process annealing path, obtain thermal imaging data in the high-temperature stage and stress wave signals in the medium-low temperature stage. Among them, the process annealing path refers to the process of annealing after large drawing, medium drawing, and small drawing of the copper wire.

[0099] It can be understood that in heavy drawing, a copper rod with a relatively large diameter (such as 8.00 mm) is drawn through a wire drawing die to reduce its diameter to a relatively thick copper wire range (such as 1.15 - 3.00 mm); in medium drawing, the copper wire that has undergone heavy drawing is further drawn to further reduce its diameter to a finer copper wire range (such as 0.40 - 1.40 mm); in light drawing, the copper wire that has undergone medium drawing is finally drawn to obtain a copper wire with a very fine diameter (such as 0.10 - 0.30 mm).

[0100] Exemplarily, it can be in the high - temperature stage, using a high - resolution thermal imaging camera to continuously capture infrared images of the copper wire from multiple angles or along different positions of the copper wire; in the medium - and low - temperature stages, when the temperature of the copper wire drops to a certain extent, a stress wave sensor is used to detect the stress change inside the copper wire and capture the stress wave reflection or scattering signals caused by defects (such as cracks, inclusions, etc.). For example, in the high - temperature stage, the resolution of the thermal imaging camera is 640x480 pixels, and the frame rate is 30 frames per second. The infrared images captured by the thermal imaging camera are converted into a temperature matrix, where each pixel corresponds to a temperature value. For example, in a certain frame of the image, the temperature value of a certain pixel is 350 °C, while the temperature values of adjacent pixels are 345 °C, which indicates that there may be a hot spot in this area; in the medium - and low - temperature stages, the signals collected by the stress wave sensor will be converted into digital signals, and through signal processing techniques, characteristic parameters such as the frequency and amplitude of the stress wave signals can be extracted. For example, if the signal collected by a certain sensor suddenly increases at a certain time point, it indicates that there may be a defect at this position.

[0101] S420, when the surface temperature is less than the preset threshold, obtain the first image.

[0102] Exemplarily, it can be when the surface temperature of the copper wire drops below the preset threshold (such as 200 °C), using a high - resolution camera to take surface images of the copper wire from multiple angles or along different positions of the copper wire.

[0103] S430, obtain abnormal hot spots and abnormal cold spots according to the thermal imaging data.

[0104] Exemplarily, the thermal imaging data can be converted into a temperature matrix, and the temperature gradient of each pixel is calculated. According to the change of the temperature gradient, abnormal hot spots and abnormal cold spots are identified. Among them, an abnormal hot spot is a point where the temperature is significantly higher than the surrounding area, and an abnormal cold spot is a point where the temperature is significantly lower than the surrounding area. For example, in a certain frame of the thermal imaging image, the temperature value of a certain pixel is 400 °C, while the temperature values of adjacent pixels are 390 °C and 395 °C respectively. Calculate the temperature gradient of this pixel to get a gradient value of 5 °C / pixel. Compare it with the temperature gradients of surrounding pixels. If the gradient value exceeds the set threshold (such as 10 °C / pixel), then mark this pixel as an abnormal hot spot.

[0105] S440. Obtain the signal mutation points based on the stress wave signals.

[0106] Exemplarily, signal processing can be performed on the stress wave signals to extract the signal mutation points. Among them, the signal mutation points correspond to the defect positions inside the copper wire, and the signal processing includes steps such as filtering, denoising, and feature extraction. Through filtering and denoising processing, the noise and interference in the signal can be removed; through feature extraction processing, the characteristic parameters such as frequency and amplitude in the signal can be extracted, and according to the changes in the characteristic parameters, the signal mutation points can be identified. For example, in the signal collected by a certain stress wave sensor, the amplitude at a certain time point suddenly increases to more than twice the original. By calculating the amplitude gradient (i.e., the amplitude change rate) at this time point and comparing it with the set threshold, if the gradient value exceeds the threshold, then this time point is marked as a signal mutation point.

[0107] S450. Obtain the defect area based on the signal mutation points, abnormal hot spots, and abnormal cold spots.

[0108] Exemplarily, the signal mutation points, abnormal hot spots, and abnormal cold spots can be marked at their spatial positions on the copper wire and connected into a defect area. Image registration and fusion techniques can also be used to register and fuse the thermal imaging data and stress wave signal data to obtain the defect area. For example, through the marking of the signal mutation points, abnormal hot spots, and abnormal cold spots, three defect positions are obtained. Using image registration technology, these three defect positions are aligned at their spatial positions on the copper wire. By calculating the relative distance and angular relationship between these three defect positions, these three defect positions are connected into a complete defect area.

[0109] S460. Obtain the defect image based on the first image and the defect area.

[0110] Exemplarily, the first image can be superimposed or fused with the defect area, and image enhancement techniques (such as contrast enhancement, sharpening, etc.) are used to obtain the defect image. Among them, the defect image is used to reflect the defect positions and morphologies on the surface of the copper wire. For example, the first image is a grayscale image, and the defect area is a binary image (i.e., only including two colors, black and white). Through image fusion technology, the first image is superimposed with the defect area (adjusting parameters such as the transparency or color of the defect area).

[0111] S470. Input the defect image into the classification model to obtain the defect type.

