Cable core wire eccentricity automatic detection and compensation method and system based on image recognition

Abstract

The invention discloses a cable core wire eccentricity automatic detection compensation method and system based on image recognition, and belongs to the technical field of quality control in cable manufacturing, and the method comprises the steps: collecting an image of a cable in operation in real time in a cross section direction, extracting a boundary contour of a core wire and a sheath, and calculating a core wire eccentricity vector; inputting the eccentricity vector into a neural network identification model, predicting an eccentricity trend at the next moment, and comparing the eccentricity trend with a preset threshold value; if the eccentricity exceeds the limit, calculating compensation displacement for adjusting the position of a mold core or a guide wheel based on a current and predicted eccentricity vector difference in combination with equipment structure fluctuation characteristics and displacement response characteristics, and further updating an extrusion path and continuing closed-loop control; the method has the advantages of being high in predictability, high in response precision and suitable for complex working conditions, high-precision real-time regulation and control of core wire eccentricity can be achieved, and the consistency and the yield of fine cable products are improved.

Description

Technical Field

[0001] This invention relates to the field of quality control technology in cable manufacturing, specifically to an automatic detection and compensation method and system for cable core eccentricity based on image recognition. Background Technology

[0002] In automated cable manufacturing, the misalignment of the cable core (conductor) relative to the outer sheath or insulation layer is a key factor affecting the electrical performance and mechanical strength of the product. Especially in high-speed extrusion production, uncontrollable factors such as mechanical vibration, raw material fluctuations, and die offset often cause the core to shift off-center in the cross-section, affecting its tensile strength and dielectric uniformity, ultimately leading to a decrease in yield.

[0003] Existing methods for detecting cable eccentricity mostly employ physical sensors such as laser displacement and ultrasonic thickness measurement to obtain the boundary positions of each layer of the cable. However, these methods have the following significant drawbacks: Limited spatial resolution: It is difficult to detect minute deviations (such as within 0.05mm); Strong environmental dependence: It is sensitive to differences in cable surface finish and color, and is prone to false detection; Lack of real-time feedback mechanism: Detection and correction are disconnected, and production errors cannot be compensated in time; It cannot adapt to different cable structures: Changes in core material, sheath color, and cross-section will interfere with identification.

[0004] Especially when producing ultra-fine signal cables with a diameter of ≤1.2mm and core conductor material of multi-strand stranded copper wire, traditional methods cannot achieve high-precision eccentricity detection and effective compensation, resulting in a product yield decrease of more than 20%. Summary of the Invention

[0005] The purpose of this invention is to provide an automatic detection and compensation method and system for cable core eccentricity based on image recognition, so as to solve the shortcomings in the prior art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: an automatic detection and compensation method for cable core eccentricity based on image recognition, comprising: Real-time images of the cable in operation are acquired in the cross-sectional direction, including grayscale or color information of the core boundary and the sheath boundary; Edge extraction and boundary segmentation are performed on the real-time image to identify the core wire region contour and the sheath region contour respectively. The centroid coordinates of the core wire and the center coordinates of the sheath are calculated to obtain the core wire eccentricity vector ΔP. The core wire eccentricity vector ΔP is input into the neural network recognition model to fit and predict the eccentricity trend, output the eccentricity vector ΔP' expected at the next moment, and determine whether it exceeds the preset threshold ε. If ΔP'>ε, then the compensation displacement ΔX is calculated based on ΔP and ΔP'. The compensation displacement ΔX is used to adjust the current position of the die core or guide wheel of the cable extruder to counteract the core wire offset trend. After performing the compensation displacement ΔX, update the extrusion path and continue acquiring new images.

[0007] Preferably, obtaining the core wire eccentricity vector ΔP includes: Perform grayscale conversion and adaptive filtering on real-time images; Based on the processed image, the first closed contour of the core wire region and the second closed contour of the sheath region are obtained respectively. The centroid coordinates of the core wire region are calculated based on the first closed contour, the center coordinates of the sheath region are calculated based on the second closed contour, and the core wire eccentricity vector ΔP is determined by the vector difference between the centroid coordinates and the center coordinates.

