An electrolytic capacitor pinch defect detection method, system, device and storage medium

Abstract

The application provides an electrolytic capacitor clamping defect detection method, system, device and storage medium, and relates to the technical field of industrial quality inspection.The method comprises the following steps: inputting an image of a component to be detected into a semantic segmentation model, extracting a rubber tube contour through the semantic segmentation model;obtaining a rubber tube outer circle, a rubber tube circumscribed rectangle or a rubber tube contour intersection combination based on the rubber tube contour, wherein the rubber tube contour intersection combination comprises two intersection points of a rubber tube contour center line and the rubber tube contour;calculating the roundness of the rubber tube outer circle, the length-width ratio of the rubber tube circumscribed rectangle or the distance of the rubber tube contour intersection combination to obtain a rubber tube contour calculation result; and determining whether the electrolytic capacitor component has a clamping defect based on the rubber tube contour calculation result.The application is based on deep learning technology and engineering processing technology, and comprehensively determines the actual structure of the electrolytic capacitor, thereby solving the problem of missed detection in the existing electrolytic capacitor clamping defect detection.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of industrial quality inspection, in particular to an electrolytic capacitor clamping defect detection method, system, device and storage medium. BACKGROUND

[0002] Industrial circuit board production and manufacturing is complex and tedious, and various product defects are easily introduced during the manufacturing process. As a common component on industrial circuit boards, electrolytic capacitors are ubiquitous on the core circuit boards of various electrical equipment. Therefore, the quality level of electrolytic capacitors directly affects or determines whether the core circuit board and even the electrical equipment is qualified and the service life. Among them, the clamping deformation type defect of electrolytic capacitor as a common component production and manufacturing defect, although it may not affect the working performance of electrolytic capacitor, but it has potential influence on the service life, and further affects the service life of electrical equipment.

[0003] In the electrolytic capacitor appearance quality inspection link, there are a large number of component quality inspection tasks. At present, the industry can only adopt manual sampling inspection method to carry out batch quality evaluation. Due to the influence of personal factors and environmental factors, manual sampling inspection has the problems of low efficiency, low accuracy and missed detection, which also represents a certain degree of safety hazard for subsequent equipment assembly. SUMMARY

[0004] In order to solve the problems of low efficiency, low accuracy and missed detection in the existing electrolytic capacitor clamping defect detection, the present application provides an electrolytic capacitor clamping defect detection method, system, device and storage medium.

[0005] In a first aspect, an electrolytic capacitor clamping defect detection method is provided in the present application. The method comprises the following steps:

[0006] Input the component image to be detected into a semantic segmentation model, and extract the rubber tube contour through the semantic segmentation model, wherein the component image to be detected is the top image or bottom image of the electrolytic capacitor component;

[0007] Obtain the rubber tube outside circle, rubber tube circumscribed rectangle or rubber tube contour intersection combination based on the rubber tube contour, wherein the rubber tube contour intersection combination includes two intersection points of the rubber tube contour center line and the rubber tube contour;

[0008] Calculate the roundness of the rubber tube outside circle, the aspect ratio of the rubber tube circumscribed rectangle or the distance of the rubber tube contour intersection combination to obtain the rubber tube contour calculation result;

[0009] Determine whether the electrolytic capacitor component has a clamping defect based on the rubber tube contour calculation result.

[0010] In the above embodiments, the deep learning technology and engineering processing technology are used to detect the pinch defect of the electrolytic capacitor according to the component image of the top or bottom of the electrolytic capacitor. Compared with the traditional manual sampling detection method, the application can realize no missed detection and has the advantages of detection speed and detection accuracy, and can face large-scale detection tasks.

[0011] As some optional embodiments of the present application, the semantic segmentation model is trained based on a deep learning network model.

[0012] As some optional embodiments of the present application, the training process of the semantic segmentation model is as follows:

[0013] Collect historical component images, and perform contour labeling on the historical component images to form a segmentation training set, wherein the historical component images are top images or bottom images of electrolytic capacitor components;

[0014] Perform image enhancement processing on the historical component images in the segmentation training set, and input the deep learning network model for image feature extraction and iterative training to form a semantic segmentation model.

