PDF全文下载：2017030106
张岩，李涛，李庆领*
(青岛科技大学 机电工程学院，山东 青岛266061）
摘要:提出一种基于全变分模型图像分解的轮胎胎侧缺陷检测方法，将轮胎胎侧X射线图像分解为纹理分量和平滑分量，使得胎侧橡胶和胎侧帘线的成像分量分解，从而可以单独对这两种结构中的缺陷进行分别检测。利用对胎侧帘线结构统计特征和帘线缺陷特征表达，分析和设计了帘线缺陷算法，有效的检测帘线断开缺陷、帘线间距缺陷以及帘线弯曲等缺陷。实验结果表明：该算法可以通过胎侧结构的图像分解获得其成像，简化了检测算法的复杂度、提高了检测精度。
关键词:子午线轮胎； 缺陷检测；X射线图像； 全变分； 图像分解
中图分类号：TP 29文献标志码：A
引用格式：张岩，李涛，李庆领.基于全变分模型的子午线轮胎X射线图像胎侧缺陷自动检测方法［J］.青岛科技大学学报(自然科学版), 2017, 38(3): 106-112.
ZHANG Yan， LI Tao， LI Qingling.Automatic radial tire sidewall defect detection in tire X-ray images based on total variation model［J］. Journal of Qingdao University of Science and Technology（Natural Science Edition), 2017, 38(3): 106-112.
Automatic Radial Tire Sidewall Defect Detection in Tire X-ray Images Based on Total Variation Model
ZHANG Yan， LI Tao， LI Qingling
(College of Electromechanical Engineering, Qingdao University of Science and Technology, Qingdao 266061， China）
Abstract:In this work, a tire sidewall defect detection method was proposed based on image total variation decomposition model. Tire sidewall X-ray images were obtained and decomposed into texture components and smooth components such that they can be detected without interference. Tire cord defect detection algorithm was designed based on the statistical characteristics of tire cord structure and cord defects. Experimental results show that the proposed algorithm can simplify complexity of the detection algorithm and improve detection accuracy effectively by virtue of tire image component decomposition.
Key words: radial tire; defect detection; X-ray image; total variation; image decomposition
收稿日期：2016-07-01
基金项目：山东省自然科学基金项目（ZR2014FL021）；青岛市自主创新计划项目（15-9-1-83-jch）；青岛科技大学引进人才科研启动基金项目（010022671）.
作者简介：张岩（1980-），男，博士，副研究员.*通信联系人.
文章编号：16726987（2017）03010607； DOI： 10.16351/j.1672-6987.2017.03.019
