基于语义分割的扫描电镜图像中纳米复合材料团聚体识别

IF 3.8 Q2 ENGINEERING, ELECTRICAL & ELECTRONIC IET Nanodielectrics Pub Date : 2022-03-04 DOI:10.1049/nde2.12034
Yu Bai, Yan Wang, Dayuan Qiang, Xin Yuan, Jiehui Wu, Weilong Chen, Sai Zhang, Yanru Zhang, George Chen
{"title":"基于语义分割的扫描电镜图像中纳米复合材料团聚体识别","authors":"Yu Bai,&nbsp;Yan Wang,&nbsp;Dayuan Qiang,&nbsp;Xin Yuan,&nbsp;Jiehui Wu,&nbsp;Weilong Chen,&nbsp;Sai Zhang,&nbsp;Yanru Zhang,&nbsp;George Chen","doi":"10.1049/nde2.12034","DOIUrl":null,"url":null,"abstract":"<p>The agglomeration causes a significant challenge to nanodielectrics. Identification of agglomerates in scanning electron microscopy (SEM) images is an important step of solving this issue. Motivated by the fast development of image recognition in computer vision, we propose a new approach for agglomerates identification in SEM images of nanodielectrics by semantic segmentation, which is more efficient and accurate than traditional methods. Three models based on convolutional neural networks are investigated in this work, namely pixel blocks classification network, full convolutional segmentation network employed with data augmentation and unsupervised self-encoding network. All three networks can preliminarily identify agglomerates of spherical silica-based blend polyethylene nanocomposites. The mean intersection over union (mIoU) of pixel blocks classification network is 0.843 and it takes 25 s to process an image. Full convolutional segmentation network only needs 0.059 s to process a sample, with a mIoU of 0.777. Unsupervised self-encoding network can reach a mIoU of 0.747 at a speed of 5.806 s. According to the amount of data sets, and requirements for different speed and accuracy, three kinds of networks can be flexibly selected.</p>","PeriodicalId":36855,"journal":{"name":"IET Nanodielectrics","volume":"5 2","pages":"93-103"},"PeriodicalIF":3.8000,"publicationDate":"2022-03-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ietresearch.onlinelibrary.wiley.com/doi/epdf/10.1049/nde2.12034","citationCount":"0","resultStr":"{\"title\":\"Identification of nanocomposites agglomerates in scanning electron microscopy images based on semantic segmentation\",\"authors\":\"Yu Bai,&nbsp;Yan Wang,&nbsp;Dayuan Qiang,&nbsp;Xin Yuan,&nbsp;Jiehui Wu,&nbsp;Weilong Chen,&nbsp;Sai Zhang,&nbsp;Yanru Zhang,&nbsp;George Chen\",\"doi\":\"10.1049/nde2.12034\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p>The agglomeration causes a significant challenge to nanodielectrics. Identification of agglomerates in scanning electron microscopy (SEM) images is an important step of solving this issue. Motivated by the fast development of image recognition in computer vision, we propose a new approach for agglomerates identification in SEM images of nanodielectrics by semantic segmentation, which is more efficient and accurate than traditional methods. Three models based on convolutional neural networks are investigated in this work, namely pixel blocks classification network, full convolutional segmentation network employed with data augmentation and unsupervised self-encoding network. All three networks can preliminarily identify agglomerates of spherical silica-based blend polyethylene nanocomposites. The mean intersection over union (mIoU) of pixel blocks classification network is 0.843 and it takes 25 s to process an image. Full convolutional segmentation network only needs 0.059 s to process a sample, with a mIoU of 0.777. Unsupervised self-encoding network can reach a mIoU of 0.747 at a speed of 5.806 s. According to the amount of data sets, and requirements for different speed and accuracy, three kinds of networks can be flexibly selected.</p>\",\"PeriodicalId\":36855,\"journal\":{\"name\":\"IET Nanodielectrics\",\"volume\":\"5 2\",\"pages\":\"93-103\"},\"PeriodicalIF\":3.8000,\"publicationDate\":\"2022-03-04\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://ietresearch.onlinelibrary.wiley.com/doi/epdf/10.1049/nde2.12034\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"IET Nanodielectrics\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://onlinelibrary.wiley.com/doi/10.1049/nde2.12034\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q2\",\"JCRName\":\"ENGINEERING, ELECTRICAL & ELECTRONIC\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"IET Nanodielectrics","FirstCategoryId":"1085","ListUrlMain":"https://onlinelibrary.wiley.com/doi/10.1049/nde2.12034","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"ENGINEERING, ELECTRICAL & ELECTRONIC","Score":null,"Total":0}
引用次数: 0

摘要

这种结块现象给纳米电介质带来了巨大的挑战。扫描电镜(SEM)图像中团聚体的识别是解决这一问题的重要步骤。在计算机视觉图像识别技术快速发展的背景下,本文提出了一种基于语义分割的纳米电介质扫描电镜图像团聚体识别新方法,该方法比传统方法更高效、更准确。本文研究了基于卷积神经网络的三种模型,即像素块分类网络、数据增强的全卷积分割网络和无监督自编码网络。这三种网络都可以初步识别球形硅基共混聚乙烯纳米复合材料的团聚体。像素块分类网络的平均交联数(mIoU)为0.843,处理一幅图像所需时间为25 s。全卷积分割网络处理一个样本只需要0.059 s, mIoU为0.777。无监督自编码网络以5.806 s的速度达到0.747的mIoU。根据数据集的数量,以及对不同速度和精度的要求,可以灵活选择三种网络。
本文章由计算机程序翻译,如有差异,请以英文原文为准。

摘要图片

查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
Identification of nanocomposites agglomerates in scanning electron microscopy images based on semantic segmentation

The agglomeration causes a significant challenge to nanodielectrics. Identification of agglomerates in scanning electron microscopy (SEM) images is an important step of solving this issue. Motivated by the fast development of image recognition in computer vision, we propose a new approach for agglomerates identification in SEM images of nanodielectrics by semantic segmentation, which is more efficient and accurate than traditional methods. Three models based on convolutional neural networks are investigated in this work, namely pixel blocks classification network, full convolutional segmentation network employed with data augmentation and unsupervised self-encoding network. All three networks can preliminarily identify agglomerates of spherical silica-based blend polyethylene nanocomposites. The mean intersection over union (mIoU) of pixel blocks classification network is 0.843 and it takes 25 s to process an image. Full convolutional segmentation network only needs 0.059 s to process a sample, with a mIoU of 0.777. Unsupervised self-encoding network can reach a mIoU of 0.747 at a speed of 5.806 s. According to the amount of data sets, and requirements for different speed and accuracy, three kinds of networks can be flexibly selected.

求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
IET Nanodielectrics
IET Nanodielectrics Materials Science-Materials Chemistry
CiteScore
5.60
自引率
3.70%
发文量
7
审稿时长
21 weeks
期刊最新文献
A combined technique for power transformer fault diagnosis based on k-means clustering and support vector machine Improvement in non-linear electrical conductivity of silicone rubber by incorporating zinc oxide fillers and grafting small polar molecules Traditional fault diagnosis methods for mineral oil-immersed power transformer based on dissolved gas analysis: Past, present and future Enhanced thermal conductivity and self-healing property of PUDA/boron nitride micro-sheets composites with a small number of graphene nano-platelets Improving the dielectric properties of polypropylene for metallised film capacitors based on cyclic olefin copolymer blending
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1