[0015] As some optional embodiments of the present application, the process of extracting the rubber tube contour by the semantic segmentation model is as follows:

[0016] Input the component image to be detected into the semantic segmentation model to obtain the coordinate information of the rubber tube contour;

[0017] Perform image segmentation on the component image to be detected according to the coordinate information of the rubber tube contour to obtain the rubber tube contour.

[0018] In the above embodiments, the semantic segmentation model is first trained, and then the contour extraction is performed based on the trained semantic segmentation model to realize the output of the rubber tube contour.

[0019] As some optional embodiments of the present application, the process of obtaining the rubber tube outer circle, the rubber tube circumscribed rectangle or the rubber tube contour intersection combination based on the rubber tube contour is as follows:

[0020] Obtain the rubber tube outer circle based on the coordinate information of the rubber tube contour, and calculate the area and perimeter of the rubber tube outer circle;

[0021] Set a plurality of rubber tube circumscribed rectangles with different angles based on the coordinate information of the rubber tube contour, and obtain the length and width of each rubber tube circumscribed rectangle with different angles;

[0022] Obtain the center of the rubber tube contour based on the coordinate information of the rubber tube contour, set a plurality of line segments with different angles through the center of the rubber tube contour, and obtain the rubber tube contour intersection combination of the plurality of line segments with different angles and the outer side of the rubber tube contour.

[0023] As some optional embodiments of the present application, the roundness of the outer circular ring of the rubber tube, the length-width ratio of the outer rectangular of the rubber tube, or the distance of the intersection point combination of the rubber tube profile is calculated to obtain the rubber tube profile calculation result as follows:

[0024] The roundness of the outer circular ring of the rubber tube is calculated based on the area and the circumference of the outer circular ring of the rubber tube, and the roundness of the outer circular ring of the rubber tube is taken as the first rubber tube profile calculation result;

[0025] The length-width ratio of the outer rectangular of the rubber tube is calculated based on the length and the width of the outer rectangular of the rubber tube at several different angles, and the length-width ratio of the outer rectangular of the rubber tube at several different angles is taken as the second rubber tube profile calculation result;

[0026] The distance of the intersection point combination of the rubber tube profile is calculated based on the intersection point combination of the rubber tube profile at several different angles, and the ratio of the minimum distance and the maximum distance of the intersection point combination of the rubber tube profile is taken as the third rubber tube profile calculation result.

[0027] As some optional embodiments of the present application, the process of determining whether the electrolytic capacitor component has a pinch defect based on the rubber tube profile calculation result is as follows:

[0028] The first rubber tube profile calculation result, the second rubber tube profile calculation result, or the third rubber tube profile calculation result is selected as the judgment basis;

[0029] If the first rubber tube profile calculation result is less than the first threshold value, it is determined that the electrolytic capacitor component has a pinch defect;

[0030] If the length-width ratio of the outer rectangular of the rubber tube in the second rubber tube profile calculation result is less than the second threshold value, it is determined that the electrolytic capacitor component has a pinch defect;

[0031] If the third rubber tube profile calculation result is less than the third threshold value, it is determined that the electrolytic capacitor component has a pinch defect.

[0032] In the above embodiments, by combining the structure of the rubber tube profile for determination, the presence or absence of a pinch defect at the top or bottom of the electrolytic capacitor can be quickly and accurately determined.

[0033] In a second aspect, the present application provides an electrolytic capacitor pinch defect detection system, which comprises:

[0034] A profile extraction unit is configured to input an image of a component to be detected into a semantic segmentation model, and extract a rubber tube profile by the semantic segmentation model, wherein the image of the component to be detected is a top image or a bottom image of an electrolytic capacitor component;

[0035] The feature acquisition unit acquires the rubber tube outside circle, the rubber tube circumscribed rectangle or the rubber tube contour intersection combination based on the rubber tube contour, wherein the rubber tube contour intersection combination comprises two intersection points of the rubber tube contour center line and the rubber tube contour;

[0036] The feature operation unit is used for calculating the roundness of the rubber tube outside circle, the length-width ratio of the rubber tube circumscribed rectangle or the distance of the rubber tube contour intersection combination to obtain the rubber tube contour calculation result.

