With a dramatic increase in car ownership, traffic accidents are becoming more and more frequent, with the death rate from flat tire accidents being extremely high. Tires may be scratched during use and bulging caused by factors such as long-distance driving or excessive temperature can affect the health of the tire and the safety of the vehicle. Therefore, it is necessary to carry out nondestructive testing of tires. Most current tires are tested for defects in the tire production process before they leave the factory. The lack of effective detection methods for in-use tire defects, and the use of manual testing is time-consuming and labor-intensive and discriminatory standards, it is not easy to guarantee the accurateness and obtiveness of the test findings. As a result, this study suggests a YOLOX-based methodology for tire detection of defects, using a self-made defective tire dataset for experimental studies. According to the test data, it can classify and locate tires' defective parts with accuracy of 93.87%, effectively detect tire cracks and bulges, and serve as a reliable ground for the use of working tires.
Research on YOLOX-based tire defect detection method
2022-10-28
4894987 byte
Conference paper
Electronic Resource
English
Residual YOLOX-based Ship Object Detection Method
IEEE | 2022
|YOLOX-based dynamic vehicle target detection
IEEE | 2022
|