Defect detection method and device for epoxy insulating material and computer equipment

Abstract

The embodiment of the invention discloses a defect detection method and device for an epoxy insulating material and computer equipment. The method comprises the following steps: acquiring a sample surface image of an epoxy insulating material sample; preprocessing the sample surface image to obtain a processed target sample surface image; inputting the surface image of the target sample into a preset surface defect detection model to obtain surface image defect data for the epoxy insulating material sample; performing internal defect detection on the target defect sample according to the surface image defect data to obtain internal defect data, and judging the target defect sample according to the surface image defect data; and obtaining a defect detection report for the epoxy insulating material sample according to the surface image defect data and the internal defect data. According to the method, surface image defect data and internal defect data are combined, comprehensive defect diagnosis is carried out on the epoxy insulating material sample, surface defects are not limited, meanwhile, the internal defects can be accurately detected, and comprehensive analysis and evaluation on the material defects are ensured.

Description

Technical Field

[0001] The present invention relates to defect detection technology, and in particular to a defect detection method, device and computer equipment for epoxy insulating materials. Background Art

[0002] Accurate defect diagnosis is essential in the application of epoxy insulation materials, especially in high-voltage power equipment. Epoxy insulation materials are widely used in the power industry due to their excellent electrical insulation properties and mechanical strength. However, these materials are prone to surface and internal defects such as cracks, bubbles or foreign matter embedding during production and use. These defects may affect the insulation properties of the material, increase the risk of equipment failure, and even lead to serious safety accidents.

[0003] Traditional defect detection methods usually focus on the detection of surface defects, such as visual inspection or using simple instruments. Although these methods are effective in some cases, they cannot accurately detect tiny defects inside the material. The detection of internal defects usually requires more complex technologies, such as X-ray or CT scanning, which are costly, complicated to operate, and cannot achieve large-scale rapid detection.

[0004] Additionally, existing technologies often struggle to rapidly test large numbers of products while maintaining high accuracy. Summary of the invention

[0005] The main purpose of the present invention is to provide a defect detection method, device and computer equipment for epoxy insulating materials, which are used to solve the technical problems that epoxy insulating materials cannot be fully (surface and internal) defect detection and the defect accuracy rate is not high.

[0006] To achieve the above objectives, the present application provides, in a first aspect, a defect detection method for epoxy insulating material, the method comprising:

[0007] Acquire sample surface images of epoxy insulation material samples;

[0008] Preprocessing the sample surface image to obtain a processed target sample surface image;

[0009] Inputting the target sample surface image into a preset surface defect detection model to obtain surface image defect data for the epoxy insulation material sample;

[0010] Performing internal defect detection on a target defect sample according to the surface image defect data to obtain internal defect data, wherein the target defect sample is determined by the surface image defect data;

[0011] A defect detection report for the epoxy insulation material sample is obtained according to the surface image defect data and the internal defect data.

[0012] Optionally, the target defect sample is subjected to internal defect detection according to the surface image defect data to obtain internal defect data, and the target defect sample is determined by the surface image defect data, including: obtaining the defect area ratio of the surface image defect data; if the defect area ratio is greater than a preset threshold, determining that the epoxy insulation material sample is a target defect sample; and performing internal defect detection on the surface image defect data corresponding to the target defect sample to obtain internal defect data.

[0013] Optionally, obtaining the defect area ratio of the surface image defect data includes: obtaining at least one defect area in the surface image defect data and boundary information corresponding to each defect area in the at least one defect area; calculating the boundary information corresponding to each defect area to obtain the defect area corresponding to each defect area; counting the defect area corresponding to each defect area to obtain a total defect area, wherein the total defect area is the defect area corresponding to all defect areas in the surface image defect data; obtaining the surface image area in the surface image defect data; and obtaining the defect area ratio based on the surface image area and the total defect area.

[0014] Optionally, performing internal defect detection on the surface image defect data corresponding to the target defect sample to obtain internal defect data includes: performing internal defect detection on the surface image defect data based on ultrasound to obtain a first ultrasound and a second ultrasound, the first ultrasound being an ultrasound reflected by a non-defective area in the surface image defect data, and the second ultrasound being an ultrasound reflected by a defective area in the surface image defect data; obtaining a first time difference corresponding to the first ultrasound and a second time difference corresponding to the second ultrasound; obtaining an internal defect depth corresponding to the defective area in the surface image defect data based on a preset formula, the first time difference and the second time difference; and counting the internal defect depths corresponding to all defective areas in the surface image defect data to obtain internal defect data.