[0037] The pinch defect detection unit determines whether the electrolytic capacitor device has the pinch defect based on the rubber tube contour calculation result.

[0038] In a third aspect, the present application provides a computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor executes the electrolytic capacitor pinch defect detection method.

[0039] In a fourth aspect, the present application provides a computer readable storage medium, wherein the computer readable storage medium stores a computer program, and the computer program is executed by a processor to implement the electrolytic capacitor pinch defect detection method.

[0040] The present application has the following advantages:

[0041] The present application uses deep learning technology and engineering processing technology to realize accurate detection of electrolytic capacitor pinch defects and solve the problem of missed detection in existing electrolytic capacitor pinch defect detection.

[0042] The present application can quickly and accurately determine whether the electrolytic capacitor top or bottom has the pinch defect by combining the structure of the rubber tube contour. BRIEF DESCRIPTION OF DRAWINGS

[0043] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments. It should be understood that the following drawings only show some embodiments of the present application, and therefore should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can also be obtained without creative labor on the basis of these drawings.

[0044] Figure 1 is a step diagram of the electrolytic capacitor pinch defect detection method according to the embodiment of the present application;

[0045] Figure 2 is a real object diagram of the electrolytic capacitor pinch defect detection according to the embodiment of the present application;

[0046] Figure 3is a structural block diagram of the electrolytic capacitor pinch defect detection system according to the embodiment of the present application. DETAILED DESCRIPTION

[0047] In order to better understand the above technical solutions, the technical solutions of the present application will be described in detail below through the drawings and specific embodiments. It should be understood that the embodiments of the present application and the specific features in the embodiments are detailed descriptions of the technical solutions of the present application, and are not limitations of the technical solutions of the present application. In the case of no conflict, the technical features in the embodiments of the present application and the embodiments can be combined with each other.

[0048] It should also be understood that, in order to simplify the description of the present application disclosed and to help the understanding of at least one embodiment of the present application, sometimes various features are combined into one embodiment, figure or description thereof in the foregoing description of the embodiments of the present application. However, this disclosure method does not mean that the features required by the present application are more than the features mentioned in the claims. In fact, the features of the embodiments are less than all the features of the above-mentioned single embodiment.

[0049] Embodiment 1

[0050] The present application provides a kind of electrolytic capacitor pinch defect detection method, please refer to Figure 1 , the method comprises the following steps:

[0051] (1) the image of the component to be detected is input into a semantic segmentation model, and the rubber tube contour is extracted by the semantic segmentation model, and the image of the component to be detected is the top image or bottom image of the electrolytic capacitor component.

[0052] Wherein, the rubber tube is the rubber wrapped outside the metal shell of the electrolytic capacitor, mainly plays the role of insulation and identification, and under normal circumstances, it should be in the state of circular ring, after obtaining the segmentation mask graph of the rubber tube contour, whether the electrolytic capacitor appears pinch problem can be judged according to the connected threshold characteristics of the segmentation mask graph of the rubber tube contour.

[0053] Specifically, the semantic segmentation model is obtained based on the training of a deep learning network model, and the deep learning network model includes convolution layer, pooling layer and full connection layer, the convolution layer is mainly used for image learning, the pooling layer is used for image normalization processing, and the full connection layer is used for rubber tube contour extraction.

[0054] In the embodiment of the present application, the training process of the semantic segmentation model is as follows:

[0055] Collect historical component images, and perform contour labeling on the rubber tube of the historical component images to form a segmentation training set, wherein the historical component images are top images or bottom images of electrolytic capacitor components;

[0056] The historical component image in the segmentation training set is subjected to image enhancement processing, and is input into a deep learning network model for image feature extraction and iterative training to form a semantic segmentation model.

[0057] In the embodiment of the present application, the image data brightness is better through image enhancement processing on the collected historical component image, which is beneficial to model learning and accurate inference.

