期刊文献+

推动4C装置图像智能识别技术持续发展的思考 被引量:3

Thoughts on Promoting the Sustainable Development of Intelligent Image Identification Technology of 4C Device
下载PDF
导出
摘要 图像智能识别是4C装置发挥安全保障功能和提高检测劳效的关键影响因素,在梳理4C装置图像智能识别技术发展现状的同时,提出现有智能识别技术存在识别项点数量少和适应能力不足等问题。在对智能识别技术路线的关键要素和我国电气化铁路接触网技术特点与优势分析的基础上,提出构建一套既符合智能识别技术路线需要,又满足我国电气化铁路接触网技术特点要求,并能充分发挥我国路网规模优势的可持续发展工作路线。针对该工作路线的建立与实施,提出统一编码规则、统一算法模块接口和优化成像质量评价方法3个方面的关键工作,为下一步工作开展奠定基础。 Intelligent image identification is the key factor affecting the security function and detection efficiency of 4C device.While analyzing the development status of intelligent image identification technology of 4C device,the paper discusses the existing intelligent image identification technology problems,such as a small number of identification points and insufficient adaptability.Based on analyzing the key elements of intelligent identification technology route and the technical characteristics and advantages of the OCS of the China's electrified railway,a set of sustainable development work route is put forward,which not only meets the requirements of the intelligent identification technology route,but also meet the technical characteristics and requirements on the OCS of the China's electrified railway,and moreover,this route could give full play to the scale advantages of China's railway network.With respect to the establishment and implementation of the work route,the key works of unifying coding rules,unifying algorithm module interface and optimizing imaging quality evaluation method are proposed,laying a foundation for the next steps.
作者 盛良 张克永 张文轩 杨志鹏 SHENG Liang;ZHANG Keyong;ZHANG Wenxuan;YANG Zhipeng(Railway Infrastructure Inspection Center,China State Railway Group Co Ltd,Beijing 100081,China;Department of Track,Communication&Signaling and Power Supply,China State Railway Group Co Ltd,Beijing 100844,China;Infrastructure Inspection Research Institute,China Academy of Railway Sciences Corporation Limited,Beijing 100081,China)
出处 《中国铁路》 2020年第10期84-88,共5页 China Railway
基金 中国铁路总公司科技研究开发计划项目(P2018G002)。
关键词 6C系统 4C装置 智能识别 编码规则 软件接口 成像质量 6C system 4C device intelligent identification coding rule software interface imaging quality
  • 相关文献

参考文献5

二级参考文献105

  • 1BENGIO Y, DELALLEAU O. On the expressive power of deep archi- tectures[ C ]//Proc of the 14th International Conference on Discovery Science. Berlin : Springer-Verlag, 2011 : 18 - 36.
  • 2BENGIO Y. Leaming deep architectures for AI[ J]. Foundations and Trends in Machine Learning ,2009,2 ( 1 ) : 1-127.
  • 3HINTON G,OSINDERO S,TEH Y. A fast learning algorithm for deep belief nets [ J ]. Neural Computation ,2006,18 (7) : 1527-1554.
  • 4BENGIO Y, LAMBLIN P, POPOVICI D, et al. Greedy layer-wise training of deep networks [ C ]//Proc of the 12th Annual Conference on Neural Information Processing System. 2006:153-160.
  • 5LECUN Y, BOTTOU L, BENGIO Y, et al. Gradient-based learning ap- plied to document recognition[ J]. Proceedings of the iEEE, 1998, 86( 11 ) :2278-2324.
  • 6VINCENT P, LAROCHELLE H, BENGIO Y, et al. Extracting and composing robust features with denoising autoencoders[ C ]//Proc of the 25th International Conference on Machine Learning. New York: ACM Press ,2008 : 1096-1103.
  • 7VINCENT P, LAROCHELLE H, LAJOIE I, et aL Stacked denoising autoencoders:learning useftd representations in a deep network with a local denoising criterion [ J ]. Journal of Machine Learning Re- search ,2010,11 ( 12 ) :3371-3408.
  • 8YU Dong, DENG Li. Deep convex net: a scalable architecture for speech pattern classification [ C]//Proc of the 12th Annual Confe-rence of International Speech Comunication Association. 2011 : 2285- 2288.
  • 9POON H, DOMINGOS P. Sum-product networks:a new deep architec- ture[ C ]//Proc of IEEE Intemational Conference on Computer Vi- sion. 2011:689-690.
  • 10BENGIO Y,LECUN Y. Scaling learning algorithms towards AI[ M]// BOTTOU L,CHAPELLE O, DeCOSTE D,et al. Large-Scale Kernel Machines. Cambridge: MIT Press ,2007:321-358.

共引文献1187

同被引文献35

引证文献3

二级引证文献10

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部