Jove
Visualize
联系我们
JoVE
x logofacebook logolinkedin logoyoutube logo
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 JournalMethods CollectionsJoVE Encyclopedia of Experiments存档
教育
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab Manual教师资源中心教师网站
使用条款与条件
隐私政策
政策

相关概念视频

Imaging Biological Samples with Optical Microscopy01:18

Imaging Biological Samples with Optical Microscopy

8.8K
Optical microscopy uses optic principles to provide detailed images of samples. Antonie van Leeuwenhoek designed the first compound optical microscope in the 17th century to visualize blood cells, bacteria, and yeast cells. In 1830, Joseph Jackson Lister created an essentially modern light microscope. The 20th century saw the development of microscopes with enhanced magnification and resolution.
In optical microscopy, the specimen to be viewed is placed on a glass slide and clipped on the stage...
8.8K

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

Optical interference coatings: measurement challenge 2025 [Invited].

Applied optics·2026
Same author

Image quality evaluation for FIB-SEM images.

Journal of microscopy·2023
Same author

Optical interference coatings: measurement problem 2022 [Invited].

Applied optics·2023
Same author

Analysis of very low bacterial counts in small sample volumes using angle-resolved light scattering.

Applied optics·2023
Same author

Topography stitching in the spatial frequency domain for the representation of mid-spatial frequency errors.

Applied optics·2022
Same author

Predicting the Tensile Behaviour of Ultra-High Performance Fibre-Reinforced Concrete from Single-Fibre Pull-Out Tests.

Materials (Basel, Switzerland)·2022

相关实验视频

Updated: Jan 15, 2026

Author Spotlight: A Machine-Vision Approach to Transmission Electron Microscopy Workflows, Results Analysis and Data Management
10:23

Author Spotlight: A Machine-Vision Approach to Transmission Electron Microscopy Workflows, Results Analysis and Data Management

Published on: June 23, 2023

3.5K

简介:用于金属表面检查中的机器视觉的图像合成管道.

Juraj Fulir1,2, Natascha Jeziorski1,3, Lovro Bosnar1,2

  • 1Image Processing Department, Fraunhofer ITWM, 67663 Kaiserslautern, Germany.

Sensors (Basel, Switzerland)
|October 16, 2025
PubMed
概括

本研究引入了一种新的合成数据生成管道,用于实现现实的纹理,克服生成模型的局限性. 该方法确保了精确的控制,并为改进的视觉检查系统生成各种数据集.

关键词:
数据相似性数据相似性缺陷识别 缺陷识别 缺陷识别域名通用化域名通用化机器视觉 机器视觉 机器视觉磨削磨削的方法 磨削表面检查检查检查表面检查检查表面质地 表面质地综合数据 综合数据纹理建模 纹理建模

更多相关视频

Automatic Laser-based Geometry Capture for Finite Element Analysis of Weld Beads
07:58

Automatic Laser-based Geometry Capture for Finite Element Analysis of Weld Beads

Published on: July 25, 2025

757
High-throughput Imaging and Analysis Workflow for Evaluating Skin Cell Phenotypes and Proliferation States in Tissue Samples
11:24

High-throughput Imaging and Analysis Workflow for Evaluating Skin Cell Phenotypes and Proliferation States in Tissue Samples

Published on: October 31, 2025

654

相关实验视频

Last Updated: Jan 15, 2026

Author Spotlight: A Machine-Vision Approach to Transmission Electron Microscopy Workflows, Results Analysis and Data Management
10:23

Author Spotlight: A Machine-Vision Approach to Transmission Electron Microscopy Workflows, Results Analysis and Data Management

Published on: June 23, 2023

3.5K
Automatic Laser-based Geometry Capture for Finite Element Analysis of Weld Beads
07:58

Automatic Laser-based Geometry Capture for Finite Element Analysis of Weld Beads

Published on: July 25, 2025

757
High-throughput Imaging and Analysis Workflow for Evaluating Skin Cell Phenotypes and Proliferation States in Tissue Samples
11:24

High-throughput Imaging and Analysis Workflow for Evaluating Skin Cell Phenotypes and Proliferation States in Tissue Samples

Published on: October 31, 2025

654

科学领域:

  • 计算机视觉 计算机视觉
  • 机器学习 机器学习
  • 材料科学 材料科学 材料科学

背景情况:

  • 视觉检查的机器学习 (ML) 需要广泛,多样化的训练数据,这往往是不切实际的.
  • 现有的合成数据生成方法,主要是生成模型,面临着诸如数据短缺,幻觉和边缘病例的有限处理等挑战.
  • 在生成具有结构化模式的物理现实的纹理以进行强大的ML模型训练方面存在差距.

研究的目的:

  • 介绍一种新的合成数据生成管道,用于创建具有结构化模式的物理现实的纹理的大数据集.
  • 为了能够精确地控制纹理参数,生成各种观察到和未观察到的纹理实例.
  • 评估生成的合成数据集的质量及其预测下游ML性能的潜力.

主要方法:

  • 基于具有可解释参数的程序纹理建模的合成数据生成管道的开发.
  • 数据集的生成,包括喷砂,并行削和螺旋削纹理.
  • 通过真实和合成域之间的图像相似度量来评估数据集质量,超出最终模型性能.

主要成果:

  • 该管道成功生成了大量物理现实的纹理数据集,具有复杂的结构图案.
  • 程序方法允许精确控制纹理参数,确保边缘案例的多样性和覆盖范围.
  • 图像相似度指标表明了可以预测下游检测性能的趋势,提供了一种新的评估方法.

结论:

  • 拟议的管道为生成高质量的合成纹理数据提供了强大的解决方案,解决了当前方法的局限性.
  • 可解释的参数控制为为机器学习创建多样化和现实的数据集提供了独特的优势.
  • 这些发现表明,图像相似度指标可以作为下游任务性能的有价值预测指标,指导未来的合成数据开发.