[0008] Preferably, obtaining the first closed contour of the core wire region and the second closed contour of the sheath region further includes: identifying two closed boundaries in the image, wherein the closed contour with a small area is defined as the first closed contour of the core wire region, and the closed contour with a large area and approximately circular shape is defined as the second closed contour of the sheath region.

[0009] Preferably, the output of the predicted eccentricity vector ΔP' for the next time step includes: The core wire eccentricity vectors ΔP collected in succession are serialized according to time sequence to construct a data sequence containing at least N historical eccentricity vectors. The data sequence is then normalized and used as input samples for the neural network recognition model. A neural network recognition model for predicting eccentricity trends is constructed based on input samples. The neural network recognition model includes an input layer, at least one hidden layer, and an output layer. The model weights are trained using a backpropagation algorithm. The latest eccentricity vector data sequence is input into the trained neural network recognition model, and the predicted core eccentricity vector ΔP' corresponding to the next sampling time is output. The magnitude of the expected core wire eccentricity vector ΔP' is then compared with a preset threshold ε.

[0010] Preferably, the compensation displacement ΔX is calculated based on ΔP and ΔP', including: If ΔP'>ε, obtain the core wire eccentricity vector ΔP corresponding to the current sampling time and the eccentricity vector ΔP′ output by the neural network recognition model for the next sampling time; Based on the core wire eccentricity vector ΔP and the eccentricity vector ΔP′, a vector difference operation is performed to obtain the core wire eccentricity change vector ΔD. Based on the core wire eccentricity change vector ΔD, and combined with the structural fluctuation characteristics and displacement response characteristics of the cable extrusion equipment, the core wire eccentricity change vector ΔD is proportionally mapped and calculated to obtain the compensation displacement ΔX used to counteract the core wire eccentricity change trend.

[0011] Preferably, the structural fluctuation characteristic parameters of the cable extrusion equipment under stable operating conditions are obtained, including the periodic offset amplitude of the die core relative to the extrusion axis and the corresponding fluctuation frequency; the displacement response characteristic parameters of the cable extrusion equipment are obtained, which characterize the actual displacement tolerance of the extrusion equipment under a unit control input.

[0012] Preferably, the periodic offset amplitude is defined as the distance between the maximum and minimum positions of the mold core center within a complete fluctuation cycle; the fluctuation frequency is defined as the number of complete reciprocating offsets of the mold core per unit time.

[0013] This invention also provides an automatic detection and compensation system for cable core eccentricity based on image recognition, comprising: Image acquisition module: Acquires real-time images of the cable in operation in the cross-sectional direction, including grayscale or color information of the core wire boundary and the sheath boundary; Feature extraction module: performs edge extraction and boundary segmentation on the real-time image, identifies the core wire region contour and the sheath region contour respectively, calculates the centroid coordinates of the core wire and the center coordinates of the sheath, and obtains the core wire eccentricity vector ΔP; Neural network prediction module: Input the core wire eccentricity vector ΔP into the neural network recognition model, fit and predict the eccentricity trend, output the eccentricity vector ΔP' expected at the next moment, and determine whether it exceeds the preset threshold ε. Eccentricity compensation calculation module: If ΔP' > ε, then calculate the compensation displacement ΔX based on ΔP and ΔP'. The compensation displacement ΔX is used to adjust the current position of the wire extruder die core or guide wheel to counteract the core wire offset trend. Path update module: After performing compensation displacement ΔX, update the extrusion path and continue to acquire new images.

[0014] The technical effects and advantages provided by the present invention in the above technical solution are as follows: 1. This invention achieves high-precision identification and predictive control of the eccentricity trend of the core wire relative to the sheath by acquiring cable images in real time along the cross-sectional direction and combining edge recognition, geometric analysis, and neural network prediction. Compared with traditional methods that rely on fixed thresholds or open-loop detection, this invention can dynamically acquire the core wire offset behavior and output compensation displacement in advance before the eccentricity trend exceeds the forming tolerance, effectively suppressing the accumulation and amplification of eccentricity error during the production process and improving the real-time performance of cable concentricity control.

