Litz wire surface defect detection system and method

Abstract

The invention relates to the technical field of wire rod production surface defect detection, in particular to a Litz wire surface defect detection system and method, and the system comprises an image collection module which is used for continuously collecting high-definition images of the surface of a Litz wire; the image processing and analyzing module is used for receiving and processing the image data transmitted by the image acquisition module; and the control and execution module is used for receiving and comparing the data transmitted by the image processing and analysis module and carrying out abnormity alarm, trend early warning and data recording. According to the method, the width, the pitch and the angle of the Litz wire are synthesized into a twisting quality index (TQI) for judgment by adopting a mechanical vision detection matching algorithm, an image is analyzed by utilizing a mode of combining a morphological opening operation and a PCA dimension reduction algorithm, high-robustness and high-precision detection is realized, quantitative detection of the pitch is realized, and the detection precision of the Litz wire is improved. The problems that existing visual detection is poor in stability and low in precision, and the pitch is difficult to quantify are solved.

Description

Technical Field

[0001] This invention relates to the field of surface defect detection technology in wire production, specifically to a surface defect detection system and method for Leeds wire. Background Technology

[0002] Litz wire is a conductor made of multiple independently insulated fine wires twisted together according to specific rules. Its core value lies in reducing the skin effect and proximity effect under high-frequency alternating current. The performance of Litz wire, such as AC resistance (ACR), is extremely sensitive to the stranding structure, which includes the size of each wire, the stranding angle, and the pitch. Even with the same material, different stranding methods, pitches, and angles can result in vastly different high-frequency performance.

[0003] In existing production processes, quality control of the aforementioned key process parameters mainly relies on manual visual observation and random sampling. This method is highly subjective, has poor accuracy, is inefficient, carries the risk of missed detections, and is also time-sensitive. By the time problems are discovered, a large amount of waste yarn has often been produced, increasing costs. Common visual inspection methods, due to the articulated nature of Litz wire, suffer from poor stability and difficulty in quantifying the pitch when inspecting its dimensions, stranding angle, and pitch. Summary of the Invention

[0004] The purpose of this invention is to provide a surface defect detection system and method for Litz lines, which solves the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: A Leeds line surface defect detection system, comprising, Image acquisition module, used to continuously acquire high-definition images of the surface of the Litz line; The image processing and analysis module is used to receive and process image data transmitted by the image acquisition module; The control and execution module is used to receive and compare the data transmitted by the image processing and analysis module, and to perform anomaly alarms, trend warnings, and data recording. The image acquisition module, image processing and analysis module, and control and execution module are all connected via signal transmission lines. The image processing and analysis module includes an image preprocessing unit, a feature extraction unit, and a parameter calculation unit. The control and execution module includes a real-time alarm unit, a trend warning unit, and a data recording and output unit.

[0006] Preferably, the image acquisition module includes two high-brightness linear light sources. The high-brightness linear light sources are installed on one side of the Litz line in a low-angle grazing illumination manner. A high-definition line scan camera is fixedly installed on the opposite side of the high-brightness linear light source relative to the Litz line. The line scan direction of the high-definition line scan camera is perpendicular to the movement direction of the Litz line.

[0007] Preferably, the image processing and analysis module includes an industrial control computer and a monitoring display. The industrial control computer is equipped with an image processing and analysis algorithm to receive image data transmitted from the camera and process it in real time. The monitoring display is used to display real-time images, detection data, trend charts, and alarm information.

[0008] Preferably, the image preprocessing unit is used to filter and denoise the original image and enhance its contrast. The feature extraction unit uses image binarization and morphological opening algorithms to accurately segment each independent wire. The parameter calculation unit uses the PCA dimensionality reduction algorithm to perform principal axis analysis on each segmented wire region, thereby accurately calculating the line width, angle and pitch of each wire, and calculating the rate of change of the above parameters based on continuous frame images.

[0009] Preferably, the control and execution module includes an audible and visual alarm and a control cabinet, wherein the audible and visual alarm is used to emit audible and visual signals when an abnormality is detected.