[0112] Exemplarily, the defective image can be input into a pre-trained classification model to obtain the defect type. Among them, the classification model is constructed using deep learning algorithms, such as convolutional neural networks (CNNs), etc. During the training process, a large number of defective images are used as input data, and the corresponding defect types are labeled as output data to train and optimize the parameters of the classification model. For example, the classification model includes five common defect types: cracks, inclusions, oxidation, corrosion, and deformation. After the defective image is input into the classification model, the classification model will output a probability distribution vector indicating the probability that the defective image belongs to each defect type, and the defect type with the highest probability is selected as the final recognition result.

[0113] Through the above steps S410 to S470, by combining the thermal imaging data and the stress wave signal data, it is possible to more comprehensively capture and identify the defect information inside the copper wire, thereby improving the detection accuracy and precision. By adopting multi-source data fusion and image enhancement technologies, the missed detection cases caused by image blurring or noise interference can be reduced, thereby lowering the missed detection rate. By using automated and intelligent technical means, it is possible to achieve rapid identification and classification of the defects on the copper wire surface, thereby improving the detection efficiency and reducing the labor cost. It is not only applicable to the process annealing path but also applicable to the post-processing annealing path, and can meet the detection requirements under different production conditions. By identifying and classifying the defect types, data support can be provided for subsequent annealing optimization, thereby further improving the quality and performance of the copper wire.

[0114] In a possible implementation, please refer to Figure 3 , S400, determining the first defect condition of the copper wire according to the defect identification stage further includes:

[0115] S401, for the post-processing annealing path, obtain the three-dimensional point cloud data of the copper wire before the final annealing. Among them, the three-dimensional point cloud data includes multiple segments of data of the copper wire at different angles, and the three-dimensional point cloud data includes three-dimensional coordinates and a second image.

[0116] Exemplarily, in the post-processing annealing path, in order to obtain the three-dimensional point cloud data of the copper wire, a high-precision three-dimensional scanning device, such as a structured light scanner or a laser scanner, is used to scan the copper wire from multiple angles. During the scanning process, the three-dimensional scanning device emits light and receives the reflected light. By calculating the propagation time and angle of the light, the three-dimensional coordinates of each point on the copper wire are obtained. At the same time, a high-resolution camera can be used to capture the second image of the copper wire. For example, when using a laser scanner to scan the copper wire, the accuracy of the scanner is 0.1 mm and the scanning speed is 1000 points per second. During the scanning process, multiple segments of data of the copper wire at different angles are obtained, and each segment of data contains the three-dimensional coordinates of thousands of points. For example, a point in a certain segment of data may have the following coordinates: (x = 10.5 mm, y = 20.3 mm, z = 30.2 mm).

[0117] S402: Register the three-dimensional coordinates to obtain a three-dimensional mesh model

[0118] Exemplarily, the three-dimensional coordinates in the three-dimensional point cloud data can be registered (such as algorithms like the Iterative Closest Point (ICP) algorithm) into a unified coordinate system to construct a three-dimensional mesh model. For example, during the registration process, there are two segments of point cloud data that need to be registered. The first segment of data contains 1000 points, and the second segment of data contains 800 points. The ICP algorithm is used for registration, and the number of iterations is set to 100 times, and the error threshold is set to 0.05 mm. After iterative calculation, the registered three-dimensional coordinates are obtained, and a three-dimensional mesh model is constructed based on these coordinates. Among them, each mesh vertex in the three-dimensional mesh model corresponds to a three-dimensional coordinate point, and the edges of the mesh connect adjacent vertices.

[0119] S403. Map the second image onto the three-dimensional mesh model to obtain a three-dimensional model.

[0120] Exemplarily, through a texture mapping algorithm, the second image can be mapped onto the three-dimensional mesh model to obtain a three-dimensional model with surface features such as color and texture. For example, there is a high-resolution second image that needs to be mapped onto the three-dimensional mesh model. The resolution of the second image is 2048x2048 pixels, and each pixel contains RGB color information. The texture mapping algorithm is used to map the image onto the three-dimensional mesh model, and mapping parameters such as the offset and scaling ratio of the texture coordinates are set. After mapping calculation, a three-dimensional model with surface features such as color and texture is obtained. Among them, each mesh vertex in the three-dimensional model corresponds to a pixel point in the image, and the faces of the mesh display information such as the color and texture in the image.

[0121] S404. Determine the defect position of the copper wire based on the three-dimensional model. Among them, the defect position is the coordinate position of the defect of the copper wire on the three-dimensional model.

[0122] Exemplarily, computer vision and image processing techniques (including algorithms such as edge detection, shape analysis, texture analysis, etc.) can be used to identify abnormal regions or feature points in the three-dimensional model, thereby detecting the defect positions of the copper wires.

[0123] Optionally, refer to Figure 3 , S404, determining the defect positions of the copper wires based on the three-dimensional model, including:

[0124] S4041, determining the bending points of the copper wires according to the three-dimensional model.