[0058] In the embodiment of the present application, the process of extracting the rubber tube contour by the semantic segmentation model is as follows:

[0059] The component image to be detected is input into the semantic segmentation model to obtain coordinate information of the rubber tube contour;

[0060] The component image to be detected is subjected to image segmentation according to the coordinate information of the rubber tube contour to obtain the rubber tube contour, and the rubber tube contour

[0061] (2) There are mainly three specific ways to determine whether the electrolytic capacitor appears the pinch problem according to the connectivity threshold characteristics of the segmentation mask map of the rubber tube contour, please refer to Figure 2 :

[0062] The first way is to obtain the rubber tube outer circular ring based on the rubber tube contour, and calculate the roundness of the rubber tube outer circular ring to obtain the rubber tube contour calculation result, and determine whether the electrolytic capacitor component exists the pinch defect based on the rubber tube contour calculation result;

[0063] Specifically, first, the rubber tube outer circular ring is obtained based on the coordinate information of the rubber tube contour, and the area and perimeter of the rubber tube outer circular ring are calculated; then the roundness e of the rubber tube outer circular ring is calculated based on the area area and the perimeter perimeter of the rubber tube outer circular ring, and the roundness of the rubber tube outer circular ring is taken as the first rubber tube contour calculation result; if the first rubber tube contour calculation result, i.e. the roundness e is less than the first threshold value (normally, the roundness e of the complete circular ring is between 0.9-1, and in the pinch case, the roundness e of the circular ring will become smaller, and the smaller the roundness e, the more irregular the circular ring), it is determined that the electrolytic capacitor component exists the pinch defect, at the same time, the first threshold value can be set between 0.9-1, which can be adjusted according to the actual situation, and is not limited.

[0064] Specifically, the calculation formula of the roundness e of the rubber tube outer circular ring is as follows:

[0065]

[0066] Wherein, area represents the area of the rubber tube outer circular ring, and perimeter represents the perimeter of the area of the rubber tube outer circular ring.

[0067] The second method is based on the rubber tube contour to obtain a rubber tube circumscribed rectangle, to calculate the length-width ratio of the rubber tube circumscribed rectangle, to obtain a rubber tube contour calculation result, and to determine whether the electrolytic capacitor device has a pinch defect based on the rubber tube contour calculation result.

[0068] Specifically, first, a plurality of different angle rubber tube circumscribed rectangles are set based on the coordinate information of the rubber tube contour, then the length and width of the plurality of different angle rubber tube circumscribed rectangles are obtained respectively, and the length-width ratios of the plurality of different angle rubber tube circumscribed rectangles are calculated based on the length and width of the plurality of different angle rubber tube circumscribed rectangles respectively, the length-width ratios of the plurality of different angle rubber tube circumscribed rectangles are taken as the second rubber tube contour calculation result, and if the length-width ratio of the rubber tube circumscribed rectangle in the second rubber tube contour calculation result is less than the second threshold value (in a normal case, the rubber tube circumscribed rectangle is a square, so the length-width ratio is close to 1, and in a pinch case, the length-width ratio of the rubber tube circumscribed rectangle becomes smaller), it is determined that the electrolytic capacitor device has a pinch defect.

[0069] Specifically, the angle of the rubber tube circumscribed rectangle can be set between 0°-180°, the second threshold value can be set between 0.8-1, and the related parameters can be set according to the actual situation, without limitation.

[0070] The third method is based on the rubber tube contour to obtain a rubber tube contour intersection combination, to calculate the distance of the rubber tube contour intersection combination, to obtain a rubber tube contour calculation result, and to determine whether the electrolytic capacitor device has a pinch defect based on the rubber tube contour calculation result.

[0071] Specifically, first, the rubber tube contour center is obtained based on the coordinate information of the rubber tube contour, a plurality of different angle line segments are set through the rubber tube contour center, and a plurality of different angle line segment and rubber tube contour intersection combinations outside the rubber tube contour are obtained; then the distance of the plurality of different angle line segment and rubber tube contour intersection combinations outside the rubber tube contour is calculated respectively, and the ratio of the minimum distance to the maximum distance of the rubber tube contour intersection combination is taken as the third rubber tube contour calculation result, and if the third rubber tube contour calculation result is less than the third threshold value (in a normal case, the ratio of the minimum distance to the maximum distance is close to 1, and in a pinch case, the ratio of the minimum distance to the maximum distance becomes smaller), it is determined that the electrolytic capacitor device has a pinch defect.