[0010] A detection method based on a Litz line surface defect detection system includes the following steps: S201: Image acquisition, using a line scan camera to continuously acquire surface images of the Litz line during production. S202: Image preprocessing, which involves preprocessing the acquired image, including filtering and contrast adjustment; S203: Image binarization, performing binarization on the preprocessed image to separate the foreground and background of the Litz line; S204: Morphological opening operation, performs morphological opening operation on the binarized image to eliminate minor noise and smooth the boundary of each line, accurately segmenting each line. S205: Feature extraction and parameter calculation. For each segmented line region, the PCA algorithm is applied to calculate its main direction, which is the real-time twisting angle of the line. Based on the pixel area and calibration coefficient of each line, its actual width is calculated. In multiple consecutive frames of images, by tracking the center point of the same line, its center line spatial waveform is drawn, and the average pitch is calculated by performing spectral analysis on the waveform. S206: Calculate the stranding quality index based on the calculation results of multiple consecutive frames of images, and analyze the line width uniformity, angle consistency and pitch stability. S207: Result judgment and output, compares the real-time calculated stranding quality index with the preset threshold, and performs real-time alarm and trend warning; S208: Record data and output reports, store all image and parameter data, and generate detection reports.

[0011] Specifically, the image binarization uses Otsu's method for adaptive threshold binarization, converting the image into a black-and-white binary image to separate the foreground and background.

[0012] Specifically, the morphological opening operation is an operation of erosion followed by dilation using a specific structuring element to eliminate noise and accurately segment each line. The specific structuring element is a 3x3 pixel rectangular structuring element.

[0013] Specifically, the calculation of the stranding quality index is based on data collected over a period of time to calculate line width uniformity, angle consistency, and pitch stability, and then substituting these values ​​into the stranding quality index formula according to preset weighting coefficients. Where TQA represents the stranding quality index, SW represents line width uniformity, SA represents angle consistency, and SP represents pitch stability. These are the weighting coefficients.

[0014] Specifically, the formula for calculating the line width uniformity is as follows: , where k W CV is a coefficient. W To measure the width of each strand on a continuous section of Leeds line, calculate the coefficient of variation (CV) of all strand widths. W =Standard deviation / Mean; The formula for calculating the angular consistency is as follows: , where k A σ is a coefficient. A The standard deviation of the angle; The formula for calculating the pitch stability is as follows: , where k P CV is a coefficient. P is the coefficient of variation of the pitch values ​​along a continuous segment of the Leeds line.

[0015] By employing the above technical solution, the present invention has at least the following beneficial effects: This invention employs a mechanical vision detection algorithm to combine the width, pitch, and angle of the Litz wire into a twist quality index (TQI) for judgment. Furthermore, it utilizes a combination of morphological opening operation and PCA dimensionality reduction algorithm to analyze the image, achieving highly robust and high-precision detection. It also enables the quantitative detection of the pitch, solving the problems of poor stability, low accuracy, and difficulty in quantifying the pitch in existing visual detection methods.

[0016] This invention uses machine vision inspection combined with algorithms to achieve full inspection, and can alarm when defects occur. Furthermore, it can provide early warnings by analyzing changing trends, enabling operators to intervene in advance, changing "post-event remediation" to "pre-event prevention", reducing the generation of waste wire from the source and greatly reducing production costs. Attached Figure Description

[0017] The accompanying drawings, which are provided to further illustrate the invention, constitute a part of this application: Figure 1 This is a schematic diagram of the system working structure of the present invention; Figure 2 This is a schematic diagram of the detection method of the present invention; Figure 3 This is an example flowchart of the image processing algorithm of the present invention. In the diagram: 101, Litz line; 102, high-brightness linear light source; 103, high-definition line scan camera; 104, industrial computer; 105, monitoring display; 106, audible and visual alarm; 107, control cabinet; 108, signal transmission line; Detailed Implementation 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, and 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.

[0018] Please see Figure 1 The present invention discloses a surface defect detection system for Litz wire, which is installed on the Litz wire 101 production line, wherein the production direction of Litz wire 101 is as follows: Figure 1 As indicated by the middle arrow.

[0019] A surface defect detection system for the Litz line includes an image acquisition module, an image processing and analysis module, and a control and execution module.

