A detection method, device, equipment and medium for a thin-film capacitor

Abstract

The present application relates to the field of image processing, and in particular, to a detection method, device, equipment and medium for thin film capacitors. The method includes: obtaining a first image and a standard image of the thin film capacitor collected by an X-ray imaging system, where the X-ray intensities of the first image and the standard image are the same; determining the similarity between the first image and the standard image; if the similarity is less than a preset similarity threshold, obtaining an X-ray image sequence of the thin film capacitor, and the X-ray image sequence sequentially includes multiple second images with increasing X-ray intensities from small to large; determining potential difference regions of the thin film capacitor according to adjacent second images with X-ray intensities; for each potential difference region, determining a third image in the second image corresponding to the potential difference region; performing anomaly detection on the third image to obtain a first detection result. The present application can improve the detection efficiency and accuracy.

Description

Technical Field

[0001] The present application relates to the field of image processing, and in particular, to a method, device, equipment, and medium for detecting thin film capacitors. Background Art

[0002] Thin film capacitors, as prominent products in the field of electronic components, are widely used in various electronic devices and undertake key functions such as energy storage and filtering. With the rapid development of technology, the volume of electronic devices is shrinking day by day, while the performance requirements are becoming increasingly stringent, which poses unprecedented challenges to the quality and reliability of thin film capacitors.

[0003] A thin film capacitor includes a metallized film, lead electrodes, an insulating medium, and a housing. Currently, the detection means for thin film capacitors are relatively single and mainly rely on visual inspection systems. These systems capture the appearance images of capacitors through high-resolution cameras and use advanced image processing algorithms to carefully analyze the images to identify possible defects such as scratches, stains, and deformations on the surface of the capacitors. However, the accuracy of detecting capacitor defects is insufficient, and only surface defects can be detected. For internal defects, related technologies still rely on manual inspection, and sampling inspection is carried out for a certain batch, with low inspection efficiency and low inspection accuracy. Summary of the Invention

[0004] The purpose of the present application is to provide a method, device, equipment, and medium for detecting thin film capacitors, which can improve the detection efficiency and accuracy.

[0005] In a first aspect, a method for detecting a thin film capacitor is provided, including:

[0006] Obtaining a first image and a standard image of the thin film capacitor collected by an X-ray imaging system, where the X-ray intensities of the first image and the standard image are the same;

[0007] Determining the similarity between the first image and the standard image;

[0008] If the similarity is less than a preset similarity threshold, obtaining an X-ray image sequence of the thin film capacitor, where the X-ray image sequence sequentially includes multiple second images with increasing X-ray intensities from small to large;

[0009] Determining a potential difference region of the thin film capacitor according to adjacent second images with different X-ray intensities;

[0010] For each potential difference region, determining a third image in the second image corresponding to the potential difference region;

[0011] Performing anomaly detection on the third image to obtain a first detection result.

[0012] Through the above technical solution, a first image and a standard image of the thin-film capacitor are obtained under the condition of the same X-ray intensity collected by the X-ray imaging system. By determining the similarity between the first image and the standard image, it can be preliminarily judged whether there is an abnormality in the capacitor. If the similarity is lower than the preset threshold, an X-ray image sequence is further obtained for detailed analysis. This progressive detection method is both efficient and accurate, avoiding unnecessary comprehensive detection and improving the detection efficiency and accuracy.

[0013] In an implementable manner, obtaining the X-ray image sequence of the thin-film capacitor includes: obtaining an initial X-ray image sequence of the thin-film capacitor, where the initial X-ray image sequence sequentially includes multiple initial second images with X-ray intensity increasing from small to large.

[0014] For each initial second image, obtain the outer package X-ray image of the outer package of the thin-film capacitor at the X-ray intensity corresponding to this initial second image; remove the outer package information in this initial second image according to the outer package X-ray image to obtain the second image corresponding to this initial second image.

[0015] After removing the outer package information of all initial second images, multiple second images are obtained, which constitute the X-ray image sequence of the thin-film capacitor.

[0016] Through the above technical solution, when obtaining the X-ray image sequence, first obtain the initial X-ray image sequence, and remove the outer package information for each initial second image, effectively eliminating the influence of the outer package on the internal detection and providing clear and interference-free image data for the subsequent determination of potential difference regions.

[0017] In an implementable manner, determining the potential difference region of the thin-film capacitor according to the second images with adjacent X-ray intensities includes:

[0018] Identify the first feature points of the first target image and the second feature points of the second target image in the second images with adjacent X-ray intensities.

[0019] Perform feature point matching according to the descriptors of the first feature points and the descriptors of the second feature points to determine the pairs of feature points with successful matching.

[0020] Align the first target image and the second target image according to the pairs of feature points with successful matching.

[0021] Calculate the gray difference value of each pixel point of the registered first target image and the second target image to obtain a difference image.

[0022] Obtain a binary image according to the difference image using a gray difference value threshold.

[0023] Determine the potential difference region of the thin film capacitor based on the binary image.

[0024] Through the above technical solution, when determining the potential difference region, the feature points in the adjacent X-ray intensity second image are identified and matched, the images are aligned, and the gray difference value is calculated to obtain the difference image; the difference image is binarized by using the gray difference value threshold, so as to clearly identify the potential difference region, which not only improves the detection accuracy, but also effectively reduces the possibility of false alarms and missed detections.

