Application of Automatic Image Recognition Algorithm in Nondestructive Testing of Automobile Tires
CHEN Jin'e
WU Shibin
WANG Yongcheng
Abstract：To address various defects in the production process of the tire sidewall area,an image processing and recognition method for non-destructive testing was designed.Based on the DR imaging principle,the image features of typical defects,including cord cracking and cord bending,were analyzed.A segmentation algorithm combining mean correction and Otsu thresholding,as well as an iterative refinement method based on structural elements,was proposed.By constructing different recognition criteria for different categories,the classification and detection of four types of microscopic and macroscopic defects,including foreign material inclusion and cord fracture,were realized.The algorithm achieved an overall defect recognition probability of 94.3％,with a missed detection rate below 8.7％.The results demonstrated that the designed image processing and recognition system can effectively implement automated non-destructive detection of tire sidewall defects.This study provided a valuable reference for ensuring the safety of automobile tires.
Keywords：automobile tireDR imagethreshold segmentationimage refinementdefects identification
Publication Date：2024-07-20
Online Publishing Date：2025-08-15(First online date of this platform, not the publication date of the document)
Pages：7( 60-66 )