计算机应用 ›› 2021, Vol. 41 ›› Issue (7): 1939-1946.DOI: 10.11772/j.issn.1001-9081.2020091488

所属专题: 人工智能

• 人工智能 • 上一篇    下一篇

基于改进Faster R-CNN的轮胎缺陷检测方法

吴则举, 焦翠娟, 陈亮   

  1. 青岛理工大学 信息与控制工程学院, 山东 青岛 266520
  • 收稿日期:2020-09-24 修回日期:2020-12-22 出版日期:2021-07-10 发布日期:2021-01-26
  • 通讯作者: 吴则举
  • 作者简介:吴则举(1980-),男,山东青岛人,副教授,博士,CCF会员,主要研究方向:机器视觉、水下三维重建;焦翠娟(1995-),女,山东日照人,硕士研究生,主要研究方向:机器视觉;陈亮(1996-),男,山东青岛人,硕士研究生,主要研究方向:机器视觉。
  • 基金资助:
    山东省重点研发计划(公益类专项)项目(2018GGX101040);青岛市源头创新计划应用基础研究项目(18-2-2-62-jch)。

Tire defect detection method based on improved Faster R-CNN

WU Zeju, JIAO Cuijuan, CHEN Liang   

  1. School of Information and Control Engineering, Qingdao University of Technology, Qingdao Shandong 266520, China
  • Received:2020-09-24 Revised:2020-12-22 Online:2021-07-10 Published:2021-01-26
  • Supported by:
    This work is partially supported by the Key Research and Development Program of Shandong Province (Public Welfare Project) (2018GGX101040), the Applied Basic Research Program of Qingdao (18-2-2-62-jch).

摘要: 轮胎生产过程中出现的胎侧异物、胎冠异物、气泡、胎冠开根以及胎侧开根等缺陷会影响轮胎出厂后的使用,所以出厂使用前需要对每条轮胎进行无损检测。为了实现在工业中对于轮胎缺陷进行自动检测,提出了一种基于改进Faster R-CNN的轮胎缺陷自动检测方法。首先,在预处理阶段,用直方图均衡化方法对轮胎图象的灰度进行拉伸,提高数据集的对比度,使图像目标和背景的灰度值产生明显差异;其次,为提高轮胎缺陷位置检测和识别的准确率,对Faster R-CNN结构进行改进,即把ZF卷积神经网络中第三层的卷积特征和第五层的卷积特征结合后输出,并将其作为区域建议网络层的输入;然后,在RoI pooling层之后引入在线难例挖掘(OHEM)算法,使轮胎缺陷检测的准确率得到进一步的提高。实验结果表明,改进后的Faster R-CNN的轮胎缺陷检测方法可以准确地分类和定位轮胎X射线图像缺陷,平均测试准确率可以达到95.7%。此外,还可以通过对网络进行微调来获得新的检测模型以检测其他类型的缺陷。

关键词: Faster R-CNN, 轮胎缺陷检测, ZF卷积神经网络, 在线难例挖掘

Abstract: The defects such as sidewall foreign matter, crown foreign body, air bubble, crown split and sidewall root opening that appear in the process of tire production will affect the use of tires after leaving factory, so it is necessary to carry out nondestructive testing on each tire before leaving the factory. In order to achieve automatic detection of tire defects in industry, an automatic tire defect detection method based on improved Faster Region-Convolutional Neural Network (Faster R-CNN) was proposed. Firstly, at the preprocessing stage, the gray level of tire image was stretched by the histogram equalization method to enhance the contrast of the dataset, resulting in a significant difference between gray values of the image target and the background. Secondly, to improve the accuracy of position detection and identification of tire defects, the Faster R-CNN structure was improved. That is the convolution features of the third layer and the convolution features of the fifth layer in ZF (Zeiler and Fergus) convolutional neural network were combined together and output as the input of the regional proposal network layer. Thirdly, the Online Hard Example Mining (OHEM) algorithm was introduced after the RoI (Region-of-Interesting) pooling layer to further improve the accuracy of defect detection. Experimental results show that the tire X-ray image defects can be classified and located accurately by the improved Faster R-CNN defect detection method with average test recognition of 95.7%. In addition, new detection models can be obtained by fine-tuning the network to detect other types of defects..

Key words: Faster Region-Convolutional Neural Network (Faster R-CNN), tire defect detection, ZF (Zeiler and Fergus) convolutional neural network, Online Hard Example Mining (OHEM)

中图分类号: