Communication line manufacturing defect detection method and equipment

Abstract

The invention is suitable for the technical field of cable defect detection, and particularly relates to a communication line manufacturing defect detection method and device, and the method comprises the steps: obtaining the optical detection data and point cloud data of a to-be-detected communication line; compensating the optical detection data and the point cloud data according to interference information of an environment where the communication line to be detected is located, obtaining compensated optical detection data and compensated point cloud data, and determining an interference level; determining an attenuation factor according to the compensated optical detection data, determining a curvature factor according to the compensated point cloud data, and determining a defect score based on the attenuation factor and the curvature factor; determining a trend index according to the historical defect score and the defect score of the communication line to be detected; and obtaining a manufacturing defect detection result of the to-be-detected communication line based on the defect score, the trend index and the interference level. The method can improve the defect detection precision and identify the quality problem in advance.

Description

Technical Field

[0001] This application belongs to the technical field of cable defect detection, and particularly relates to a method and equipment for detecting manufacturing defects of communication lines. Background Art

[0002] A communication line (communication cable) is a cable used to transmit electrical signals, data, or optical signals, and its structure includes a conductor, an insulating layer, a shielding layer, and an outer sheath. The conductor is responsible for the conduction of electric current or optical signals, the insulating layer isolates the contact between conductors or with the external environment, the shielding layer is used to reduce electromagnetic interference, and the outer sheath provides mechanical protection for the cable. Communication cables are widely used in fields such as telephone, Internet, television signal transmission, as well as data centers, enterprise networks, etc., ensuring the high-speed, stable, and secure operation of modern communication systems.

[0003] In the prior art, when detecting manufacturing defects of communication lines, it is easily affected by interference such as environmental light, background noise, vibration, etc., resulting in inaccurate detection data. Moreover, traditional detection methods often rely on single visual detection or optical detection, and it is difficult to accurately distinguish material defects and surface texture changes, resulting in the inability to comprehensively evaluate defects.

[0004] In summary, when detecting manufacturing defects of communication lines, there is a problem of insufficient defect detection accuracy. Summary of the Invention

[0005] The embodiments of this application provide a method and equipment for detecting manufacturing defects of communication lines, which can solve the problem of insufficient defect detection accuracy in the related art when detecting manufacturing defects of communication lines.

[0006] In a first aspect, the embodiments of this application provide a method for detecting manufacturing defects of communication lines, including:

[0007] Obtain optical detection data and point cloud data of the communication line to be tested; wherein, the optical detection data is obtained through a multispectral imaging system, the multispectral imaging system includes visible light imaging and near-infrared imaging, and the point cloud data is obtained through laser scanning;

[0008] According to the interference information of the environment where the communication line to be tested is located, compensate the optical detection data and the point cloud data to obtain compensated optical detection data and compensated point cloud data, and determine the interference level;

[0009] Determine an attenuation factor according to the compensated optical detection data, and determine a curvature factor according to the compensated point cloud data, and based on the attenuation factor and the curvature factor, determine the defect score of the communication line to be tested; wherein, the attenuation factor is used to characterize the attenuation of the scattering intensity in the defect area of the communication line to be tested, and the curvature factor is used to characterize the surface curvature distribution of the communication line to be tested.

[0010] Determine a trend index based on the historical defect score and the defect score of the communication line to be tested; wherein, the historical defect score is the defect score of the communication lines in consecutive historical batches, and the trend index is an index used to quantify the direction and speed of the change in the quality of the communication line;

[0011] Obtain the manufacturing defect detection result of the communication line to be tested based on the defect score of the communication line to be tested, the trend index, and the interference level.

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

[0013] The communication line manufacturing defect detection method provided by the present application first obtains the optical detection data (obtained through a multispectral imaging system) and point cloud data (obtained through laser scanning) of the communication line to be tested, then compensates the optical detection data and point cloud data according to the interference information of the environment where the communication line to be tested is located to obtain the compensated optical detection data and compensated point cloud data, and determines the interference level. Then, determine the attenuation factor (the attenuation of the scattering intensity in the defect area of the communication line to be tested) according to the compensated optical detection data, and determine the curvature factor (the surface curvature distribution of the communication line to be tested) according to the compensated point cloud data, and determine the defect score of the communication line to be tested based on the attenuation factor and the curvature factor. Then, determine the trend index according to the historical defect score (the defect score of the communication lines in consecutive historical batches) and the defect score of the communication line to be tested. Finally, obtain the manufacturing defect detection result of the communication line to be tested based on the defect score, trend index, and interference level of the communication line to be tested. This method takes into account the interference factors of the environment where the communication line (communication cable) is located. By compensating the optical detection data and point cloud data, the influence of environmental interference on the detection result can be effectively reduced, and the compensated data can more truly reflect the state of the communication line to be tested, thereby improving the accuracy of defect detection. This method combines the comprehensive analysis of defect scores, trend indices, and interference levels. The obtained manufacturing defect detection result can comprehensively and accurately reflect the quality state of the communication line to be tested, and the interference level can provide a quantitative index regarding the environmental impact, further ensuring the credibility of the final detection result. This method can achieve real-time monitoring and quality assessment in the production process, help quickly discover potential defects in the communication line, identify quality problems in advance, reduce the flow of unqualified products into the market, and at the same time reduce the subsequent maintenance and repair costs caused by quality problems.

[0014] In a second aspect, an embodiment of the present application provides a communication line manufacturing defect detection device, including:

[0015] An acquisition unit for acquiring optical detection data and point cloud data of a communication line to be measured; wherein, the optical detection data is acquired by a multi-spectral imaging system, the multi-spectral imaging system includes visible light imaging and near-infrared imaging, and the point cloud data is acquired by laser scanning;

[0016] A compensation unit for compensating the optical detection data and the point cloud data according to the interference information of the environment where the communication line to be measured is located, obtaining the compensated optical detection data and the compensated point cloud data, and determining the interference level;

[0017] A defect score determination unit for determining an attenuation factor according to the compensated optical detection data, and determining a curvature factor according to the compensated point cloud data, and based on the attenuation factor and the curvature factor, determining the defect score of the communication line to be measured; wherein, the attenuation factor is used to characterize the attenuation of the scattering intensity in the defect area of the communication line to be measured, and the curvature factor is used to characterize the surface curvature distribution of the communication line to be measured;

[0018] A trend index determination unit for determining a trend index according to historical defect scores and the defect score of the communication line to be measured; wherein, the historical defect scores are the defect scores of communication lines in historical consecutive batches, and the trend index is an index used to quantify the direction and speed of the change in the quality of the communication line;

[0019] A defect detection result determination unit for obtaining the manufacturing defect detection result of the communication line to be measured based on the defect score of the communication line to be measured, the trend index and the interference level.

[0020] In a third aspect, an embodiment of the present application provides a communication line manufacturing defect detection device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, and when the processor executes the computer program, it implements the method according to any one of the embodiments in the first aspect.

[0021] It can be understood that the beneficial effects of the above second aspect to the third aspect can refer to the relevant descriptions in the first aspect above, and will not be repeated here. Description of the Drawings

[0022] 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 embodiments or the description of the prior art. Obviously, the following drawings 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.

[0023] Figure 1 It is a schematic flowchart of a communication line manufacturing defect detection method provided by an embodiment of the present application;

[0024] Figure 2 It is a schematic diagram of the implementation process for compensating optical detection data and point cloud data in the communication line manufacturing defect detection method provided by the embodiments of the present application;

[0025] Figure 3 It is a schematic diagram of the implementation process for determining the defect score of the communication line to be measured in the communication line manufacturing defect detection method provided by the embodiments of the present application;

[0026] Figure 4 It is a schematic diagram of the structure of the communication line manufacturing defect detection device provided by the embodiments of the present application;

[0027] Figure 5 It is a schematic diagram of the structure of the communication line manufacturing defect detection equipment provided by the embodiments of the present application. Detailed implementation manners

[0028] In the following description, specific details such as specific system structures and technologies are presented 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 clearly 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.

[0029] It should be understood that when used in the specification and appended claims of the present application, 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.

[0030] It should also be understood that the term " / and / " as used in the specification and appended claims of the present application refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0031] As used in the specification and appended claims of the present application, the term "if" can be interpreted as "when", "once", "in response to determining", or "in response to detecting" according to the context. Similarly, the phrase "if determined" or "if [the described condition or event] is detected" can be interpreted as meaning "once determined", "in response to determining", "once [the described condition or event] is detected", or "in response to detecting [the described condition or event]" according to the context.

[0032] In addition, in the description of the specification and appended claims of the present application, the terms "first", "second", "third", etc. are only used for distinguishing descriptions and cannot be understood as indicating or implying relative importance.

[0033] References to "one embodiment" or "some embodiments" etc. described in the specification of this application mean 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. The terms "comprising", "including", "having" and their variants all mean "including but not limited to", unless otherwise specifically emphasized.

[0034] In the related art, when detecting manufacturing defects of communication lines, it is easily affected by environmental light, background noise, vibration, etc., resulting in inaccurate detection data. Moreover, traditional detection methods often rely on single visual detection or optical detection, and it is difficult to accurately distinguish material defects from surface texture changes, resulting in an inability to comprehensively evaluate defects.