[0015] Optionally, before performing internal defect detection on the surface image defect data according to ultrasound to obtain the first ultrasound and the second ultrasound, the method also includes: acquiring a target defect area in a dense area in the surface image defect data; taking the target defect area as a starting point, performing internal defect detection on the defect areas in the surface image defect data one by one according to ultrasound.

[0016] Optionally, obtaining a defect detection report for the epoxy insulation material sample based on the surface image defect data and the internal defect data includes: performing regional data division on the surface image defect data and the internal defect data to obtain at least one set of defect region data, wherein the defect region data is each defect region in the surface image defect data and internal defect data corresponding to each defect region; performing data analysis on the at least one set of defect region data to obtain a defect detection report for the epoxy insulation material sample.

[0017] Optionally, the preprocessing of the sample surface image to obtain a processed target sample surface image includes: performing image enhancement processing on the sample surface image to obtain a first surface image; and performing standardization processing on the first surface image to obtain a processed target sample surface image.

[0018] On the other hand, the present application also provides a defect detection device for epoxy insulation material, comprising:

[0019] An acquisition module, used for acquiring a sample surface image of an epoxy insulation material sample;

[0020] A processing module, used for preprocessing the sample surface image to obtain a processed target sample surface image;

[0021] A detection module, used for inputting the surface image of the target sample into a preset surface defect detection model to obtain surface image defect data for the epoxy insulation material sample;

[0022] The detection module is further used to perform internal defect detection on the target defect sample according to the surface image defect data to obtain internal defect data, and the target defect sample is determined by the surface image defect data;

[0023] A determination module is used to obtain a defect detection report for the epoxy insulation material sample according to the surface image defect data and the internal defect data.

[0024] On the other hand, the present application also provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the first aspect and any possible implementation method thereof.

[0025] On the other hand, the present application further provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the processor executes each step of the method described in the first aspect.

[0026] The present application provides a defect detection method for epoxy insulation materials, firstly, collecting a sample surface image of an epoxy insulation material sample, secondly, preprocessing the sample surface image to obtain a processed target sample surface image, then inputting the target sample surface image into a preset surface defect detection model to obtain surface image defect data for the epoxy insulation material sample, performing internal defect detection on the target defect sample according to the surface image defect data to obtain internal defect data, the target defect sample is determined by the surface image defect data, and finally, obtaining a defect detection report for the epoxy insulation material sample according to the surface image defect data and the internal defect data. The method can quickly and accurately identify the surface defects of the epoxy insulation material sample through image processing technology and a preset surface defect detection model, and then further perform internal defect detection on the material determined as the target defect sample. This staged detection method greatly improves the efficiency and accuracy of defect detection of epoxy insulation materials. That is, by combining the surface image defect data and the internal defect data, a comprehensive defect diagnosis can be performed on the epoxy insulation material sample, which is not limited to surface defects, but also can accurately detect internal defects, ensuring a comprehensive analysis and evaluation of material defects. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0028] in:

[0029] Figure 1 A schematic flow chart of a defect detection method for epoxy insulation material provided in an embodiment of the present application;

[0030] Figure 2 A schematic diagram of an ultrasonic detection scenario provided in an embodiment of the present application;

[0031] Figure 3 A schematic structural diagram of a defect detection device for epoxy insulating material provided in an embodiment of the present application;

[0032] Figure 4 A schematic diagram of the structure of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0033] In order to enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present application.

[0034] The terms "first", "second", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes steps or units that are not listed, or optionally includes other steps or units inherent to these processes, methods, products or devices.

[0035] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0036] The technical solution of the present application can be applied to defect detection scenarios of various epoxy insulating materials.

[0037] The embodiments of the present application are described below in conjunction with the drawings in the embodiments of the present application.

[0038] See also Figure 1 , is a flow chart of a defect detection method for epoxy insulation material provided in an embodiment of the present application, such as Figure 1 As shown, the method includes:

[0039] 101. Collect sample surface images of epoxy insulation material samples.

[0040] Among them, the above-mentioned epoxy insulation material samples can be multiple or one, and epoxy resin insulation material samples of different sizes and shapes need to be cast according to the actual process.

[0041] For example, a two-stage curing process can be used, with one stage of curing at 130°C for 10 hours and the second stage of curing at 130°C for 18 hours. The shapes of epoxy resin insulation material samples include cubes, cuboids, curved disks, cylinders, and cones, and the sample thickness is within 90 mm. In this way, epoxy resin insulation material samples with different physical properties can be prepared to meet different testing requirements.

[0042] Among them, in the process of collecting, a high-resolution camera is required to obtain the surface image of the epoxy insulation material sample. The sample surface can be photographed or scanned at multiple angles to obtain comprehensive surface image data. The shooting angle and number of shots are not limited here.

