A detection method and detection device for welding defects of welding wire

Abstract

The invention relates to a method and a device for detecting welding defects of a bonding wire. The method comprises: firstly, an image of a bonding wire is obtained from a vertical angle; preprocessing images, judging whether the fuse is disconnected or not, if yes, converting an angle acquisition image, extracting an ROI image of a disconnection area at a vertical angle, performing image matching on the image acquired at the second angle and the ROI image of the bonding wire disconnection area at the vertical angle after image preprocessing, and further judging whether the fuse is disconnected truly or not. The method can correctly judge whether the fuse is disconnected and whether the twisted bonding wire is subjected to pseudo soldering or not, and has good adaptability. According to the method, a judgment conclusion at a vertical angle is taken as a main point, judgment at other angles is taken as an auxiliary point, and judgment results at other angles are added, so that the qualification rate of welding wires of the lithium battery module can be improved, and the working efficiency of the whole lithium battery module is ensured.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a detection method and a detection device for welding defects of welding wires. Background technique [0002] With the continuous improvement of science and technology, electric vehicles have become a development trend of modern vehicles, and batteries are the most important and expensive components of electric vehicles. As the energy storage system of electric vehicles, lithium battery modules bear the burden of the entire system. The quality of the energy source directly determines the life of the entire system, so it is particularly important to study the welding wire defects of lithium batteries to ensure the quality of the entire battery pack. At present, there is no detection method that can detect lithium battery module fuse defects from multiple angles in this field. [0003] Wire bonding is to form an electrical connection by bonding the two ends of a metal wir...

AI Technical Summary

Problems solved by technology

These defects can be avoided by some machine adjustments, such as reducing the welding pressure, reducing the welding time, etc., but the success rate cannot be achieved 100%, so it is necessary to use the judgment of the machine or human to eliminate the defective welding wire, but manual inspection The ability of the machine is limited, and it cannot continue to work for a long time, and the detection efficiency is not high. Therefore, it has become a trend to replace manual detection with machines.

[0004] Most of the machine detection technologies now detect the object under test at a fixed angle, but the result of machine detection depends on the collection of images by the machine, and the quality of the image often depends on environmental factors, such as lighting, so machine detection also has Certain limitations, this limitation cannot be completely overcome, so the present invention proposes a multi-angle detection method, which can better improve the detection efficiency

Images