[0125] Exemplarily, the three-dimensional model can be preprocessed to extract the geometric information of the copper wires, such as the starting point, ending point, and intermediate point coordinates of the line segments; calculate the angle or curvature change between adjacent line segments, set a threshold, and when the angle or curvature change exceeds this threshold, it is considered that there is a bending point at that place; according to all the identified bending points, obtain the set of bending points of the copper wires. For example, there is a three-dimensional model containing 1000 points, and these points form the continuous path of the copper wire. Extract the coordinates of these points, calculate the line segments between adjacent points, calculate the angle between each pair of adjacent line segments, and set an angle threshold, such as 15 degrees. Traverse all line segment pairs and find that there are three places where the angle exceeds 15 degrees, and these three places are identified as bending points.

[0126] S4042, determining the copper wire segments of the copper wires according to the bending points. Among them, each copper wire segment contains at most one bending point.

[0127] Exemplarily, according to the coordinates of the bending points, the points in the three-dimensional model can be sorted, and the points between adjacent bending points are divided into a copper wire segment; for the case of only one bending point, the bending point and a part of its adjacent points are divided into a separate copper wire segment; for a straight line segment without a bending point, it is treated as a separate copper wire segment. For example, the copper wire in the three-dimensional model contains 1000 points, and there are three bending points, located at the 200th, 500th, and 800th points respectively. The line segments between these points are divided into four copper wire segments: the first copper wire segment contains the points from the 1st point to the 300th point, the second copper wire segment contains the points from the 301st point to the 600th point, the third copper wire segment contains the points from the 601st point to the 900th point, and the fourth copper wire segment contains the points from the 801st point to the 1000th point.

[0128] S4043, detecting the eddy current signals of each copper wire segment one by one.

[0129] Exemplarily, the eddy current detection probe can be placed on the copper wire segment, and an alternating magnetic field is applied; measure and record the change of the eddy current signal; move the eddy current detection probe to the next copper wire segment and repeat the above steps to obtain the eddy current signals of each copper wire segment.

[0130] S4044, determine the defective section of the copper wire based on each eddy current signal.

[0131] Exemplarily, the eddy current signals of each copper wire segment can be preprocessed, such as filtering and denoising; the features of the eddy current signals can be extracted, such as peak values, valley values, average values, etc.; the features of the eddy current signals are compared with the preset defect features, and when the features match, it is considered that there is a defect in the copper wire segment. For example, the peak value of each eddy current signal is extracted as the feature value, and a peak value threshold is set. When the peak value exceeds the peak value threshold, it is considered that there is a defect in the copper wire segment.

[0132] S4045, obtain the ultrasonic signal of the defective section.

[0133] Exemplarily, an ultrasonic detection probe can emit ultrasonic waves into the object to be detected, and the automatic identification device for surface defects after copper wire drawing receives the reflected ultrasonic signals.

[0134] S4046, obtain the defect position based on the ultrasonic signal and the three-dimensional model.

[0135] Exemplarily, the ultrasonic signal can be preprocessed and analyzed to extract the reflected echo of the defect; according to the time parameter of the reflected echo and the propagation speed of ultrasonic waves in copper, the depth of the defect is calculated; combined with the geometric information of the defective section in the three-dimensional model, the position of the defect in the three-dimensional space is determined. For example, the propagation speed of ultrasonic waves in copper is 5000 m / s. By calculating the ratio of the time parameter of the reflected echo to the propagation speed, the depth of the defect is obtained as 2 mm. Combined with the geometric information of the second copper wire segment in the three-dimensional model, the position of the defect in the three-dimensional space is determined as (x = 450 mm, y = 0 mm, z = 2 mm).

[0136] Through the above steps S4041 to S4046, combined with a three-dimensional model and various non-destructive testing technologies such as eddy current testing and ultrasonic testing, the bending points and defective sections of the copper wire, as well as the position of the defect in the three-dimensional space, can be determined more accurately. It is applicable not only to the straight segments of copper wire but also to copper wires with complex shapes having bending points. Through segment-by-segment detection and analysis, it is possible to ensure a comprehensive inspection of the entire copper wire without missing any potential defects. Compared with traditional detection methods, it does not require destructive testing of the copper wire, so it can reduce the detection cost and extend the service life of the copper wire.

[0137] Optionally, please refer to Figure 3 , S404, the method further includes:

[0138] S40401, determine whether the shape of the copper wire conforms to the standard range according to the three-dimensional point cloud data. Among them, the shape includes diameter and length.

[0139] Exemplarily, the acquired three-dimensional point cloud data can be preprocessed (including noise removal, filtering, and smoothing), and the point cloud data can be converted into a format suitable for shape analysis (such as PointXYZ or PointXYZRGB in PCL (Point Cloud Library)). Features related to the shape of the copper wire (including the diameter and length of the copper wire, etc.) can be extracted from the preprocessed point cloud data. Algorithms in the PCL library (such as radius filtering, statistical filtering, etc.) are used to remove outliers and noise points, and the diameter of the copper wire is estimated by calculating the statistical characteristics (such as mean, standard deviation, etc.) of the point cloud data. The length of the copper wire is estimated by traversing the points in the point cloud data and calculating the distance between adjacent points. The extracted features are compared with a preset standard range. For example, the diameter of the copper wire is d, the length is l, the standard range is diameter d_std ± Δd, length l_std ± Δl, diameter deviation = |d - d_std|, length deviation = |l - l_std|. If the diameter deviation is less than or equal to Δd (such as 0.2 mm), and the length deviation is less than or equal to Δl (such as 1 m), it is considered that the shape of the copper wire conforms to the standard range.