[0043] Optionally, the collected surface image is saved in a digital format, such as JPEG, PNG or TIFF, etc., which is not limited here.

[0044] For example, the resolution of high-resolution cameras must reach 4K (4096x2160) or above. High resolution helps capture subtle defects. It uses industrial-grade CCD sensors, which have better image quality and a wider dynamic range. It uses "low distortion" lenses to avoid the impact of image distortion on defect detection.

[0045] It can be seen that in this embodiment, by collecting high-quality surface images of epoxy insulation material samples, an accurate data basis is provided for subsequent defect detection and analysis.

[0046] 102. Preprocess the sample surface image to obtain a processed target sample surface image.

[0047] The above preprocessing may include but is not limited to grayscale, noise reduction, contrast enhancement, pixel value normalization and unified image size.

[0048] Among them, when the above preprocessing is grayscale processing, the color image is converted into a grayscale image. During the grayscale process, the color value of each pixel will be converted into a grayscale value according to a specific formula, usually based on the weighted average calculation of the RGB color model, thereby reducing the interference of color information on subsequent processing.

[0049] Among them, when the above preprocessing is noise reduction processing, a specific algorithm is used to reduce the noise in the image. Noise is inevitable in the process of image acquisition and processing, and it will interfere with the clarity and accuracy of the image. Commonly used noise reduction methods include median filtering, Gaussian filtering, etc. These methods can effectively reduce random noise while retaining the main features of the image, that is, noise reduction processing helps to reduce noise interference in the image, making the detection of defects more accurate.

[0050] Among them, when the above preprocessing is contrast enhancement processing, the features in the image are enhanced by adjusting the brightness and contrast of the image, so that the difference between the target and the background in the image is more obvious. In this solution, it helps to highlight the defects in the image and improve the accuracy and reliability of defect detection. Contrast enhancement can be achieved through technologies such as histogram equalization and local contrast enhancement.

[0051] Among them, when the above preprocessing is pixel value normalization processing, the pixel value range of the image is adjusted from the original dynamic range to the range of 0-1. Normalization processing can standardize the pixel values ​​of the image, so that the image data is processed on the same scale, and make the subsequent feature extraction and model training more stable.

[0052] Among them, when the above preprocessing is to unify the image size, the image size is adjusted to a uniform size to facilitate feature extraction and model training. Unifying the image size is usually achieved through image scaling technology, and an appropriate interpolation method (such as nearest neighbor interpolation, bilinear interpolation, etc.) can be selected to ensure the quality of the image scaling process. Unifying the image size helps improve the efficiency of model training and ensures that the feature extraction results between different images are comparable.

[0053] The target sample surface image is a result image obtained during the preprocessing of the epoxy insulation material sample surface image.

[0054] In one embodiment, the preprocessing of the sample surface image to obtain a processed target sample surface image includes: performing image enhancement processing on the sample surface image to obtain a first surface image; and performing standardization processing on the first surface image to obtain a processed target sample surface image.

[0055] The above-mentioned image enhancement processing may include but is not limited to grayscale, noise reduction, contrast enhancement, etc. For details of grayscale, noise reduction, and contrast enhancement, please refer to the explanation in 102.

[0056] The above-mentioned standardization processing may include but is not limited to pixel value normalization and unified image size. For specific pixel value normalization and unified image size, please refer to the explanation in 102.

[0057] It can be seen that, in this embodiment, through the above preprocessing steps, a processed surface image of the target sample can be obtained, which provides accurate and high-quality image data for subsequent defect detection and analysis.

[0058] 103. Input the target sample surface image into a preset surface defect detection model to obtain surface image defect data for the epoxy insulation material sample.

[0059] Among them, the above-mentioned preset surface defect detection model can be a DETR (Detection Transformer) model, which is not limited here.

[0060] Specifically, the above DETR model consists of three parts: the backbone network, the Transformer attention module, and the classification prediction. Among them, the backbone network uses ResNet-50 as the feature extraction network. The preprocessed image is input through the backbone network ResNet-50 to extract the feature map of the image. ResNet-50, as a feature extraction network, can effectively capture the key information in the image. The Transformer attention module is specifically used to add position encoding to the feature map obtained in the previous step, and then the encoder token sparsification is performed using the scoring network to reduce the computational cost by selecting a part of the tokens that are most relevant to object detection, and the decoder cross-attention map is used to determine the importance of each input token. This step is implemented by the decoder of the DETR model, and the number of decoder layers matches the number of encoder layers, usually 6 layers. In this way, the model can more effectively focus on important feature areas in the image, and apply auxiliary detection losses to the selected tokens in the encoder, which helps to improve detection performance while minimizing computational overhead. Classification prediction is specifically used to predict various objects in the final output image, including their categories and bounding boxes. This process involves classifying and locating defects based on the features processed by the Transformer module. The number of object queries is set based on the maximum number of defects expected to be detected to optimize performance.

[0061] The parameters of the DETR model are as follows: the backbone network uses ResNet-50 as the feature extraction network; the number of encoder layers is 6; the number of decoder layers is 6, matching the number of encoder layers to optimize performance. Token sparsification rate: 30% to 50%, adjusted according to the actual data set. Number of object queries: set according to the maximum number of defects expected to be detected.

[0062] Among them, the above-mentioned surface image defect data may include but is not limited to at least one defect area and the boundary information corresponding to each defect area in the at least one defect area, the boundary information may be a boundary box, boundary size, boundary area, etc., surface image area, category, etc.

[0063] It can be seen that the preset surface defect detection model in this embodiment can effectively identify defects in the surface image of epoxy insulation material samples, such as fine cracks, scratches, dents or other surface irregularities, which not only improves the detection speed, but also significantly improves the recognition ability of small-size and complex-shaped defects.

[0064] 104. Perform internal defect detection on a target defect sample according to the surface image defect data to obtain internal defect data, wherein the target defect sample is determined by the surface image defect data.

[0065] Among them, the above steps can be divided into the first step in the specific implementation: surface image defect data analysis, that is, analyzing the surface image defect data, evaluating the severity and distribution of the defects, and judging which samples may have internal defects based on the surface defect data. These samples are called target defect samples. The second step: select samples with higher defect rates in the surface defect data as target defect samples, that is, these samples may have internal defects and need further internal testing to confirm. The third step: perform internal defect detection on the target defect samples to obtain defect information inside the samples. Internal defect detection may include ultrasonic detection, X-ray detection, CT scanning and other technologies. The fourth step: obtain the defect data inside the sample through internal defect detection, including the location, size, type and other information of the defects. The above internal defect data is crucial for evaluating the quality and safety performance of the samples.

[0066] It can be seen that in this embodiment, by marking products with higher defect rates in the surface defect detection stage, internal defect detection technology is further used to detect internal defects, which can penetrate the interior of the material and accurately locate and evaluate internal defects such as bubbles, cracks or impurities.

[0067] 105. Obtain a defect detection report for the epoxy insulation material sample according to the surface image defect data and the internal defect data.

[0068] Among them, the above defect detection report not only summarizes the type, location, size and potential impact of the defects, but also provides an assessment of the impact on material properties and improvement suggestions.

[0069] Specifically, in the defect inspection report, the surface image defect data and the internal defect data are fused, that is, matched and integrated, to ensure that each surface defect corresponds to the corresponding internal inspection result. Through data fusion, the connection between the sample surface and internal defects can be established, providing more accurate information for the report. Based on the fused data, a defect inspection report for the epoxy insulation material sample is generated.

[0070] Among them, defect detection reports can be displayed visually, using charts, 3D models or other visualization tools to intuitively present the information in the defect detection report to users. Visualization helps users better understand the location and distribution of defects and their possible impact on material properties.

[0071] It can be seen that the present embodiment combines surface and internal defect detection technologies to achieve a comprehensive diagnosis of epoxy insulation material defects, which not only improves the detection efficiency but also reduces the failure rate and safety risks of the product through accurate diagnosis.

[0072] In one embodiment, the defect detection report for the epoxy insulation material sample is obtained based on the surface image defect data and the internal defect data, including: performing regional data division on the surface image defect data and the internal defect data to obtain at least one set of defect region data, wherein the defect region data is each defect region in the surface image defect data and internal defect data corresponding to each defect region; performing data analysis on the at least one set of defect region data to obtain the defect detection report for the epoxy insulation material sample.

[0073] Among them, regional data division means dividing the sample surface into at least one group of defect areas according to the position and shape of each defect in the surface image defect data, ensuring that each defect area corresponds to one or more internal defect detection results to form a group of defect area data. For each defect area, the corresponding internal defect data is collected, including the location, size, type and other information of the defect.

[0074] The at least one set of defect area data is a set of comprehensive defect area data created for each defect area, that is, each set of data describes the surface features and internal features of a defect area in detail.

[0075] The data analysis may be an in-depth analysis of at least one set of defect area data to evaluate the severity and distribution of each defect area and its impact on sample performance. The analysis may include calculation and comparison of parameters such as defect size, shape, density, depth, etc.

[0076] It can be seen that in this example, the surface image defect data and the internal defect data are divided by region to facilitate detailed analysis of each defect region. This regional data division helps to more accurately locate and evaluate defects, thereby improving the quality and reliability of defect detection reports.

