• 首页
  • 期刊简介
  • 编委会
  • 投稿指南
  • 收录情况
  • 杂志订阅
  • 联系我们
引用本文:刘 玄,文 勇,梁建烈,马 坤.无监督学习在相图测定中识别合金三相平衡点的研究[J].软件工程,2023,(7):21-24.【点击复制】
【打印本页】   【下载PDF全文】   【查看/发表评论】  【下载PDF阅读器】  
←前一篇|后一篇→ 过刊浏览
分享到: 微信 更多
无监督学习在相图测定中识别合金三相平衡点的研究
刘 玄1, 文 勇2, 梁建烈3, 马 坤3
(1.广西民族大学电子信息学院, 广西 南宁 530006;
2.广西民族大学人工智能学院, 广西 南宁 530006;
3.广西民族大学材料与环境学院, 广西 南宁 530006)
2020210854001104@stu.gxmzu.edu.cn; wenyong@gxmzu.edu.cn; Liangjianlie@gxun.edu.cn; mkkxjh@163.com
摘 要: 传统的扫描电镜分析方法,合金物相的三相平衡点均为手工标记。针对扫描电镜金相分析中常碰到的物相分析问题,提出了一种物相三相平衡点的自动识别方法,实现平衡点的智能化标注,提升实验效率。该方法分为三相分割和三相平衡点识别两个阶段。在分割阶段提出一种无监督分割方法完成三种物相分割,在平衡点识别阶段设计识别算法对分割结果的平衡点区域进行识别。实验结果表明,分割方法在三元合金相图数据集上的平均交并比(Mean Intersection over Union, MIoU)达到49.86%,三相平衡点平均检出率达62%。
关键词: 三相平衡点识别;三元合金相图;无监督语义分割;识别算法
中图分类号: TP391    文献标识码: A
基金项目: 2020年广西民族大学研究生教育创新计划项目(gxun-chxps202085).
Research on Unsupervised Learning for Identification of Three-Phase Equilibrium Point of Alloy in Phase Diagram Determination
LIU Xuan1, WEN Yong2, LIANG Jianlie3, MA Kun3
(1.College of Electronic Information, Guangxi Minzu University, Nanning 530006, China;
2.School of Artificial Intelligence, Guangxi Minzu University, Nanning 530006, China;
3.School of Materials and Enviroment, Guangxi Minzu University, Nanning 530006, China)
2020210854001104@stu.gxmzu.edu.cn; wenyong@gxmzu.edu.cn; Liangjianlie@gxun.edu.cn; mkkxjh@163.com
Abstract: In traditional SEM (Scanning Electron Microscopy) analysis method, the three-phase equilibrium points of alloy phases are manually marked. In order to address common phase analysis problems in scanning electron microscopy metallographic analysis, this paper proposes a method for automatic identification of three-phase equilibrium points to achieve intelligent mark of equilibrium points and improve experimental efficiency. This method is divided into two stages: three-phase segmentation and three-phase equilibrium point recognition. In the segmentation stage, an unsupervised segmentation method is proposed to complete three-phase segmentation. In the equilibrium point recognition stage, a recognition algorithm is designed to identify the equilibrium point area of the segmentation results. The experimental results show that the average Mean Intersection over Union (MIoU) of the segmentation method on the ternary alloy phase diagram dataset reaches 49.86% , and the average detection rate of the three-phase equilibrium point reaches 62% .
Keywords: three-phase equilibrium point recognition; ternary alloy phase diagram; unsupervised semantic segmentation; recognition algorithm


版权所有:软件工程杂志社
地址:辽宁省沈阳市浑南区新秀街2号 邮政编码:110179
电话:0411-84767887 传真:0411-84835089 Email:semagazine@neusoft.edu.cn
备案号:辽ICP备17007376号-1
技术支持:北京勤云科技发展有限公司

用微信扫一扫

用微信扫一扫