[0025] In an implementable manner, before obtaining the binary image by using the gray difference value threshold according to the difference image, it further includes:

[0026] Determine the local mean and local standard deviation of the window corresponding to each pixel point of the difference image;

[0027] Determine the gray difference value threshold corresponding to each pixel point according to the local mean and local standard deviation of the window corresponding to each pixel point.

[0028] Through the above technical solution, the local mean and local standard deviation of the window corresponding to each pixel point of the difference image are calculated, and the threshold is dynamically adjusted according to the local mean and local standard deviation to adapt to the feature changes of different regions, further improving the accuracy and reliability of the binary image.

[0029] In an implementable manner, it further includes:

[0030] Obtain the CCD image of the thin film capacitor;

[0031] Perform an appearance inspection on the thin film capacitor according to the CCD image to obtain a second inspection result;

[0032] If the second inspection result is that the appearance of the thin film capacitor is abnormal, a prompt message is generated, and the prompt message includes the abnormal type and the abnormal position.

[0033] Through the above technical solution, an appearance inspection link is added. By obtaining the CCD image of the thin film capacitor and performing an appearance inspection, abnormal phenomena on the surface of the capacitor can be found in time. When the appearance inspection result is abnormal, a prompt message is generated in time.

[0034] In an implementable manner, if the second inspection result is that the appearance of the thin film capacitor is abnormal, the abnormal type is bulging, and the first inspection result is an internal abnormality, the method further includes:

[0035] Obtain the first position information point set of the positioning mark of the third image in the X-ray coordinate system and the second position information point set of the positioning mark of the CCD image in the CCD coordinate system;

[0036] Register the third image and the CCD image according to the first set of position information points and the second set of position information points;

[0037] Determine the comprehensive abnormal information of the thin-film capacitor according to the potential difference area of the registered third image and the area corresponding to the potential difference area in the registered CCD image, where the comprehensive abnormal information includes: the type of abnormality and the position of the abnormality.

[0038] Through the above technical solution, in the case where a bulge is found in the appearance inspection and an abnormality is detected inside, the third image and the CCD image are registered, and the comprehensive abnormal information is determined by combining the information of the potential difference area and the corresponding area; the organic combination of the internal and appearance inspection results is realized, providing more comprehensive and accurate abnormal information for maintenance personnel and helping to quickly locate the problem.

[0039] In a feasible manner, the abnormal detection of the third image to obtain the first detection result includes:

[0040] Perform bubble detection on the third image to obtain the first sub-inspection result;

[0041] Perform crack detection on the third image to obtain the second sub-inspection result;

[0042] Perform impurity detection on the third image to obtain the third sub-inspection result;

[0043] Determine the first inspection result according to the first sub-inspection result, the second sub-inspection result and the third sub-inspection result.

[0044] Through the above technical solution, when performing abnormal detection on the third image, perform bubble, crack and impurity detections respectively, and comprehensively determine the first inspection result according to the detection results, improving the accuracy and comprehensiveness of the detection.

[0045] In a second aspect, a detection device for a thin-film capacitor is provided, including:

[0046] An acquisition module, configured to acquire a first image and a standard image of the thin-film capacitor collected by an X-ray imaging system, where the X-ray intensities of the first image and the standard image are the same;

[0047] A similarity determination module, configured to determine the similarity between the first image and the standard image;

[0048] The acquisition module is further configured to, if the similarity is less than a preset similarity threshold, acquire an X-ray image sequence of the thin-film capacitor, where the X-ray image sequence sequentially includes multiple second images with increasing X-ray intensities from small to large;

[0049] A potential difference area determination module, configured to determine a potential difference area of the thin film capacitor according to a second image of adjacent X-ray intensities;

[0050] A third image determination module, configured to determine a third image in the second image corresponding to the potential difference area for each potential difference area;

[0051] An anomaly detection module, configured to perform anomaly detection on the third image to obtain a first detection result.

[0052] In a third aspect, an electronic device is provided, which includes:

[0053] One or more processors;

[0054] A memory;

[0055] One or more applications, where one or more applications are stored in the memory and are configured to be executed by one or more processors, and one or more programs are configured to: execute operations corresponding to the method shown in any possible implementation manner of the first aspect.

[0056] In a fourth aspect, a computer-readable storage medium is provided, which stores at least one instruction, at least one segment of program, a code set or an instruction set, and at least one instruction, at least one segment of program, a code set or an instruction set is loaded and executed by a processor to implement the method shown in any possible implementation manner of the first aspect.

[0057] In a fifth aspect, a computer program product is provided, including a computer program, where the computer program implements the method shown in any possible implementation manner of the first aspect when being executed by a processor.

[0058] In summary, the present application includes at least one of the following beneficial technical effects:

[0059] 1. The first image and the standard image of the thin film capacitor under the same condition of X-ray intensity collected by the X-ray imaging system are obtained; by determining the similarity between the first image and the standard image, it can be preliminarily judged whether there is an anomaly in the capacitor; if the similarity is lower than a preset threshold, an X-ray image sequence is further obtained for detailed analysis. This progressive detection method is both efficient and accurate, avoiding unnecessary comprehensive detection and improving the detection efficiency.

