Search

CN-116468680-B - Component reverse pole defect detection method, system, equipment and storage medium

CN116468680BCN 116468680 BCN116468680 BCN 116468680BCN-116468680-B

Abstract

The invention provides a method, a system, equipment and a storage medium for detecting the reverse electrode defect of a component, which relate to the technical field of electronic component defect detection, and the method comprises the steps of inputting a component image p 1 to be detected into a pin screening model M, and outputting a pin positioning frame Bbox through the pin screening model M; and calculating the length and the position of the pin based on the pin outline drawing and the pin endpoint coordinates, and checking the calculation result of the length and the position of the pin with preset information to obtain the detection result of the reverse electrode defect of the component. The invention realizes the detection of the pin position and the length of the capacitor element based on the deep learning and image processing technology, and solves the problems of low efficiency, low accuracy and omission of the existing detection of the inverse defect of the capacitor element.

Inventors

  • Request for anonymity
  • Request for anonymity
  • Request for anonymity

Assignees

  • 成都数之联科技股份有限公司

Dates

Publication Date
20260505
Application Date
20230331

Claims (8)

  1. 1. The method for detecting the reverse electrode defect of the component is characterized by comprising the following steps of: step 1, inputting a component image p 1 to be detected into a pin screening model M, and outputting a pin positioning frame Bbox through the pin screening model M; step 2, carrying out pixel processing and contour extraction processing on the pin positioning frame Bbox to obtain a pin contour map and pin endpoint coordinates; the processes of pixel processing and contour extraction processing are carried out on the pin positioning frame Bbox to obtain a pin contour map and pin endpoint coordinates are as follows: step 2.1, carrying out pixel processing and pin positioning processing on the pin positioning frame Bbox, and carrying out contour extraction processing on the pin positioning frame Bbox after the pin positioning processing to obtain a pin contour map and the left endpoint coordinate of a first pin And right endpoint coordinates Left end point coordinates of the second pin And right endpoint coordinates ; Step 2.2 left endpoint coordinates based on the first root pin And right endpoint coordinates Rectangular box B 1 is constructed and based on the left endpoint coordinates of the second root pin And right endpoint coordinates Constructing a rectangular frame B 2 ; Step 3, calculating the length and the position of the pin based on the pin profile and the pin endpoint coordinates, and checking the calculation result of the length and the position of the pin with preset information to obtain a detection result of the reverse electrode defect of the component; And (3) calculating the length and the position of the pin based on the pin profile and the pin endpoint coordinates, and checking the calculation result of the length and the position of the pin with preset information to obtain the detection result of the reverse electrode defect of the component, wherein the flow is as follows: step 3.1 if the left endpoint ordinate y 1 of the first leg is greater than the left endpoint ordinate of the second leg Judging that the first pin is positioned above the second pin, otherwise, judging that the second pin is positioned above the first pin to realize pin position relation judgment; Step 3.2, calculating the pin length L 1 of the first pin and the pin length L 2 of the second pin based on the rectangular frame B 1 and the rectangular frame B 2 ; Step 3.3, checking based on the pin position relation judging result, the pin length L 1 of the first pin and the pin length L 2 of the second pin to obtain a component reverse electrode defect detecting result; Based on the pin position relation determination result, the pin length L 1 of the first pin and the pin length L 2 of the second pin, checking to obtain a component antipole defect detection result, the flow is as follows: If the preset information is that the positive electrode is located above the second pin, judging that the component has no antipole defect when the first pin is located above the second pin and the pin length L 1 of the first pin is larger than the pin length L 2 of the second pin, otherwise, judging that the component has antipole defect; and 3.31, if the preset information is that the positive electrode is positioned below, judging that the component has no antipole defect when the first pin is positioned below the second pin and the pin length L 1 of the first pin is larger than the pin length L 2 of the second pin, otherwise, judging that the component has antipole defect.
  2. 2. The method for detecting the reverse electrode defect of the component of claim 1, wherein the pin screening model M is obtained based on training of a deep learning network model.
  3. 3. The method for detecting the reverse electrode defect of the component of claim 2, wherein the deep learning network model comprises a convolution layer, a pooling layer and a full connection layer.
  4. 4. The method for detecting the reverse electrode defect of the component of claim 2, wherein the training process of the pin screening model M is as follows: Step 1.1, collecting a historical component image p 0 , and labeling pins of the component image p 0 ; And 1.2, inputting the component image p 0 after pin labeling into a deep learning network model for image feature extraction and iterative training to obtain a pin screening model M.
  5. 5. The method for detecting the reverse defect of the component according to claim 1, wherein the process of outputting the pin positioning frame Bbox through the pin screening model M is as follows: Step 1.3, inputting a component image p 1 to be detected into a pin screening model M to output pin positioning information; And 1.4, performing image cutting on the component image p 1 according to the pin positioning information to obtain a pin positioning frame Bbox.
  6. 6. A component antipodal defect detection system for use in carrying out the method of claim 1, the system comprising: The pin positioning frame acquisition unit is used for inputting the component image p 1 to be detected into a pin screening model M, and outputting a pin positioning frame Bbox through the pin screening model M; the pin profile image processing unit is used for carrying out pixel processing and profile extraction processing on the pin positioning frame Bbox so as to obtain a pin profile image and pin endpoint coordinates; and the antipole defect detection unit is used for calculating the length and the position of the pin based on the pin profile diagram and the pin endpoint coordinates and checking the calculation result of the length and the position of the pin with preset information to obtain the detection result of the antipole defect of the component.
  7. 7. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements a method for detecting a component counter defect as claimed in any one of claims 1 to 5 when the computer program is executed by the processor.
  8. 8. A computer readable storage medium, wherein a computer program is stored on the computer readable storage medium, and when executed by a processor, the computer program implements a method for detecting a component inverse polarity defect according to any one of claims 1 to 5.

