Wheel-rail contact area image segmentation method based on Deeplabv3+

Author:

Zheng Yunshui1,Cai Mengyao1

Affiliation:

1. Lanzhou Jiaotong University

Abstract

Abstract

The contact curve can be extracted from the wheel-rail contact image to better study the wheel-rail contact relationship. To solve this problem, an improved Deeplabv3 + semantic segmentation model is proposed. Firstly, the feature extraction network in the original Deeplabv3 + model was optimized to lightweight MobileNetv2 to reduce the number of model parameters. Secondly, the original parallel atrous spatial pyramid pooling was optimized into a closely connected structure to improve the network extraction capability. Finally, the dual attention mechanism was added after the backbone network was extracted to improve the accuracy of the model. Experiments on wheel-rail contact image dataset showed that the improved model has good real-time performance, and the mean pixel accuracy is 88.72%, the mean intersection over union is 85.84%, and the F1-score is 91.48%, which are 2.29%, 2.63% and 2.15% higher than the original model, respectively. The method can be used as an effective method for wheel-rail contact image segmentation.

Publisher

Springer Science and Business Media LLC

Reference20 articles.

1. 1. ZHAI W M, ZHAO C F, XIA H et al. Basic scientific issues on dynamic performance evolution of high-speed railway infrastructure and service safety [J]. Science in China: Technical Sciences,2014,44(07):645–660. Yang G W, Wei Y J, Zhao G L, et al. Key mechanical problems of high-speed trains [J]. Advances in Mechanics,2015,45(00): 217–460.

2. 2. ZHAI W M, CHEN G. Method and criteria for evaluation of wheel derailment based on wheel vehicle rise [J]. Journal of the Railway Society,2001(02): 17–26.

3. 3. ZENG Q Y, XIANG J, ZHOU Z H, et al. Theory and Application of Train derailment Analysis [M]. Hunan: Central South University Press, 2006: 93–112.

4. 4. ZHONG Y H. Effect of wheel abnormal wear on dynamic derailment of high speed train [D]. Chengdu: Southwest Jiaotong University of China, 2017:10–22.

5. 5. MATSUMOTO A, MICHITSUJI Y, YOSUKE I, et al. Safety measures against flange-climb derailment in sharp curve-considering friction coefficient between wheel and rail –.[J].Wear,2019,Vol.432–433: 202931.

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3