The efficient detection of tire sidewall defects is crucial for ensuring safety and quality control in manufacturing. Traditional inspection is slow and inconsistent, while automated methods fail to address the complexity and coexistence of multiple tire sidewall defects. To alleviate those shortcomings, this study develops a curvature-based three-dimensional (3D) defect detection system that leverages the inherent rotational symmetry of tire sidewalls, allowing for more accuracy and efficiency in detecting intricate tire sidewall defects. Firstly, a defect detection system is developed that collects the three-dimensional data of tires, enabling precise quality assessments and facilitating accurate defect identification. Secondly, a dataset encompassing various types of intricate tire sidewall defects is constructed. This study leverages normal vectors and surface variation features to conduct an in-depth analysis of the complex three-dimensional shapes of tire sidewalls, while incorporating optimized curvature calculations that significantly enhance detection accuracy and algorithm efficiency. Moreover, the approach enables the simultaneous detection of intricate defect types, such as scratches, transportation damage, and cuts, thereby improving the comprehensiveness and accuracy of the detection process. The experimental results demonstrate that the system achieves a detection accuracy of 95.3%, providing crucial technical support for tire quality control.
Read full abstract