Tire X-ray defect detection method for carrying out re-ranking by utilizing background features

Abstract

The invention provides a tire X-ray defect detection method for carrying out re-ranking by utilizing background features. When defect detection is carried out after a model is trained, a to-be-detected picture is input into the model, and a plurality of pictures without any defect are randomly selected and are input into the model; when feature extraction is carried out on the to-be-detected picture, features of corresponding positions in the picture without any defect are also extracted; and the similarity between the defect feature vector and the background feature vector is calculated, candidate boxes are reranked according to the similarity, and a final detection result is output. The method has the following advantage: 1) the problems of low efficiency, high labor cost and the like inthe tire quality inspection process caused by human factors can be avoided through tire X-ray flaw detection based on deep learning; and 2) the information of the picture without defects is fully utilized, and the probability of the candidate boxes is corrected to a certain extent.

Description

technical field [0001] The invention relates to the technical fields of computer vision and industrial detection, in particular to a tire X-ray defect detection method using background features for reranking. Background technique [0002] Tires are an important pillar of my country's national economy, and tires in my country can be divided into bias-ply tires and radial-ply tires according to their carcass. Radial tires are widely used due to the advantages of small flow resistance, long service life, and good shock absorption performance. However, radial tires have very high production requirements, and their manufacturing process is more complicated than bias-ply tires. In the production process of radial tires, it is very susceptible to the influence of mechanical equipment, production process and other external environments, and there will be some defects such as impurities, thin tires, air bubbles, etc. These defects will affect the quality of tires and further affect ...

AI Technical Summary

Problems solved by technology

However, radial tires have very high production requirements, and their manufacturing process is more complicated than that of bias tires

In the production process of radial tires, it is very susceptible to the influence of mechanical equipment, production process and other external environments, and there will be some defects such as impurities, thin tires, air bubbles, etc. These defects will affect the quality of tires and further affect the safety of traffic driving

Images