| 基于深度学习的铁路关键部件缺陷检测研究 |
| |
| 引用本文： | 赵冰,代明睿,李平,马小宁,吴艳华.基于深度学习的铁路关键部件缺陷检测研究[J].铁道学报,2019(8):67-73. |
| |
| 作者姓名： | 赵冰 代明睿 李平 马小宁 吴艳华 |
| |
| 作者单位： | 中国铁道科学研究院研究生部;中国铁道科学研究院铁路大数据研究与应用创新中心 |
| |
| 基金项目： | 国家重点研发计划(2018YFB1201403);中国铁路总公司科技研究开发计划(2017J003-D) |
| |
| 摘 要： | 关键部件缺陷图像自动检测对于复兴号动车组运营维护意义重大,但目前主要依靠专业人员对检测图像进行分析,耗费大量人力、物力,造成检测周期长,检测准确率无法保证。提出一种结合部件检测与缺陷分类流程的双通道缺陷检测框架MCDDF(Multi-channel Defect Detection Framework),部件检测通道基于目标检测算法实现动车组关键部件定位,定位后的关键部件经裁剪进行超分辨率提升,传入缺陷分类通道基于迁移学习方法实现缺陷类别的准确分类,结合两通道信息实现缺陷检测任务。实验分析两通道的性能提升方法,对比MCDDF与传统基于目标检测方法在铁路关键部件缺陷图像上的检测效果,验证了MCDDF方法的有效性。
|
| 关 键 词： | 缺陷检测 卷积神经网络 目标检测 缺陷分类 图像超分辨率 |
Research on Defect Detection of Railway Key Components Based on Deep Learning |
| |
| ZHAO Bing,DAI Mingrui,LI Ping,MA Xiaoning,WU Yanhua.Research on Defect Detection of Railway Key Components Based on Deep Learning[J].Journal of the China railway Society,2019(8):67-73. |
| |
| Authors: | ZHAO Bing DAI Mingrui LI Ping MA Xiaoning WU Yanhua |
| |
| Institution: | (Department of Postgraduates,China Academy of Railway Sciences,Beijing 100081,China;Railway Big Data Research and Application Innovation Center,China Academy of Railway Sciences,Beijing 100081,China) |
| |
| Abstract: | ZHAO Bing;DAI Mingrui;LI Ping;MA Xiaoning;WU Yanhua(Department of Postgraduates,China Academy of Railway Sciences,Beijing 100081,China;Railway Big Data Research and Application Innovation Center,China Academy of Railway Sciences,Beijing 100081,China) |
| |
| Keywords: | defect detection convolutional neural network object detection detect classification image super resolution |
| 本文献已被 CNKI 维普 等数据库收录！ |
|