[0060] 2. When obtaining the X-ray image sequence, the initial X-ray image sequence is first obtained, and the outer packaging information is removed for each initial second image, effectively eliminating the influence of the outer packaging on the internal detection and providing clear and interference-free image data for the subsequent determination of potential difference areas. Description of the Drawings

[0061] Figure 1It is a schematic flowchart of a detection method for a thin-film capacitor provided by an embodiment of the present application;

[0062] Figure 2 It is a schematic flowchart of a detection process for a thin-film capacitor provided by an embodiment of the present application;

[0063] Figure 3 It is a schematic structural diagram of a detection device for a thin-film capacitor provided by an embodiment of the present application;

[0064] Figure 4 It is a schematic structural diagram of an electronic device provided by an embodiment of the present application. Specific embodiments

[0065] This specific embodiment is only an interpretation of the present application and does not limit the present application. Those skilled in the art can make modifications without creative contributions to this embodiment after reading this specification, but as long as they are within the scope of the claims of the present application, they are protected by the patent law.

[0066] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. Apparently, the described embodiments are some, but not all, of the embodiments of the present application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application without making creative efforts belong to the scope of protection of the present application.

[0067] In addition, the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " in this article generally represents an "or" relationship between the associated objects before and after, unless otherwise specified.

[0068] Specifically, an embodiment of the present application provides a detection method for a thin-film capacitor. As Figure 1 shown, the method provided in the embodiment of the present application can be executed by an electronic device, which can be a server or a terminal device. Among them, the server can be an independent physical server, a server cluster or a distributed system composed of multiple physical servers, or a cloud server providing cloud computing services. The terminal device can be a smart phone, a tablet computer, a notebook computer, a desktop computer, etc., but is not limited thereto. The terminal device and the server can be directly or indirectly connected through wired or wireless communication methods, and the embodiment of the present application does not limit this. The method includes:

[0069] S101. Obtain the first image and the standard image of the thin film capacitor collected by the X-ray imaging system;

[0070] The X-ray intensities of the first image and the standard image are the same;

[0071] Among them, the X-ray imaging system is a device that uses X-rays for imaging, and can generate an internal structure image of the thin film capacitor by emitting X-rays and capturing the attenuation of the X-rays after passing through the object. Both the first image and the standard image are used to represent the images of the thin film capacitor collected by the X-ray imaging system. Among them, the first image refers to the image to be analyzed currently, and the standard image refers to the standard reference image without defects, which is the image of a capacitor with the same model size, etc. as the current thin film capacitor, and the X-ray intensities of the two are the same.

[0072] In the thin film capacitor production line, quality inspection can be performed. Specifically, the quality of the thin film capacitor is determined by the detection method provided in the embodiments of the present application. X-ray imaging detection is a non-destructive detection method that can intuitively display the internal structure of the capacitor. Specifically, when the capacitor production is completed, the mechanical arm / conveyor device on the automated production line is used to place it under the X-ray imaging system, adjust the X-ray intensity to a predetermined value, and collect the first image. This predetermined value can be set by technicians according to experience, and the embodiments of the present application do not limit it further. For the standard image, the capacitor image corresponding to the current thin film capacitor can be called from the database.

[0073] S102. Determine the similarity between the first image and the standard image;

[0074] Among them, the similarity is used to measure the similarity degree between two images. In the quality inspection of the thin film capacitor, the image of the defect-free capacitor is used as the standard image, and the similarity between the currently collected first image and the standard image is calculated; if the similarity is very high, it indicates that the current capacitor is very likely to be defect-free; if the similarity is low, it may indicate that the capacitor has defects.

[0075] It should be noted that when calculating the similarity, the first image and the standard image need to be registered, and the background image needs to be removed. At the same time, the image acquisition environment and camera parameters of the two images are the same.

[0076] In some possible cases, the first image and the standard image can also be preprocessed, such as denoising, enhancing contrast, etc., to improve the image quality, and then the similarity calculation is performed. In an implementable manner, key features in the first image and the standard image are extracted, such as edges, textures, shapes, etc. that can reflect the internal structure or appearance state of the capacitor; a feature matching algorithm, such as SIFT (Scale-Invariant Feature Transform), SURF (Speeded-Up Robust Features), is used to match the features in the first image with the features in the standard image; the similarity between the two images is evaluated according to the number of feature point matches. In another implementable manner, histogram equalization processing is performed on the first image and the standard image respectively; SSIM (Structural Similarity Index) or MSE (Mean Squared Error) is used to calculate the similarity between the first image and the standard image. It can be understood that other methods can also be used to determine the similarity, which is not limited here.

[0077] In this step, by comparing the similarity between the currently acquired image and the standard image, it can be preliminarily determined whether there are defects or abnormalities in the capacitor; if there are no abnormalities, it is determined that the detection result is normal, and if it is determined that there are abnormalities, it is necessary to further determine whether there are abnormalities and the situation of the abnormalities according to S103 - S106; this can greatly improve the detection accuracy and detection effect.

[0078] S103. If the similarity is less than the preset similarity threshold, then obtain the X-ray image sequence of the thin film capacitor,

[0079] The X-ray image sequence successively includes multiple second images with increasing X-ray intensity from small to large;

[0080] Among them, the preset similarity threshold is a preset value that can be set by the user according to actual needs. Exemplarily, it can be 90%, 95%, 98%, and is used to determine whether the similarity between the current image and the standard image meets the quality requirements. If the similarity is lower than this threshold, it is considered that there is a difference between the current image and the standard image, and it is necessary to further verify whether there are defects. If the similarity is not less than this threshold, it is considered that the current image is the same as the standard image and there are no defects.