[0035] To solve the above problems, the embodiments of the present application provide a method and device for detecting manufacturing defects of communication lines. In this method, first, the optical detection data (obtained by a multispectral imaging system) and point cloud data (obtained by laser scanning) of the communication line to be tested are acquired. Then, according to the interference information of the environment where the communication line to be tested is located, the optical detection data and point cloud data are compensated to obtain the compensated optical detection data and compensated point cloud data, and the interference level is determined. Then, the attenuation factor (the attenuation of the scattering intensity in the defect area of the communication line to be tested) is determined according to the compensated optical detection data, and the curvature factor (the surface curvature distribution of the communication line to be tested) is determined according to the compensated point cloud data. Based on the attenuation factor and curvature factor, the defect score of the communication line to be tested is determined. Then, according to the historical defect score (the defect scores of communication lines in historical consecutive batches) and the defect score of the communication line to be tested, the trend index is determined. Finally, based on the defect score, trend index, and interference level of the communication line to be tested, the manufacturing defect detection result of the communication line to be tested is obtained. This method takes into account the interference factors of the environment where the communication line (communication cable) is located. By compensating the optical detection data and point cloud data, the influence of environmental interference on the detection result can be effectively reduced, and the compensated data can more truly reflect the state of the communication line to be tested, thereby improving the accuracy of defect detection. By combining the comprehensive analysis of the defect score, trend index, and interference level, the obtained manufacturing defect detection result can comprehensively and accurately reflect the quality state of the communication line to be tested, and the interference level can provide a quantitative index regarding the environmental impact, further ensuring the credibility of the final detection result. This method can realize real-time monitoring and quality assessment in the production process, help quickly discover potential defects of communication lines, identify quality problems in advance, reduce the flow of unqualified products into the market, and at the same time reduce the subsequent maintenance and repair costs caused by quality problems.

[0036] The method for detecting manufacturing defects of communication lines provided by the embodiments of the present application can be applied to a device for detecting manufacturing defects of communication lines. At this time, the device for detecting manufacturing defects of communication lines is the execution subject of the method for detecting manufacturing defects of communication lines provided by the embodiments of the present application. The embodiments of the present application do not impose any restrictions on the specific type of the device for detecting manufacturing defects of communication lines.

[0037] Exemplarily, the communication line manufacturing defect detection device may include a multispectral imaging device, a laser scanning device, and a control device communicatively connected to the multispectral imaging device and the laser scanning device. The multispectral imaging device is capable of capturing image data of a communication cable through spectra of different bands (such as visible light and near-infrared spectra), and may include a multispectral camera, an imaging spectrometer, a spectrometer, etc.; the laser scanning device is a device capable of scanning the surface of the communication cable to obtain point cloud data, and may include a 3D laser scanner, a handheld laser scanning device, a lidar (LiDAR) system, etc.; the control device is a device capable of controlling the multispectral imaging device, the laser scanning device, and performing data processing, and may be a tablet computer, an augmented reality (AR) / virtual reality (VR) device, a laptop computer, an ultra-mobile personal computer (UMPC), a netbook, a personal digital assistant (PDA), a desktop computer, a smart large screen, a computer, a laptop computer, a handheld computing device, a satellite wireless device, a wireless modem card, a customer premise equipment (CPE), and / or other devices for communicating on a wireless system, and next-generation communication systems, for example, a mobile terminal in a 5G network or a mobile terminal in a future evolved public land mobile network (PLMN).

[0038] To better understand the communication line manufacturing defect detection method provided by the embodiments of the present application, the following provides an exemplary introduction to the specific implementation process of the communication line manufacturing defect detection method provided by the embodiments of the present application.

[0039] Figure 1 The schematic flowchart of the communication line manufacturing defect detection method provided by the embodiments of the present application is shown. The communication line manufacturing defect detection method includes:

[0040] S100, obtaining optical detection data and point cloud data of a communication line to be measured. Among them, the optical detection data is obtained through a multispectral imaging system, the multispectral imaging system includes visible light imaging and near-infrared imaging, and the point cloud data is obtained through laser scanning.

[0041] It can be understood that visible light imaging can use a visible light camera or imaging device (such as an ordinary RGB camera), and the visible light camera or imaging device can capture the visible light band data (400nm to 700nm) of the communication line; near-infrared imaging can use a near-infrared camera to capture near-infrared data in the wavelength range of 700nm to 2500nm. The near-infrared camera is equipped with an infrared sensor and can obtain images in low-light conditions or in environments where visible light cannot penetrate.

[0042] Exemplarily, the multispectral imaging device can be fixed at a suitable position, such as mounted on a tripod, a mobile platform, or using a drone (UAV) for aerial photography, to ensure that the multispectral imaging device can completely cover all parts of the communication line. The specific installation method depends on the test environment of the communication line.

[0043] A multispectral imaging system can be used to collect image data of the communication line to be tested from the visible light and near-infrared bands respectively. Appropriate shooting parameters can be selected according to the characteristics of the communication line and the requirements of the target to be tested, such as exposure time, camera resolution, aperture, etc., and different filters can be switched to capture images in different bands respectively.

[0044] The obtained image data can be preprocessed, such as denoising, color correction, image fusion, etc. Especially in near-infrared images, light and reflection may cause certain interference. Combining visible light and near-infrared images to generate multispectral data (optical detection data) for subsequent analysis and defect detection.

[0045] It can be understood that a laser scanner emits laser light and measures the time it takes for the laser to reflect back from the surface of an object, thereby obtaining the three-dimensional coordinate points on the surface of the communication line. If the communication line to be tested is located in a small area or requires high-precision detection, a terrestrial laser scanner (such as FARO Focus, Leica RTC360) can be selected. Such scanners obtain the surface data of an object by emitting laser pulses. For the acquisition of point cloud data in a small area and local region, a handheld laser scanner (such as FARO Freestyle3D) can also effectively collect point cloud data. If the communication line to be tested is long and requires large-scale scanning, a mobile laser scanning system (such as Leica Pegasus, RIEGL VMX series) can be used. These devices can perform laser scanning on platforms such as vehicles and drones.

[0046] Exemplarily, the laser scanning device can be installed at an appropriate position so that the device can scan the surface of the communication line to be tested and the possible defect areas. If it is a handheld scanner, it can be directly moved along the surface of the communication line for scanning; if a mobile laser scanning system is used, the device can be installed on a mobile platform or a drone and scanned along the communication line.

[0047] The resolution, scanning angle, and scanning frequency of the laser scanner can be set to obtain high-quality point cloud data. The laser scanner emits laser pulses and measures the time it takes for the pulses to reflect back. Based on the reflected time, the three-dimensional coordinates of each laser point are calculated to form point cloud data. During the scanning process, device jitter or occlusion should be minimized to ensure the integrity and accuracy of the acquired data.

[0048] The obtained point cloud data can be post-processed, such as noise filtering, data registration, coordinate system transformation, etc. Point cloud processing software (such as CloudCompare, Autodesk ReCap, Leica Cyclone, etc.) can be used to visualize, analyze, and extract defects from the point cloud data. By analyzing the point cloud data, information such as the geometric shape, curvature, and surface defects of the communication line can be identified.

[0049] Through this step, an optical detection data can be obtained using a multispectral imaging system, and point cloud data can be obtained using a laser scanning device. Both will provide key bases for subsequent defect identification and quality assessment.

[0050] S200. According to the interference information of the environment where the communication line to be measured is located, compensate the optical detection data and the point cloud data to obtain the compensated optical detection data and the compensated point cloud data, and determine the interference level.

[0051] It can be understood that the interference information (interference source) can include: lighting interference, such as strong light, reflected light, shadows, etc. that affect the imaging quality of the optical imaging system; atmospheric interference, such as weather factors like temperature, humidity, and wind speed that affect the accuracy of optical detection and laser scanning; surface interference, such as pollutants, surface stains, or texture changes that affect the accuracy of the optical sensor; device errors, such as calibration errors and noise of the optical imaging system and laser scanner.

[0052] Exemplarily, compensating the optical detection data includes: using image processing algorithms (such as histogram equalization) to balance the brightness and contrast of the image and reduce the influence caused by lighting changes; compensating for the intensity change of the light source by correcting the light source (such as using a light source sensor to monitor the ambient light change); detecting and compensating for shadow areas, and using an edge-based shadow removal algorithm to correct the shadow part in the image.

[0053] The visible light image and the near-infrared image can be registered and then image fusion can be performed. By optimizing the image registration method (such as a feature-based matching algorithm), the spatial position deviation during imaging can be eliminated, and the image can be denoised using filters (such as Gaussian filtering and mean filtering) to reduce the influence caused by sensor noise or external environmental noise.

[0054] An environmental compensation model can be established according to changes in the environment (such as temperature, humidity, weather changes, etc.). Environmental data is collected through sensors, and the compensation amount is predicted by combining existing models and applied to the correction of optical detection data. Parameters such as the exposure time and gain of the sensor can be automatically adjusted according to changes in external interference sources.

