A component reverse pole defect detection method, system, device and storage medium

Abstract

This invention provides a method, system, device, and storage medium for detecting reverse polarity defects in electronic components, relating to the field of electronic component defect detection technology. The method includes the following steps: inputting an image p1 of the component to be detected into a pin selection model M, and outputting a pin positioning box Bbox through the pin selection model M; performing pixel processing and contour extraction on the pin positioning box Bbox to obtain a pin contour map and pin endpoint coordinates; calculating the pin length and position based on the pin contour map and pin endpoint coordinates, and verifying the calculated pin length and position results against preset information to obtain the component reverse polarity defect detection result. This invention, based on deep learning and image processing technology, realizes the detection of the pin position and length of capacitor components, solving the problems of low efficiency, low accuracy, and missed detection in existing capacitor component reverse polarity defect detection methods.

Description

Technical Field

[0001] This invention relates to the field of electronic component defect detection technology, and more specifically, to a method, system, device, and storage medium for detecting reverse polarity defects in electronic components. Background Technology

[0002] Industrial circuit board manufacturing is complex and cumbersome, and various product defects can easily be introduced during the manufacturing process. Capacitors, as common components on industrial circuit boards, are ubiquitous on the core circuit boards of various electrical devices. Therefore, the quality level of capacitors directly affects or determines the qualification of the core circuit board and even the electrical equipment. Reverse polarity of capacitors, a common manufacturing defect, poses a significant safety hazard to electrical equipment, and in severe cases, can even cause the core circuit board and electrical equipment to burn out, leading to disastrous consequences.

[0003] In the quality inspection of electronic components, the industry currently faces a huge number of component quality inspection tasks and can only use manual sampling to conduct batch quality assessments. Since manual sampling is greatly affected by personal and environmental factors, it suffers from low efficiency, low accuracy, and missed inspections, which also means that it poses certain safety hazards to subsequent equipment assembly. Summary of the Invention

[0004] To address the problems of low efficiency, low accuracy, and missed detection in existing capacitor component reverse polarity defect detection methods, this invention provides a method, system, device, and storage medium for detecting component reverse polarity defects.

[0005] In a first aspect, embodiments of the present invention provide a method for detecting reverse polarity defects in electronic components, the method comprising the following steps:

[0006] Step 1: Input the image p1 of the component to be inspected into the pin screening model M, and output the pin positioning box Bbox through the pin screening model M;

[0007] Step 2: Perform pixel processing and contour extraction on the pin positioning box (Bbox) to obtain the pin contour map and pin endpoint coordinates;

[0008] Step 3: Calculate the pin length and position based on the pin outline diagram and pin endpoint coordinates, and verify the pin length and position calculation results with preset information to obtain the component reverse polarity defect detection results.

[0009] In the above embodiments, deep learning and image processing technologies are used to perform reverse polarity defect detection based on images of capacitor components. Compared with the traditional manual sampling method, the present invention has the dual advantages of detection speed and detection accuracy. The detection speed is about 20ms, which has a huge advantage when facing large-scale detection tasks.

[0010] As some optional embodiments of this application, the pin selection model M is obtained by training a deep learning network model.

[0011] As some optional embodiments of this application, the deep learning network model includes convolutional layers, pooling layers, and fully connected layers.

[0012] As some optional embodiments of this application, the training process of the pin selection model M is as follows:

[0013] Step 1.1: Acquire historical component images p0 and label the pins of component images p0;

[0014] Step 1.2: Input the component image p0 after pin labeling into the deep learning network model for image feature extraction and iterative training to obtain the pin selection model M.

[0015] As some optional embodiments of this application, the process of outputting the pin positioning box Bbox through the pin screening model M is as follows:

[0016] Step 1.3: Input the image p1 of the component to be inspected into the pin screening model M to output pin positioning information;

[0017] Step 1.4: Crop the component image p1 according to the pin positioning information to obtain the pin positioning box Bbox.