[0020] The image acquisition module utilizes two high-brightness linear light sources 102, mounted on one side of the Litz wire 101 using a low-angle grazing illumination method to highlight the texture and contours of the wire surface, producing a clear contrast between light and shadow. Two high-definition line scan cameras 103 are then mounted opposite the high-brightness linear light sources 102, facing the Litz wire 101. Their linear scanning direction is perpendicular to the movement direction of the Litz wire 101, allowing for continuous acquisition of high-definition images of the surface, thus completing the continuous image acquisition of the Litz wire 101. It is important to note that the resolution of the high-definition line scan cameras 103 must be no less than 4K, and the scanning line frequency must match the production line speed.

[0021] The image processing and analysis module includes an industrial computer 104 and a monitoring display 105. The industrial computer 104 has built-in image processing and analysis algorithms, receives image data transmitted from the camera, and processes it in real time. The image data processing by the industrial computer 104 is divided into three steps: an image preprocessing unit, a feature extraction unit, and a parameter calculation unit.

[0022] Image preprocessing unit: preprocesses the transmitted image, including filtering and noise reduction, and contrast enhancement; Feature extraction unit: Using image binarization and morphological opening algorithms, each individual wire is accurately segmented; Parameter calculation unit: Based on the PCA dimensionality reduction algorithm, the principal axis analysis is performed on each segmented line region to accurately calculate the line width, angle and pitch of each line, and the rate of change of the above parameters is calculated based on continuous frame images.

[0023] The monitoring display 105 is used to display real-time images, detection data, trend charts and alarm information, making it convenient for operators to view and operate the data in real time.

[0024] The control and execution module includes an audible and visual alarm 106 and a control cabinet 107. The audible and visual alarm 106 emits audible and visual signals to alert the operator when an anomaly is detected. Different audible and visual signals are generated depending on the type of anomaly; that is, the audible and visual signals for warnings and alarms are different. The control cabinet 107 is an electrical control cabinet used to control related electrical equipment such as the camera position adjustment stepper motor, switching power supply, controller, stranding machine, and tension controller. This allows control of the electrical equipment through the control cabinet 107, enabling adjustments to the Litz line 101 production line based on warning signals. This allows operators to intervene in advance, transforming "post-event remediation" into "pre-event prevention," reducing waste yarn generation at the source and significantly lowering production costs.

[0025] Please see Figure 1 The high-definition line scan camera 103 and the industrial control computer 104 are connected via signal transmission line 108, allowing the industrial control computer 104 to receive raw images. The industrial control computer 104 is also connected to the monitoring display 105 and the audible and visual alarm 106 via signal transmission line 108, for displaying and alarming the processed data. The industrial control computer 104, control cabinet 107, and high-definition line scan camera 103 are connected via signal transmission line 108, enabling the industrial control computer 104 to control the control cabinet 107, and then using the control cabinet 107 to control and adjust the production of various workpieces and the position of the high-definition line scan camera 103.

[0026] Please see Figure 2 A detection method based on a Litz line surface defect detection system includes the following steps: S201: Image Acquisition. A line scan camera continuously acquires surface images of the Litz wire 101 during its production process. Low-angle grazing illumination is used to highlight the texture and contours of the wire surface, improving the clarity of the original image. Please refer to Figure (a): Original grayscale image, which shows the original grayscale image of the Litz wire 101 acquired by the line scan camera. Multiple strands of wire are clearly visible, with some areas exhibiting adhesion, noise, or uneven lighting.

[0027] S202: Image preprocessing, which preprocesses the acquired image by filtering and denoising, and enhancing contrast. S203: Image binarization. The preprocessed image is binarized using Otsu's method with adaptive thresholding to convert it into a black-and-white binary image, separating the foreground and background. Please refer to sub-image (b): the image after binarization, which shows the binary image after thresholding. Liz line 101 (foreground) is white, and the background is black. At this point, there are still some small noise points and slight adhesion between the strands. S204: Morphological opening operation, using 3x3 pixel rectangular structuring elements for erosion followed by dilation to eliminate noise and precisely segment each line; see sub-image (c): image after morphological opening operation, showing the binary image after morphological "opening" processing. Compared with image (b), it can be seen that small noise points have been eliminated, the originally slightly adhered lines are clearly separated, and the boundaries of each line become very smooth and independent.