[0081] The second image is a single image in the X-ray image sequence. The multiple second images successively show the internal structure of the capacitor under gradually increasing X-ray intensity. By comparing these images, the change situation of the internal structure of the capacitor can be determined, so as to judge whether there are defects.

[0082] The X-ray image sequence can show the internal structure of the capacitor at different X-ray intensities, and the settings for a series of X-ray intensities can be set by the user based on experience. Specifically, for each type of thin-film capacitor, technicians will set multiple groups of X-ray intensities according to the thin-film capacitor, then collect and compare the images, and then determine a series of X-ray intensities corresponding to each type of thin-film capacitor; and construct the corresponding relationship between the type and the X-ray intensity series. Therefore, in the embodiments of the present application, determine the type of the current thin-film capacitor, and based on the type of the current thin-film capacitor and this corresponding relationship, determine the corresponding X-ray intensity series; then based on the X-ray intensity series, control the X-ray imaging system to collect the X-ray image sequence of the thin-film capacitor, so as to obtain the X-ray image sequence of the thin-film capacitor. Specifically, automatically adjust the X-ray intensity of the X-ray imaging system to obtain the first second image with a lower X-ray intensity; gradually increase the X-ray intensity and sequentially collect the subsequent second images until the predetermined X-ray intensity range is reached; arrange all the collected second images in ascending order of X-ray intensity to form an X-ray image sequence.

[0083] S104. Determine the potential difference region of the thin-film capacitor according to the second images with adjacent X-ray intensities;

[0084] The second images with adjacent X-ray intensities are any two images that are adjacent in X-ray intensity, and the X-ray intensity of the previous image is less than that of the subsequent image, showing the internal structure of the capacitor under different X-ray penetration capabilities.

[0085] The potential difference region refers to the region of difference or change in the internal structure of the capacitor that appears due to the change in X-ray intensity in adjacent X-ray images, which may reflect defects inside the capacitor, such as bubbles, cracks, and impurities.

[0086] S105. For each potential difference region, determine the third image in the second image corresponding to the potential difference region;

[0087] The third image refers to the image selected from the second image corresponding to the potential difference region. The third image can be the image with the highest clarity in the second image corresponding to the potential difference region in general.

[0088] S106. Perform anomaly detection on the third image to obtain the first detection result.

[0089] After obtaining the third image corresponding to each potential difference region, if there is only one potential difference region, directly perform anomaly detection on the third image to determine its defect type and location. If there are at least two potential difference regions, determine the distance between the potential difference regions. If the distance is less than the distance threshold, which can be set by the technician according to actual needs or experience, it is determined that the two potential difference regions need to be merged. If the distance is not less than the distance threshold, there is no need to merge. After determining the potential difference regions that need to be merged, perform fusion of the potential difference regions according to the third images corresponding to the potential difference regions that need to be merged. When performing fusion, the pixel mean difference of the non-difference regions can be used to adjust one of the third images and then perform fusion, which can retain more details.

[0090] Exemplarily, for a thin film capacitor, it is photographed according to the first X-ray intensity, the second X-ray intensity, the third X-ray intensity, and the fourth X-ray intensity to obtain the first image, the second image, the third image, and the fourth image; the first X-ray intensity is less than the second X-ray intensity is less than the third X-ray intensity is less than the fourth X-ray intensity; then the first image and the second image are compared to obtain the first potential difference region; the second image and the third image are compared to obtain the second potential difference region; at this time, the ranges of the first potential difference region and the second potential difference region are not close, and the fault types such as bubbles, cracks, and impurities can be calculated respectively based on the first potential difference region; if the ranges of the first potential difference region and the second potential difference region are close, it may be the same fault type such as a bubble, and the first potential difference region and the second potential difference region are merged, and then the images corresponding to the two regions are used together to determine the difference type.

[0091] It can be seen that in the embodiment of the present application, the first image and the standard image of the thin film capacitor under the condition of the same X-ray intensity collected by the X-ray imaging system are obtained; by determining the similarity between the first image and the standard image, it can be preliminarily judged whether there is an anomaly in the capacitor; if the similarity is lower than the preset threshold, an X-ray image sequence is further obtained for detailed analysis. This progressive detection method is both efficient and accurate, avoiding unnecessary comprehensive detection and improving the detection efficiency.

[0092] Further, perform anomaly detection on the third image to obtain the first detection result, including: perform bubble detection on the third image to obtain the first sub-check result; perform crack detection on the third image to obtain the second sub-check result; perform impurity detection on the third image to obtain the third sub-check result; determine the first check result according to the first sub-check result, the second sub-check result, and the third sub-check result.

[0093] Among them, the bubbles usually appear as circular or oval bright areas in the image, the cracks appear as slender dark lines or fractured areas, and the impurities appear as bright spots, dark spots or irregular shapes in the image. Therefore, extract the feature information in the third image, compare the extracted bubble features with the preset bubble feature library to determine whether there are bubbles and the number, size and position of the bubbles; use a classification algorithm to classify the extracted crack features to determine the type, length and severity of the cracks; determine the number and distribution of impurities through algorithms such as threshold segmentation and clustering analysis, so as to obtain the first inspection result.

[0094] Of course, for the quality assessment of the thin film capacitor, it can be further determined whether it is a qualified product according to the user's requirements.