[0055] The optical detection data obtained after the above compensation steps will be closer to the actual state of the communication line, removing the influence of environmental interference and being able to more accurately reflect the true situation of the communication line.

[0056] Compensation for point cloud data includes: Atmospheric refraction and weather conditions can cause changes in the propagation speed of laser signals, thus affecting the accuracy of point clouds. Compensation can be carried out by using an atmospheric refraction model in combination with real-time environmental conditions (such as temperature, humidity, air pressure, etc.). According to weather information (such as haze, rain, snow, etc.), the distance values in the point cloud data can be adjusted and processed using a compensation algorithm to reduce errors caused by weather interference.

[0057] The intensity values of the point cloud data obtained by a laser scanner may be affected by changes in surface reflectivity (such as the color, material, and illumination angle of the object surface). The influence of reflectivity changes on the results can be reduced by standardizing or normalizing the reflectivity during data processing. The angle between the laser beam and the object surface affects the reflection intensity. Therefore, the reflection intensity of the point cloud data can be compensated according to the scanning angle to reduce errors caused by the angle.

[0058] For point cloud data from multiple scanning positions, point cloud registration algorithms (such as the ICP algorithm) can be used to match the data at different scanning positions to eliminate errors caused by changes in the device position. Point cloud filtering methods (such as statistical filters, Voxel Grid filters) can be used to remove noise and smooth the point cloud to eliminate local noise caused by environmental interference.

[0059] The compensated point cloud data will eliminate errors caused by environmental interference (such as atmospheric refraction, reflectivity changes, etc.), and the obtained point cloud will be more accurate and can more accurately represent the three-dimensional shape of the communication line.

[0060] Determination of the interference level includes: The interference degree of the environment can be evaluated by collecting real-time environmental data (such as temperature, humidity, air pressure, light intensity, etc.). Based on the amplitude and frequency of environmental changes, the intensity of interference can be determined.

[0061] By evaluating the change in data quality before and after compensation, the influence of interference can be quantified. For example, standard image quality evaluation metrics (such as PSNR, SSIM, etc.) are used to evaluate optical data, and point cloud quality evaluation methods (such as point cloud density, accuracy, noise level) are used to evaluate the quality of point cloud data.

[0062] Based on historical data and real-time environmental monitoring data, an interference model can be established to quantify the impact of interference. By calculating the data error before and after compensation and comparing it with the set threshold, the interference level can be determined, and the interference level can be divided into multiple levels, such as low interference, medium interference, and high interference.

[0063] The optical detection data and point cloud data after compensation will be more accurate, which is helpful for subsequent defect detection and quality assessment.

[0064] In a possible implementation, please refer to Figure 2 , S200, compensating the optical detection data according to the interference information of the environment where the communication line to be measured is located to obtain the compensated optical detection data, including: S210, calculating the dust attenuation coefficient according to the dust concentration in the interference information; S220, calculating the refractive index correction parameter according to the temperature and humidity in the interference information; S230, compensating the optical detection data according to the dust attenuation coefficient and the refractive index correction parameter to obtain the compensated optical detection data.

[0065] It can be understood that dust or particulate matter will cause light scattering, resulting in the attenuation of the light intensity passing through the multispectral imaging system. When the dust concentration is high, the scattering effect is more significant, resulting in blurred images and decreased contrast. Therefore, the brightness and contrast of the image can be compensated according to the dust attenuation coefficient. Changes in temperature and humidity will cause changes in the refractive index of air, resulting in image distortion or distortion, especially in long-distance imaging. To eliminate this effect, the image can be optically corrected according to the refractive index correction parameter.

[0066] Exemplarily, the dust attenuation coefficient can be calculated according to the dust concentration and the wavelength of light. The calculation formula of the dust attenuation coefficient is α dust = 0.15·C dust ·λ -1.2 , where α dust is the dust attenuation coefficient, representing the attenuation intensity of dust on light; C dust represents the dust concentration, with the unit of hm -3 ; λ represents the wavelength of light, with the unit of nm.

[0067] The refractive index correction parameter can be calculated according to the real-time temperature and relative humidity of the environment where the communication line to be measured is located. The calculation formula of the refractive index correction parameter is Δn = 2.7×10 -7 ·RH·T -0.4 , where Δn represents the refractive index correction parameter, RH represents the relative humidity, and T represents the temperature.

[0068] It can be understood that the goal of optical detection data compensation is to address the degradation of image quality caused by environmental factors (such as dust, temperature and humidity changes, etc.) and limitations of the device itself (such as optical interference, lens distortion, etc.) during the imaging process.

[0069] Exemplarily, the dust attenuation factor can be calculated based on the dust attenuation coefficient and the distance from the pixel point in the image to the observation point. For each pixel in the image, the corresponding brightness value can be adjusted according to the dust attenuation factor corresponding to each pixel, that is, the light intensity after dust attenuation compensation is , where I 补 represents the light intensity after dust attenuation compensation of a certain pixel, I 初 represents the initial light intensity of this pixel, represents the dust attenuation factor, d1 represents the distance from this pixel point in the image to the observation point. If the pixel pitch of the image is a unit distance, the distance d1 is the attenuation effect corresponding to the physical position of each pixel. Through the calculation formula of the light intensity after dust attenuation compensation, the brightness attenuation caused by dust in the image will be compensated, and the image will restore its original brightness level.

[0070] In practical applications, when the dust concentration is high, it may cause a reduction in the overall contrast of the image. Therefore, on the basis of attenuation compensation, the contrast of the image can be further enhanced (such as using histogram equalization or other contrast enhancement algorithms) to restore the details of the image.

[0071] The refractive index correction parameter can be used to correct the propagation path of light in the image. The propagation path of each pixel in the image can be corrected through an optical model (such as an optical transfer function or a beam propagation model). Under a simplified model, the light intensity of each pixel in the image will be adjusted accordingly according to the change of the refractive index correction parameter. If it is a simple image distortion or perspective error, the correction can be achieved through some geometric transformations. For example, use perspective transformation (such as bilinear interpolation) to adjust the pixel position and restore the correct shape of the image. The refractive index correction parameter can be used to re-correct the image to facilitate the true reflection of the optical path in the image. This process can be completed through optical inversion technology or directly by correcting the light intensity distribution of the image.

[0072] After compensation, further enhancement operations can be performed on the image. The multi-scale Retinex algorithm can be used for global and local brightness adjustment of the image, which is beneficial to the brightness and contrast uniformity of the image at different scales, and further reduces the visual distortion caused by dust and refraction.

[0073] Through the above steps, optical images after dust attenuation and refractive index correction compensation can be obtained. These optical images will be closer to the true optical signals, reducing the influence of environmental factors (such as dust, temperature and humidity changes) on the image quality, restoring the brightness, contrast and geometric accuracy of the images, making the images more conform to the actual optical transmission situation, and improving the accuracy of subsequent image analysis and detection.

[0074] In a possible implementation, please refer to Figure 2 , S200, according to the interference information of the environment where the communication line to be measured is located, compensate the point cloud data to obtain the compensated point cloud data, including: S201, according to the dust attenuation coefficient, compensate the echo intensity of the point cloud data to obtain the first point cloud data, and use the interpolation algorithm to interpolate and repair the first point cloud data to obtain the second point cloud data; S202, according to the refractive index correction parameter, calculate the point cloud offset, and according to the point cloud offset, compensate the point cloud coordinates of the second point cloud data to obtain the compensated point cloud data.

[0075] It can be understood that the dust attenuation coefficient will affect the intensity of the echo signal, so the echo intensity in the point cloud data can be compensated to restore the original value of the echo intensity.

[0076] Exemplarily, the initial echo intensity of each point can be extracted from the point cloud data, and the echo intensity is based on the intensity of the laser reflected echo. The dust attenuation coefficient and the distance from each point in the point cloud to the sensor can be used to calculate the point cloud attenuation factor. Compensate the initial echo intensity of each point according to the point cloud attenuation factor, and the calculation formula of the compensated echo intensity is A 补 = A 初 × e -αdust·d2 , where, A 补 represents the compensated echo intensity of a certain point, A 初 represents the initial echo intensity of this point, e -αdust·d2 represents the point cloud attenuation factor, and d2 represents the distance from each point in the point cloud to the sensor. The echo intensity of each point in the point cloud data is compensated by the above method to obtain the first point cloud data, where the echo intensity of each point in the first point cloud data has been corrected.

[0077] It can be understood that due to the influence of dust or other factors, the first point cloud data may have missing points or noise. To repair these problems, the interpolation algorithm can be used to fill in the missing data points.

[0078] Interpolation algorithms can include linear interpolation, which is suitable for relatively uniform spaces; nearest neighbor interpolation, which is suitable for point cloud data with large gaps; K-nearest neighbor interpolation (KNN), which can perform interpolation based on the surrounding k points; and spline interpolation, which is suitable for complex point cloud data and provides smooth transitions.

[0079] Exemplarily, an appropriate interpolation algorithm can be selected. Based on the existing first point cloud data, the interpolation algorithm is used to repair missing or incomplete point cloud data. The interpolation can predict based on the data of the surrounding known points to generate new supplementary points. The point cloud data after interpolation repair is the second point cloud data, which includes the supplementary missing points and the repaired data points.