[0015] 2. This invention establishes a proportional compensation model oriented towards the physical characteristics of the equipment by introducing a mapping relationship between the core wire eccentricity variation vector and the structural response characteristics of the equipment, making the calculation of the compensation displacement ΔX more accurate and controllable. Combining structural fluctuation parameters and displacement response parameters, the compensation calculation not only considers the change in eccentricity trend but also takes into account the control sensitivity and stability of the equipment itself, thus improving the compensation accuracy. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0017] Figure 1 This is a flowchart of the method of the present invention.

[0018] Figure 2 This is a flowchart of the system modules of the present invention. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] Example 1, please refer to Figure 1 As shown in this embodiment, the automatic detection and compensation method for cable core eccentricity based on image recognition includes: Real-time images of the cable in operation are acquired in the cross-sectional direction, including grayscale or color information of the core wire boundary and the sheath boundary. Edge extraction and boundary segmentation are performed on the real-time images to identify the contours of the core wire region and the sheath region, respectively. The centroid coordinates of the core wire and the center coordinates of the sheath are calculated to obtain the core wire eccentricity vector ΔP. The core wire eccentricity vector ΔP is input into a neural network recognition model to fit and predict the eccentricity trend, output the eccentricity vector ΔP' expected at the next moment, and determine whether it exceeds a preset threshold ε. If ΔP'>ε, a compensation displacement ΔX is calculated based on ΔP and ΔP'. The compensation displacement ΔX is used to adjust the current position of the cable extruder die core or guide wheel to counteract the core wire offset trend. After executing the compensation displacement ΔX, the extrusion path is updated, and new images are acquired again.

[0021] This includes acquiring real-time images of the cable in operation along its cross-section, containing grayscale or color information of the core wire boundary and sheath boundary, specifically including: In this embodiment, a non-contact image of the high-speed cable is first acquired. The image acquisition direction is perpendicular to the cable axis, i.e., the cross-sectional direction, in order to fully capture the relative positional relationship between the core wire and the sheath in the cross-section.

[0022] The image acquisition device can be a high-frame-rate industrial camera, equipped with a telecentric lens and a high-brightness coaxial light source to ensure clear, distortion-free images of the cable cross-section even in motion. The acquisition frequency can be dynamically adjusted according to the cable's running speed to ensure equidistant images under different speed conditions.

[0023] In the acquired images, the core wire typically appears as an inner layer of bright or dark areas, while the sheath appears as an outer closed ring-shaped area. Due to differences in color, reflectivity, or texture between the core wire and sheath materials, distinct grayscale or color boundaries are visible in the images. This information can serve as a basis for boundary recognition in subsequent image processing algorithms.

[0024] To enhance boundary recognition, image preprocessing operations can be performed, including grayscale conversion, noise reduction, and contrast enhancement, to make the brightness gradient between the core wire edge and the sheath edge more obvious, thereby improving the accuracy of subsequent edge detection and geometric fitting.

[0025] In this embodiment, the cross-sectional image acquired in the above manner can not only reflect the core wire eccentricity, but also assist in judging key quality parameters such as core wire shape and sheath roundness.

[0026] Edge extraction and boundary segmentation are performed on the real-time image to identify the outline of the core wire region and the outline of the sheath region, respectively. The centroid coordinates of the core wire and the center coordinates of the sheath are calculated to obtain the core wire eccentricity vector ΔP.

[0027] In this embodiment, to accurately extract the boundary features between the core wire region and the sheath region in the cable cross-section image, the acquired real-time image is first subjected to grayscale conversion and adaptive filtering. Grayscale conversion uses the standard RGB weighted average method to convert the three-channel color image into a single-channel grayscale image. Subsequently, an adaptive median filtering algorithm based on local pixel mean is used to smooth the grayscale image. This algorithm constructs a 5×5 window centered on each pixel and adjusts the median calculation range to effectively remove random noise and preserve edge gradient change regions.

[0028] After image preprocessing, the Canny edge detection algorithm is used to perform edge extraction on the grayscale image. This algorithm calculates the first-order derivative gradients of the image in the horizontal and vertical directions, and performs preliminary smoothing of the image using a Gaussian filter. After determining the gradient magnitude and direction, edge filtering is performed using a set dual threshold.

