首页> 外文期刊>生産研究 >Preliminary Study of Ceiling Damage Detection System Using Image Database by Deep Learning Approach (Convolutional Neural Networks)
【24h】

Preliminary Study of Ceiling Damage Detection System Using Image Database by Deep Learning Approach (Convolutional Neural Networks)

机译:使用深度学习方法(卷积神经网络)的图像数据库检测天花板伤害的系统的初步研究

获取原文
获取原文并翻译 | 示例
       

摘要

Convolutional neural networks approach is showing as good as or even better accuracy in classification jobs than that human does by researchers in recent years. When the job becomes more abstractive, like evaluating dangerous degree in the ceilings and detect where it is likely to fail, ConvNets is also a comprehensive competent approach to aid human, especially to non-professionals. The ConvNets is capable of telling the notion of damage from intact through all the layers and filters. The architecture in this report is also possible to grasp important features that reveal safety in ceilings. This can further support the human-DL interaction.
机译:卷积神经网络方法在分类工作中显示出的准确性与近年来研究人员所能达到的甚至更好。当工作变得更加抽象时,例如评估天花板的危险程度并检测出可能发生故障的地方,ConvNets也是一种综合胜任的方法,可以帮助人类,尤其是对非专业人士。 ConvNets能够说出所有层和过滤器的完好无损。本报告中的体系结构也有可能掌握揭示天花板安全性的重要特征。这可以进一步支持人与DL的交互。

著录项

  • 来源
    《生産研究》 |2017年第6期|39-43|共5页
  • 作者

    Pujin WANG; Kenichi KAWAGUCHI;

  • 作者单位

    Graduate School of Engineering, The University of Tokyo;

    Graduate School of Engineering, The University of Tokyo;

  • 收录信息
  • 原文格式 PDF
  • 正文语种 eng
  • 中图分类
  • 关键词

  • 入库时间 2022-08-18 03:50:04

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号