[0080] It can be understood that in the actual application of point clouds, the refractive index will cause the spatial position of the point cloud to shift. Therefore, the offset of each point can be calculated according to the refractive index correction parameter, and the point cloud coordinates can be adjusted.

[0081] Exemplarily, the spatial offset of each point can be calculated according to the refractive index correction parameter. Here, it is assumed that the coordinates of a certain point are (x, y, z), the spatial offset of this point is Δx = Δn·x, Δy = Δn·y, Δz = Δn·z, and the corrected coordinates of this point are x1 = x + Δx, y1 = y + Δy, z1 = z + Δz. Through this step, the coordinates of each point can be corrected, compensating for the spatial offset caused by the refractive index.

[0082] Through the above steps, the compensated point cloud data is obtained, including the echo intensity after dust attenuation correction, the point cloud data after interpolation repair, and the spatial coordinates after refractive index correction, which can effectively correct the influence of environmental factors on the point cloud data.

[0083] In a possible implementation, S200, determining the interference level, includes: S10, normalizing the dust attenuation coefficient and the refractive index correction parameter to obtain the normalized dust attenuation coefficient and the normalized refractive index correction parameter; S20, calculating the weighted sum of the weight corresponding to the dust attenuation coefficient and the normalized dust attenuation coefficient and the weight corresponding to the refractive index correction parameter and the normalized refractive index correction parameter to obtain the interference level.

[0084] Exemplarily, the dust attenuation coefficient can be mapped to the maximum value of 0.8 for normalization. The normalization formula for the dust attenuation coefficient is where α1 represents the normalized dust attenuation coefficient, The function is to map α dust to between 0 and 1 for normalization.

[0085] In most actual environments, the dust concentration and the characteristics of the particles cause the dust attenuation coefficient not to exceed 0.8. Generally speaking, when light passes through dust with a very high concentration, the attenuation coefficient will become very large, but it will not completely disappear or increase infinitely. 0.8 can be considered as an attenuation coefficient close to the limit. By setting the maximum value to 0.8, extreme results caused by excessive attenuation (such as an environment where light cannot penetrate at all) can be avoided. Mapping α dust to 0.8 (i.e., the maximum value is 1) is for the convenience of comparison with other parameters and weighted summation. As a physically reasonable upper limit value, 0.8 can ensure that the normalized parameters will not be too extreme and avoid unreasonable calculation results.

[0086] The absolute value of the refractive index correction parameter can be mapped to the maximum value of 5×10 -6 for normalization. The normalization formula for the refractive index correction parameter is where Δn1 represents the normalized refractive index correction parameter, The function is to map Δn to between 0 and 1 for normalization.

[0087] In the actual environment, the change in the refractive index generally does not exceed 5×10 -6 this value. For example, the influence of aerosols or dust on the refractive index of light is usually very small. 5×10 -6 as the maximum value can cover most actual situations. 5×10 -6 can be determined through a large number of experimental or observational data analyses. The influence of dust particles in nature or the atmosphere on the refractive index of light is relatively small. Therefore, 5×10 -6 this value is set as an upper limit. Similarly, setting the maximum value to 5×10 -6 is to normalize the refractive index correction parameter so that it can be used for consistent weighted sum calculations with other parameters (such as the dust attenuation coefficient). If the change in the refractive index is large, the normalization process can make quantities with different units comparable.

[0088] To calculate the interference level, weights corresponding to the normalized dust attenuation coefficient and the normalized refractive index correction parameter can be set, and the sum of the weight corresponding to the normalized dust attenuation coefficient and the weight corresponding to the normalized refractive index correction parameter is 1. For example, if it is considered that the dust attenuation has a greater impact, the weight corresponding to the normalized dust attenuation coefficient can be set to 0.6, and the weight corresponding to the normalized refractive index correction parameter can be set to 0.4. The weights can be adjusted according to experimental data or experience. After the weights are set, the normalized dust attenuation coefficient and the normalized refractive index correction parameter can be weighted and summed to obtain the interference level. The calculation formula is E = ω1·α1 + ω2·Δn1, where ω1 represents the weight corresponding to the normalized dust attenuation coefficient, and ω2 represents the weight corresponding to the normalized refractive index correction parameter.

[0089] Through normalization and weight weighting, different parameters (dust attenuation coefficient and refractive index correction parameter) can be compared and integrated on a unified scale, thus simplifying the interference calculation in complex environments. According to different environments and requirements, the influence of the dust attenuation coefficient and the refractive index correction parameter can be flexibly evaluated by adjusting the weights.

[0090] S300, determine the attenuation factor based on the compensated optical detection data, and determine the curvature factor based on the compensated point cloud data, and determine the defect score of the communication line to be measured based on the attenuation factor and the curvature factor. The attenuation factor is used to characterize the attenuation of the scattering intensity in the defect area of the communication line to be measured, and the curvature factor is used to characterize the surface curvature distribution of the communication line to be measured.

[0091] It can be understood that the scattering intensity refers to the intensity of light reflected back after interacting with the object surface. If there are defects (such as cracks, stains, etc.) on the object surface, it will cause the scattering intensity of light to decrease. The level of the attenuation factor reflects the attenuation of the scattering intensity in the defect area (such as scratches, cracks, etc.) of the communication line to be measured. The attenuation factor can be defined as the attenuation amount of the reflection intensity in the defect area relative to the normal area.

[0092] Exemplarily, in the compensated optical image (optical detection data), select the area where defects may exist. The defect area may exhibit a lower reflection intensity, representing the attenuation of scattered light. The reflection intensity of each pixel point can be calculated through the intensity value of the image. For visible light and near-infrared imaging, the reflection intensities of two bands can be calculated respectively. By comparing the reflection intensity of the target area with the reflection intensity of a normal area (an area without defects), calculate their ratio. For example, a defect-free reference area can be selected as a benchmark, calculate the average reflection intensity of this area, and then compare it with the average reflection intensity of the area to be measured. The calculation formula Among them, α represents the attenuation factor, I2 represents the average reflection intensity of the defective area, and I1 represents the average radiation intensity of the normal area. The attenuation factor is less than 1, and the smaller the attenuation factor, the more severe the defect and the more the intensity of the reflected light attenuates. The attenuation factor can not only be used for single-point defect detection, but also identify large-scale scattering intensity changes by performing regional analysis on the entire surface of the communication line, thereby locating possible defective areas.

[0093] It can be understood that the curvature factor can reflect the local geometric shape of the surface of the communication line to be measured, including areas where deformation or bending occurs on the surface of the communication line to be measured. The curvature factor can quantify the degree of concavity and convexity of the surface of the communication line to be measured.

[0094] Exemplarily, the least squares method can be used to fit the local surface of the point cloud and calculate the Gaussian curvature value on the surface. Each point of the point cloud data has a corresponding Gaussian curvature value, representing the local bending degree of the surface at that point. The calculation formula of Gaussian curvature: Among them, K represents the Gaussian curvature, and R represents the radius of curvature at that point. It is possible to judge whether there are areas of excessive bending or deformation on the surface of the communication line to be measured according to the Gaussian curvature value. A larger curvature value represents obvious defects or irregular deformations (such as cracks, depressions, etc.) on the surface, while a smaller curvature value indicates that the surface of the communication line is relatively flat.

[0095] The defect degree of the communication line to be measured can be more comprehensively evaluated by combining the attenuation factor and the curvature factor. Set the weights of the attenuation factor and the curvature factor according to the actual situation (for example, it can be optimized and adjusted through historical data), and use the weighted average method to calculate the final defect score.

[0096] By compensating the optical detection data and the point cloud data, the influence of environmental interference on the detection result can be effectively reduced. The compensated data can more truly reflect the state of the communication line to be measured, thereby improving the accuracy of defect detection. The attenuation factor can accurately describe the change in the scattering intensity of the defective area of the communication line, and the curvature factor can effectively characterize the distribution of the surface curvature of the communication line. Combining these two factors for comprehensive scoring can more comprehensively evaluate the quality of the communication line to be measured, and provide a basis for subsequent quality control and maintenance.

[0097] In a possible implementation, please refer to Figure 3, S300, determining the attenuation factor based on the compensated optical detection data, including: S310, calculating the current attenuation rate according to the compensated optical detection data, where the current attenuation rate is used to characterize the attenuation rate of the communication line to be measured; S320, calculating the attenuation deviation based on the current attenuation rate and the reference attenuation rate, where the reference attenuation rate is calculated based on the optical detection data of the standard communication line corresponding to the communication line to be measured; S330, calculating the weighted average of the signal-to-noise ratio data and the attenuation deviation in the compensated optical detection data to obtain the attenuation factor.

