利用多尺度变压器网络自动识别电子衍射中心

IF 2.1 3区工程技术 Q2 MICROSCOPY

Ultramicroscopy Pub Date : 2024-01-24 DOI:10.1016/j.ultramic.2024.113926

Mengshu Ge , Yue Pan , Xiaozhi Liu , Zhicheng Zhao , Dong Su

{"title":"利用多尺度变压器网络自动识别电子衍射中心","authors":"Mengshu Ge , Yue Pan , Xiaozhi Liu , Zhicheng Zhao , Dong Su","doi":"10.1016/j.ultramic.2024.113926","DOIUrl":null,"url":null,"abstract":"<div><p>Selected area electron diffraction (SAED) is a widely used technique for characterizing the structure and measuring lattice parameters of materials. An autonomous analytic method has become an urgent demand for the large-scale SAED data produced from in-situ experiments. In this work, we realize the automatic processing for center identification with a proposed deep segmentation model named the multi-scale Transformer (MS-Trans) network. This algorithm enables robust segmentation of the central spots by combining a novel gated axial-attention module and multi-scale feature fusion. The proposed MS-Trans model shows high precision and robustness, enabling autonomous processing of SAED patterns without any prior knowledge. The application on in-situ SAED data of the oxidation process of FeNi alloy demonstrates its capability of implementing autonomous quantitative processing.</p><p>© 2017 Elsevier Inc. All rights reserved.</p></div>","PeriodicalId":23439,"journal":{"name":"Ultramicroscopy","volume":"259 ","pages":"Article 113926"},"PeriodicalIF":2.1000,"publicationDate":"2024-01-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Automatic center identification of electron diffraction with multi-scale transformer networks\",\"authors\":\"Mengshu Ge , Yue Pan , Xiaozhi Liu , Zhicheng Zhao , Dong Su\",\"doi\":\"10.1016/j.ultramic.2024.113926\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><p>Selected area electron diffraction (SAED) is a widely used technique for characterizing the structure and measuring lattice parameters of materials. An autonomous analytic method has become an urgent demand for the large-scale SAED data produced from in-situ experiments. In this work, we realize the automatic processing for center identification with a proposed deep segmentation model named the multi-scale Transformer (MS-Trans) network. This algorithm enables robust segmentation of the central spots by combining a novel gated axial-attention module and multi-scale feature fusion. The proposed MS-Trans model shows high precision and robustness, enabling autonomous processing of SAED patterns without any prior knowledge. The application on in-situ SAED data of the oxidation process of FeNi alloy demonstrates its capability of implementing autonomous quantitative processing.</p><p>© 2017 Elsevier Inc. All rights reserved.</p></div>\",\"PeriodicalId\":23439,\"journal\":{\"name\":\"Ultramicroscopy\",\"volume\":\"259 \",\"pages\":\"Article 113926\"},\"PeriodicalIF\":2.1000,\"publicationDate\":\"2024-01-24\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Ultramicroscopy\",\"FirstCategoryId\":\"5\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S0304399124000056\",\"RegionNum\":3,\"RegionCategory\":\"工程技术\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q2\",\"JCRName\":\"MICROSCOPY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Ultramicroscopy","FirstCategoryId":"5","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0304399124000056","RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"MICROSCOPY","Score":null,"Total":0}

引用次数: 0

摘要

选区电子衍射（SAED）是一种广泛应用于表征材料结构和测量晶格参数的技术。对于原位实验产生的大规模 SAED 数据，迫切需要一种自主分析方法。在这项工作中，我们利用一种名为多尺度变压器（MS-Trans）网络的深度分割模型，实现了中心识别的自动处理。该算法结合了新颖的门控轴向注意力模块和多尺度特征融合，能够对中心点进行稳健的分割。所提出的 MS-Trans 模型具有高精度和鲁棒性，能够在没有任何先验知识的情况下自主处理 SAED 模式。在铁镍合金氧化过程的原位 SAED 数据上的应用证明了其实现自主定量处理的能力。保留所有权利。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

查看原文本刊更多论文

Automatic center identification of electron diffraction with multi-scale transformer networks

Selected area electron diffraction (SAED) is a widely used technique for characterizing the structure and measuring lattice parameters of materials. An autonomous analytic method has become an urgent demand for the large-scale SAED data produced from in-situ experiments. In this work, we realize the automatic processing for center identification with a proposed deep segmentation model named the multi-scale Transformer (MS-Trans) network. This algorithm enables robust segmentation of the central spots by combining a novel gated axial-attention module and multi-scale feature fusion. The proposed MS-Trans model shows high precision and robustness, enabling autonomous processing of SAED patterns without any prior knowledge. The application on in-situ SAED data of the oxidation process of FeNi alloy demonstrates its capability of implementing autonomous quantitative processing.

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

Ultramicroscopy 工程技术-显微镜技术

CiteScore

4.60

自引率

13.60%

发文量

117

审稿时长

5.3 months

期刊介绍： Ultramicroscopy is an established journal that provides a forum for the publication of original research papers, invited reviews and rapid communications. The scope of Ultramicroscopy is to describe advances in instrumentation, methods and theory related to all modes of microscopical imaging, diffraction and spectroscopy in the life and physical sciences.