[0095] Furthermore, the thin film capacitor has an outer package, which is prone to cause noise interference during X-ray imaging. Therefore, the noise can be removed to better distinguish the internal structure. Obtain the X-ray image sequence of the thin film capacitor, including:

[0096] Obtain the initial X-ray image sequence of the thin film capacitor. The initial X-ray image sequence successively includes multiple initial second images with increasing X-ray intensity from small to large; for each initial second image, obtain the outer package X-ray image of the outer package of the thin film capacitor at the X-ray intensity corresponding to the initial second image; remove the outer package information in the initial second image according to the outer package X-ray image to obtain the second image corresponding to the initial second image; after removing the outer package information of all initial second images, obtain multiple second images, which constitute the X-ray image sequence of the thin film capacitor.

[0097] The initial X-ray image sequence refers to the image set of the thin film capacitor at different X-ray intensities directly collected by X-ray imaging. The initial second image refers to a single image in the initial X-ray image sequence. The outer package refers to the packaging material outside the thin film capacitor. Correspondingly, the outer package X-ray image refers to the X-ray image that only contains the outer package information of the thin film capacitor at the X-ray intensity.

[0098] Specifically, convert the gray values of each pixel point of the initial first image and the outer package X-ray image into corresponding energy matrices, calculate the matrix difference between the initial first image and the outer package X-ray image, and convert the matrix difference into the corresponding gray value to obtain a gray image and multiple second images. The influence of the outer package on the image is eliminated through the energy matrix, making the second image more accurate and more conducive to the detection of the capacitor.

[0099] Further, based on the second image of adjacent X-ray intensities, determine the potential difference region of the thin film capacitor, including: identifying the first feature points of the first target image and the second feature points of the second target image in the second image of adjacent X-ray intensities; performing feature point matching based on the descriptors of the first feature points and the descriptors of the second feature points to determine the successfully matched feature point pairs; aligning the first target image and the second target image according to the successfully matched feature point pairs; calculating the gray difference of each pixel point of the registered first target image and the second target image to obtain a difference image; obtaining a binary image based on the difference image using a gray difference threshold; and determining the potential difference region of the thin film capacitor according to the binary image.

[0100] In the embodiment of the present application, the second image of adjacent X-ray intensities includes a first target image and a second target image. In the first target image and the second target image, a feature point detection algorithm is used to identify key feature points, such as edges, corners, etc. Calculate descriptors for each feature point, and match the feature points in the first target image with the feature points in the second target image by comparing the similarity of the descriptors (such as using Euclidean distance, Hamming distance, etc.). The successfully matched feature point pairs represent the same or similar structural points in the two images. Use the successfully matched feature point pairs to align the first target image and the second target image through a transformation matrix (such as affine transformation, perspective transformation, etc.). The purpose of alignment is to ensure that the same structural points in the two images are consistent in spatial position. In the registered images, calculate the gray value difference of each pixel point to obtain a difference image, where each pixel value in the difference image represents the gray difference between the two images at that position. Binarize the pixel values in the difference image according to a threshold. Pixel values greater than the threshold are set to white, and pixel values less than or equal to the threshold are set to black. The white area represents the area where differences may exist in the thin film capacitor. Perform morphological operations (such as erosion, dilation, opening operation, and closing operation) on the binary image to remove small isolated points and fill small holes to obtain the potential difference region.

[0101] Further, before obtaining the binary image based on the difference image using the gray difference threshold, it further includes:

[0102] Determine the local mean and local standard deviation of the window corresponding to each pixel point of the difference image;

[0103] Determine the gray difference threshold corresponding to each pixel point according to the local mean and local standard deviation of the window corresponding to each pixel point.

[0104] In the embodiment of the present application, when calculating the binarization threshold of each pixel in the difference image, a dynamic threshold method based on local statistical characteristics can be adopted.

[0105] Specifically, calculate the local mean and standard deviation of each pixel, and dynamically calculate a threshold for each pixel according to the local mean and standard deviation.

[0106] Among them, for each pixel (i, j) in the difference image, we define a small window W(i, j) centered on this pixel, and the size of the window can be selected as needed (such as 3x3, 5x5, etc.). The calculation formula for the local mean μ(i, j) is:

[0107] , where I(x, y) is the pixel value at coordinates (x, y) in the difference image, is the number of pixels in the window W(i, j).

[0108] The calculation formula for the local standard deviation σ(i, j) is: .

[0109] According to the local mean μ(i, j) and the local standard deviation σ(i, j), dynamically calculate a threshold T(i, j) for each pixel. Specifically, T(i, j) = μ(i, j) + α × σ(i, j), where α is an adjustable parameter that can be customized by the user.

[0110] Furthermore, referring to Figure 2 , it further includes: obtaining a CCD image of the thin-film capacitor; performing an appearance inspection on the thin-film capacitor according to the CCD image to obtain a second inspection result; if the second inspection result indicates that the appearance of the thin-film capacitor is abnormal, generating a prompt message, and the prompt message includes the type of abnormality and the location of the abnormality.

[0111] In the embodiments of the present application, the CCD image specifically refers to an image of the thin-film capacitor captured by a CCD camera. The shooting angle should be the same as that of the first image.

[0112] The second inspection result refers to the evaluation result of the appearance quality of the thin-film capacitor obtained through the appearance inspection, including the type of abnormality and the location of the abnormality.

[0113] For the appearance inspection, the cases of appearance abnormality include but are not limited to: bulging, cracking, scratching, denting, unclear marking. Bulging may be caused by internal cracks, impurities, etc., and among them, the crack may be that the electrolytic paper of the pin-type thin-film capacitor is punctured.

