A Hybrid Template Match Approach Based on Wavelet Analysis and Threshold Segmentation for Detecting Tire Surface Wear

At present, study on performances of tyre is mainly focused on such characteristics as pressure, temperature and frictional coefficient, etc, however, with the neglect of texture analysis on tyre surface. In fact, during the course of driving, the latter has recorded large amount of information about acting features from environment, vehicle and tire self, whose forces together act on formation of various wear. So viewing from tire surface image, a hybrid template match, incorporating with threshold segmentation and wavelet analysis is proposed to detect wearing states of tyre and explore its corresponding formation mechanism in this paper. The former two provide their respectively dealt inter and intra contour as sample to be matched, and matching principle is in terms of maximum likelihood presented as well. The instance shows that it is fused with threshold´s stability, wavelet-multi scalability, simplified in working amount, and enhances the ability to extract local detailed features for keeping border continuous, more important, graphic analysis is stimulant to improve the existed monitoring methods on tire.