[0140] S40402, if the shape conforms to the standard range, it is determined that there is no shape defect in the copper wire.

[0141] It can be understood that if the shape conforms to the standard range, then there is no shape defect in the copper wire.

[0142] S40403, if the shape does not conform to the standard range, it is determined that there is a shape defect in the copper wire.

[0143] It can be understood that if the shape does not conform to the standard range, then there is a shape defect in the copper wire.

[0144] S40404, mark the three-dimensional model according to the shape defect.

[0145] Exemplarily, the three-dimensional model can be marked according to the position where the shape defect of the copper wire appears by adding annotations, changing colors or textures, etc. in the three-dimensional model.

[0146] Through the above steps S40401 to S40404, using the three-dimensional point cloud data, the shape characteristics of the copper wire, including key parameters such as diameter and length, can be accurately detected, so as to achieve an accurate judgment of the shape defect of the copper wire. Through steps such as preprocessing, feature extraction, and shape comparison, a large amount of three-dimensional point cloud data can be efficiently processed, and the rapid detection of the shape defect of the copper wire can be realized. By marking the defective part in the three-dimensional model, the shape defect of the copper wire can be intuitively displayed, providing convenience for subsequent processing and analysis.

[0147] S405, determine the first defect situation according to the defect position.

[0148] Exemplarily, the defect texture can be determined according to the defect position, and the first defect situation can be determined according to the defect position and the defect texture.

[0149] Through the above steps S401 to S405, the three-dimensional point cloud data and the second image of the copper wire are obtained, and a three-dimensional model is constructed, which can more accurately identify and analyze the defect position and characteristics on the copper wire. By constructing a three-dimensional grid model and mapping the second image, a three-dimensional model with surface features such as color and texture is obtained, realizing the three-dimensional visualization of the copper wire defects, which helps to more intuitively understand the shape and distribution of the defects. By calculating and analyzing parameters such as the size, depth, and quantity of the defects, the severity and distribution of the defects can be determined, and decision-making support can be provided for subsequent repair or replacement, which helps to reduce the scrap rate and cost in the production process.

[0150] Optionally, please refer to Figure 3 , S405, determining the first defect situation according to the defect position includes:

[0151] S4051, determining the defect texture according to the defect position. Among them, the defect texture includes grayscale, structure, and direction.

[0152] Exemplarily, the grayscale mean and variance of the pixels in the defect area can be calculated to obtain the grayscale in the defect texture; the Canny edge detection algorithm can be used to extract the edges of the defects, and parameters such as the length, curvature, and direction of the edges are calculated to obtain the structure in the defect texture; the Gabor filter is applied to filter the defect area, and the response intensity of the filtered image is calculated to obtain the direction in the defect texture.

[0153] S4052, growing outward according to the defect position and the defect texture to obtain the first defect situation.

[0154] Exemplarily, algorithms for region growing or expanding can be performed on the image or three-dimensional point cloud data according to the defect position and the defect texture to obtain the first defect situation.

[0155] Through the above steps S4051 to S4052, by deeply analyzing the texture features such as the grayscale, structure, and direction of the defects, the characteristics and behaviors of the defects can be more accurately described, providing strong support for subsequent processing and analysis. Using the region growing algorithm and preset growth conditions, the process of the defect growing outward from the initial position can be realistically simulated, which helps to better understand the diffusion and evolution mechanism of the defects in the actual material. By simulating the growth process of the defects, a comprehensive evaluation of the first defect situation can be obtained, including information such as the shape, size, and position of the defects, providing an important reference for subsequent quality control and improvement.

[0156] Exemplarily, please refer to Figure 3, S4052, grow outward according to the defect location and defect texture to obtain the first defect situation, including:

[0157] S40521, determine the seed region according to the defect location and defect texture.

[0158] Exemplarily, one or more seed points can be manually or automatically selected according to the defect location and defect texture to determine the seed region. For example, there is an image containing defects, and its grayscale value range is 0 - 255. The image is grayscaled using the weighted average method, Gaussian filtering is applied to remove noise, the Canny operator is used for edge detection to obtain a binary edge image, and by traversing the edge image, the pixel points that meet specific texture features are found as seed points. For example, a grayscale value threshold and a gradient change threshold can be set, and the pixel points that satisfy both conditions are used as seed points.

[0159] S40522, obtain the defect type according to the seed region and the classification model.

[0160] Exemplarily, the features of the seed region (such as grayscale value, texture features, shape features, etc.) can be extracted and input into a pre-trained classification model. The classification model will output one or more possible defect types according to the input features. The classification model can be a support vector machine (SVM), decision tree, convolutional neural network (CNN), etc.

[0161] S40523, calculate the local gradient direction according to the seed region and the defect type.

[0162] Exemplarily, an image gradient calculation algorithm (such as the Sobel operator) can be used to calculate the local gradient direction of this region. Among them, the local gradient direction represents the direction in which the grayscale value changes fastest in the image and is consistent with the edge direction of the defect. For example, the Sobel operator is used to calculate the gradient direction. The Sobel operator is applied to the seed region to obtain the horizontal and vertical gradient components Gx and Gy, and the gradient direction θ is calculated according to Gx and Gy using the formula θ = arctan(Gy / Gx), and the gradient direction is quantized, dividing it into several discrete direction intervals (such as 0°, 45°, 90°, 135°, etc.).

[0163] S40524, grow the seed region outward along the local gradient direction.

[0164] Exemplarily, a region growing algorithm can be used to expand the defect region. Among them, the region growing algorithm is an iterative process. According to certain growth rules (such as gray-scale similarity, gradient direction consistency, etc.), adjacent pixel points are gradually added to the defect region. In each iteration, the pixel points on the edge of the current defect region are checked, and it is judged whether they can be added to the defect region according to the growth rules. For example, using a region growing algorithm based on gray-scale similarity and gradient direction consistency, an empty set of defect regions is initialized, and the seed points are added to this set to start the iterative process. In each iteration, a pixel point is taken out from the set of defect regions as the current point, and its adjacent pixel points (such as 8-neighborhood or 4-neighborhood) are checked. For each adjacent pixel point, the gray-scale difference and gradient direction difference between it and the current point are calculated. If the gray-scale difference is less than a certain threshold and the gradient direction difference is also within the allowable range, then this adjacent pixel point is added to the set of defect regions. Repeat this process until no new pixel points can be added to the set of defect regions.

[0165] S40525, when the growth process reaches the preset maximum number of growth times or the number of newly added pixel points is less than the quantity threshold, stop growing to obtain the defect region.

[0166] Exemplarily, it can be that after each iteration ends, it is checked whether it meets the condition of reaching the preset maximum number of growth times or the number of newly added pixel points is less than the quantity threshold. If one of the conditions is met, stop the growth process and output the final defect region. For example, the maximum number of growth times is set to 1000 times, and the quantity threshold is set to 10 pixel points. After each iteration ends, it is checked whether the current iteration number reaches the maximum number of growth times and whether the number of newly added pixel points is less than the quantity threshold. If the current iteration number reaches 1000 times or the number of newly added pixel points is less than 10, stop the growth process and output the final defect region. Otherwise, continue with the next iteration.

[0167] Through the above steps S40521 to S40525, combined with technical means such as morphological processing, edge detection, classification models, and region growing algorithms, accurate identification and positioning of defects can be achieved. At the same time, by controlling the growth conditions and stop rules, the stability and reliability of the algorithm are ensured. It can automatically identify and classify defect types, and perform region growth according to defect characteristics, avoiding the subjectivity and uncertainty of manual judgment. This improves the intelligent level of defect detection and reduces the need for manual intervention.

[0168] S500, obtain the annealing optimization parameters according to the first defect situation. Among them, the annealing optimization parameters include temperature and time.

[0169] Exemplarily, the annealing process can be simulated and analyzed by numerical simulation methods (such as the finite element method, boundary element method, etc.), and the annealing parameters can be adjusted according to the first defect situation. Machine learning algorithms (such as deep neural networks (DNN), etc.) can also be used to train and learn the first defect situation and annealing parameters, establish a prediction model, input the first defect situation into the prediction model, and output appropriate annealing parameters.

[0170] In a possible implementation, please refer to Figure 4 , S500, to obtain annealing optimization parameters according to the first defect situation, including:

[0171] S510, obtain historical parameters. Among them, the historical parameters include defect types, annealing parameters, and wire drawing parameters.

[0172] Exemplarily, historical parameters can be obtained from channels such as historical production records, quality inspection reports, and equipment logs.

[0173] S520, establish a defect correlation matrix according to the historical parameters. Among them, the defect correlation matrix is used to reflect the sensitivity of defect types to annealing parameters and wire drawing parameters.

[0174] Exemplarily, statistical software (such as SPSS, R language, etc.) can be used to perform correlation analysis on the historical parameters, calculate the correlation coefficients between defect types and annealing parameters, wire drawing parameters, and construct a defect correlation matrix according to the results of the correlation analysis. Among them, the rows of the defect correlation matrix represent defect types, the columns represent annealing parameters and wire drawing parameters, and the elements in the defect correlation matrix represent the correlation coefficients between defect types and corresponding parameters.

[0175] S530, according to the first defect situation and the defect correlation matrix, use a multi-objective optimization function to obtain annealing optimization parameters.

[0176] Exemplarily, according to the first defect situation, the optimization objectives can be determined, such as reducing the number of crack defects or reducing the severity of crack defects. According to the optimization objectives and the defect correlation matrix, a multi-objective optimization function can be constructed, including annealing parameters as independent variables and optimization objectives as dependent variables (or objective function values), and a multi-objective optimization algorithm (such as genetic algorithm, particle swarm algorithm, etc.) can be used to solve the optimization problem to obtain annealing parameters.

[0177] Through the above steps S510 to S530, the annealing parameters are optimized, the number and severity of defects such as cracks are reduced, and the overall quality of the product is improved. While optimizing the annealing parameters, the factor of production efficiency is also considered, and the decrease in production efficiency caused by too long annealing time is avoided. By establishing a defect correlation matrix and an optimization function, we can more accurately predict and control the defect situation in the production process, enhancing the controllability of the production process.

[0178] S600. Obtain the second defect condition of the copper wire according to the first defect condition and the annealing optimization parameters. Among them, the second defect condition is used to reflect the surface defects of the copper wire after wire drawing and annealing.

[0179] Exemplarily, the copper wire can be annealed according to the annealing optimization parameters, and the surface defect condition after annealing can be detected according to the first defect condition to obtain the second defect condition of the copper wire.

[0180] In a possible implementation manner, please refer to Figure 4 , S600. Obtain the second defect condition of the copper wire according to the first defect condition and the annealing optimization parameters, including:

[0181] S610. For the process annealing path, after annealing according to the annealing optimization parameters, obtain the second defect condition according to the first defect condition.

[0182] Exemplarily, the copper wire can be annealed according to the annealing optimization parameters (such as annealing temperature, annealing time, etc.) obtained through multi-objective optimization before. After annealing is completed, defect detection of the copper wire can be carried out by methods such as visual inspection, X-ray detection, and ultrasonic detection to obtain the second defect condition, and evaluate whether annealing effectively reduces the defects in the first defect condition and what new defects may have been generated. Compare the second defect condition with the first defect condition to analyze the effect of the annealing optimization parameters (including calculating the reduction of the number of defects, the reduction of the severity of defects, etc.).

[0183] Through the above step S610, a quantitative evaluation of the annealing effect under the process annealing path is realized, which helps to verify the effectiveness of the annealing optimization parameters and provides data support for subsequent optimization.

[0184] Optionally, please refer to Figure 4 , after S610 obtains the second defect condition according to the first defect condition, the method further includes:

[0185] S6101. Adjust the wire drawing parameters during drawing and small drawing of the copper wire according to the second defect condition and the defect correlation matrix.

[0186] Exemplarily, the second defect condition can be analyzed to obtain the main factors causing these defects, including wire drawing speed, wire drawing tension, wear degree of the wire drawing die, etc. Determine the sensitivity relationship between the wire drawing parameters and various defect types according to the defect correlation matrix, and adjust the wire drawing parameters (such as wire drawing speed, wire drawing tension, etc.). For example, if the second defect condition shows a relatively large number of cracks, refer to the defect correlation matrix, analyze and determine that it is caused by too fast wire drawing speed, reduce the wire drawing speed from the original 15 m / s to 12 m / s, and re-conduct trial production.

[0187] Through the above step S6101, the precise adjustment of the wire drawing parameters is achieved according to the second defect situation and the defect correlation matrix, which helps to further reduce the number of defects on the copper wire and improve the overall quality of the product. At the same time, the refined control ability of the copper wire production process is enhanced.

[0188] In a possible implementation, please refer to Figure 4 , S600, obtaining the second defect situation of the copper wire according to the first defect situation and the annealing optimization parameters, further includes:

[0189] S601, for the post-treatment annealing path, after annealing according to the annealing optimization parameters, obtaining the second defect situation according to the first defect situation.

[0190] Exemplarily, the copper wire can be subjected to post-treatment annealing according to the annealing optimization parameters. After annealing is completed, defect detection is performed to obtain the second defect situation, and the second defect situation is compared with the first defect situation to evaluate whether the annealing effectively reduces the defects in the first defect situation and whether new defects are generated.

[0191] Through the above step S601, the quantitative evaluation of the annealing effect under the post-treatment annealing path is achieved, which helps to verify the effectiveness of the annealing optimization parameters in different production stages and provides data support for subsequent optimization. At the same time, the quality control ability of the entire copper wire production process is also improved.

[0192] It should be understood that the magnitudes of the sequence numbers of the above steps in the embodiments do not mean the order of execution. The execution order of each process should be determined according to its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.

[0193] Corresponding to the method for automatically identifying surface defects after copper wire drawing described in the above embodiments, the embodiments of the present application further provide an apparatus for automatically identifying surface defects after copper wire drawing. Each module of the apparatus can implement each step of the method for automatically identifying surface defects after copper wire drawing. Figure 5 The structural block diagram of the apparatus for automatically identifying surface defects after copper wire drawing provided by the embodiments of the present application is shown. For the sake of convenience of description, only the parts related to the embodiments of the present application are shown.

[0194] Refer to Figure 5 , the apparatus includes:

[0195] A monitoring module, configured to monitor the surface temperature of the copper wire in real time after wire drawing;

[0196] A temperature module, configured to establish a temperature decay model according to the surface temperature;

[0197] A defect identification module, configured to divide the defect identification stage according to a temperature decay model; wherein, the defect identification stage includes a high-temperature stage and a medium-low temperature stage;

[0198] A first defect module, configured to determine a first defect condition of the copper wire according to the defect identification stage; wherein, the first defect condition is used to reflect the surface defects of the copper wire after wire drawing and before annealing, and the surface defects include a defect area and a defect type;

[0199] An annealing optimization module, configured to obtain annealing optimization parameters according to the first defect condition; wherein, the annealing optimization parameters include temperature and time;

[0200] A second defect module, configured to obtain a second defect condition of the copper wire according to the first defect condition and the annealing optimization parameters; wherein, the second defect condition is used to reflect the surface defects of the copper wire after wire drawing and annealing.

[0201] It should be noted that for the information interaction, execution process, etc. between the above modules, since they are based on the same concept as the method embodiment of the present application, their specific functions and the technical effects brought about can be specifically referred to in the method embodiment part, and will not be elaborated here.

[0202] Those skilled in the art can clearly understand that for the convenience and conciseness of description, only the above division of each functional unit and module is used as an example. In actual applications, the above functions can be allocated to different functional units and modules according to needs, that is, the internal structure of the device is divided into different functional units or modules to complete all or part of the functions described above. Each functional unit and module in the embodiment can be integrated into a processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above integrated unit can be implemented in the form of hardware or in the form of a software functional unit. In addition, the specific names of each functional unit and module are only for the convenience of mutual distinction and do not limit the protection scope of the present application. The specific working processes of the units and modules in the above device can refer to the corresponding processes in the foregoing method embodiment and will not be elaborated here.

[0203] An embodiment of the present application further provides an automatic surface defect identification device for copper wire after wire drawing, Figure 6 which is a schematic structural diagram of an automatic surface defect identification device for copper wire after wire drawing provided in an embodiment of the present application. As Figure 6 shown, the automatic surface defect identification device 6 for copper wire after wire drawing in this embodiment includes: at least one processor 60 ( Figure 6 only one is shown here), at least one memory 61 ( Figure 6only one is shown) and a computer program 62 stored in the at least one memory 61 and executable on the at least one processor 60. When the processor 60 executes the computer program 62, the automatic surface defect recognition device 6 for copper wires after wire drawing implements the steps in any of the above embodiments of the automatic surface defect recognition method for copper wires after wire drawing, or the automatic surface defect recognition device 6 for copper wires after wire drawing implements the functions of each module / unit in the above device embodiments.

[0204] Exemplarily, the computer program 62 can be divided into one or more modules / units. The one or more modules / units are stored in the memory 61 and executed by the processor 60 to complete this application. The one or more modules / units can be a series of computer program instruction segments capable of performing specific functions, and these instruction segments are used to describe the execution process of the computer program 62 in the automatic surface defect recognition device 6 for copper wires after wire drawing.

[0205] The automatic surface defect recognition device 6 for copper wires after wire drawing can be a computing device such as a desktop computer, a notebook, a palm computer, and a cloud server. The automatic surface defect recognition device for copper wires after wire drawing can include, but is not limited to, a processor 60 and a memory 61. Those skilled in the art can understand that Figure 6 merely examples of the automatic surface defect recognition device 6 for copper wires after wire drawing, which do not constitute a limitation on the automatic surface defect recognition device 6 for copper wires after wire drawing. It may include more or fewer components than shown in the figure, or combine some components, or different components. For example, it may also include input / output devices, network access devices, a bus, etc.

[0206] The processor 60 can be a central processing unit (CPU), and the processor 60 can also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc.

[0207] The memory 61 can be an internal storage unit of the automatic surface defect recognition device 6 after copper wire drawing in some embodiments, such as the hard disk or memory of the automatic surface defect recognition device 6 after copper wire drawing. The memory 61 can also be an external storage device of the automatic surface defect recognition device 6 after copper wire drawing in some other embodiments, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, a Flash Card, etc. equipped on the automatic surface defect recognition device 6 after copper wire drawing. Further, the memory 61 can also include both the internal storage unit and the external storage device of the automatic surface defect recognition device 6 after copper wire drawing. The memory 61 is used to store an operating system, application programs, a BootLoader, data, and other programs, such as the program code of the computer program, etc. The memory 61 can also be used to temporarily store the data that has been output or will be output.

[0208] An embodiment of the present application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in any of the above method embodiments are implemented.

[0209] An embodiment of the present application provides a computer program product. When the computer program product runs on the automatic surface defect recognition device after copper wire drawing, the automatic surface defect recognition device after copper wire drawing implements the steps in any of the above method embodiments.

[0210] When the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, to implement all or part of the processes in the above-described embodiment methods of the present application, a computer program can be used to instruct the relevant hardware to complete. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of the above-described method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file, or some intermediate form, etc. The computer-readable medium can at least include: any entity or device that can carry the computer program code to the automatic surface defect recognition device after copper wire drawing, a recording medium, a computer memory, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), an electrical carrier signal, a telecommunication signal, and a software distribution medium. For example, a USB flash drive, a mobile hard disk, a magnetic disk, or an optical disc, etc. In some jurisdictions, according to legislation and patent practice, the computer-readable medium cannot be an electrical carrier signal and a telecommunication signal.

[0211] In the above embodiments, the descriptions of each embodiment have their own emphases. For the parts not detailed or recorded in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0212] Those of ordinary skill in the art can realize that the units and algorithm steps of the examples described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. A professional technician can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present application.

[0213] In the embodiments provided by the present application, it should be understood that the disclosed automatic surface defect recognition device and method after copper wire drawing can be implemented in other ways. For example, the above-described embodiment of the automatic surface defect recognition device after copper wire drawing is merely illustrative. For example, the division of the modules or units is only a logical function division, and there can be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection to each other can be through some interfaces, and the indirect coupling or communication connection of devices or units can be in an electrical, mechanical, or other form.