[0098] Exemplarily, construction of the reference attenuation rate: A defect-free cable certified by the laboratory (such as a sample fully inspected by X-ray and ultrasonic waves) can be used. Optical data, including the intensity of light at different wavelengths, is collected on the standard communication line (defect-free cable certified by the laboratory). Based on the attenuation characteristics of this standard communication line, an initial standard attenuation curve (reference attenuation rate) is established, that is where, α ref (λ) represents the reference attenuation rate, that is, the actual attenuation of the standard communication line at wavelength λ; I λ represents the light intensity of the standard communication line at wavelength λ, and I λ0 represents the light intensity of the standard communication line at the reference wavelength λ0, and d represents the length of the standard communication line. It can be recalibrated before each batch of production or after equipment maintenance to avoid sensor drift caused by long-term use, which is beneficial for the attenuation rate model to accurately reflect the optical characteristics of the cable.

[0099] During normal production, the reference curvature can be continuously optimized through intelligent algorithms: The test area is divided into multiple independent areas (such as each area is a 10mm×10mm grid) for independent measurement. Only when no defects are detected in N consecutive frames (such as N = 50) within the area, are its data allowed to participate in the update of the reference attenuation rate. The reference attenuation rate is recalculated based on the updated test data, and the relevant model is adjusted. It is also possible to test the impedance, capacitance, etc. of the cable section passed through optical detection. If the electrical parameters are normal, the accuracy of the optical reference data is confirmed in reverse; Regular sampling is carried out for microscopic observation or tensile testing to correct the reference attenuation rate.

[0100] The above steps ensure the accuracy and stability of the reference attenuation rate, enabling the reference attenuation rate to reflect the optical performance in the real environment.

[0101] Process of calculating the attenuation factor: The current attenuation rate can be calculated according to the above formula for the reference attenuation rate, that is where, α c (λ) represents the current attenuation rate, that is, the actual attenuation of the communication line to be measured at wavelength λ; I1 λ represents the light intensity of the communication line to be measured at wavelength λ, and I1 λ0It represents the optical intensity of the communication line to be measured at the reference wavelength λ0, and d represents the length of the communication line to be measured.

[0102] The relative attenuation deviation can be calculated from the current attenuation rate and the reference attenuation rate to analyze the attenuation of the defective area of the communication line to be measured. The formula for calculating the relative attenuation deviation is

[0103] It can be understood that 400nm to 900nm is the wavelength range of visible light, from violet (about 400nm) to red (about 700nm) and approaching near-infrared (up to 900nm). In the wavelength range of 400nm to 900nm, the relative attenuation deviation at each wavelength can be calculated and weighted according to the corresponding signal-to-noise ratio to obtain the attenuation factor. The signal-to-noise ratio is used to describe the quality of the signal. The higher the signal-to-noise ratio, the clearer the signal and the smaller the noise interference.

[0104] Exemplarily, the relative attenuation deviation between the communication line to be measured and the standard sample communication line at each wavelength (400nm - 900nm) can be calculated according to the above method. The weight corresponding to the relative attenuation deviation at each wavelength can be obtained through the ratio between the signal-to-noise ratio corresponding to each wavelength and the sum of the signal-to-noise ratios corresponding to all wavelengths. The attenuation factor can be obtained by weighted summation of the relative attenuation deviation corresponding to the weight at each wavelength and the relative attenuation deviation corresponding to each wavelength, that is where α represents the attenuation factor, SNR(λ) represents the signal-to-noise ratio corresponding to a certain wavelength, represents the sum of the signal-to-noise ratios corresponding to all wavelengths.

[0105] During the manufacturing process of the communication line, the communication line may have increased signal attenuation due to manufacturing defects (such as uneven fiber surface, loose connection points, material defects, etc.). The attenuation factor can help detect these defects. The attenuation factor can help discover problems in the manufacturing process by evaluating the attenuation of different regions of the communication line. For example, if there is abnormal attenuation in certain regions, it may indicate material defects or fiber quality problems in that region. By detecting changes in the attenuation factor, it can help accurately locate the regions with larger attenuation in the communication line, and these regions are potential defective areas. The value of the attenuation factor can be used as a standard for real-time quality control during the production process. If the attenuation factor exceeds the preset range during the production process, it can be considered that the communication line has defects or quality problems, and adjustments or re-production can be carried out in a timely manner. Therefore, the attenuation factor is mainly used to detect the optical attenuation problem inside the communication line, and to locate possible manufacturing defects such as material non-uniformity and connection problems by identifying regions with abnormal attenuation.

[0106] In a possible implementation, please refer to Figure 3, S300. Determine the curvature factor based on the compensated point cloud data, including: S301. Calculate the Gaussian curvature of each point cloud family according to the compensated point cloud data; S302. Based on the Gaussian curvature of each point cloud family and the reference curvature, calculate the absolute value of the difference between the Gaussian curvature of each point cloud family and the reference curvature, and mark the point cloud family with an absolute value greater than 0.15 times the reference curvature as abnormal; where the reference curvature is obtained by calculating the Gaussian curvature based on the point cloud data of the standard communication line corresponding to the communication line to be measured. The marking value corresponding to the marked abnormal point cloud family is 1, and the marking value corresponding to the marked normal point cloud family is 0; S303. Based on the Gaussian curvature of each point cloud family, calculate the total sum of the absolute values of the curvatures of all point cloud families, and determine the weight of each point cloud family based on the Gaussian curvature of each point cloud family and the total sum of the absolute values of the curvatures; S304. Obtain the curvature factor based on the weight of each point cloud family, the marking value of each point cloud family, and the scanning area of the compensated point cloud data.

[0107] Exemplarily, construction of the reference curvature: The theoretical curvature (Gaussian curvature) can be calculated according to cable design parameters (such as diameter, material elastic modulus, etc.). Gaussian curvature is a scalar that describes the curvature of a surface and is applicable to describing the curvature of a two-dimensional surface. The calculation formula is where z(x, y) represents the height function of the surface, are second-order partial derivatives, representing the curvature in the x and y directions respectively. The calculation results (theoretical curvature) can be verified through experiments. Use a high-precision 3D scanner (such as a CMM measuring instrument) to scan the standard communication line to obtain its curvature distribution histogram, and determine the normal fluctuation range (such as ±5%).

[0108] The ICP algorithm can be used to align and compare the scan data of the same communication line segment at different time points to exclude instantaneous interference. Use control charts (such as X-bar charts) to monitor the stability of the production process in real time. Calculating statistical quantities (such as mean, standard deviation) during the production process can be used to control the fluctuation of curvature. When the curvature exceeds the control limit (UCL or LCL), it indicates that the production process may be abnormal, and the update of the reference curvature can be paused. Clustering algorithms such as DBSCAN can be used to identify defective areas based on the density of curvature data points, generate a binary mask for the defective areas, and automatically exclude these areas in the calculation of reference data.

[0109] Curvature factor calculation process: The Gaussian curvature K of each point cloud family in the compensated point cloud data can be calculated according to the above formula for theoretical curvature. According to the marking formula, the point cloud family with an absolute value of the difference between the Gaussian curvature and the reference curvature greater than 0.15 times the reference curvature can be marked as 1 (abnormal), and other point cloud families can be marked as 0 (normal). The marking formula is where Flag i represents the marking value of the i-th point cloud family, and Ki represents the Gaussian curvature corresponding to the i-th point cloud family, K ref represents the reference curvature.

[0110] Weights can be assigned to each point cloud family based on the Gaussian curvature of each point cloud family and the sum of the absolute values of the Gaussian curvatures of all point cloud families. The weight calculation formula is where ω i represents the weight of the i-th point cloud family, |K i | represents the absolute value of the Gaussian curvature of the i-th point cloud family, represents the sum of the absolute values of the Gaussian curvatures of all point cloud families.

[0111] The curvature factor can be calculated based on the weight, marked value, and scanning area of each point cloud family, that is where C represents the curvature factor and A represents the scanning area. The curvature factor obtained through this calculation formula can be used to evaluate the overall quality of the scanning area. The closer the value of the curvature factor is to 1, the smaller the impact of the abnormal point cloud family on the overall area and the better the area quality; while the lower the value of the curvature factor, the greater the impact of the abnormal point cloud family and the worse the quality.

[0112] During the manufacturing process of communication lines, irregularities, bends, or compressions on the cable surface may affect the propagation of optical signals. The curvature factor can help identify these irregularities. The curvature factor can help detect whether the geometry of the communication line surface meets the standards. Areas with abnormal curvature changes may indicate defects on the communication line surface, such as unevenness, bends, or improper pressure. For possible surface defects (such as indentations, bends, wear, etc.) during the manufacturing process, the curvature factor can effectively reflect them. Abnormal curvature factor values are related to deformations or defects on the communication line surface. The curvature factor can be adaptively adjusted according to real-time scanning data to dynamically detect the surface state of the communication line, which is very effective for discovering defects caused by improper operations or equipment failures. Therefore, the curvature factor is mainly used to detect the surface quality and geometry of communication lines, and to discover possible surface defects, such as bends, indentations, or surface unevenness, by evaluating curvature changes.

[0113] In a possible implementation, please refer to Figure 3 , S300, to determine the defect score of the communication line to be measured based on the attenuation factor and the curvature factor, including: S101, calculate the inverse attenuation factor according to the attenuation factor; S102, perform a weighted sum of the inverse attenuation factor and its corresponding weight, and the curvature factor and its corresponding weight to obtain the defect score of the communication line to be measured.