[0077] This method can quickly and accurately identify the surface defects of epoxy insulation material samples through image processing technology and a preset surface defect detection model. Subsequently, internal defect detection is further performed on the materials determined as target defect samples. This phased detection method greatly improves the efficiency and accuracy of defect detection of epoxy insulation materials. That is, by combining surface image defect data and internal defect data, a comprehensive defect diagnosis of epoxy insulation material samples can be performed, which is not limited to surface defects, but also can accurately detect internal defects, ensuring a comprehensive analysis and evaluation of material defects.

[0078] In one embodiment, the target defect sample is subjected to internal defect detection based on the surface image defect data to obtain internal defect data, and the target defect sample is determined by the surface image defect data, including: obtaining the defect area ratio of the surface image defect data; if the defect area ratio is greater than a preset threshold, determining that the epoxy insulation material sample is a target defect sample; and performing internal defect detection on the surface image defect data corresponding to the target defect sample to obtain internal defect data.

[0079] Among them, the above-mentioned defect area ratio refers to the ratio of the total defect area to the total area of ​​the target sample surface image. The total defect area is calculated by calculating the area of ​​each defect area and adding up these areas to obtain the total defect area.

[0080] The above-mentioned preset threshold value may be determined based on historical data, industry standards or expert experience, for example, may be 20%, which is not a sole limitation here.

[0081] Among them, the above-mentioned target defect samples are samples whose defect area ratio is greater than the preset threshold, indicating that there may be more serious internal defects and internal testing is required for confirmation.

[0082] The above-mentioned internal defect data includes information such as the location, size, and type of the defect.

[0083] It can be seen that in this embodiment, the proportion of defective areas is used to effectively screen out samples that may have internal defects. For those samples with a high proportion of surface defective areas and internal defects, further analysis and processing are required to ensure high performance and reliability of detection.

[0084] In one embodiment, the defect area ratio of the surface image defect data is obtained, including: obtaining at least one defect area in the surface image defect data and boundary information corresponding to each defect area in the at least one defect area; calculating the boundary information corresponding to each defect area to obtain the defect area corresponding to each defect area; counting the defect area corresponding to each defect area to obtain the total defect area, and the total defect area is the defect area corresponding to all defect areas in the surface image defect data; obtaining the surface image area in the surface image defect data; and obtaining the defect area ratio based on the surface image area and the total defect area.

[0085] The boundary information may be a boundary box, a boundary size, a boundary area, etc., which is used to define the exact location and range of the defect.

[0086] Specifically, for each identified defect area, the preset surface defect detection model will provide a bounding box, which accurately defines the location of the defect area. The bounding box is usually composed of four coordinates, which describe the precise location of the defect area in the image.

[0087] Specifically, the boundary size is the size information of the defect area, that is, the size of the boundary box.

[0088] Specifically, the boundary area can be calculated by a preset surface defect detection model to determine the area of ​​the defect region, which is achieved by analyzing the number of pixels within the boundary box. The boundary area information helps to assess the impact of defects on overall material properties.

[0089] The surface image area is the area information of the entire surface image, specifically the total area within the image capture range, which can be calculated by the size and resolution of the image.

[0090] The at least one defective region may include different types of defects such as cracks, scratches, and pits.

[0091] The calculation of the defect area may involve a triangle area formula or other geometric calculation methods. Specifically, the boundary information of the defect area may be used to calculate the area of ​​each defect area through image processing software or algorithms, that is, counting and calculating the area of ​​pixels within the boundary to determine the actual size of the defect.

[0092] The areas of all identified defect regions are summed up to obtain the total defect area, which represents the sum of the areas of all defect regions on the sample surface.

[0093] The defect area ratio refers to the ratio of the total defect area to the surface image area, that is, the defect area ratio is obtained by dividing the total defect area by the total area of ​​the surface image. This ratio reflects the distribution degree of defects on the entire sample surface.

[0094] It can be seen that in this example, by obtaining the defect area ratio of the surface image defect data of the epoxy insulation material sample, this ratio can be used as an indicator to judge the severity of the sample defects and help determine which samples require further internal defect detection.

[0095] In one embodiment, the internal defect detection is performed on the surface image defect data corresponding to the target defect sample to obtain the internal defect data, including: performing internal defect detection on the surface image defect data based on ultrasound to obtain a first ultrasound and a second ultrasound, the first ultrasound being an ultrasound reflected from a non-defective area in the surface image defect data, and the second ultrasound being an ultrasound reflected from a defective area in the surface image defect data; obtaining a first time difference corresponding to the first ultrasound and a second time difference corresponding to the second ultrasound; obtaining an internal defect depth corresponding to the defective area in the surface image defect data based on a preset formula, the first time difference and the second time difference; and counting the internal defect depths corresponding to all defective areas in the surface image defect data to obtain the internal defect data.