[0029] After edge extraction, a contour tracking algorithm is used to identify two closed boundaries in the image. The smaller closed contour is defined as the first closed contour of the core region, and the larger, approximately circular closed contour is defined as the second closed contour of the sheath region. Specifically, this includes: The binary image after edge extraction is regarded as a two-dimensional grid composed of foreground pixels (value 1) and background pixels (value 0). The connectivity judgment of 8 neighboring pixels is used, that is, each foreground pixel is connected to the pixels in the 8 directions around it to identify continuous edge structures.

[0030] Starting from the top left corner of the image, scan row by row. When the first foreground pixel is encountered, record its position as the initial edge point P0, and initialize the current contour point set to empty.

[0031] Starting from the initial point P0, search for the next foreground pixel in its neighborhood in eight clockwise directions. Record its position and mark it as "visited". Continue repeating this process with P1 as the center until the search returns to the starting point P0, forming a closed boundary. All tracking points are stored sequentially in the current contour point set.

[0032] Continue scanning the entire image, searching for unvisited foreground pixels as new starting points, and perform step 3 for the second round of contour tracking. Each time a closed contour is identified, it is stored as an independent boundary.

[0033] All extracted closed contours are sorted according to their area. Those with smaller areas and higher boundary curvature are identified as the first closed contour of the core region; those with larger areas and approximately circular edges are identified as the second closed contour of the sheath region.

[0034] Based on the obtained first closed contour, the centroid coordinates of the core region are determined using a centroid calculation method. This method averages the coordinates of all pixels within the closed region, where the average x-coordinate is the centroid x-coordinate and the average y-coordinate is the centroid y-coordinate, ultimately forming the core centroid coordinates (X1, Y1). For the second closed contour of the sheath region, a least-squares circle fitting algorithm is used to construct a fitted circle model. This algorithm is based on the general equation of a circle, constructing a ternary nonlinear equation system containing the center coordinates (X2, Y2) and radius R, and iteratively solving it using the Levenberg-Marquardt optimization method to finally determine the center coordinates (X2, Y2) of the sheath region.

[0035] The core wire eccentricity vector ΔP is represented by the coordinate difference between the two geometric centers, that is, by the two-dimensional vector difference formed by the coordinates of the core wire centroid (X1, Y1) and the coordinates of the sheath center (X2, Y2). This eccentric vector not only reflects the spatial offset of the core wire relative to the sheath, but can also be used to assess the direction and magnitude of the eccentricity, providing quantitative input for subsequent compensation control.

[0036] The core wire eccentricity vector ΔP is input into the neural network recognition model to fit and predict the eccentricity trend, output the eccentricity vector ΔP' expected at the next moment, and determine whether it exceeds the preset threshold ε.

[0037] Following the sampling time sequence, the continuously acquired core wire eccentricity vectors ΔP are grouped into a historical data sequence, denoted as S={ΔP1, ΔP2, ..., ΔPn}, where n is the sequence length, and each ΔP contains two components in the X and Y directions. To improve the adaptability of the neural network recognition model to inputs of different orders of magnitude, min-max normalization is performed on each component in S. The normalization formula is: Normalized value = (Original value - Minimum value) / (Maximum value - Minimum value). The normalized data range is 0 to 1, and the normalized sequence serves as the input sample for the neural network recognition model.

[0038] The neural network recognition model is a feedforward fully connected neural network, comprising an input layer, two hidden layers, and an output layer. The number of nodes in the input layer equals the total number of components in each normalized sample; the hidden layers use the ReLU activation function, and the output layer uses a linear activation function to generate predicted values. The network uses mean squared error as the loss function and employs backpropagation combined with gradient descent for parameter training. The training samples are normalized historical eccentricity sequences, labeled with the value of the eccentricity vector ΔP observed at the next time step. The training process alternates between the training and validation sets until convergence, yielding a final model whose prediction accuracy meets the set error range.