[0028] S205: Feature extraction and parameter calculation. For each segmented line region, the PCA algorithm is applied to calculate its principal direction, which is the real-time twist angle of the line. For each connected region (i.e., a single line), the coordinates of all its pixels are extracted, the covariance matrix of the point set is calculated, and then the eigenvalues ​​and eigenvectors of the matrix are solved. The direction of the eigenvector corresponding to the largest eigenvalue is the principal direction (angle) of the line. Based on the pixel area and calibration coefficient of each line, its actual width is calculated. In multiple consecutive frames of images, by tracking the center point of the same line, its centerline spatial waveform is depicted, and the average pitch is calculated by performing spectral analysis on the waveform. Please refer to sub-figure (d): PCA principal axis analysis results. On the image after the opening operation, the principal axis calculated by PCA is plotted for each identified independent connected component (each line). The direction of this principal axis precisely represents the twist angle of that line. Multiple principal axes at different angles are clearly shown in the figure.

[0029] S206: Calculate the stranding quality index (TQI) based on the calculation results of multiple consecutive frames of images, and analyze the line width uniformity, angle consistency and pitch stability.

[0030] The formula for calculating the linewidth uniformity is as follows: , where k W CV is a coefficient. W To measure the width of each strand on a continuous section of Leeds Line 101, calculate the coefficient of variation (CV) of all strand widths. W =Standard deviation / Mean.

[0031] S W This involves scoring the uniformity of line width, where k W It is a magnification factor used to amplify CV to between 0 and 1. When CV is 0, S W =1; the larger the CV, the greater the S W The lower the value, the minimum is 0.

[0032] The formula for calculating the angular consistency is as follows: , where k A σ is a coefficient. A Let be the standard deviation of the angle.

[0033] S A The score is for angular consistency, where k A It is a magnification factor used to increase σ A Magnified to 0 to 1, when σ A When S is 0, W =1; σ A The larger S is A The lower the value, the minimum is 0.

[0034] The formula for calculating the pitch stability is as follows: , where k P CV is a coefficient. P It is the coefficient of variation of the pitch values ​​on a continuous segment of the Lids Line 101.

[0035] S P This pertains to the evaluation of pitch stability, where k P This is the magnification factor, used to increase the CV. P Magnification, when CV P When S is 0, P =1, CV P The larger S is P The lower the value, the minimum is 0.

[0036] The formula for calculating TQI is: When using multiple weighting coefficients to calculate TQI, the calculation is customized based on the performance sensitivity of different Litz 101 models. For example, for high-frequency applications, the weights for angle consistency and pitch stability can be set higher, thereby improving the flexibility and versatility of the method.

[0037] Furthermore, the use of a weighted geometric mean instead of an arithmetic mean for judgment makes the TQI extremely sensitive to any single defect. This means that any imbalance in any factor during the production process of the Leeds Line 101 will cause the TQI to drop rapidly, thus triggering an alarm or warning, thereby improving the sensitivity of the algorithm.

[0038] S207: Result judgment and output, compare the real-time calculated stranding quality index (TQI) with the preset threshold, and perform real-time alarm and trend warning; TQI's output range is between 0 and 1. A TQI ≥ 0.9 is considered excellent. A value of 0.8 ≤ TQI < 0.9 is considered good. A TQI of 0.7 ≤ TQI < 0.8 is considered acceptable, but close monitoring is still necessary. A TQI < 0.7 indicates a failure and triggers an alarm.

[0039] Continuous downward trend: Even if the TQI is currently qualified, if its value continues to decline, it can trigger a trend warning.

[0040] S208: Record data and output reports, store all image and parameter data, and generate detection reports.

[0041] Specifically, TQI can quickly determine the real-time quality of Litz 101 during the production process, and by analyzing the formula, it can quickly identify the factors affecting the quality of Litz 101 during the production process. By adjusting the production line, it can quickly make adjustments to avoid the generation of waste yarn.

[0042] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0043] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A surface defect detection system for Litz lines, characterized in that, include, Image acquisition module for continuously acquiring high-definition images of the surface of the Litz line (101); The image processing and analysis module is used to receive and process image data transmitted by the image acquisition module; The control and execution module is used to receive and compare the data transmitted by the image processing and analysis module, and to perform anomaly alarms, trend warnings, and data recording. The image acquisition module, image processing and analysis module, and control and execution module are all connected by a signal transmission line (108). The image processing and analysis module includes an image preprocessing unit, a feature extraction unit, and a parameter calculation unit. The control and execution module includes a real-time alarm unit, a trend warning unit, and a data recording and output unit.