[0018] In the above embodiments, the present invention first trains the pin selection model M, and then performs pin positioning based on the trained pin selection model M to realize the pin position detection output of components.

[0019] As some optional embodiments of this application, the process of performing pixel processing and contour extraction processing on the pin positioning box (Bbox) to obtain the pin contour map and pin endpoint coordinates is as follows:

[0020] Step 2.1: Perform pixel processing and pin positioning processing on the pin positioning box Bbox, and perform contour extraction processing on the pin positioning box Bbox after pin positioning processing to obtain the pin contour map, the left endpoint coordinates (x1, y1) and right endpoint coordinates (x2, y2) of the first pin, and the left endpoint coordinates (x′1, y′1) and right endpoint coordinates (x′2, y′2) of the second pin;

[0021] Step 2.2: Construct rectangle B1 based on the left endpoint coordinates (x1, y1) and right endpoint coordinates (x2, y2) of the first pin, and construct rectangle B2 based on the left endpoint coordinates (x′1, y′1) and right endpoint coordinates (x′2, y′2) of the second pin.

[0022] In the above embodiments, by constructing rectangle B1 and rectangle B2, the length relationship between the first pin and the second pin can be quickly predicted.

[0023] As some optional implementations of this application, the process of calculating the pin length and position based on the pin outline diagram and pin endpoint coordinates, and verifying the calculated pin length and position results with preset information to obtain the component reverse polarity defect detection results is as follows:

[0024] Step 3.1: If the ordinate y1 of the left endpoint of the first pin is greater than the ordinate y′1 of the left endpoint of the second pin, then the first pin is determined to be above the second pin; otherwise, the second pin is determined to be above the first pin, so as to realize the pin position relationship determination.

[0025] Step 3.2: Calculate the pin length L1 of the first pin and the pin length L2 of the second pin based on rectangles B1 and B2;

[0026] Step 3.3: Verify the results based on the pin position relationship determination results, the pin length L1 of the first pin, and the pin length L2 of the second pin to obtain the component reverse polarity defect detection results.

[0027] As some optional implementations of this application, the process for obtaining the component reverse polarity defect detection result by verifying the pin position relationship determination result, the pin length L1 of the first pin, and the pin length L2 of the second pin is as follows:

[0028] Step 3.31: If the preset information is that the positive terminal is on top, then when the first pin is above the second pin and the pin length L1 of the first pin is greater than the pin length L2 of the second pin, it is determined that the component does not have a reverse polarity defect; otherwise, it is determined that the component has a reverse polarity defect.

[0029] Step 3.31: If the preset information is that the positive terminal is below, then when the first pin is below the second pin and the pin length L1 of the first pin is greater than the pin length L2 of the second pin, it is determined that the component does not have a reverse polarity defect; otherwise, it is determined that the component has a reverse polarity defect.

[0030] In the above embodiments, the present invention can quickly and accurately determine whether a component has a reverse polarity defect by judging the length and spatial position relationship of the pins.

[0031] In a second aspect, the present invention provides a component reverse polarity defect detection system, the system comprising:

[0032] A pin positioning box acquisition unit is used to input the image p1 of the component to be detected into the pin screening model M, and output the pin positioning box Bbox through the pin screening model M.

[0033] A pin contour image processing unit is used to perform pixel processing and contour extraction processing on the pin positioning box Bbox to obtain a pin contour image and pin endpoint coordinates.

[0034] The reverse polarity defect detection unit calculates the pin length and position based on the pin profile diagram and pin endpoint coordinates, and verifies the calculated pin length and position results with preset information to obtain the component reverse polarity defect detection results.

[0035] In a third aspect, the present invention provides a computer device including a memory, a processor, and a computer program stored in the memory and executable on the processor, the processor performing the aforementioned method for detecting reverse polarity defects in components.

[0036] In a fourth aspect, the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the component reverse polarity defect detection method.

[0037] The beneficial effects of this invention are as follows:

[0038] This invention uses deep learning and image processing technology to achieve accurate detection of component pins, solving the problem of missed detection in the existing detection of reverse polarity defects in capacitor components.