[0039] The newly acquired n sets of core wire eccentricity vector data sequences are input into the neural network recognition model that has been trained above. The model automatically outputs the expected core wire eccentricity vector ΔP' corresponding to the next sampling time. Its form is a two-dimensional vector ΔP'=(X', Y'), which corresponds to the expected offset of the core wire in the X and Y directions, respectively.

[0040] The modulus of the expected core wire eccentricity vector ΔP' is calculated using the Euclidean norm formula: modulus = ... The modulus value is compared with a preset threshold ε. The threshold ε is the maximum allowable core wire offset set according to the product quality standard, in millimeters. If the modulus value of ΔP' is greater than the threshold ε, it is determined that the current core wire eccentricity trend has exceeded the allowable range, and the subsequent compensation displacement calculation step needs to be performed.

[0041] If ΔP'>ε, then the compensation displacement ΔX is calculated based on ΔP and ΔP'. The compensation displacement ΔX is used to adjust the current position of the die core or guide wheel of the cable extruder to counteract the core wire offset trend.

[0042] When the magnitude of the predicted core wire eccentricity vector ΔP′ output by the neural network recognition model is greater than a preset threshold ε, it is considered that the current core wire has a significant eccentricity trend, and compensation displacement calculation is required for control. In this embodiment, the calculation process of the compensation displacement ΔX includes the following steps: The core wire eccentricity vector ΔP corresponding to the current sampling time is collected, and the expected core wire eccentricity vector ΔP′ corresponding to the next sampling time is output by the neural network recognition model. ΔP and ΔP′ are both two-dimensional vectors, representing the positional offset of the core wire in the horizontal (X direction) and vertical (Y direction) directions, respectively, in millimeters.

[0043] Subtracting the current eccentricity vector ΔP from the expected eccentricity vector ΔP′ yields the core wire eccentricity change vector ΔD, i.e. The ΔD is a two-dimensional vector, whose X-direction component represents the predicted offset increment of the core wire in the horizontal direction, and whose Y-direction component represents the predicted offset increment of the core wire in the vertical direction.

[0044] Under stable equipment operation conditions, based on high-precision displacement sensors and structural analysis methods, the offset behavior of the die core relative to the extrusion axis over a period of time is collected, and its periodic offset amplitude and corresponding fluctuation frequency are extracted to construct a set of structural fluctuation characteristic parameters. The periodic offset amplitude is defined as the distance between the maximum and minimum positions of the die core center within a complete fluctuation cycle; the fluctuation frequency is defined as the number of complete reciprocating offsets of the die core per unit time, in Hertz.

[0045] Under known control input conditions, the displacement response of the extrusion equipment is measured to establish a mapping relationship between input and response. The displacement response characteristic parameter is defined as the actual displacement tolerance of the moving parts of the equipment under the action of a unit control command (such as a unit voltage or step pulse), i.e., the average displacement, in millimeters per unit input.

[0046] Divide each component of the core wire eccentricity variation vector ΔD by the structural fluctuation characteristic amplitude in its corresponding direction, and multiply by the unit displacement gain coefficient defined in the response characteristic parameters to obtain the compensation displacements ΔXx and ΔXy in the corresponding directions, ultimately forming the compensation displacement ΔX = (ΔXx, ΔXy). This displacement is used to adjust the position of the guide device or mold core to counteract the expected offset trend of the core wire.

[0047] It should be noted that in this embodiment, the unit displacement gain coefficient is denoted as Kdir, where dir represents the direction (X or Y direction), and is defined as the stable effective displacement generated by the guide component of the extrusion equipment in that direction under a unit control input. Its calculation formula is: Unit displacement gain coefficient Kdir = Effective displacement / Control input intensity, where: the effective displacement is the steady-state displacement value generated by the extrusion equipment in the corresponding direction, recorded by a measuring device (such as a high-precision displacement sensor), under a single control input, and the unit is millimeters; the control input intensity is the amplitude of the control signal applied to the servo driver or stepper motor, and the unit can be voltage (volts), pulse count, or PWM duty cycle percentage, depending on the specific control method.

[0048] After performing the compensation displacement ΔX, update the extrusion path and continue acquiring new images.