[0096] Among them, the ultrasonic internal defect detection technology uses the ultrasonic reflection method to detect internal defects of epoxy insulation materials. The basic principle of ultrasonic reflection method for defect detection is: after the ultrasonic wave is emitted, defects larger than the detection sensitivity will reflect the ultrasonic wave, and the propagation time of the reflected wave is related to the defect depth. The amplitude of the reflected wave is usually related to the attenuation of the material and the equivalent reflection area of ​​the defect. The ultrasonic reflection method uses 1MHz, 1.5MHz, 2.5MHz, and 5MHz ultrasonic probes, and uses different gains (dB) to inspect epoxy insulation materials multiple times to ensure that defects of different sizes at different locations are detected. The principle is as follows: the lower the ultrasonic frequency, the greater the depth and depth range of the defect detection, but the lower the detection sensitivity. The greater the gain, the greater the detection depth, but too large a gain will produce noise and waveform oscillation.

[0097] For example, see Figure 2 , Figure 2 This is a schematic diagram of an ultrasonic testing scenario, which depicts a schematic diagram of an ultrasonic testing device. In this device, there are three main components, Ultrasonic pulse generator: This is a rectangular device with a display, keyboard and some control buttons. It generates ultrasonic pulses, which are transmitted by the probe. Probe (marked "T / R", which stands for transmit / receive): Connected to the ultrasonic pulse generator, it sends ultrasonic waves to the material being tested (in this case, the epoxy sample) and receives the echoes reflected back. This probe is responsible for detecting defects or taking measurements within the epoxy sample. Oscilloscope: This is a rectangular device with a display and some side buttons, which is connected to the probe through wires. The oscilloscope displays the received ultrasonic signal, usually in the form of a waveform or A-scan, which can be analyzed to evaluate the properties of the sample or identify defects. The PC (which can be a personal computer) is connected to the oscilloscope via a USB cable, indicating that the oscilloscope may be connected to a computer interface for further processing, analysis or display of data.

[0098] The first time difference refers to the time difference between the emission and reception of the first ultrasonic wave, while the second time difference refers to the time difference between the emission and reception of the second ultrasonic wave. These two time differences reflect the propagation time of the ultrasonic wave inside the sample and are related to the medium (defect or non-defect area) encountered by the ultrasonic wave.

[0099] The preset formula is as follows:

[0100]

[0101] In the above preset formula, D is the internal defect depth position, d is the thickness of the epoxy insulation material sample, T1 is the first time difference, and T2 is the second time difference.

[0102] Among them, the internal defect data includes the depth position information of each defect area, which is used to evaluate the distribution and severity of the internal defects of the sample.

[0103] It can be seen that in this example, this internal defect detection method based on ultrasonic reflection can effectively conduct a comprehensive quality assessment of epoxy insulation material samples. This method can not only identify subsurface defects, but also provide the specific location and size of the defects.

[0104] In one embodiment, before performing internal defect detection on the surface image defect data according to ultrasound to obtain the first ultrasound and the second ultrasound, the method also includes: acquiring a target defect area in a dense area in the surface image defect data; taking the target defect area as a starting point, performing internal defect detection on the defect areas in the surface image defect data one by one according to ultrasound.

[0105] The dense area refers to the concentrated area of ​​defect areas in the surface image defect data, that is, the place where the defects are most concentrated, which may contain more potential internal defects.

[0106] Ultrasonic testing is performed by sending ultrasonic pulses to the material and receiving the reflected signal. Changes in the reflected signal can reveal the defects inside the material. Ultrasonic testing can provide detailed information about the depth, size, shape and nature of the defects. By starting from the target defect area, detection resources can be allocated more efficiently, giving priority to those areas that are most likely to show significant problems.

[0107] After the target defect area is inspected, the inspection process will continue in a predetermined order, gradually covering other defect areas on the sample surface. This sequential execution method helps ensure that each defect area is fully inspected, while also improving the efficiency of inspection.

[0108] Among them, in the process of performing internal defect detection on the defect areas in the surface image defect data in sequence according to ultrasound with the target defect area as the starting point, a breadth-first search algorithm can be specifically used. That is, starting from the starting point, adjacent defect areas are detected one by one to ensure that all areas with concentrated surface defects are detected. Breadth-first search is a graph search algorithm that starts from a node, visits all adjacent nodes first, and then expands outward layer by layer to more distant nodes. In this scenario, this means that the detection will start from the defect concentration area and gradually expand outward to ensure that the surrounding areas are also inspected.