[0214] The unit described as a separation component may or may not be physically separated. The component shown as a unit may or may not be a physical unit, that is, it may be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0215] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the various embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. An automatic recognition method for surface defects after copper wire drawing, characterized in that, including: Real-time monitoring of the surface temperature of the copper wire after wire drawing; Establishing a temperature decay model based on the surface temperature; Dividing the defect identification stage according to the temperature decay model; wherein, the defect identification stage includes a high-temperature stage and a medium-low temperature stage; Determining the first defect condition of the copper wire according to the defect identification stage; wherein, the first defect condition is used to reflect the surface defects of the copper wire after wire drawing and before annealing, and the surface defects include defect regions and defect types; Obtaining annealing optimization parameters according to the first defect condition; wherein, the annealing optimization parameters include temperature and time; Obtaining the second defect condition of the copper wire according to the first defect condition and the annealing optimization parameters; wherein, the second defect condition is used to reflect the surface defects of the copper wire after wire drawing and annealing.

2. The automatic identification method for surface defects after copper wire drawing according to claim 1, wherein, The determining the first defect condition of the copper wire according to the defect identification stage includes: For the process annealing path, obtaining thermal imaging data in the high-temperature stage and stress wave signals in the medium-low temperature stage; wherein, the process annealing path refers to the process of annealing after large drawing, medium drawing, and small drawing of the copper wire; When the surface temperature is less than a preset threshold, obtaining a first image; Obtaining abnormal hot spots and abnormal cold spots according to the thermal imaging data; Obtaining signal mutation points according to the stress wave signals; Obtaining the defect region according to the signal mutation points, the abnormal hot spots, and the abnormal cold spots; Obtaining a defect image according to the first image and the defect region; Inputting the defect image into a classification model to obtain the defect type.

3. The automatic identification method for surface defects of copper wire after wire drawing according to claim 2, characterized in that, The obtaining the annealing optimization parameters according to the first defect condition includes: Obtaining historical parameters; wherein, the historical parameters include defect types, annealing parameters, and wire drawing parameters; Establishing a defect correlation matrix according to the historical parameters; wherein, the defect correlation matrix is used to reflect the sensitivity of defect types to annealing parameters and wire drawing parameters; Using a multi-objective optimization function to obtain the annealing optimization parameters according to the first defect condition and the defect correlation matrix.

4. The automatic identification method for surface defects after copper wire drawing according to claim 3, characterized in that, The obtaining the second defect condition of the copper wire according to the first defect condition and the annealing optimization parameters includes: For the process annealing path, after annealing according to the annealing optimization parameters, obtaining the second defect condition according to the first defect condition; And / or, after obtaining the second defect condition according to the first defect condition, the method further includes: Adjusting the wire drawing parameters during medium drawing and small drawing of the copper wire according to the second defect condition and the defect correlation matrix.

5. The automatic identification method for surface defects after copper wire drawing according to claim 1, wherein, The determining the first defect condition of the copper wire according to the defect identification stage further includes: For the post-treatment annealing path, obtaining three-dimensional point cloud data of the copper wire before final annealing; wherein, the three-dimensional point cloud data includes multiple segments of data of the copper wire at different angles, and the three-dimensional point cloud data includes three-dimensional coordinates and a second image; Registering the three-dimensional coordinates to obtain a three-dimensional mesh model; Mapping the second image onto the three-dimensional mesh model to obtain a three-dimensional model; Determine the defect position of the copper wire based on the three-dimensional model; wherein, the defect position is the coordinate position of the defect of the copper wire on the three-dimensional model; Determine the first defect condition according to the defect position.

6. The automatic identification method for surface defects after copper wire drawing according to claim 5, wherein, The determining the defect position of the copper wire based on the three-dimensional model includes: Determine the bending points of the copper wire according to the three-dimensional model; Determine the copper wire segments of the copper wire according to the bending points; wherein, each copper wire segment contains at most one bending point; Detect the eddy current signals of each copper wire segment one by one; Determine the defective segments of the copper wire according to the eddy current signals; Obtain the ultrasonic signals of the defective segments; Obtain the defect position according to the ultrasonic signals and the three-dimensional model.

7. The automatic identification method for surface defects of copper wires after wire drawing according to claim 5, wherein, The method further includes: Judge whether the shape of the copper wire conforms to the standard range according to the three-dimensional point cloud data; wherein, the shape includes diameter and length; If the shape conforms to the standard range, determine that there is no shape defect in the copper wire; If the shape does not conform to the standard range, determine that the copper wire has the shape defect; Mark the three-dimensional model according to the shape defect.

8. The automatic identification method for surface defects of copper wire after wire drawing according to claim 5, characterized in that, The determining the first defect condition according to the defect position includes: Determine the defect texture according to the defect position; wherein, the defect texture includes gray scale, structure and direction; Grow outward according to the defect position and the defect texture to obtain the first defect condition.

9. The automatic identification method for surface defects after copper wire drawing according to claim 5, characterized in that, The growing outward according to the defect position and the defect texture to obtain the first defect condition includes: Determine the seed region according to the defect position and the defect texture; Obtain the defect type according to the seed region and the classification model; Calculate the local gradient direction according to the seed region and the defect type; Grow the seed region outward along the local gradient direction; When the growth process reaches the preset maximum number of growth times or the number of newly added pixel points is less than the quantity threshold, stop growing to obtain the defect region.

10. The automatic recognition method for surface defects after copper wire drawing according to claim 5, characterized in that, The obtaining the second defect condition of the copper wire according to the first defect condition and the annealing optimization parameters further includes: For the post-processing annealing path, after annealing according to the annealing optimization parameters, obtain the second defect condition according to the first defect condition.

Images