[0114] Exemplarily, the value obtained by subtracting the attenuation factor from 1 can be used as the inverse attenuation factor. Subtracting the attenuation factor from 1 is used here to reverse the impact of the attenuation factor on the scoring, such that a higher attenuation factor (i.e., greater attenuation, more severe defects) corresponds to a higher defect score, while a lower attenuation factor (i.e., less attenuation, better quality) corresponds to a lower defect score. Through this step, when the attenuation factor is large, the score will be high, enabling the score to reflect the quality issues of the communication line, that is, a communication line with poor quality has a high defect score, and a communication line with good quality has a low defect score. In the weighted summation of defect scores, the attenuation factor and the curvature factor will be combined with certain weights. By reversing the influence of the attenuation factor (i.e., using 1 minus the attenuation factor), the weighted summation can more reasonably reflect the comprehensive quality of the communication line to be tested. When the attenuation factor is large (i.e., severe attenuation), the corresponding defect score will be high, indicating that the quality of the communication line is poor; while the curvature factor reflects the surface quality and plays a balancing role.

[0115] The weights corresponding to the inverse attenuation factor and the curvature factor can be set according to historical data or experience. For example, the weight corresponding to the inverse attenuation factor is 0.5, the weight corresponding to the curvature factor is 0.5, or the weight corresponding to the inverse attenuation factor is 0.6, and the weight corresponding to the curvature factor is 0.4. After the weights are set, the defect score of the communication line to be tested can be calculated based on the inverse attenuation factor and the curvature factor, that is, S = w1·(1 - α) + w2·C, where S represents the defect score of the communication line to be tested, w1 represents the weight corresponding to the inverse attenuation factor, and w2 represents the weight corresponding to the curvature factor.

[0116] The attenuation factor provides an internal quality indicator for optical signal transmission, while the curvature factor provides a surface quality indicator for the physical form. By combining the attenuation factor and the curvature factor to calculate the defect score, the manufacturing quality of the communication line to be tested can be comprehensively detected, potential defects can be discovered in a timely manner, and the performance and reliability of the communication line to be tested can be ensured.

[0117] S400. Determine the trend index based on the historical defect score and the defect score of the communication line to be tested. Among them, the historical defect score is the defect score of the communication lines in historical consecutive batches, and the trend index is an indicator used to quantify the direction and speed of the change in the quality of the communication line.

[0118] Exemplarily, the defect scores of the communication lines in historical consecutive batches can be recorded as a sequence, that is, H = {S1, S2,..., S n}, where S i represents the defect score of the i-th batch. Two methods can be used to calculate the trend index: Method 1 is the simple linear method, and the linear regression method can be used to fit the historical scores to obtain the trend of score change. It can be assumed that the historical score sequence H = {S1, S2,..., S n} The corresponding time points are \(t = \{1, 2, \ldots, n\}\), then the linear fitting equation of this historical score sequence is \(S\) t \(= at + b\), where \(a\) is the slope, representing the change rate of the defect score (i.e., the quality change trend), and \(b\) represents the intercept (which can be ignored). If \(a\gt0\), it means the quality is continuously deteriorating (the score is rising); if \(a\lt0\), it means the quality is continuously improving (the score is falling); if \(a = 0\), it means the quality is stable. In this method, the trend index can be defined as the degree of deviation of the current score from the historical linear trend, that is \(T\) represents the trend index, \(S\) n+1 represents the defect score of the communication line to be measured, represents the defect score predicted by the linear fitting equation, represents the deviation of the defect score of the communication line to be measured from the trend line, \(\lambda_1\) and \(\lambda_2\) represent the weight coefficients, controlling the influence of the two parts of the historical trend and the deviation of the latest score. Both \(\lambda_1\) and \(\lambda_2\) take 0.5. If \(T\gt0\), it means the quality has a deteriorating trend (the score is on the high side); if \(T\lt0\), it means the quality has an improving trend (the score is on the low side).

[0119] Suppose the historical score sequence is \(H=\{32, 35, 37, 39, 42\}\), and the current defect score \(S\) n+1 is 47. Through the linear fitting equation, \(a\) is calculated to be 2.5 and \(b\) is 29.5. Then the predicted defect score can be calculated as The trend index \(T = 0.5\times2.5+0.5\times(47 - 44.5)=2.5\), indicating that the quality of the communication line to be measured has significantly deteriorated.

[0120] Method 2 is the moving average trend comparison. The moving average value of the historical score sequence can be calculated, that is where \(\overline{H}\) represents the average value of the historical score sequence. In this method, the trend index can be defined as the relative change rate of the defect score of the communication line to be measured with respect to the average value of the historical score sequence, that is If \(T\gt0\), it means the current score is higher than the historical average and the quality is deteriorating; if \(T\lt0\), it means the current score is lower than the historical average and the quality is improving.

[0121] Suppose the historical score sequence is \(H=\{32, 35, 37, 39, 42\}\), and the current defect score \(S\) n+1 is 47. The average value of the historical score sequence is calculated as The trend index indicating that the defect score of the communication line to be measured has increased by 27% compared to the average value of the historical score sequence, and the quality of the communication line to be measured has significantly deteriorated.

[0122] This step provides strong data support for future communication line quality prediction. The trend index can quantify the change direction and speed of communication line quality, predict possible quality problems in advance, and provide a decision-making basis for production management, which helps with quality control and production optimization.

[0123] In a possible implementation, in S400, according to the historical defect scores and the defect scores of the communication line to be tested, determine the trend index, including: in S410, according to a sliding window with a set length of T, obtain the first T historical defect scores, and construct a score sequence based on the first T historical defect scores and the defect scores of the communication line to be tested; in S420, based on the score sequence, use a temporal convolutional network to calculate the trend state probability of the score sequence; in S430, determine the initial trend index according to the trend state probability; in S440, correct the initial trend index according to the interference level to obtain the trend index.

[0124] It can be understood that the trend index is an indicator used to quantify the change direction and speed of communication line quality. By analyzing historical and current defect scores, abnormal fluctuations (such as sudden defects) in the production process can be identified.

[0125] Exemplarily, the length T of the sliding window can be set to 60 (60 production batches). The purpose of the sliding window is to capture the defect scores of T historical batches. The defect scores of the first T batches can be extracted from historical data, and combined with the defect scores of the first T batches and the defect scores of the current communication line to be tested, a score sequence is constructed, that is, X = [S t-T ,S t-T+1 ,...,S t-1 ,S]. This score sequence contains the historical defect scores of the first T batches and the defect scores of the current communication line to be tested, and is used for subsequent trend calculation.

[0126] Based on the score sequence, a temporal convolutional network (TCN) can be used to calculate the trend state probability of the score sequence. The temporal convolutional network is a network structure for processing time series data. It learns the temporal features in the time series through convolutional layers, and then predicts the trend of the sequence. The input dimension of the TCN network is a univariate time series (score sequence). The number of channels in the convolutional layer of the TCN network gradually expands from 64 to 128 to 256 to extract deep temporal features. The 4-head multi-head attention mechanism of the TCN network captures the dependencies at different time scales. The TCN network outputs the probabilities of various trend states (stable, fluctuating, sudden change), and these trend states reflect the expected changes in the quality of the communication line to be tested in the future.

[0127] The initial trend index can be calculated based on the probabilities corresponding to the stable state, fluctuating state, and mutation state output by the TCN network, i.e., T = 0.5×P1 + 1.2×P2 + 2.5×P3, where T represents the initial trend index, P1 represents the probability corresponding to the stable state, P2 represents the probability corresponding to the fluctuating state, and P3 represents the probability corresponding to the mutation state. The stable state means that the quality of the communication line remains stable, with little change and no obvious faults or quality fluctuations. Therefore, the contribution of the stable state to the trend index is relatively small. So, a lower coefficient of 0.5 can be assigned to the stable state in the initial trend index calculation formula, indicating that the growth or change of the trend index in the stable state is relatively gentle. The fluctuating state indicates that the quality of the communication line has undergone moderate changes, which may be due to some minor defects or quality fluctuations, resulting in unstable performance of the communication line. Therefore, the coefficient of the fluctuating state reflects the moderate impact of quality fluctuations on the trend of the communication line, and a higher coefficient of 1.2 can be assigned. The mutation state means that the quality of the communication line has changed drastically, which may be due to sudden fluctuations or changes caused by major defects or faults. Since the mutation state indicates that the communication line has serious quality problems or faults, leading to a sharp decline in the performance of the communication line, its impact on the trend index is the greatest. To reflect this impact, the coefficient of the mutation state can be assigned a value of 2.5. Here, the coefficients (0.5, 1.2, 2.5) in front of each trend state category represent the degree of influence of different trend states on the final trend index. These coefficients are not limited to the values given in this application and can be set according to actual experience, experimental data, or system requirements.