[0109] It can be seen that the method of combining surface image analysis and ultrasonic testing in this example can effectively identify and evaluate the internal defects of epoxy insulation material samples, which not only improves the accuracy of detection but also optimizes the detection process.

[0110] As another aspect of the embodiment of the present application, the embodiment of the present application provides a defect detection device for epoxy insulating materials. The defect detection device for epoxy insulating materials can be a software module, and the software module includes a number of instructions stored in a memory, and the processor can access the memory and call the instructions for execution to complete the defect detection method for epoxy insulating materials described in the above-mentioned various embodiments.

[0111] See also Figure 3 , Figure 3 Schematic diagram of a defect detection device for epoxy insulation material provided in an embodiment of the present application. Figure 3 As shown, the defect detection device of the epoxy insulation material comprises:

[0112] The acquisition module 301 is used to acquire a sample surface image of the epoxy insulation material sample;

[0113] The processing module 302 is used to pre-process the sample surface image to obtain a processed target sample surface image;

[0114] A detection module 303, used to input the target sample surface image into a preset surface defect detection model to obtain surface image defect data for the epoxy insulation material sample;

[0115] The detection module 303 is further used to perform internal defect detection on the target defect sample according to the surface image defect data to obtain internal defect data, and the target defect sample is determined by the surface image defect data;

[0116] The determination module 304 is used to obtain a defect detection report for the epoxy insulation material sample according to the surface image defect data and the internal defect data.

[0117] This method can quickly and accurately identify the surface defects of epoxy insulation material samples through image processing technology and a preset surface defect detection model. Subsequently, internal defect detection is further performed on the materials determined as target defect samples. This phased detection method greatly improves the efficiency and accuracy of defect detection of epoxy insulation materials. That is, by combining surface image defect data and internal defect data, a comprehensive defect diagnosis of epoxy insulation material samples can be performed, which is not limited to surface defects, but also can accurately detect internal defects, ensuring a comprehensive analysis and evaluation of material defects.

[0118] It can be understood that the relevant contents of each module in the above device have been described in detail in the aforementioned method embodiment, and the details can be referred to the contents in the method embodiment; that is, the defect detection device for epoxy insulation material provided in the present application can be performed as follows: Figure 1 Any steps in the illustrated embodiment will not be described in detail here.

[0119] See also Figure 4 , Figure 4 4 is a schematic diagram of a computer device provided in an embodiment of the present application. The computer device includes one or more processors 41 and a memory 42. The memory 42 is connected to the one or more processors 41, for example, connected to the processor 41 via a bus.

[0120] The processor 41 is configured to support the computer device to perform the corresponding functions in the method in the above method embodiment. The processor 41 can be a central processing unit (CPU), a network processor (NP), a hardware chip or any combination thereof. The above hardware chip can be an application specific integrated circuit (ASIC), a programmable logic device (PLD) or a combination thereof. The above PLD can be a complex programmable logic device (CPLD), a field-programmable gate array (FPGA), a generic array logic (GAL) or any combination thereof.

[0121] The memory 42 is used to store program codes, etc. The memory 42 may include a volatile memory (VM), such as a random access memory (RAM); the memory 42 may also include a non-volatile memory (NVM), such as a read-only memory (ROM), a flash memory, a hard disk drive (HDD) or a solid-state drive (SSD); the memory 42 may also include a combination of the above-mentioned types of memories.

[0122] The memory 42 can be used to store non-volatile software programs, non-volatile computer executable programs and modules, such as program instructions / modules corresponding to the defect detection method for epoxy insulating materials in the embodiment of the present application. The processor 41 executes various functional applications and data processing of the defect detection method for epoxy insulating materials and the defect detection device for epoxy insulating materials by running the non-volatile software programs, instructions and modules stored in the memory 42, that is, realizes the functions of each module or unit of the defect detection method for epoxy insulating materials and the defect detection device for epoxy insulating materials provided in the above method embodiment.

[0123] The memory 42 may include a program storage area and a data storage area, wherein the program storage area may store an operating system and an application required by at least one function. The data storage area may store data created according to the use of the defect detection device for epoxy insulating materials, etc. In some embodiments, the memory 42 may optionally include a memory 42 remotely arranged relative to the processor 41, and these remote memories 42 may be connected to the defect detection device for epoxy insulating materials via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0124] The one or more modules are stored in the memory 42, and when executed by the one or more processors 41, execute the defect detection method of epoxy insulation material in any of the above method embodiments, for example, execute the method steps described in the above method embodiments to realize the functions of the modules described in the above device embodiments.

[0125] In one embodiment, a computer-readable storage medium is further provided. The computer-readable storage medium stores a computer program. When the computer program is executed by a processor, the processor executes any step in the above method embodiment.