[0114] In the embodiment of the present application, an anomaly recognition network model is used to automatically classify and locate CCD images. Specifically, a large number of CCD images of thin-film capacitors are obtained as training data, and the anomaly types and locations therein are labeled; the initial model is used to learn and train the training data to obtain an anomaly recognition network model. The CCD image of the thin-film capacitor is input into the anomaly recognition network model for prediction. If the prediction result is an anomaly, a prompt message including the anomaly type and location is generated according to the output of the model.

[0115] Further, if the second detection result is that the appearance of the thin-film capacitor is abnormal, the anomaly type is bulging, and the first detection result is an internal anomaly, the method further includes:

[0116] Obtain the first position information point set of the positioning mark of the third image in the x-ray coordinate system and the second position information point set of the positioning mark of the CCD image in the CCD coordinate system;

[0117] Register the third image and the CCD image according to the first position information point set and the second position information point set;

[0118] Determine the comprehensive anomaly information of the thin-film capacitor according to the potential difference region of the registered third image and the region corresponding to the potential difference region in the registered CCD image. The comprehensive anomaly information includes: the anomaly type and the anomaly location.

[0119] Among them, if the anomaly type is bulging, it may be that the electrolytic paper of the lead-pin type thin-film capacitor is punctured and the situation is relatively serious. Of course, there may be other situations. At this time, further verification is required. Therefore, the third image and the CCD image are registered according to the first position information point set and the second position information point set; according to the position and shape of the potential difference region in the third image, the corresponding region is found in the registered CCD image; combining the internal and appearance detection results, the comprehensive anomaly information of the thin-film capacitor is determined, including the anomaly type and the anomaly location.

[0120] Specifically, in one realizable way of the process of determining the comprehensive abnormal information, according to the potential difference region of the registered third image and the region corresponding to the potential difference region in the registered CCD image, the image is sent to the technician's client device so that the technician can check and confirm the abnormal type and abnormal location; the information fed back by the technician's client device is used as the comprehensive abnormal information of this region. In another realizable way, according to the potential difference region of the registered third image and the region corresponding to the potential difference region in the registered CCD image, a recognition model is used for detection to determine the comprehensive abnormal information of the thin film capacitor, wherein the recognition model is trained based on multiple x-ray images and multiple CCD images, and each pair of x-ray images and CCD images are marked with the abnormal type and abnormal location.

[0121] Further, after determining the comprehensive abnormal information of the thin film capacitor, an abnormal report can be generated, including the abnormal type, abnormal location, possible failure reasons, and recommended repair measures, so as to guide the technician to quickly and accurately locate and solve the problem.

[0122] In the embodiment of the present application, a detection device for a thin film capacitor is provided. Refer to Figure 3 , including:

[0123] An acquisition module 210, configured to acquire a first image and a standard image of the thin film capacitor collected by the X-ray imaging system, where the X-ray intensities of the first image and the standard image are the same;

[0124] A similarity determination module 220, configured to determine the similarity between the first image and the standard image;

[0125] The acquisition module 210 is further configured to, if the similarity is less than a preset similarity threshold, acquire an X-ray image sequence of the thin film capacitor, and the X-ray image sequence sequentially includes multiple second images with increasing X-ray intensities from small to large;

[0126] A potential difference region determination module 230, configured to determine the potential difference region of the thin film capacitor according to adjacent second images with X-ray intensities;

[0127] A third image determination module 240, configured to, for each potential difference region, determine a third image in the second image corresponding to the potential difference region;

[0128] An abnormal detection module 250, configured to perform abnormal detection on the third image to obtain a first detection result.

[0129] In one realizable way, the acquisition module 210 is further configured to: acquire an initial X-ray image sequence of the thin film capacitor, and the initial X-ray image sequence sequentially includes multiple initial second images with increasing X-ray intensities from small to large;

[0130] For each initial second image, obtain the outer package X-ray image of the outer package of the thin-film capacitor at the X-ray intensity corresponding to the initial second image; remove the outer package information in the initial second image according to the outer package X-ray image to obtain the second image corresponding to the initial second image;

[0131] After completing the removal of the outer package information of all the initial second images, multiple second images are obtained, constituting the X-ray image sequence of the thin-film capacitor.

[0132] In an implementable manner, the potential difference region determination module 230 is configured to:

[0133] Identify the first feature points of the first target image and the second feature points of the second target image in the second images with adjacent X-ray intensities;

[0134] Perform feature point matching according to the descriptors of the first feature points and the descriptors of the second feature points to determine the successfully matched feature point pairs;

[0135] Align the first target image and the second target image according to the successfully matched feature point pairs;

[0136] Calculate the gray difference value of each pixel point of the registered first target image and the second target image to obtain a difference image;

[0137] Obtain a binary image according to the difference image using the gray difference value threshold;

[0138] Determine the potential difference region of the thin-film capacitor according to the binary image.

[0139] In an implementable manner, it further includes:

[0140] The threshold determination module is configured to determine the local mean and local standard deviation of the window corresponding to each pixel point of the difference image;

[0141] Determine the gray difference value threshold corresponding to each pixel point according to the local mean and local standard deviation of the window corresponding to each pixel point.

[0142] In an implementable manner, it further includes:

[0143] The CCD image detection module is configured to: obtain the CCD image of the thin-film capacitor;

[0144] Perform an appearance detection on the thin-film capacitor according to the CCD image to obtain a second detection result;

[0145] If the second detection result is that the appearance of the thin-film capacitor is abnormal, generate a prompt message, and the prompt message includes the abnormal type and the abnormal position.

