Search

CN-115170478-B - Electronic element visual recognition method and system

CN115170478BCN 115170478 BCN115170478 BCN 115170478BCN-115170478-B

Abstract

The invention provides an electronic component visual identification method which comprises the following steps of S1, establishing a learning module for counting parameters before and after core-spun-infusion, counting parameters before and after dry core-spun-infusion, S2, judging whether the learned quantity is satisfied, entering step S3, and returning to step S1, wherein the step S3 is not satisfied, extracting a threshold value of the counted parameters, and detecting according to the extracted threshold value. According to the invention, the learning module for counting the parameters before and after the core-spun-infusion is built, the parameters before and after the core-spun-infusion is counted, and the core-spun-infusion detection is carried out according to the counted parameters, so that the quality of the product can be ensured, and the loss caused by unqualified rejection of the product is avoided.

Inventors

  • GUO MINGPENG
  • WANG GUILIN
  • HUANG GUANGXIAN

Assignees

  • 湖南云眼智能装备有限公司

Dates

Publication Date
20260512
Application Date
20220610

Claims (5)

  1. 1. A method for visually identifying an electronic component, comprising the steps of: s1, establishing a learning module for counting parameters before and after the core-spun infusion, and counting parameters before and after the dry core-spun infusion; s2, judging whether the learned quantity is satisfied, wherein the step S3 is satisfied, and the step S1 is not satisfied; S3, extracting a threshold value of the counted parameter, and detecting according to the extracted threshold value, wherein in S1, when the learning module counts the parameters before and after the core-spun-lace liquid, the method comprises the following steps: S11, inputting pictures before and after core package dropping; S12, extracting the position of the core pack body; s13, acquiring the area of the core pack body; s14, extracting the white position of the core pack; s15, acquiring the area of a white area; s16, extracting the bottom position of the core pack; s17, acquiring the dripping area of the core bag; in the step S3, when detecting according to the extracted threshold value, the method includes the following steps: S31, inputting core package pictures before and after dropping; s32, ROI operation; s33, searching a core pack body; s34, angle detection operation; s35, searching the position of the white area region before dripping; s36, detecting abnormal picture sequence; s37, drip level extraction operation; s38, drip level detection operation; In the step S34, during the angle detection operation, boundary points on the left and right sides of the contour are extracted, the boundary points are fitted into a straight line, the slope of the straight line is obtained, and the angle deviation on the left and right sides is obtained according to the slope and compared with an angle deviation threshold; Detecting the maximum area of the core pack body during angle detection operation, and judging that the core pack angle is abnormal if the area of the core pack body is larger than the area threshold of the core pack body and the angle deviation of the left side and the right side is larger than the angle deviation threshold; if the area of the core bag body is smaller than the area threshold value of the core bag body, judging that the angle deviation of any one of the left side and the right side is larger than the angle deviation threshold value, and judging that the core bag angle is abnormal; in the step S36, during the operation of detecting abnormal picture sequence, the maximum area of the white area of the core package is obtained, and then compared with the threshold value of the white area of the core package, if the maximum area is smaller than the threshold value, the abnormal picture sequence is indicated; In the step S38, during the operation of detecting the liquid drop volume, whether the bottom of the core bag is free of liquid drop or has small liquid drop amount is detected, and the liquid free of liquid drop or the small liquid drop amount is removed; And comparing the bottom liquid drop level obtained in the step S38 with a liquid drop level threshold value, and if the liquid drop level is less than the threshold value, the core bag liquid drop is abnormal.
  2. 2. The method for visual recognition of electronic components according to claim 1, wherein in the step S13, when the area of the core pack body is obtained, the maximum area of the outline is obtained through contourArea functions, and the maximum area is used as a learning parameter of the area of the core pack body.
  3. 3. The method for visual recognition of electronic components according to claim 1, wherein in S15, when the area of the white area is obtained, the area of the outline of the white area is obtained by contourArea functions and a maximum value is selected, and the maximum value is used as a learning value of the area of the white area of the core package.
  4. 4. An electronic component visual recognition system for implementing the electronic component recognition method of any one of claims 1-3, comprising a learning module, a detection module, and a determination module: the learning module is used for counting parameters before and after the liquid is dispensed from the dry core package; the detection module is used for detecting the dry core package according to the parameter threshold before and after the dry core package is spotted; the judging module is used for judging whether core package inclination abnormality, picture sequence abnormality and core package dropping abnormality occur after the core package is dropped with liquid.
  5. 5. An electronic device comprising a processor and a memory storing computer readable instructions which, when executed by the processor, perform the steps of the method of any of claims 1-3.