[0126] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0127] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0128] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A defect detection method for epoxy insulation material, characterized in that: The method comprises: Acquire sample surface images of epoxy insulation material samples; Preprocessing the sample surface image to obtain a processed target sample surface image; Inputting the target sample surface image into a preset surface defect detection model to obtain surface image defect data for the epoxy insulation material sample; Performing internal defect detection on a target defect sample according to the surface image defect data to obtain internal defect data, wherein the target defect sample is determined by the surface image defect data; A defect detection report for the epoxy insulation material sample is obtained according to the surface image defect data and the internal defect data.

2. The defect detection method of epoxy insulation material according to claim 1, characterized in that: The step of performing internal defect detection on a target defect sample according to the surface image defect data to obtain internal defect data, wherein the target defect sample is determined by the surface image defect data, comprises: Obtaining a defect area ratio of the surface image defect data; If the defect area ratio is greater than a preset threshold, the epoxy insulation material sample is determined to be a target defect sample; Internal defect detection is performed on the surface image defect data corresponding to the target defect sample to obtain internal defect data.

3. The defect detection method of epoxy insulation material according to claim 2, characterized in that: The obtaining of the defect area ratio of the surface image defect data includes: Acquire at least one defective area in the surface image defect data and boundary information corresponding to each defective area in the at least one defective area; Calculating the boundary information corresponding to each defect area to obtain the defect area corresponding to each defect area; Counting the defect area corresponding to each defect area to obtain a total defect area, where the total defect area is the defect area corresponding to all defect areas in the surface image defect data; Acquire the surface image area in the surface image defect data; The defect area ratio is obtained according to the surface image area and the total defect area.

4. The defect detection method of epoxy insulation material according to claim 2, characterized in that: The performing internal defect detection on the surface image defect data corresponding to the target defect sample to obtain internal defect data includes: Performing internal defect detection on the surface image defect data according to the ultrasonic wave to obtain a first ultrasonic wave and a second ultrasonic wave, wherein the first ultrasonic wave is an ultrasonic wave reflected from a non-defective area in the surface image defect data, and the second ultrasonic wave is an ultrasonic wave reflected from a defective area in the surface image defect data; Acquire a first time difference corresponding to the first ultrasonic wave and a second time difference corresponding to the second ultrasonic wave; Obtaining the internal defect depth corresponding to the defect area in the surface image defect data according to a preset formula, the first time difference, and the second time difference; The internal defect depths corresponding to all defect areas in the surface image defect data are counted to obtain internal defect data.

5. The defect detection method of epoxy insulation material according to claim 4, characterized in that: Before performing internal defect detection on the surface image defect data according to the ultrasonic wave to obtain the first ultrasonic wave and the second ultrasonic wave, the method further includes: Acquire a target defect area in a dense area in the surface image defect data; Taking the target defect area as a starting point, internal defect detection is performed on the defect areas in the surface image defect data in sequence according to ultrasound.

6. The defect detection method of epoxy insulation material according to claim 1, characterized in that: The defect detection report for the epoxy insulation material sample is obtained according to the surface image defect data and the internal defect data, including: Performing region data division on the surface image defect data and the internal defect data to obtain at least one set of defect region data, wherein the defect region data is each defect region in the surface image defect data and internal defect data corresponding to each defect region; Data analysis is performed on the at least one set of defect area data to obtain a defect detection report for the epoxy insulation material sample.

7. The defect detection method of epoxy insulation material according to claim 1, characterized in that: The preprocessing of the sample surface image to obtain a processed target sample surface image includes: Performing image enhancement processing on the sample surface image to obtain a first surface image; The first surface image is subjected to standardization processing to obtain a processed target sample surface image.

8. A defect detection device for epoxy insulating material, characterized in that: The device comprises: An acquisition module, used for acquiring a sample surface image of an epoxy insulation material sample; A processing module, used for preprocessing the sample surface image to obtain a processed target sample surface image; A detection module, used for inputting the surface image of the target sample into a preset surface defect detection model to obtain surface image defect data for the epoxy insulation material sample; The detection module is further used to perform internal defect detection on the target defect sample according to the surface image defect data to obtain internal defect data, and the target defect sample is determined by the surface image defect data; A determination module is used to obtain a defect detection report for the epoxy insulation material sample according to the surface image defect data and the internal defect data.

9. A computer device, characterized in that: The method comprises a memory and a processor, wherein the memory is connected to the processor, and the processor is used to execute one or more computer programs stored in the memory. When the processor executes the one or more computer programs, the computer device implements the method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the processor is caused to perform the steps of the method according to any one of claims 1 to 7.

Images