Research on tire internal defect identification method based on deep learning

Jiaqi  Chen; Aijuan  Li; Te  Wang; Xibo  Wang

doi:10.59400/mea.v2i2.1495

Research on tire internal defect identification method based on deep learning

Jiaqi Chen School of Automotive Engineering, Shan Dong Jiaotong University, Jinan 250357, China
Aijuan Li School of Automotive Engineering, Shan Dong Jiaotong University, Jinan 250357, China
Te Wang School of Automotive Engineering, Shan Dong Jiaotong University, Jinan 250357, China
Xibo Wang School of Automotive Engineering, Shan Dong Jiaotong University, Jinan 250357, China

Ariticle ID: 1495

31 Views, 19 PDF Downloads

DOI: https://doi.org/10.59400/mea.v2i2.1495

Keywords: tire; internal defect; deep learning; dataset training; YOLOv5

Abstract

As an important part of automobile, the safety and durability of tire have attracted more and more attention. Tire defect detection is an important link to ensure tire quality, while traditional detection methods have problems such as low efficiency, high false detection rate and high labor intensity. Therefore, this study aims to develop an efficient and accurate tire defect identification and classification technique to improve the efficiency and accuracy of tire inspection. In this paper, based on YOLO (You Only Look Once) v5 algorithm, tire defect recognition and classification are studied. Firstly, the data sets containing various types of tire defects were collected and sorted out, and the data sets were preprocessed. Then, by constructing, training and optimizing the YOLOv5 tire defect recognition model, the fast and accurate recognition of tire defects was realized. Finally, the performance of the model was evaluated through experiments and compared with the existing methods. The experimental results show that the tire defect recognition and classification method based on YOLOv5 proposed in this study has high accuracy. Compared with traditional methods, this method has a significant improvement in detection speed and accuracy.

References

[1] Qin Z, Song Y, Li M, et al. Research on Quality Problems and Countermeasures of Automobile Tire. Theory and Research. 2012; 48(9): 25-27.

[2] Cui M, Yang H, Xie S, et al. Design of Automobile Tire Wear Detection System Based on Machine Vision. Manufacturing Automation. 2020; 42(01): 10-14.

[3] Li M, Jiang J. Tire Defect Detection Algorithm Based on Deep Learning. Information Technology and Informatization. 2021; (2): 52-53.

[4] Wang Z, Ren A, Li S. Detection System of Tire Tread Wear Degree. Computer System Application. 2021; 30(06): 88-93.

[5] Yang H, Wang Q, Xue M. Selection and Use Effect Analysis of Construction Machinery Tires. Rubber & Plastic Technology & Equipment. 2018; 44(11): 19-23.

[6] Huang Z, Zhang Y. A Survey of Object Detection Based on Deep Learning. Science and Technology Information. 2023; 21(24): 13-16.

[7] Zeng Z. Application Progress of Deep Learning in the field of Computer Vision. Computer Products and Circulation. 2020; (01): 230.

[8] Duan X, Zhou Y, Tian D. et al. Applications of Deep Learning in the field of autonomous driving. Unmanned System Technology. 2021; 4(06): 1-27.

[9] Zhang Y, Dong Z, Meng X. Research on Personalized Advertisement recommendation System and its application. Journal of Computer Science. 2021; 44(03): 531-563.

[10] Liu J, Jiang X, Ding Y, et al. Application of Face recognition Technology Based on Deep Learning in Computer Room Monitoring. Information Technology and Informatization. 2023; (11): 213-216.

[11] Wu S, Xu Y, Zhao D. A Survey of Object Detection Based on Deep Convolutional Networks. Pattern Recognition and Artificial Intelligence. 2018; 31(04): 335-346.

[12] Krizhevsky A, Sutskever I, Hinton GE. ImageNet classification with deep convolutional neural networks. Communications of the ACM. 2017; 60(6): 84-90. doi: 10.1145/3065386

[13] Abhronil S, Yuting Y, Robert W et al. Going Deeper in Spiking Neural Networks: VGG and Residual Architectures. Frontiers in neuroscience. 2019: 1395.

[14] He K, Zhang X, Ren S, et al. Deep Residual Learning for Image Recognition. In: Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). doi: 10.1109/cvpr.2016.90

[15] Zhou F, Jin L, Dong J. A Survey on Convolutional Neural Networks. Journal of Computer Science. 2017; 40(06): 1229-1251.

[16] Li B, Liu K, Gu Z, et al. A Survey on Convolutional Neural Networks. Computer Times. 2021; (04): 8-12, 17.

