Search research articles

关于 JoVE

概览领导团队博客 JoVE 帮助中心

作者

出版流程编辑委员会范围与政策同行评审常见问题投稿

图书馆员

用户评价订阅访问资源图书馆顾问委员会常见问题

研究

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments 存档

教育

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual 教师资源中心教师网站

使用条款与条件

相关概念视频

您也可能阅读

相关文章

通过共同作者、期刊和引用图与本文相关的文章。

排序

Same author

All-Day Freshwater Harvesting Using Solar Auto-Tracking Assisted Selective Solar Absorption and Radiative Cooling.

Materials (Basel, Switzerland)·2025

Same author

Deep Reinforcement Learning-Guided Inverse Design of Transparent Heat Mirror Film for Broadband Spectral Selectivity.

Materials (Basel, Switzerland)·2025

Same author

Reinforcement learning-based inverse design of composite films for spacecraft smart thermal control.

Physical chemistry chemical physics : PCCP·2025

Same author

VO<sub>2</sub>-Based Spacecraft Smart Radiator with High Emissivity Tunability and Protective Layer.

Nanomaterials (Basel, Switzerland)·2024

Same author

Preparation of Thermochromic Vanadium Dioxide Films Assisted by Machine Learning.

Nanomaterials (Basel, Switzerland)·2024

Same author

Design and Implementation of Electrochromic Smart Windows with Self-Driven Thermoelectric Power Generation.

Nanomaterials (Basel, Switzerland)·2024

Search research articles

相关实验视频

Updated: Sep 19, 2025

Picometer-Precision Atomic Position Tracking through Electron Microscopy

Picometer-Precision Atomic Position Tracking through Electron Microscopy

Published on: July 3, 2021

在SEM图像中测量纳米粒子大小的深度学习方法.

Tingwang Tao¹, Haining Ji^1,2, Bin Liu^1,2

¹School of Physics and Optoelectronics, Xiangtan University Xiangtan 411105 China.

|June 16, 2025

概括

本研究介绍了一种使用改进的U-Net模型进行纳米粒子大小测量的自动化方法. 该技术提高了小或低对比度粒子的精度,提高了纳米粒子分析的效率.

科学领域:

材料科学材料科学材料科学
纳米技术纳米技术
图像分析图像分析

背景情况:

精确的纳米粒子尺寸分布对于材料性能和应用至关重要.
在SEM图像中手动测量纳米粒子尺寸是低效的,容易出现错误.
现有的自动化方法难以处理小颗粒,低对比度和尺度尺度校准.

研究的目的:

从SEM图像开发自动化,准确和高效的纳米粒子尺寸测量方法.
克服纳米粒子分析中现有的语义细分模型的局限性.
通过自动尺度表识别实现精确的像素到物理尺寸的转换.

主要方法:

开发了一种改进的U-Net模型,包括注意力机制 (CBAM) 和残余网络 (ResNet50).
集成了一个自动标尺识别算法,用于准确的像素到物理尺寸转换.
该模型在SEM图像上进行训练和评估,用于纳米粒子细分和尺寸测量.

主要成果:

增强的U-Net模型在测试组件上实现了高性能,IOU为87.79%,F1得分为93.50%.
在自动和手动尺寸测量之间观察到0.91的强烈斯皮尔曼相关系数.
颗粒大小测量的平均相对误差低至4.25%,显示出高精度和稳定性.

更多相关视频

Nanoparticle Tracking Analysis for the Quantification and Size Determination of Extracellular Vesicles

Nanoparticle Tracking Analysis for the Quantification and Size Determination of Extracellular Vesicles

Published on: March 28, 2021

Studying the Effects of Temperature on the Nucleation and Growth of Nanoparticles by Liquid-Cell Transmission Electron Microscopy

Studying the Effects of Temperature on the Nucleation and Growth of Nanoparticles by Liquid-Cell Transmission Electron Microscopy

Published on: February 17, 2021

相关实验视频

Last Updated: Sep 19, 2025

Picometer-Precision Atomic Position Tracking through Electron Microscopy

Picometer-Precision Atomic Position Tracking through Electron Microscopy

Published on: July 3, 2021

Nanoparticle Tracking Analysis for the Quantification and Size Determination of Extracellular Vesicles

Nanoparticle Tracking Analysis for the Quantification and Size Determination of Extracellular Vesicles

Published on: March 28, 2021

Studying the Effects of Temperature on the Nucleation and Growth of Nanoparticles by Liquid-Cell Transmission Electron Microscopy

Studying the Effects of Temperature on the Nucleation and Growth of Nanoparticles by Liquid-Cell Transmission Electron Microscopy

Published on: February 17, 2021

结论:

拟议的自动化方法显著提高了纳米粒子大小测量的准确性和效率.
注意力机制和自动尺度尺度校准的整合解决了纳米粒子分析的关键挑战.
这种可靠的自动化工具为研究和工程应用的纳米粒子表征提供了便利.