[0039] This invention uses pin outline diagrams and pin endpoint coordinates to comprehensively determine pin length and position, thus enabling rapid and accurate determination of reverse polarity defects in capacitor components. Attached Figure Description

[0040] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0041] Figure 1 This is a flowchart illustrating the steps of the component reverse polarity defect detection method according to an embodiment of the present invention.

[0042] Figure 2 This is a flowchart of a component reverse polarity defect detection method according to an embodiment of the present invention.

[0043] Figure 3 This is a schematic diagram of the pin positioning frame and pin outline according to an embodiment of the present invention.

[0044] Figure 4 This is a schematic diagram of the positioning frames for the first and second pins according to an embodiment of the present invention. Detailed Implementation

[0045] To better understand the above technical solutions, the technical solutions of the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present invention and the specific features in the embodiments are detailed descriptions of the technical solutions of the present invention, rather than limitations on the technical solutions of the present invention. In the absence of conflict, the embodiments of the present invention and the technical features in the embodiments can be combined with each other.

[0046] It should also be understood that, in order to simplify the description of the invention and thus aid in the understanding of at least one embodiment, multiple features may sometimes be grouped into a single embodiment, drawing, or description thereof in the foregoing description of the embodiments of the invention. However, this method of disclosure does not imply that the subject matter of the invention requires more features than those mentioned in the claims. In fact, the embodiments contain fewer features than all the features of the single embodiment disclosed above.

[0047] Example 1

[0048] This invention provides a method for detecting reverse polarity defects in electronic components. Please refer to [link / reference]. Figure 1 , Figure 2 The method includes the following steps:

[0049] Step 1: Input the image p1 of the component to be inspected into the pin screening model M, and output the pin positioning box Bbox through the pin screening model M;

[0050] The pin selection model M is trained based on a deep learning network model, which includes convolutional layers, pooling layers, and fully connected layers. The convolutional layers are mainly used for image learning, the pooling layers are used for image normalization, and the fully connected layers are used for pin location processing.

[0051] In this embodiment of the invention, the training process of the pin selection model M is as follows:

[0052] Step 1.1: Collect historical component images p0 and label the pins of component images p0.

[0053] Step 1.2: Input the component image p0 after pin labeling into the deep learning network model for image feature extraction and iterative training to obtain the pin selection model M.

[0054] In this embodiment of the invention, the acquired component image p0 is first reduced in size and filled. The main reason is that the pins are continuous targets and occupy a large proportion of the image. Based on the relevant theories and practical experience of convolutional neural networks, using a reduced and filled image will be more conducive to model learning and accurate reasoning.

[0055] In this embodiment of the invention, the process of outputting the pin positioning box Bbox through the pin selection model M is as follows:

[0056] Step 1.3: Input the image p1 of the component to be inspected into the pin screening model M to output pin positioning information.

[0057] Step 1.4: Crop the component image p1 according to the pin positioning information to obtain the pin positioning box Bbox.

[0058] Step 2: Perform pixel processing and contour extraction on the pin positioning box (Bbox) to obtain the pin contour map and pin endpoint coordinates;

[0059] In the embodiments of the present invention, please refer to Figure 3 The process of performing pixel processing and contour extraction on the pin positioning box (Bbox) to obtain the pin contour map and pin endpoint coordinates is as follows:

[0060] Step 2.1: Perform pixel processing and pin positioning processing on the pin positioning box Bbox, and perform contour extraction processing on the pin positioning box Bbox after pin positioning processing to obtain the pin contour map, the left endpoint coordinates (x1, y1) and right endpoint coordinates (x2, y2) of the first pin, and the left endpoint coordinates (x′1, y′1) and right endpoint coordinates (x′2, y′2) of the second pin;

[0061] Step 2.2: Construct rectangle B1 based on the left endpoint coordinates (x1, y1) and right endpoint coordinates (x2, y2) of the first pin, and construct rectangle B2 based on the left endpoint coordinates (x′1, y′1) and right endpoint coordinates (x′2, y′2) of the second pin. Please refer to [link to relevant documentation]. Figure 4 .