[17] Liu Y, Wang Y, Wang Y, et al. Image recognition based on deep learning convolutional neural network. Information and Communication. 2019; (07): 7-8.

[18] Zheng X. Tire Band Layer Defect Detection Based on X-ray Imaging and Texture Analysis. Jinan University; 2019.

[19] Wang CY, Mark Liao HY, Wu YH, et al. CSPNet: A New Backbone that can Enhance Learning Capability of CNN. In: Proceedings of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW). doi: 10.1109/cvprw50498.2020.00203

[20] Zhang Y, Guo W, Cai Z, et al. Remote sensing image object detection based on joint multi-scale and attention mechanism. Journal of Zhejiang University: Engineering Science Edition. 2022; 56(11): 2215–223.

[21] Zheng Z, Wang P, Liu W, et al. Distance-IoU Loss: Faster and Better Learning for Bounding Box Regression. In: Proceedings of the AAAI Conference on Artificial Intelligence. doi: 10.1609/aaai.v34i07.6999

[22] Ai X. Research on Automatic Defect Detection Algorithm Based on Tire X-ray Image. Harbin Engineering University; 2019.

[23] Chen M. Research on Tire Defect detection Algorithm Based on X-ray Image. Shanghai Jiaotong University; 2020.

Published

2024-10-25

How to Cite

Chen, J., Li, A., Wang, T., & Wang, X. (2024). Research on tire internal defect identification method based on deep learning. Mechanical Engineering Advances, 2(2), 1495. https://doi.org/10.59400/mea.v2i2.1495

Download Citation

Issue

Vol. 2 No. 2 (2024)

Section

Article

This work is licensed under a Creative Commons Attribution 4.0 International License.

In the paper, the MATLAB-Simulink model of simulation of operation of the fast light innovative regional train from “Serbian Railways” in traction and braking mode is exposed where changes are observed: stator currents of three-phase traction motors, traction electric motor speeds and electric multiple units, electromagnetic torque on the rotor shaft of the traction electric motor, and direct current bus voltage. The model allowed review of the listed parameters for: different allowed values of contact network voltage and total voltage distortion at the place of connection of the electric multiple unit to the contact network; different mechanical loads of the electric multiple unit and traction electric motor; and different train speeds and rotations of traction electric motors. Appropriate conclusions were made through the analysis of the simulation results obtained.

This paper defines vehicles and buildings as main sources of United Kingdom (UK) carbon dioxide (CO₂) and seeks to cut such emissions using green hydrogen made from combined wind and solar energy. Combustion vehicles powered by fossil petroleum emit near half of UK climate-warming CO₂ while buildings heated by natural gas provide a third. First, current UK grid problems are defined: Electricity, gas and petroleum grids. Refueling green vehicles has been a particular problem. Then experiments on the private wire community of Keele University show how green hydrogen could integrate both green vehicles and buildings. Next, the model supply chain is planned and tested. Finally, experiments and calculations are outlined, analyzing the optimum system design criteria proposed. We conclude that economic green hydrogen can displace petroleum in vehicles, while powering buildings instead of natural gas. Also, the prospect in 2024 is that profits can be made all along the green hydrogen supply chain, such that new businesses involved in local private clean communities can cost less than the National Grid monopoly and other dominant fossil energy companies.

Presently, dynamic impacts occurrences caused by car crashes have been frequently reported. Statistical results from different articles indicate that vehicle crashworthiness of bridge pier supersedes the other events. However, majority of the published articles focusses on sustainability and finding severity of distressed pier due to impact correlating faster construction methods, like accelerated bridge construction (ABC). This article is an attempt to examine post impacted behavior of a commonly used connector in ABC for resisting short duration shock. Static and dynamic performance analyses of a connector embedded within a coupler system has been examined. A representative ABC pier with the standardized and selected material properties collected from manufacturer’s data has been utilized. Coupler composite materials consisting of a hollow uniform cross-section cast iron cylinder filled up with specified concrete grout conforms higher strength. The performance examination has been conducted by numerical modeling using finite element method (FEM). A commercially used software ANSYS has been utilized for carrying out the simulations. To investigate the post impact dynamic behavior, mesh-independent studies seems inevitable and hence are executed to evaluate dynamic impact factor (DIF). Sensitivity studies are carried out for validating results in precision. The DIF of the main reinforcing tensile steel embedded into the grout placed inside the connector has been determined. Results captured from simulations to identify material properties at plastic stage in sustaining such load have been conducted. The results provide significant information that aids opting for selecting suitable connectors for attaining help the design offices.