[0072] Specifically, the angle of the line segment can be set between 0°-180°, the third threshold value can be set between 0.8-1, and the related parameters can be set according to the actual situation, without limitation; preferably, the angle of the line segment is set to 0°, 60°, and 90°.

[0073] The first mode of the three modes has the smallest calculation amount, but the roundness cannot accurately reflect the degree of pinch, and is not sensitive to the pinch feedback; the second mode can more accurately reflect the pinch condition, and more accurate results can be obtained by increasing the number of different angle outer rectangles of the rubber tube; the third mode does not need to solve the outer rectangle frame, and the same can increase different angle line segments to obtain more accurate results, and the calculation amount is relatively small, that is, the actual calculation mode can be selected according to the actual situation (efficiency and accuracy); at the same time, the three modes can also be combined to increase the accuracy of defect detection.

[0074] To sum up, the embodiment of the application first trains the semantic segmentation model through historical component images, and extracts the image contour of the component image to be detected through the trained semantic segmentation model to obtain the rubber tube contour; and then determines the pinch defect of the electrolytic capacitor based on the rubber tube contour. That is, based on deep learning technology and engineering processing technology, and combined with the actual structure of the electrolytic capacitor, a comprehensive determination is made, and the problem of missing detection in the existing electrolytic capacitor pinch defect detection is solved.

[0075] Embodiment 2

[0076] The application provides an electrolytic capacitor pinch defect detection system, please refer to Figure 3 , the system and the method of embodiment 1 one by one, the system comprises:

[0077] The contour extraction unit is used for inputting the component image to be detected into the semantic segmentation model, and extracting the rubber tube contour through the semantic segmentation model, wherein the component image to be detected is the top image or the bottom image of the electrolytic capacitor component;

[0078] The feature acquisition unit acquires the rubber tube outside circle, the rubber tube outer rectangle or the rubber tube contour intersection combination based on the rubber tube contour, wherein the rubber tube contour intersection combination includes two intersection points of the rubber tube contour center line and the rubber tube contour;

[0079] The feature operation unit is used for calculating the roundness of the rubber tube outside circle, the length-width ratio of the rubber tube outer rectangle or the distance of the rubber tube contour intersection combination to obtain the rubber tube contour calculation result;

[0080] The pinch defect detection unit determines whether the electrolytic capacitor component has a pinch defect based on the rubber tube contour calculation result.

[0081] Embodiment 3

[0082] The application provides a computer device, which comprises a memory and a processor, the memory stores a computer program, and the computer program performs the electrolytic capacitor clamping defect detection method in embodiment 1 when the processor runs.

[0083] The computer device provided in the embodiment can implement the method in embodiment 1, and details are not described herein again to avoid repetition.

[0084] Embodiment 4

[0085] The application provides a computer readable storage medium, which stores a computer program, and the computer program implements the electrolytic capacitor clamping defect detection method in embodiment 1 when executed by a processor.

[0086] The computer readable storage medium provided in the embodiment can implement the method in embodiment 1, and details are not described herein again to avoid repetition.

[0087] The processor can be a central processing unit (CPU), and can also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor.

[0088] The memory can be used to store the computer program and / or modules, and the processor realizes various functions of the electrolytic capacitor clamping defect detection system in the application by running or executing data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application required by a function (such as a sound playing function, an image playing function, etc.), etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, for example, a hard disk, a memory, a plug-in hard disk, an intelligent memory card, a secure digital card, a flash memory card, at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0089] If the electrolytic capacitor pinch defect detection system is implemented in the form of a software function unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the above-mentioned embodiment methods can also be stored in a computer-readable storage medium through a computer program. When the processor executes the computer program, the steps of the above-mentioned various method embodiments can be implemented. The computer program includes computer program code, object code form, executable files, or some intermediate forms, etc. The computer-readable medium can include any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory, random access memory, point carrier signal, telecommunication signal, and software distribution medium, etc. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction.

[0090] The present application has described the basic concept, and it is obvious that the above detailed disclosure is only as an example and does not constitute a limitation on the specification for those skilled in the art. Although it is not explicitly stated here, those skilled in the art can make various modifications, improvements and corrections to the specification. Such modifications, improvements and corrections are suggested in the specification, so such modifications, improvements and corrections still belong to the spirit and scope of the exemplary embodiments of the specification.

