2023年 03期

基于YOLOX模型的轮胎X射线图像中0°带束层接头检测

0° Belt Joint Detection in Tire X-ray Images Based on YOLOX Model


摘要(Abstract):

针对现有轮胎X射线图像中0°带束层接头检测算法易受轮胎自身花纹影响,造成较严重的误报和漏报的问题,提出基于YOLOX模型的轮胎X射线图像中0°带束层接头检测方法;利用投影直方图定位带束层区域,缩小检测范围,提高检测速度;在自建数据集上测试所提方法的检测性能。结果表明,所提出的方法能够准确定位0°带束层接头,误报率、漏报率分别小于1.15%、 0.59%。

关键词(KeyWords): 轮胎;0°带束层接头缺陷;目标检测;区域分割

基金项目(Foundation): 山东省重点研发计划项目(2017CXGC0810)

作者(Author): 史建杰,李金屏,赵建玉

DOI: 10.13349/j.cnki.jdxbn.20230320.001

参考文献(References):

[1] 逄增治,郑修楠,李金屏.全钢子午线轮胎X光图像的缺陷检测研究现状[J].智能系统学报,2019,14(4):793.

[2] 孙虹霞.轮胎X光图像缺陷检测算法研究[D].合肥:中国科学技术大学,2021.

[3] 康宇豪.子午线轮胎胎体帘线缺陷视觉检测方法研究[D].沈阳:沈阳工业大学,2020.

[4] LIN C W,CHEN G,ZHANG Y X,et al.Automatic detection of shoulder bending defects in tire X-ray images[C]//2020 International Conference on Computer Engineering and Application (ICCEA),March 18-20,2020,Guangzhou,China.New York:IEEE,2020:877.

[5] 于向茹,丁健配,李金屏.轮胎帘线交叉重叠缺陷检测[J].济南大学学报(自然科学版),2017,31(6):494.

[6] 逄增治.子午线轮胎X光图像钢丝圈缺陷检测方法研究[D].济南:济南大学,2020.

[7] ZHAO G,QIN S Y.High-precision detection of defects of tire texture through X-ray imaging based on local inverse difference moment features[J].Sensors,2018,18(8):2524.

[8] 林丽红,马铁军,徐培.Gabor变换在轮胎X光图像处理的应用[J].机械与电子,2016,34(4):59.

[9] REN S Q,HE K M,GIRSHICK R,et al.Faster R-CNN:towards real-time object detection with region proposal networks[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2019,39(6):1137.

[10] REDMON J,FARHADI A.YOLOv3:an incremental improvement[EB/OL].(2018-04-08) [2021-09-27].https://arxiv.org/pdf/1804.02767.pdf.

[11] 崔雪红.基于深度学习的轮胎缺陷无损检测与分类技术研究[D].青岛:青岛科技大学,2018.

[12] 吴则举,焦翠娟,陈亮.基于改进Faster R-CNN的轮胎缺陷检测方法[J].计算机应用,2021,41(7):1939-1946.

[13] CHEN J Y,LI Y W,ZHAO J X.X-ray of tire defects detection via modified faster R-CNN[C]//2019 2nd International Conference on Safety Produce Informatization(ⅡCSPI),November 28-30,2019,Chongqing,China.New York:IEEE,2019:257.

[14] ZHU Q D,AI X T.The defect detection algorithm for tire X-ray images based on deep learning[C]//2018 IEEE 3rd International Conference on Image,Vision and Computing (ICIVC),June 27-29,2018,Chongqing,China.New York:IEEE,2018:138.

[15] 陈梦焱.基于X光图像的轮胎缺陷检测算法研究[D].上海:上海交通大学,2020.

[16] 陈亮,白文涛.基于Efficient-Net的轮胎X光片缺陷检测技术研究[J].沈阳理工大学学报,2021,40(2):8.

[17] TAN M X,LE Q V.EfficientNet:rethinking model scaling for convolutional neural networks[EB/OL].(2020-09-11) [2021-09-27].https://arxiv.org/pdf/1905.11946.pdf.

[18] WANG R,GUO Q,LU S M,et al.Tire defect detection using fully convolutional network[J].IEEE Access,2019(7):43502.

[19] 王任.基于深度卷积网络的轮胎缺陷检测方法研究[D].济南:山东财经大学,2020.

[20] 林佳佳,吴则举,刘中冬.轮胎X射线0号带束层接头检测定位量化算法的研究[J].科学技术与工程,2016,16(25):121.

[21] 林佳佳.基于轮胎X光图像的0号带束层缺陷检测算法研究[D].青岛:青岛科技大学,2017.

[22] 张元刚,刘中华.基于GLCM算法的轮胎0°带束层接头缺陷检测[J].橡胶工业,2018,65(12):1402.

[23] BOCHKOVSKIY A,WANG C Y,LIAO H Y M.YOLOv4:optimal speed and accuracy of object detection[EB/OL].(2020-04-23) [2021-09-27].https://arxiv.org/pdf/2004.10934.pdf.

[24] GLENN J,ALEX S,JIRKA B,et al.Ultralytics/YOLOv5:v5.0-YOLOv5-P6 1280 models,AWS,Supervisely and YouTube integrations[EB/OL].(2021-04-11) [2021-09-27].https://zenodo.org/record/4679653.

[25] TAN M X,PANG R M,LE Q V.EfficientDet:scalable and efficient object detection[C]//2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR),June 13-19,2020,Seattle,WA,USA.New York:IEEE,2020:19874970.