[0146] In an implementable manner, if the second detection result is that the appearance of the thin-film capacitor is abnormal, the abnormal type is bulging, and the first detection result is internal abnormality, the device further includes:

[0147] An integrated detection module, configured to: obtain a first set of position information points of the positioning mark of the third image in the x-ray coordinate system, and a second set of position information points of the positioning mark of the CCD image in the CCD coordinate system;

[0148] Register the third image and the CCD image according to the first set of position information points and the second set of position information points;

[0149] Determine the comprehensive abnormality information of the thin-film capacitor according to the potential difference region of the registered third image and the region corresponding to the potential difference region in the registered CCD image, where the comprehensive abnormality information includes: the abnormal type and the abnormal position.

[0150] In an implementable manner, an abnormality detection module 250 is configured to:

[0151] Perform bubble detection on the third image to obtain a first sub-inspection result;

[0152] Perform crack detection on the third image to obtain a second sub-inspection result;

[0153] Perform impurity detection on the third image to obtain a third sub-inspection result;

[0154] Determine the first inspection result according to the first sub-inspection result, the second sub-inspection result, and the third sub-inspection result.

[0155] In the embodiments of the present application, an electronic device is provided, as Figure 4 shown, Figure 4 The electronic device 300 shown includes: a processor 301 and a memory 303. Among them, the processor 301 and the memory 303 are connected, such as connected through a bus 302. Optionally, the electronic device 300 may further include a transceiver 304. It should be noted that in practical applications, the transceiver 304 is not limited to one, and the structure of the electronic device 300 does not constitute a limitation to the embodiments of the present application.

[0156] The processor 301 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute various exemplary logical blocks, modules, and circuits described in connection with the disclosure of this application. The processor 301 may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0157] The bus 302 may include a path for transmitting information between the above components. The bus 302 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. The bus 302 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 4 only a thick line is used to represent it in the figure, but it does not mean that there is only one bus or one type of bus.

[0158] The memory 303 may be a ROM (Read Only Memory) or other type of static storage device that can store static information and instructions, a RAM (Random Access Memory) or other type of dynamic storage device that can store information and instructions, or it may also be an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory), or other optical disc storage, optical disc storage (including compact discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), magnetic storage media, or other magnetic storage devices, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto.

[0159] The memory 303 is used to store the application program code for executing the solution of this application, and is controlled by the processor 301 to execute. The processor 301 is used to execute the application program code stored in the memory 303 to implement the content shown in the foregoing method embodiments.

[0160] Among them, the electronic device includes but is not limited to: mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Tablet Computers), PMPs (Portable Multimedia Players), in-vehicle terminals (such as in-vehicle navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 4 The shown electronic device is only an example and should not impose any restrictions on the functions and usage scope of the embodiments of this application.

[0161] The embodiments of this application provide a computer-readable storage medium, on which a computer program is stored. When it runs on a computer, it enables the computer to execute the corresponding content in the foregoing method embodiments.

[0162] The embodiments of this application provide a computer program product, including a computer program, which when executed by a processor, implements the corresponding content in the foregoing method embodiments.

[0163] It should be understood that although the steps in the flowchart of the accompanying drawings are shown in sequence according to the indication of the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless there is a clear description in this article, the execution of these steps has no strict order restriction, and they can be executed in other orders. Moreover, at least a part of the steps in the flowchart of the accompanying drawings may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same moment, but can be executed at different moments, and their execution order is not necessarily sequential, but can be executed alternately or alternately with at least a part of other steps or sub-steps or stages of other steps.

[0164] The above are only some implementation manners of this application. It should be noted that for those of ordinary skill in the art, without departing from the principle of this application, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of this application.

Claims

1. A method for detecting an aluminum electrolytic capacitor, characterized in that: include: Acquire a first image and a standard image of an aluminum electrolytic capacitor acquired by an X-ray imaging system, wherein the X-ray intensity of the first image and the standard image is the same; Determining a similarity between the first image and the standard image; If the similarity is less than a preset similarity threshold, an X-ray image sequence of the aluminum electrolytic capacitor is obtained, wherein the X-ray image sequence includes a plurality of second images with X-ray intensities from small to large in sequence; the X-ray intensity series of the aluminum electrolytic capacitor is determined according to the type of the aluminum electrolytic capacitor and a pre-established correspondence between the type and the X-ray intensity series; determining a potential difference region of the aluminum electrolytic capacitor based on a second image of adjacent X-ray intensities; For each potential difference region, determine a third image in the second image corresponding to the potential difference region, wherein the third image is an image with the highest definition in the second image corresponding to the potential difference region; Performing abnormality detection on the third image to obtain a first detection result; Among them, if the potential difference area of ​​the third image is one, the third image is directly detected for abnormalities; if there are at least two potential difference areas, the distance between the potential difference areas is determined, and if the distance is less than the distance threshold, it is determined that the two potential difference areas need to be merged; the potential difference areas are fused according to the third images corresponding to the potential difference areas to be merged; when fusion is performed, one of the third images is adjusted according to the difference in the pixel means of the non-difference areas, and then the fusion is performed.