Claims

1. A method of detecting a pinch defect of an electrolytic capacitor, characterized by, The method comprises the following steps: The image of the component to be detected is input into a semantic segmentation model, and the rubber tube contour is extracted by the semantic segmentation model, wherein the image of the component to be detected is the top image or the bottom image of the electrolytic capacitor component; Based on the rubber tube contour, a rubber tube outside circle, a rubber tube circumscribed rectangle or a rubber tube contour intersection combination is obtained, wherein the rubber tube contour intersection combination comprises two intersection points of the rubber tube contour center line and the rubber tube contour; The roundness of the rubber tube outside circle, the length-width ratio of the rubber tube circumscribed rectangle or the distance of the rubber tube contour intersection combination is calculated to obtain a rubber tube contour calculation result; Based on the rubber tube contour calculation result, it is determined whether the electrolytic capacitor component has a pinch defect; The first method is based on the rubber tube contour to obtain a rubber tube outside circle, and the roundness of the rubber tube outside circle is calculated to obtain a rubber tube contour calculation result, and based on the rubber tube contour calculation result, it is determined whether the electrolytic capacitor component has a pinch defect, and the specific process is as follows: First, the rubber tube outside circle is obtained based on the coordinate information of the rubber tube contour, and the area and the perimeter of the rubber tube outside circle are calculated; then the roundness e of the rubber tube outside circle is calculated based on the area area and the perimeter perimeter of the rubber tube outside circle, and the roundness of the rubber tube outside circle is taken as the first rubber tube contour calculation result; if the first rubber tube contour calculation result, i.e. the roundness e, is less than a first threshold value, it is determined that the electrolytic capacitor component has a pinch defect; The second method is based on the rubber tube contour to obtain a rubber tube circumscribed rectangle, and the length-width ratio of the rubber tube circumscribed rectangle is calculated to obtain a rubber tube contour calculation result, and based on the rubber tube contour calculation result, it is determined whether the electrolytic capacitor component has a pinch defect, and the specific process is as follows: First, a plurality of rubber tube circumscribed rectangles with different angles are set based on the coordinate information of the rubber tube contour, and then the lengths and widths of the plurality of rubber tube circumscribed rectangles with different angles are obtained, and the length-width ratios of the plurality of rubber tube circumscribed rectangles are calculated based on the lengths and widths of the plurality of rubber tube circumscribed rectangles with different angles, and the length-width ratios of the plurality of rubber tube circumscribed rectangles are taken as the second rubber tube contour calculation result; if the length-width ratio of the rubber tube circumscribed rectangle in the second rubber tube contour calculation result is less than a second threshold value, it is determined that the electrolytic capacitor component has a pinch defect; The third method is based on the rubber tube contour to obtain a rubber tube contour intersection combination, and the distance of the rubber tube contour intersection combination is calculated to obtain a rubber tube contour calculation result, and based on the rubber tube contour calculation result, it is determined whether the electrolytic capacitor component has a pinch defect, and the specific process is as follows: First, the rubber tube contour center is obtained based on the coordinate information of the rubber tube contour, a plurality of line segments with different angles are set through the rubber tube contour center, and a rubber tube contour intersection combination of the plurality of line segments with different angles and the rubber tube contour outside the rubber tube contour is obtained; then the distance of the rubber tube contour intersection combination is calculated based on the rubber tube contour intersection combination of the plurality of line segments with different angles and the rubber tube contour outside the rubber tube contour, and the ratio of the minimum distance and the maximum distance of the rubber tube contour intersection combination is taken as the third rubber tube contour calculation result; if the third rubber tube contour calculation result is less than a third threshold value, it is determined that the electrolytic capacitor component has a pinch defect.

2. The method of claim 1, wherein: The semantic segmentation model is trained based on a deep learning network model.

3. The method of claim 2, wherein: The training process of the semantic segmentation model is as follows: Collect historical component images, and perform contour labeling on the historical component images to form a segmentation training set, wherein the historical component images are top images or bottom images of electrolytic capacitor components; Perform image enhancement processing on the historical component images in the segmentation training set, and input a deep learning network model for image feature extraction and iterative training to form a semantic segmentation model.