[0062] Specifically, the pixel processing procedure for the pin positioning frame (Bbox) is as follows:

[0063] Step 2.11: Perform grayscale processing on the pin positioning box Bbox to obtain the pin grayscale image Bbox1;

[0064] Step 2.12: Preset the binarization segmentation threshold, and perform image binarization processing on the pin grayscale image Bbox1 based on the binarization segmentation threshold.

[0065] Step 3: Calculate the pin length and position based on the pin outline diagram and pin endpoint coordinates, and verify the pin length and position calculation results with preset information to obtain the component reverse polarity defect detection results.

[0066] In this embodiment of the invention, the process for verifying the pin length and position calculation results with preset information is as follows:

[0067] Step 3.1: If the ordinate y1 of the left endpoint of the first pin is greater than the ordinate y′1 of the left endpoint of the second pin, then the first pin is determined to be above the second pin; otherwise, the second pin is determined to be above the first pin, thus realizing the pin position determination.

[0068] Step 3.2: Calculate the pin length L1 of the first pin and the pin length L2 of the second pin based on rectangles B1 and B2; where pin length L1 = |x1 - x2| and pin length L2 = |x′1 - x′2|.

[0069] Meanwhile, in order to eliminate the influence of pin crossing, bending and other factors on pin length determination, pin length can also be determined by predicting the actual distance.

[0070] Specifically, the process of determining pin length through actual distance prediction is as follows:

[0071] Step 3.21: Based on Step 2.1, using the pin positioning processing results, obtain the number of all endpoints of the first and second pins respectively. If the number of endpoints is greater than the threshold, obtain the coordinates of all endpoints of the first pin, i.e., the coordinates of the left endpoint (x1, y1), the coordinates of the middle endpoint (x3, y3), (x4, y4)...(x...). n ,y n The coordinates of the right endpoint (x2, y2); and the coordinates of all endpoints of the second pin, namely the coordinates of the left endpoint (x′1, y′1), the middle endpoint (x′3, y′3), (x′4, y′4)...(x′1, y′1), and the coordinates of the middle endpoint (x′3, y′3), (x′4, y′4)...(x′1, y′1). m ,y′ m )y and the right endpoint coordinates (x′2, y′2). If the number of endpoints is less than the threshold, then the pin length L1 = |x1-x2| and the pin length L2 = |x′1-x′2| are predicted in advance.

[0072] Step 3.22: If the number of endpoints is greater than the threshold, obtain the distance between adjacent endpoints of the first pin, and sum all the distances between adjacent endpoints to obtain the pin length L1 of the first pin and the pin length L2 of the second pin. The threshold can be set to 3.

[0073] The pin length of the first pin is:

[0074]

[0075] The pin length of the second pin:

[0076]

[0077] Step 3.3: Verify based on the pin position determination result, the pin length L1 of the first pin and the pin length L2 of the second pin to obtain the component reverse polarity defect detection result.

[0078] Specifically, the process for verifying the reverse polarity defect detection results of components based on the pin position determination result, the pin length L1 of the first pin, and the pin length L2 of the second pin is as follows:

[0079] Step 3.31: If the preset information is that the positive terminal is on top, then when the first pin is above the second pin and the pin length L1 of the first pin is greater than the pin length L2 of the second pin, it is determined that the component does not have a reverse polarity defect; otherwise, it is determined that the component has a reverse polarity defect.

[0080] Step 3.31: If the preset information is that the positive terminal is below, then when the first pin is below the second pin and the pin length L1 of the first pin is greater than the pin length L2 of the second pin, it is determined that the component does not have a reverse polarity defect; otherwise, it is determined that the component has a reverse polarity defect.

[0081] The preset information can be obtained by image recognition based on a neural network model to determine the relative position of the negative electrode and thus the position of the positive electrode. At the same time, the preset information can also be obtained by other means, which are not limited in this embodiment of the invention.

[0082] In this embodiment of the invention, since the pins may be crossed or bent, the determination of the pin position relationship is generally based on the coordinates of the left endpoint, and the left end of the pin will hardly be subject to any related interference.

[0083] In this embodiment, the pin selection model M is first trained using historical component images p0. The trained pin selection model M then performs pin selection on the component image p1 to be detected, obtaining pin location boxes (Bboxes). Next, image processing and contour extraction are performed on the pin location boxes (Bboxes) to obtain the pin contour map and pin endpoint coordinates. Finally, the reverse polarity defect of the component is determined based on the pin contour map and pin endpoint coordinates. That is, by combining deep learning and image processing technologies with the structure of the capacitor component for comprehensive judgment, accurate pin detection and positional relationship determination are achieved, solving the problem of missed detections in existing capacitor component reverse polarity defect detection methods.

[0084] Example 2

[0085] This invention provides a component reverse polarity defect detection system, the system comprising:

[0086] A pin positioning box acquisition unit is used to input the image p1 of the component to be detected into the pin screening model M, and output the pin positioning box Bbox through the pin screening model M.

[0087] A pin contour image processing unit is used to perform pixel processing and contour extraction processing on the pin positioning box Bbox to obtain a pin contour image and pin endpoint coordinates.

[0088] The reverse polarity defect detection unit calculates the pin length and position based on the pin profile diagram and pin endpoint coordinates, and verifies the calculated pin length and position results with preset information to obtain the component reverse polarity defect detection results.

[0089] Example 3

[0090] The present invention provides a computer device, the computer device including a memory and a processor, the memory storing a computer program, the computer program executing a component reverse polarity defect detection method as described in Embodiment 1 when the processor is running.

[0091] The computer device provided in this embodiment can implement the method described in Embodiment 1. To avoid repetition, it will not be described again here.

[0092] Example 4

[0093] This invention provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements a component reverse polarity defect detection method as described in Embodiment 1.

[0094] The computer-readable storage medium provided in this embodiment can implement the method described in Embodiment 1. To avoid repetition, it will not be described again here.

[0095] The processor can be a central processing unit (CPU), or other general-purpose processors, digital signal processors, application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor.

[0096] The memory can be used to store the computer program and / or modules. The processor implements various functions of the component reverse polarity defect detection system of the invention by running or executing the data stored in the memory. The memory may mainly include a program storage area and a data storage area. The program storage area may store the operating system, at least one application program required for a function (such as sound playback function, image playback function, etc.). In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, RAM, plug-in hard disk, smart memory card, secure digital card, flash memory card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0097] If a component reverse polarity defect detection system is implemented as a software functional 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 methods of the above embodiments of the present invention can also be implemented by a computer program that can be stored in a computer-readable storage medium. When executed by a processor, this computer program can implement the steps of the various method embodiments described above. The computer program includes computer program code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory, random access memory, dot carrier signals, telecommunication signals, and software distribution media, 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.

[0098] The basic concepts of this invention have been described. It is obvious to those skilled in the art that the detailed disclosure above is merely illustrative and does not constitute a limitation of this specification. Although not explicitly stated herein, those skilled in the art may make various modifications, improvements, and corrections to this specification. Such modifications, improvements, and corrections are suggested in this specification and therefore remain within the spirit and scope of the exemplary embodiments described herein.