Description

Component reverse pole defect detection method, system, equipment and storage medium Technical Field The invention relates to the technical field of defect detection of electronic components, in particular to a method, a system, equipment and a storage medium for detecting the reverse electrode defect of a component. Background The industrial circuit board is complex and tedious to manufacture, and various product defects are easily introduced in the manufacturing process. The capacitor component is used as a common component on an industrial circuit board and is ubiquitous on core circuit boards of various electrical equipment. Therefore, the quality level of the capacitor component directly influences or determines whether the core circuit board and even the electrical equipment are qualified, wherein the counter electrode of the capacitor is used as a common component production and manufacturing defect to cause a great hidden danger to the safety of the electrical equipment, and even the core circuit board and the electrical equipment can be burnt in a serious scene, so that a bad result is caused. In the electronic component quality inspection link, the current industry can only adopt a manual spot inspection method to carry out batch quality evaluation, and the problems of low efficiency, low accuracy and omission are caused because the manual spot inspection is greatly influenced by personal factors and environmental factors, which also represents a potential safety hazard for the subsequent equipment assembly and burying to a certain extent. Disclosure of Invention In order to solve the problems of low efficiency, low accuracy and omission factor in the conventional detection of the inverse electrode defect of the capacitor component, the embodiment of the invention provides a component inverse electrode defect detection method, a system, equipment and a storage medium. In a first aspect, an embodiment of the present invention provides a method for detecting a reverse polarity defect of a component, where the method includes the following steps: step 1, inputting a component image p 1 to be detected into a pin screening model M, and outputting a pin positioning frame Bbox through the pin screening model M; step 2, carrying out pixel processing and contour extraction processing on the pin positioning frame Bbox to obtain a pin contour map and pin endpoint coordinates; And 3, calculating the length and the position of the pin based on the pin profile and the pin endpoint coordinates, and checking the calculation result of the length and the position of the pin with preset information to obtain the detection result of the reverse electrode defect of the component. In the above embodiment, the method for detecting the inverse polar defect according to the capacitor component image by using the deep learning and image processing technology has the dual advantages of the detection speed and the detection precision compared with the method based on the traditional manual sampling inspection, and the detection speed is about 20ms and has great advantages when facing a large-scale detection task. As some optional embodiments of the present application, the pin screening model M is obtained based on training of a deep learning network model. As some optional embodiments of the application, the deep learning network model includes a convolution layer, a pooling layer, and a full connection layer. As some optional embodiments of the present application, the training procedure of the pin screening model M is as follows: Step 1.1, collecting a historical component image p 0, and labeling pins of the component image p 0; And 1.2, inputting the component image p 0 after pin labeling into a deep learning network model for image feature extraction and iterative training to obtain a pin screening model M. As some optional embodiments of the present application, the process of outputting the pin positioning frame Bbox by the pin screening model M is as follows: Step 1.3, inputting a component image p 1 to be detected into a pin screening model M to output pin positioning information; And 1.4, performing image cutting on the component image p 1 according to the pin positioning information to obtain a pin positioning frame Bbox. In the above embodiment, the present invention firstly trains the pin screening model M, and then performs pin positioning based on the trained pin screening model M, so as to realize the pin position detection output of the component. As some optional embodiments of the present application, the process of performing pixel processing and contour extraction processing on the pin positioning box Bbox to obtain a pin contour map and pin endpoint coordinates is as follows: Step 2.1, carrying out pixel processing and pin positioning processing on a pin positioning frame Bbox, and carrying out contour extraction processing on the pin positioning frame Bbox after the pin positioning processing to obtain