2. The method for detecting an aluminum electrolytic capacitor according to claim 1, characterized in that: The step of acquiring the X-ray image sequence of the aluminum electrolytic capacitor comprises: acquiring an initial X-ray image sequence of the aluminum electrolytic capacitor, wherein the initial X-ray image sequence sequentially comprises a plurality of initial second images with increasing X-ray intensities from small to large; For each initial second image, obtaining an outer packaging X-ray image of the outer packaging of the aluminum electrolytic capacitor at the X-ray intensity corresponding to the initial second image; removing the outer packaging information in the initial second image according to the outer packaging X-ray image to obtain a second image corresponding to the initial second image; After the outer packaging information of all the initial second images is removed, a plurality of second images are obtained, forming an X-ray image sequence of the aluminum electrolytic capacitor.

3. The method for detecting an aluminum electrolytic capacitor according to claim 1, characterized in that: Determine the potential difference area of ​​the aluminum electrolytic capacitor according to the second image of adjacent X-ray intensities, including: Identifying a first feature point of a first target image and a second feature point of a second target image in the second image of adjacent X-ray intensities; Perform feature point matching according to the descriptor of the first feature point and the descriptor of the second feature point to determine a successfully matched feature point pair; Aligning the first target image and the second target image according to the successfully matched feature point pairs; Calculate the grayscale difference of each pixel of the first target image and the second target image after registration to obtain a difference image; Obtaining a binary image using a grayscale difference threshold according to the difference image; A potential difference area of ​​the aluminum electrolytic capacitor is determined according to the binarized image.

4. The method for detecting an aluminum electrolytic capacitor according to claim 3, characterized in that: Before obtaining a binary image using a grayscale difference threshold according to the difference image, the method further includes: Determine the local mean and local standard deviation of the window corresponding to each pixel point of the difference image; The grayscale difference threshold corresponding to each pixel is determined according to the local mean and local standard deviation of the window corresponding to each pixel.

5. The method for detecting an aluminum electrolytic capacitor according to claim 1, characterized in that: Also includes: Acquiring a CCD image of the aluminum electrolytic capacitor; Performing appearance inspection on the aluminum electrolytic capacitor according to the CCD image to obtain a second inspection result; If the second detection result is that the aluminum electrolytic capacitor has an abnormal appearance, a prompt message is generated, and the prompt message includes the abnormal type and the abnormal location.

6. The method for detecting an aluminum electrolytic capacitor according to claim 5, characterized in that: If the second detection result is that the aluminum electrolytic capacitor has an appearance abnormality, the abnormality type is bulging, and the first detection result is an internal abnormality, the method further includes: Acquire a first position information point set of the positioning mark of the third image in the X-ray coordinate system, and a second position information point set of the positioning mark of the CCD image in the CCD coordinate system; registering the third image with the CCD image according to the first position information point set and the second position information point set; According to the potential difference area of ​​the registered third image and the area corresponding to the potential difference area in the registered CCD image, comprehensive abnormality information of the aluminum electrolytic capacitor is determined, and the comprehensive abnormality information includes: abnormality type and abnormality position.

7. The method for detecting an aluminum electrolytic capacitor according to claim 1, characterized in that: The performing abnormality detection on the third image to obtain a first detection result includes: Perform bubble detection according to the third image to obtain a first sub-inspection result; Perform crack detection according to the third image to obtain a second sub-inspection result; Perform impurity detection according to the third image to obtain a third sub-inspection result; A first inspection result is determined according to the first sub-inspection result, the second sub-inspection result, and the third sub-inspection result.

8. A detection device for aluminum electrolytic capacitors, characterized in that: include: An acquisition module, used to acquire a first image and a standard image of the aluminum electrolytic capacitor acquired by an X-ray imaging system, wherein the X-ray intensity of the first image and the standard image is the same; A similarity determination module, used to determine the similarity between the first image and the standard image; The acquisition module is further used to acquire an X-ray image sequence of the aluminum electrolytic capacitor if the similarity is less than a preset similarity threshold, wherein the X-ray image sequence sequentially includes a plurality of second images with X-ray intensities from small to large; the X-ray intensity series of the aluminum electrolytic capacitor is determined based on the type of the aluminum electrolytic capacitor and a pre-established correspondence between the type and the X-ray intensity series; A potential difference region determination module, used for determining the potential difference region of the aluminum electrolytic capacitor according to the adjacent second image of X-ray intensity; A third image determination module is used to determine, for each potential difference region, a third image from the second image corresponding to the potential difference region, wherein the third image is an image with the highest definition in the second image corresponding to the potential difference region; an abnormality detection module, used to perform abnormality detection on the third image to obtain a first detection result; Among them, if the potential difference area of ​​the third image is one, the third image is directly detected for abnormalities; if there are at least two potential difference areas, the distance between the potential difference areas is determined, and if the distance is less than the distance threshold, it is determined that the two potential difference areas need to be merged; the potential difference areas are fused according to the third images corresponding to the potential difference areas to be merged; when fusion is performed, one of the third images is adjusted according to the difference in the pixel means of the non-difference areas, and then the fusion is performed.

9. An electronic device, characterized in that: include: one or more processors; Memory; One or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the one or more processors, and the one or more applications are configured to: execute the steps of the detection method of the aluminum electrolytic capacitor according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The storage medium stores at least one instruction, at least one program, code set or instruction set, and the at least one instruction, at least one program, code set or instruction set is loaded by the processor and executes the steps of the detection method of the aluminum electrolytic capacitor according to any one of claims 1 to 7.

Images