The Defect Detection Algorithm for Tire X-Ray Images Based on Deep Learning

For current domestic and international tire detection systems, the software operation of them is complex and poor to be put into application. In reality, it is necessary to do the task of defect detection by observing the X-ray image of the tire with the help of human eyes. This practice is affected by some subjective factors and both the accuracy and efficiency vary from person to person without strong robustness. To tackle this issue, one detection algorithm for tire defects based on deep learning is proposed. In this case, the model is trained, learnt and tested using the collected defect samples preprocessed from tire X-ray images. The designed algorithm was verified by the developed automatic tire defect detection software, in which the desired results were obtained.