Description

Electronic element visual recognition method and system Technical Field The invention relates to the technical field of electronic element part detection, in particular to a method for detecting core-coating liquid in an electronic element by adopting visual identification. Background In the process of manufacturing an electronic component, it is necessary to perform a liquid dispensing (electrolyte) operation on a core pack inside the electronic component, the liquid dispensing operation directly affecting parameter values in terms of some physical properties of the electronic component. The invention is authorized by China, namely an X-Ray detection system and a detection method (publication number: CN 106945890B) of a lithium battery core package structure, and discloses a detection method of the lithium battery core package, wherein the lithium battery core package is fed into an X-Ray detection device through a conveying device and a loading and unloading device, the detection efficiency is improved by combining a pipelined conveying system with an inductive switch, 360-degree dead-angle-free continuous detection of the lithium battery core package can be realized by adopting a stable rotating rack, the accuracy of detection data is improved, and the detection method is flexible and quick to use by arranging different functional areas, can effectively improve the whole production process flow efficiency of a lithium battery, effectively isolates the radiation of the X-Ray by adopting a totally-enclosed protective outer cover, and has certain defects in detection including folds, twists, pole piece layers and tab heights: The existing core pack detection method has no good detection means for the liquid dropping effect of the core pack, and particularly when the core pack subjected to liquid dropping has abnormal core pack inclination, abnormal picture sequence and abnormal core pack liquid dropping, the product quality cannot be ensured, and the loss is easily caused by unqualified and scrapped products. Therefore, a visual recognition method and a visual recognition system for electronic elements are provided. Disclosure of Invention In view of the foregoing, it is desirable to provide a visual recognition method and system for electronic components, so as to solve or alleviate the technical problems in the prior art, and at least provide a beneficial choice. The technical scheme of the embodiment of the invention is realized in such a way that the electronic element visual identification method and system comprise the following steps: s1, establishing a learning module for counting parameters before and after the core-spun infusion, and counting parameters before and after the dry core-spun infusion; s2, judging whether the learned quantity is satisfied, wherein the step S3 is satisfied, and the step S1 is not satisfied; S3, extracting the threshold value of the counted parameter, and detecting according to the extracted threshold value. Further preferably, in the step S1, when the learning module counts parameters before and after the core-spun-infusion, the method comprises the following steps: S11, inputting pictures before and after core package dropping; S12, extracting the position of the core pack body; s13, acquiring the area of the core pack body; s14, extracting the white position of the core pack; s15, acquiring the area of a white area; s16, extracting the bottom position of the core pack; S17, obtaining the dripping area of the core bag. Further preferably, in the step S13, when the area of the core pack body is obtained, the maximum area of the contour is obtained through contourArea functions, and the maximum area is used as a learning parameter of the area of the core pack body. Further preferably, in the step S15, when the area of the white area is obtained, the area of the outline of the white area is obtained through contourArea functions, and a maximum value is selected, wherein the maximum value is used as a learning value of the area of the white area of the core package. Further preferably, in the step S3, the detecting according to the extracted threshold value includes the following steps: S31, inputting core package pictures before and after dropping; s32, ROI operation; s33, searching a core pack body; s34, angle detection operation; s35, searching the position of the white area region before dripping; s36, detecting abnormal picture sequence; s37, drip level extraction operation; S38, a drip level detection operation. Further preferably, in the step S34, during the angle detection operation, boundary points on the left and right sides of the contour are extracted, the boundary points are fitted into a straight line, the slope of the straight line is obtained, and the angle deviation on the left and right sides is obtained according to the slope and is compared with an angle deviation threshold; Detecting the maximum area of the core pack body during angle detection operati