Pseudo soldering detection method and system based on machine vision and deep learning, and medium

Abstract

The invention provides a pseudo soldering detection method based on machine vision and deep learning. The pseudo soldering detection method comprises the steps: 1, shooting welding spots, and obtaining original data images; 2, performing graying processing on the original data images; 3, preprocessing the images after graying processing, performing filtering processing, and eliminating noise; 4, carrying out segmentation processing on the preprocessed images, and distinguishing welding spots from a background to obtain all welding spot targets; and 5, performing pseudo soldering identificationon all the welding spot targets to complete pseudo soldering detection of the welding spots. According to the pseudo soldering detection method, automatic identification and pseudo soldering weldingspot detection of the body-in-white welding spots are achieved, and the labor cost can be saved; and the pseudo soldering recognition rate is high, and the human influence is reduced, and the industrial production efficiency and reliability can be improved.

Description

technical field [0001] The invention relates to the field of welding technology, in particular to a method, system and medium for detecting false welds based on machine vision and deep learning. Background technique [0002] With the continuous development of the automobile industry, the demand for automation and intelligence is increasing day by day, and the inspection of the solder joints of the body-in-white still mainly relies on human eyes. [0003] For the identification of false soldering, the penetrating effect of X-rays is needed to know the internal situation. [0004] However, the cost of X-ray detection is relatively high, and it is not suitable for large-scale promotion and use; and there are potential safety hazards, which affect production efficiency. [0005] Traditional visual inspection methods, such as detecting size, color, position, and comparing with given standards, are not suitable for the detection of false welding. It is difficult for humans to gi...

AI Technical Summary

Problems solved by technology

[0015] At present, the inspection of solder joints of body-in-white widely relies on human eyes, and human fatigue, sense of responsibility and lack of experience may cause misjudgment

Moreover, the management problems brought about by the large number of personnel used, the disputes between different evaluation results and the lack of evaluation standards, etc., all highlight the far-reaching impact that machine vision inspection will bring.

[0016] The technical problem of the present invention is: to overcome the deficiencies of the prior art, to propose a virtual welding detection method based on image feature extraction and machine learning based on machine vision and deep learning, to solve the problem of automatic detection of false welding of the solder joints of the body-in-white, and to realize Classification of normal / false solder joints

Images