2. The Litz line surface defect detection system according to claim 1, characterized in that, The image acquisition module includes a high-brightness linear light source (102), and there are two high-brightness linear light sources (102). The high-brightness linear light sources (102) are installed on one side of the Litz line (101) in a low-angle grazing illumination manner. A high-definition line scan camera (103) is fixedly installed on the opposite side of the high-brightness linear light source (102) relative to the Litz line (101). The line scan direction of the high-definition line scan camera (103) is perpendicular to the movement direction of the Litz line (101).

3. The Litz line surface defect detection system according to claim 1, characterized in that, The image processing and analysis module includes an industrial computer (104) and a monitoring display (105). The industrial computer (104) is equipped with an image processing and analysis algorithm, which receives image data transmitted from the camera and processes it in real time. The monitoring display (105) is used to display real-time images, detection data, trend charts and alarm information.

4. The Litz line surface defect detection system according to claim 1, characterized in that, The image preprocessing unit is used to filter and denoise the original image and enhance its contrast. The feature extraction unit uses image binarization and morphological opening algorithms to accurately segment each independent wire. The parameter calculation unit uses PCA dimensionality reduction algorithm to perform principal axis analysis on each segmented wire region, thereby accurately calculating the line width, angle and pitch of each wire, and calculating the rate of change of the above parameters based on continuous frame images.

5. The Litz line surface defect detection system according to claim 1, characterized in that, The control and execution module includes an audible and visual alarm (106) and a control cabinet (107). The audible and visual alarm (106) is used to emit an audible and visual signal when an abnormality is detected.

6. The detection method based on the Litz line surface defect detection system according to claims 1-5, characterized in that, Includes the following steps: S201: Image acquisition, using a line scan camera to continuously acquire surface images during the production process of the Litz line; S202: Image preprocessing, which involves preprocessing the acquired image, including filtering and contrast adjustment; S203: Image binarization, performing binarization on the preprocessed image to separate the foreground and background of the Litz line; S204: Morphological opening operation, performs morphological opening operation on the binarized image to eliminate minor noise and smooth the boundary of each line, accurately segmenting each line. S205: Feature extraction and parameter calculation. For each segmented line region, the PCA algorithm is applied to calculate its main direction, which is the real-time twisting angle of the line. Based on the pixel area and calibration coefficient of each line, its actual width is calculated. In multiple consecutive frames of images, by tracking the center point of the same line, its center line spatial waveform is drawn, and the average pitch is calculated by performing spectral analysis on the waveform. S206: Calculate the stranding quality index based on the calculation results of multiple consecutive frames of images, and analyze the line width uniformity, angle consistency and pitch stability. S207: Result judgment and output, compares the real-time calculated stranding quality index with the preset threshold, and performs real-time alarm and trend warning; S208: Record data and output reports, store all image and parameter data, and generate detection reports.

7. The method for detecting surface defects on a Litz line according to claim 6, characterized in that, The image binarization is performed using Otsu's method for adaptive threshold binarization, converting the image into a black and white binary image to separate the foreground and background.

8. The method for detecting surface defects on a Litz line according to claim 6, characterized in that, The morphological opening operation is an operation that uses a specific structuring element to perform erosion followed by dilation, eliminating noise and accurately segmenting each line. The specific structuring element is a 3x3 pixel rectangular structuring element.

9. A method for detecting surface defects on a Litz line according to claim 6, characterized in that, The stranding quality index is calculated based on data collected over a period of time to determine line width uniformity, angle consistency, and pitch stability, and then substituted into the stranding quality index formula according to preset weighting coefficients. Where TQA represents the stranding quality index, S W S represents line width uniformity. A Represents consistency of angle, S P Pitch stability, These are the weighting coefficients.

10. A method for detecting surface defects on a Litz line according to claim 9, characterized in that, The formula for calculating the linewidth uniformity is as follows: , where k W CV is a coefficient. W To measure the width of each strand along a continuous segment of the Leeds line (101), calculate the coefficient of variation (CV) of all strand widths. W =Standard deviation / Mean; The formula for calculating the angular consistency is as follows: , where k A σ is a coefficient. A The standard deviation of the angle; The formula for calculating the pitch stability is as follows: , where k P CV is a coefficient. P is the coefficient of variation of the pitch values ​​on a continuous segment of the Lids line (101).

Images