[0128] The initial trend index can be corrected using the environmental sensitivity coefficient, which is determined by the dominant one between the normalized dust attenuation coefficient and the normalized refractive index correction parameter, i.e., which value is larger between the normalized dust attenuation coefficient and the normalized refractive index correction parameter. The environmental sensitivity coefficient corresponding to the normalized dust attenuation coefficient and the environmental sensitivity coefficient corresponding to the normalized refractive index correction parameter can be set. The initial trend index can be corrected according to the correction formula: T1 = T×(1 + k·E), where T1 represents the corrected trend index, k represents the environmental sensitivity coefficient, and E represents the interference level. To prevent overcorrection, an upper limit can be set, i.e., T1 = min(T1, 3.0).

[0129] Assume that the environmental sensitivity coefficient k corresponding to the normalized dust attenuation coefficient is 0.6, the environmental sensitivity coefficient k corresponding to the normalized refractive index correction parameter is 0.4, the initial trend index T is 2.1, the interference level E is 0.7, and the normalized dust attenuation coefficient is dominant. Then the corrected trend index is T1 = 2.1×(1 + 0.6×0.7) = 2.982.

[0130] The trend index provides a comprehensive assessment based on historical scores and current scores, which can accurately reflect the quality trend of the communication line. It can evaluate whether there are persistent problems in the manufacturing process of the communication line or whether sudden quality fluctuations occur. Through the trend index, the long-term change trend of the communication line quality can be identified, and early signals of quality decline can be detected in a timely manner. The trend index quantifies the quality status of the communication line into a single value through comprehensive analysis of different indicators (such as attenuation factor, curvature factor, environmental factors, etc.). This value can be used to determine whether the communication line meets the quality standards, providing a data-based defect detection result instead of a subjective judgment. This quantitative defect detection method greatly improves the accuracy and reliability of detection.

[0131] For S500, based on the defect score, trend index, and interference level of the communication line to be tested, the manufacturing defect detection result of the communication line to be tested is obtained.

[0132] Exemplarily, in order to make different types of data such as the defect score, trend index, and interference level of the communication line to be tested comparable, each data can be normalized. For example, a range (such as 0 to 100) can be set to normalize the defect score, so that the higher the value, the more serious the defect; a range (such as -1 to 1) can be set for the trend index; a grade range (such as 0 to 5) can be set for the interference level, where 0 means no interference and 5 means extremely high interference.

[0133] These three standardized data can be combined, and a weighted model is used to calculate the final manufacturing defect detection result. The weighted model is R = ω1·S + ω2·T + ω3·E, where R represents the manufacturing defect detection result, S represents the defect score of the communication line to be tested, E represents the interference level, ω1, ω2, and ω3 are the weights corresponding to the defect score, trend index, and interference level of the communication line to be tested respectively, indicating the influence degree of these three data on the final detection result. The setting of the weights can be adjusted according to the actual situation and can be determined through historical data analysis, expert experience, or machine learning models. If the manufacturing defect detection result is on the low side, it can indicate that the quality of the communication line to be tested is very good, with almost no defects, and it is suitable for continued use or leaving the factory; if the manufacturing defect detection result is close to the high value, it indicates that the communication line to be tested has serious defects, which may affect its normal operation or safety, and it needs to be repaired or replaced.

[0134] The trend index can help determine the change trend of defects. If the trend index is positive and the defect score gradually increases, it indicates that the quality may continue to deteriorate and improvement is needed in the production process. When the interference level is relatively high, it may indicate measurement errors caused by environmental or equipment problems, and it is necessary to further confirm the nature of the interference and correct the data.

[0135] The manufacturing defect detection result obtained by comprehensively calculating based on the defect score, trend index, and interference level can comprehensively evaluate the quality status of the communication line to be tested. Through reasonable weight allocation and standardization processing, it is conducive to the detection result reflecting the actual quality of the communication line to be tested and helping the production and maintenance teams make optimization decisions.

[0136] In a possible implementation, S500, based on the defect score, trend index, and interference level of the communication line to be tested, obtain the manufacturing defect detection result of the communication line to be tested, including: S510, based on the defect score, trend index, and interference level of the communication line to be tested, look up the manufacturing defect type in the decision matrix; where the decision matrix is a pre-set decision table according to different defect scores, trend indexes, and interference levels; S520, based on the defect score, trend index, and interference level of the communication line to be tested, calculate the result confidence level; S530, integrate the manufacturing defect type and the result confidence level to obtain the manufacturing defect detection result of the communication line to be tested.

[0137] Exemplarily, the decision matrix is shown in the following table:

[0138]

[0139] The manufacturing defect type corresponding to the communication line to be tested can be looked up from the above decision matrix according to the defect score, trend index, and interference level. If the defect score of the communication line to be tested is less than 0.3 and the trend index is greater than 2.0, it indicates that the communication line to be tested may have serious quality problems, and production should be stopped immediately; if the defect score of the communication line to be tested is between 0.3 and 0.5, the trend index is greater than 1.5, and the interference level is less than or equal to 0.5, it means that the communication line to be tested may have potential defects and can be tracked and marked; if the defect score of the communication line to be tested is between 0.3 and 0.5, the trend index is greater than 1.5, and the interference level is between 0.5 and 1.0, it indicates that the communication line to be tested may have quality problems and can be further reviewed and confirmed; if the defect score of the communication line to be tested is greater than 0.5 and the trend index is less than 1.0, it indicates that the quality of the communication line to be tested is normal and production can continue.

[0140] The confidence level of the detection result can be calculated through the result confidence level calculation formula, and the calculation formula is where Z represents the result confidence level. Substitute the defect score, trend index, and interference level of the communication line to be tested into the result confidence level calculation formula, and the result confidence level of the communication line to be tested can be calculated. The value range of the confidence level is from 0 to 1. The closer the confidence level is to 1, the higher the reliability of the judgment result; while the confidence level is close to 0, it indicates a higher uncertainty of the result.

[0141] After determining the type of manufacturing defect, the type of manufacturing defect can be integrated with the result confidence level to make a final defect detection decision. If the result confidence level is high, the determination result is relatively reliable, and the defect type can be directly used as the basis for production decisions; if the confidence level is low, further inspection and confirmation are required, and more data or review processes may be needed to confirm the accuracy of the defect determination.

[0142] It should be understood that the magnitudes of the sequence numbers of the steps in the above embodiments do not imply the order of execution. The order of execution of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.

[0143] Corresponding to the communication line manufacturing defect detection method described in the above embodiments, an embodiment of the present application further provides a communication line manufacturing defect detection device, and each unit of the device can implement each step of the communication line manufacturing defect detection method. Figure 4 The structural block diagram of the communication line manufacturing defect detection device provided by the embodiment of the present application is shown. For the sake of convenience of description, only the parts related to the embodiment of the present application are shown.

[0144] Referring to Figure 4 , the device includes:

[0145] An acquisition unit, configured to acquire optical detection data and point cloud data of a communication line to be measured. Among them, the optical detection data is acquired by a multispectral imaging system, the multispectral imaging system includes visible light imaging and near-infrared imaging, and the point cloud data is acquired by laser scanning.

[0146] A compensation unit, configured to compensate the optical detection data and the point cloud data according to the interference information of the environment where the communication line to be measured is located, obtain the compensated optical detection data and the compensated point cloud data, and determine the interference level.

[0147] A defect score determination unit, configured to determine an attenuation factor according to the compensated optical detection data, and determine a curvature factor according to the compensated point cloud data, and determine the defect score of the communication line to be measured based on the attenuation factor and the curvature factor. Among them, the attenuation factor is used to characterize the attenuation of the scattering intensity of the defect area of the communication line to be measured, and the curvature factor is used to characterize the surface curvature distribution of the communication line to be measured.

[0148] A trend index determination unit, configured to determine a trend index according to the historical defect score and the defect score of the communication line to be measured. Among them, the historical defect score is the defect score of the communication lines in historical consecutive batches, and the trend index is an index used to quantify the direction and speed of the change in the quality of the communication line.

[0149] A defect detection result determination unit, configured to obtain the manufacturing defect detection result of the communication line to be measured based on the defect score, the trend index, and the interference level of the communication line to be measured.

[0150] It should be noted that for the information interaction, execution process, etc. between the above units, since they are based on the same concept as the method embodiments of the present application, for their specific functions and the technical effects brought, reference can be specifically made to the method embodiment part, and details will not be repeated here.

[0151] Those skilled in the art can clearly understand that for the convenience and simplicity 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 process of the units and modules in the above device can refer to the corresponding process in the foregoing method embodiments and will not be repeated here.

[0152] The embodiment of the present application also provides a communication line manufacturing defect detection device. Figure 5 It is a schematic structural diagram of a communication line manufacturing defect detection device provided by an embodiment of the present application. The communication line manufacturing defect detection device includes a multi-spectral imaging device, a laser scanning device, and a control device communicatively connected to the multi-spectral imaging device and the laser scanning device. As Figure 5 shown, the control device 6 of the communication line manufacturing defect detection device in this embodiment includes: at least one processor 60 ( Figure 5 only one is shown in the figure), at least one memory 61 ( Figure 5 only one is shown in the figure), 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 communication line manufacturing defect detection device realizes the steps in any of the above communication line manufacturing defect detection method embodiments, or the functions of each unit in the above device embodiments.