[0049] After calculating the compensation displacement ΔX, ΔX is applied as a control input to the guide mechanism or die core position adjustment mechanism of the extrusion equipment to achieve fine-tuning of the core wire's position in the cross-sectional direction. The compensation execution process employs closed-loop servo control to ensure that the actual executed displacement remains consistent with the calculated value, avoiding secondary eccentricity due to execution errors. After the compensation action is completed, the spatial position of the core wire in the cross-section tends to return to near the sheath center, thus effectively reducing eccentricity errors.

[0050] After the compensation displacement ΔX is completed, the real-time extrusion path of the cable is immediately updated, the relative position of the guide position and the center line of the core is reset, and the image acquisition process is restarted in the next sampling cycle. The cross-sectional image acquired at this time will reflect the new position of the core wire after compensation. The relative relationship between the core wire and the sheath boundary in the image is used as a new input into the edge extraction and eccentricity vector calculation process, forming a continuous closed-loop image recognition-prediction-compensation control link to ensure that eccentricity control is stable under dynamic working conditions.

[0051] This invention achieves high-precision identification and predictive control of the eccentricity trend of the core wire relative to the sheath by acquiring cable images in real time along the cross-sectional direction and combining edge recognition, geometric analysis, and neural network prediction. Compared with traditional methods that rely on fixed thresholds or open-loop detection, this invention can dynamically acquire the core wire offset behavior and output compensation displacement in advance before the eccentricity trend exceeds the forming tolerance, effectively suppressing the accumulation and amplification of eccentricity error during the production process and improving the real-time performance of cable concentricity control.

[0052] Example 2, please refer to Figure 2 As shown in this embodiment, the automatic detection and compensation system for cable core eccentricity based on image recognition includes: Image acquisition module: Acquires real-time images of the cable in operation in the cross-sectional direction, including grayscale or color information of the core wire boundary and the sheath boundary; Feature extraction module: performs edge extraction and boundary segmentation on the real-time image, identifies the core wire region contour and the sheath region contour respectively, calculates the centroid coordinates of the core wire and the center coordinates of the sheath, and obtains the core wire eccentricity vector ΔP; Neural network prediction module: Input the core wire eccentricity vector ΔP into the neural network recognition model, fit and predict the eccentricity trend, output the eccentricity vector ΔP' expected at the next moment, and determine whether it exceeds the preset threshold ε. Eccentricity compensation calculation module: If ΔP' > ε, then calculate the compensation displacement ΔX based on ΔP and ΔP'. The compensation displacement ΔX is used to adjust the current position of the wire extruder die core or guide wheel to counteract the core wire offset trend. Path update module: After performing compensation displacement ΔX, update the extrusion path and continue to acquire new images.

[0053] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. An automatic detection and compensation method for cable core eccentricity based on image recognition, characterized in that: include: Real-time images of the cable in operation are acquired in the cross-sectional direction, including grayscale or color information of the core boundary and the sheath boundary; Edge extraction and boundary segmentation are performed on the real-time image to identify the core wire region contour and the sheath region contour respectively. The centroid coordinates of the core wire and the center coordinates of the sheath are calculated to obtain the core wire eccentricity vector ΔP. The core wire eccentricity vector ΔP is input into the neural network recognition model to fit and predict the eccentricity trend, output the eccentricity vector ΔP' expected at the next moment, and determine whether it exceeds the preset threshold ε. If ΔP'>ε, then the compensation displacement ΔX is calculated based on ΔP and ΔP'. The compensation displacement ΔX is used to adjust the current position of the die core or guide wheel of the cable extruder to counteract the core wire offset trend. After performing the compensation displacement ΔX, update the extrusion path and continue acquiring new images.

2. The automatic detection and compensation method for cable core eccentricity based on image recognition according to claim 1, characterized in that: The process of obtaining the core wire eccentricity vector ΔP includes: Perform grayscale conversion and adaptive filtering on real-time images; Based on the processed image, the first closed contour of the core wire region and the second closed contour of the sheath region are obtained respectively. The centroid coordinates of the core wire region are calculated based on the first closed contour, the center coordinates of the sheath region are calculated based on the second closed contour, and the core wire eccentricity vector ΔP is determined by the vector difference between the centroid coordinates and the center coordinates.