Claims

1. A method for detecting reverse polarity defects in electronic components, characterized in that, The method includes the following steps: Step 1: Input the image p1 of the component to be inspected into the pin screening model M, and output the pin positioning box Bbox through the pin screening model M; Step 2: Perform pixel processing and contour extraction on the pin positioning box (Bbox) to obtain the pin contour map and pin endpoint coordinates; The process of performing pixel processing and contour extraction on the pin positioning box (Bbox) to obtain the pin contour map and pin endpoint coordinates is as follows: Step 2.1: Perform pixel processing and pin positioning processing on the pin positioning box (Bbox), and then perform contour extraction processing on the pin positioning box (Bbox) after pin positioning processing to obtain the pin contour image and the coordinates of the left endpoint of the first pin. and the coordinates of the right endpoint The coordinates of the left endpoint of the second pin and the coordinates of the right endpoint ; Step 2.2: Based on the coordinates of the left endpoint of the first pin and the coordinates of the right endpoint Construct rectangle B1, based on the coordinates of the left endpoint of the second pin. and the coordinates of the right endpoint Construct rectangle B2; Step 3: Calculate the pin length and position based on the pin outline diagram and pin endpoint coordinates, and verify the pin length and position calculation results with preset information to obtain the component reverse polarity defect detection results; The process for calculating pin length and position based on pin outline diagrams and pin endpoint coordinates, and then verifying the calculated pin length and position results against preset information to obtain component reverse polarity defect detection results is as follows: Step 3.1: If the y-coordinate of the left endpoint of the first pin is greater than the y-coordinate of the left endpoint of the second pin... If the first pin is above the second pin, then the second pin is above the first pin; otherwise, the second pin is above the first pin, thus determining the pin position relationship. Step 3.2: Calculate the pin length L1 of the first pin and the pin length L2 of the second pin based on rectangles B1 and B2; Step 3.3: Verify the results based on the pin position relationship determination results, the pin length L1 of the first pin, and the pin length L2 of the second pin to obtain the component reverse polarity defect detection results; The process for obtaining the component reverse polarity defect detection result based on the pin position relationship determination result, the pin length L1 of the first pin, and the pin length L2 of the second pin is as follows: Step 3.31: If the preset information is that the positive terminal is on top, then when the first pin is above the second pin and the pin length L1 of the first pin is greater than the pin length L2 of the second pin, it is determined that the component does not have a reverse polarity defect; otherwise, it is determined that the component has a reverse polarity defect. Step 3.31: If the preset information is that the positive terminal is below, then when the first pin is below the second pin and the pin length L1 of the first pin is greater than the pin length L2 of the second pin, it is determined that the component does not have a reverse polarity defect; otherwise, it is determined that the component has a reverse polarity defect.

2. The method for detecting reverse polarity defects in electronic components according to claim 1, characterized in that: The pin selection model M is obtained by training a deep learning network model.

3. The method for detecting reverse polarity defects in electronic components according to claim 2, characterized in that: The deep learning network model includes convolutional layers, pooling layers, and fully connected layers.

4. The method for detecting reverse polarity defects in electronic components according to claim 2, characterized in that: The training process for the pin selection model M is as follows: Step 1.1: Acquire historical component images p0 and label the pins of component images p0; Step 1.2: Input the component image p0 after pin labeling into the deep learning network model for image feature extraction and iterative training to obtain the pin selection model M.

5. The method for detecting reverse polarity defects in electronic components according to claim 1, characterized in that: The process of outputting the pin positioning box (Bbox) through the pin selection model M is as follows: Step 1.3: Input the image p1 of the component to be inspected into the pin screening model M to output pin positioning information; Step 1.4: Crop the component image p1 according to the pin positioning information to obtain the pin positioning box Bbox.

6. A component reverse polarity defect detection system for implementing the method of claim 1, characterized in that, The system includes: A pin positioning box acquisition unit is used to input the image p1 of the component to be detected into the pin screening model M, and output the pin positioning box Bbox through the pin screening model M. A pin contour image processing unit is used to perform pixel processing and contour extraction processing on the pin positioning box Bbox to obtain a pin contour image and pin endpoint coordinates. The reverse polarity defect detection unit calculates the pin length and position based on the pin profile diagram and pin endpoint coordinates, and verifies the calculated pin length and position results with preset information to obtain the component reverse polarity defect detection results.

7. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, it implements the component reverse polarity defect detection method according to any one of claims 1-5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the component reverse polarity defect detection method according to any one of claims 1-5.

Images