4. The method of claim 1, wherein the method further comprises: The process of extracting the rubber tube contour through the semantic segmentation model is as follows: Input the component image to be detected into the semantic segmentation model to obtain the coordinate information of the rubber tube contour; Perform image segmentation on the component image to be detected according to the coordinate information of the rubber tube contour to obtain the rubber tube contour.

5. An electrolytic capacitor dishing defect detection system, characterized by, The system comprises: A contour extraction unit configured to input the component image to be detected into the semantic segmentation model to extract the rubber tube contour through the semantic segmentation model, wherein the component image to be detected is a top image or a bottom image of an electrolytic capacitor component; A feature acquisition unit configured to acquire a rubber tube outer circle, a rubber tube circumscribed rectangle, or a rubber tube contour intersection combination based on the rubber tube contour, wherein the rubber tube contour intersection combination comprises a rubber tube contour center line and two intersection points of the rubber tube contour; A feature operation unit configured to calculate the roundness of the rubber tube outer circle, the length-width ratio of the rubber tube circumscribed rectangle, or the distance of the rubber tube contour intersection combination to obtain a rubber tube contour calculation result; A pinch defect detection unit configured to determine whether the electrolytic capacitor component has a pinch defect based on the rubber tube contour calculation result; Firstly, the rubber tube outer circle is acquired based on the coordinate information of the rubber tube contour, and the area and the circumference of the rubber tube outer circle are calculated; then the roundness e of the rubber tube outer circle is calculated based on the area area and the circumference perimeter of the rubber tube outer circle, and the roundness of the rubber tube outer circle is taken as the first rubber tube contour calculation result; if the first rubber tube contour calculation result, i.e., the roundness e, is less than a first threshold value, it is determined that the electrolytic capacitor component has a pinch defect; Secondly, the rubber tube circumscribed rectangle is acquired based on the rubber tube contour, and the length-width ratio of the rubber tube circumscribed rectangle is calculated to obtain the rubber tube contour calculation result, and it is determined whether the electrolytic capacitor component has a pinch defect based on the rubber tube contour calculation result, and the specific process is as follows: Firstly, a plurality of rubber tube circumscribed rectangles with different angles are set based on the coordinate information of the rubber tube contour, and then the lengths and widths of the plurality of rubber tube circumscribed rectangles with different angles are acquired, and the length-width ratios of the plurality of rubber tube circumscribed rectangles are calculated based on the lengths and widths of the plurality of rubber tube circumscribed rectangles with different angles, and the length-width ratios of the plurality of rubber tube circumscribed rectangles are taken as the second rubber tube contour calculation result; if the length-width ratio of the rubber tube circumscribed rectangle in the second rubber tube contour calculation result is less than a second threshold value, it is determined that the electrolytic capacitor component has a pinch defect. ​ The third method is based on the rubber tube contour to obtain a rubber tube contour intersection combination, calculate the distance of the rubber tube contour intersection combination to obtain a rubber tube contour calculation result, and determine whether the electrolytic capacitor component has a pinch defect based on the rubber tube contour calculation result. The specific process is as follows: First, the rubber tube contour center is obtained based on the coordinate information of the rubber tube contour. A plurality of line segments with different angles are set through the rubber tube contour center, and a plurality of rubber tube contour intersection combinations between the line segments with different angles and the outer side of the rubber tube contour are obtained. Then, the distances of the rubber tube contour intersection combinations are calculated based on the rubber tube contour intersection combinations between the line segments with different angles and the outer side of the rubber tube contour. The ratio of the minimum distance and the maximum distance of the rubber tube contour intersection combinations is taken as the third rubber tube contour calculation result. If the third rubber tube contour calculation result is less than a third threshold value, it is determined that the electrolytic capacitor component has a pinch defect.

6. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: The processor executes the computer program to implement the electrolytic capacitor pinch defect detection method of any one of claims 1-4.

7. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a computer program, and the computer program is executed by the processor to implement the electrolytic capacitor pinch defect detection method of any one of claims 1-4.

Images