[0153] Exemplarily, the computer program 62 may be divided into one or more modules / units, which are stored in the memory 61 and executed by the processor 60 to complete the present application. The one or more modules / units may 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 control device 6 of the communication line manufacturing defect detection device.

[0154] The control device 6 of the communication line manufacturing defect detection device may be a computing device such as a desktop computer, a notebook, a palm computer, or a cloud server. The communication line manufacturing defect detection device may include, but is not limited to, a processor 60 and a memory 61. Those skilled in the art can understand that Figure 5 merely examples of the communication line manufacturing defect detection device are provided, which do not constitute a limitation on the communication line manufacturing defect detection device. It may include more or fewer components than those shown in the figure, or combine certain components, or different components. For example, it may also include input / output devices, network access devices, a bus, etc.

[0155] The processor 60 may be a central processing unit (CPU), and the processor 60 may 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 may be a microprocessor or any conventional processor, etc.

[0156] The memory 61 may be an internal storage unit of the control device 6 of the communication line manufacturing defect detection device in some embodiments, such as a hard disk or memory of the communication line manufacturing defect detection device. The memory 61 may also be an external storage device of the communication line manufacturing defect detection device in 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 communication line manufacturing defect detection device. Further, the memory 61 may also include both an internal storage unit and an external storage device of the communication line manufacturing defect detection device. The memory 61 is used to store an operating system, application programs, a BootLoader, data, and other programs, such as program codes of the computer program. The memory 61 may also be used to temporarily store data that has been output or will be output.

[0157] An embodiment of the present application also provides a computer-readable storage medium storing a computer program, and when the computer program is executed by a processor, the steps in any of the above method embodiments are implemented.

[0158] An embodiment of the present application provides a computer program product, and when the computer program product runs on a communication line manufacturing defect detection device, the communication line manufacturing defect detection device is enabled to implement the steps in any of the above method embodiments.

[0159] If 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 such an understanding, to implement all or part of the processes in the above method embodiments of the present application, a computer program can be used to instruct relevant hardware to complete, and the computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps in the above 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 may at least include: any entity or device capable of carrying the computer program code to the communication line manufacturing defect detection device, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), 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 may not be an electrical carrier signal and a telecommunication signal.

[0160] In the above embodiments, the descriptions of the respective embodiments have their own emphases. For parts not described or recorded in a certain embodiment, reference may be made to the relevant descriptions of other embodiments.

[0161] 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. Professional technicians 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 this application.

[0162] In the embodiments provided in this application, it should be understood that the disclosed communication line manufacturing defect detection device, equipment, and method can be implemented in other ways. For example, the communication line manufacturing defect detection device and equipment embodiments described above are merely illustrative. For example, the division of the modules or units is only a logical function division, and there may 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 couplings or direct couplings or communication connections to each other can be through some interfaces, and the indirect couplings or communication connections of the devices or units can be in electrical, mechanical, or other forms.

[0163] The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they can be located in one place, or can be 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.

[0164] The above-described embodiments are only used to illustrate the technical solutions of this application and are not intended to limit them. Although this 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 for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method for detecting manufacturing defects of a communication line, characterized in that, Including: Obtain the optical detection data and point cloud data of the communication line to be tested; wherein, the optical detection data is obtained by a multispectral imaging system, the multispectral imaging system includes visible light imaging and near-infrared imaging, and the point cloud data is obtained by laser scanning; According to the interference information of the environment where the communication line to be tested is located, compensate the optical detection data and the point cloud data to obtain the compensated optical detection data and the compensated point cloud data, and determine the interference level; Determine the attenuation factor according to the compensated optical detection data, and determine the curvature factor according to the compensated point cloud data, and based on the attenuation factor and the curvature factor, determine the defect score of the communication line to be tested; wherein, the attenuation factor is used to characterize the scattering intensity attenuation of the defect area of the communication line to be tested, and the curvature factor is used to characterize the surface curvature distribution of the communication line to be tested; Determine the trend index according to the historical defect score and the defect score of the communication line to be tested; wherein, the historical defect score is the defect score of the communication lines in historical consecutive batches, and the trend index is an index used to quantify the change direction and speed of the communication line quality; Based on the defect score of the communication line to be tested, the trend index and the interference level, obtain the manufacturing defect detection result of the communication line to be tested.

2. The method for detecting manufacturing defects of communication lines according to claim 1, wherein Compensate the optical detection data according to the interference information of the environment where the communication line to be tested is located to obtain the compensated optical detection data, including: Calculate the dust attenuation coefficient according to the dust concentration in the interference information; Calculate the refractive index correction parameter according to the temperature and humidity in the interference information; Compensate the optical detection data according to the dust attenuation coefficient and the refractive index correction parameter to obtain the compensated optical detection data.

3. The communication line manufacturing defect detection method according to claim 2, characterized in that, Compensate the point cloud data according to the interference information of the environment where the communication line to be tested is located to obtain the compensated point cloud data, including: Compensate the echo intensity of the point cloud data according to the dust attenuation coefficient to obtain the first point cloud data, and use the interpolation algorithm to interpolate and repair the first point cloud data to obtain the second point cloud data; Calculate the point cloud offset according to the refractive index correction parameter, and compensate the point cloud coordinates of the second point cloud data according to the point cloud offset to obtain the compensated point cloud data.

4. The communication line manufacturing defect detection method according to claim 2, characterized in that, The determination of the interference level includes: Perform normalization processing on the dust attenuation coefficient and the refractive index correction parameter to obtain the normalized dust attenuation coefficient and the normalized refractive index correction parameter; Calculate the weighted sum of the weight corresponding to the dust attenuation coefficient and the normalized dust attenuation coefficient and the weight corresponding to the refractive index correction parameter and the normalized refractive index correction parameter to obtain the interference level.

5. The method for detecting manufacturing defects of a communication line according to claim 1, characterized in that, The determination of the attenuation factor according to the compensated optical detection data includes: Calculate the current attenuation rate according to the compensated optical detection data; wherein, the current attenuation rate is used to characterize the attenuation rate of the communication line to be tested; Calculate an attenuation deviation based on the current attenuation rate and a reference attenuation rate, where the reference attenuation rate is calculated based on optical detection data of a standard communication line corresponding to the communication line to be measured. Calculate a weighted average of the signal-to-noise ratio data in the compensated optical detection data and the attenuation deviation to obtain the attenuation factor.

6. The method for detecting manufacturing defects of a communication line according to claim 1, characterized in that, The determining the curvature factor according to the compensated point cloud data includes: Calculate the Gaussian curvature of each point cloud family according to the compensated point cloud data. Based on the Gaussian curvature of each point cloud family and a reference curvature, calculate the absolute value of the difference between the Gaussian curvature of each point cloud family and the reference curvature, and mark as abnormal the point cloud families whose absolute value is greater than 0.15 times the reference curvature; where the reference curvature is obtained by calculating the Gaussian curvature based on the point cloud data of a standard communication line corresponding to the communication line to be measured, the marking value corresponding to the marked abnormal point cloud family is 1, and the marking value corresponding to the marked normal point cloud family is 0. Based on the Gaussian curvature of each point cloud family, calculate the total sum of the absolute values of the curvatures of all point cloud families, and determine the weight of each point cloud family based on the Gaussian curvature of each point cloud family and the total sum of the absolute values of the curvatures. Obtain the curvature factor based on the weight of each point cloud family, the marking value of each point cloud family, and the scanned area of the compensated point cloud data.

7. The method for detecting manufacturing defects of a communication line according to claim 1, wherein The determining the defect score of the communication line to be measured based on the attenuation factor and the curvature factor includes: Calculate an inverse attenuation factor according to the attenuation factor. Perform a weighted sum of the inverse attenuation factor and the weight corresponding to the inverse attenuation factor, and the curvature factor and the weight corresponding to the curvature factor to obtain the defect score of the communication line to be measured.

8. The method for detecting manufacturing defects of a communication line according to claim 1, wherein, The determining the trend index according to the historical defect score and the defect score of the communication line to be measured includes: According to a sliding window with a set length of T, obtain the first T historical defect scores, and construct a score sequence according to the first T historical defect scores and the defect score of the communication line to be measured. Based on the score sequence, use a temporal convolutional network to calculate the trend state probability of the score sequence. Determine an initial trend index according to the trend state probability. Correct the initial trend index according to the interference level to obtain the trend index.

9. The method for detecting manufacturing defects of a communication line according to claim 1, wherein The obtaining the manufacturing defect detection result of the communication line to be measured based on the defect score of the communication line to be measured, the trend index, and the interference level includes: Based on the defect score of the communication line to be measured, the trend index, and the interference level, look up the manufacturing defect type in a decision matrix; where the decision matrix is a pre-set decision table according to different defect scores, trend indexes, and interference levels. Based on the defect score of the communication line to be measured, the trend index, and the interference level, calculate the result confidence. Integrate the manufacturing defect type and the result confidence to obtain the manufacturing defect detection result of the communication line to be measured.

10. A communication line manufacturing defect detection device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method according to any one of claims 1 to 9.

Images