3. The automatic detection and compensation method for cable core eccentricity based on image recognition according to claim 2, characterized in that: Obtaining a first closed contour of the core wire region and a second closed contour of the sheath region further includes: identifying two closed boundaries in the image, wherein a small-area closed contour is defined as the first closed contour of the core wire region, and a large-area, approximately circular closed contour is defined as the second closed contour of the sheath region.

4. The automatic detection and compensation method for cable core eccentricity based on image recognition according to claim 1, characterized in that: Output the predicted eccentricity vector ΔP' for the next time step, including: The core wire eccentricity vectors ΔP collected in succession are serialized according to time sequence to construct a data sequence containing at least N historical eccentricity vectors. The data sequence is then normalized and used as input samples for the neural network recognition model. A neural network recognition model for predicting eccentricity trends is constructed based on input samples. The neural network recognition model includes an input layer, at least one hidden layer, and an output layer. The model weights are trained using a backpropagation algorithm. The latest eccentricity vector data sequence is input into the trained neural network recognition model, and the expected core eccentricity vector ΔP' corresponding to the next sampling time is output. The magnitude of the expected core wire eccentricity vector ΔP' is then compared with a preset threshold ε.

5. The automatic detection and compensation method for cable core eccentricity based on image recognition according to claim 1, characterized in that: The compensation displacement ΔX is calculated based on ΔP and ΔP', including: If ΔP'>ε, obtain the core wire eccentricity vector ΔP corresponding to the current sampling time and the eccentricity vector ΔP′ output by the neural network recognition model for the next sampling time; Based on the core wire eccentricity vector ΔP and the eccentricity vector ΔP′, a vector difference operation is performed to obtain the core wire eccentricity change vector ΔD. Based on the core wire eccentricity change vector ΔD, and combined with the structural fluctuation characteristics and displacement response characteristics of the cable extrusion equipment, the core wire eccentricity change vector ΔD is proportionally mapped and calculated to obtain the compensation displacement ΔX used to counteract the core wire eccentricity change trend.

6. The automatic detection and compensation method for cable core eccentricity based on image recognition according to claim 5, characterized in that: Obtain structural fluctuation characteristic parameters of the cable extrusion equipment under stable operating conditions. The structural fluctuation characteristic parameters include the periodic offset amplitude of the die core relative to the extrusion axis and the corresponding fluctuation frequency. Obtain the displacement response characteristic parameters of the cable extrusion equipment, wherein the displacement response characteristic parameters characterize the actual displacement tolerance of the extrusion equipment under a unit control input.

7. The automatic detection and compensation method for cable core eccentricity based on image recognition according to claim 6, characterized in that: The periodic offset amplitude is defined as the distance between the maximum and minimum positions of the mold core center within a complete fluctuation cycle; The fluctuation frequency is defined as the number of times the mold core undergoes a complete reciprocating offset per unit time.

8. An image recognition-based automatic detection and compensation system for cable core eccentricity, used to implement the image recognition-based automatic detection and compensation method for cable core eccentricity as described in any one of claims 1-7, characterized in that: include: Image acquisition module: Acquires real-time images of the cable in operation in the cross-sectional direction, including grayscale or color information of the core wire boundary and the sheath boundary; Feature extraction module: performs edge extraction and boundary segmentation on the real-time image, identifies the core wire region contour and the sheath region contour respectively, calculates the centroid coordinates of the core wire and the center coordinates of the sheath, and obtains the core wire eccentricity vector ΔP; Neural network prediction module: Input the core wire eccentricity vector ΔP into the neural network recognition model, fit and predict the eccentricity trend, output the eccentricity vector ΔP' expected at the next moment, and determine whether it exceeds the preset threshold ε. Eccentricity compensation calculation module: If ΔP' > ε, then calculate the compensation displacement ΔX based on ΔP and ΔP'. The compensation displacement ΔX is used to adjust the current position of the wire extruder die core or guide wheel to counteract the core wire offset trend. Path update module: After performing compensation displacement ΔX, update the extrusion path and continue to acquire new images.

Images