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

Deconvolution01:20

Deconvolution

190
Deconvolution, also known as inverse filtering, is the process of extracting the impulse response from known input and output signals. This technique is vital in scenarios where the system's characteristics are unknown, and they must be inferred from the observable signals.
Deconvolution involves several mathematical techniques to derive the impulse response. One common approach is polynomial division. In this method, the input and output sequences are treated as coefficients of...
190

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

Rapid Filtration Allows Ultrafast Measurement of Antibody and Nanoparticle Capture in Porous Matrices: Implications for Lateral Flow Immunoassays and Membrane Chromatography.

ACS applied nano materials·2026
Same author

Correction: A novel technology for home monitoring of lupus nephritis that tracks the pathogenic urine biomarker ALCAM.

Frontiers in immunology·2025
Same author

Between the lines: investigating health beliefs and emotional expressions in online mental health communities.

Frontiers in psychology·2025
Same author

Lateral flow assay-based detection of nuclear fusion oncoprotein: implications for screening of acute promyelocytic leukemia.

Sensors & diagnostics·2025
Same author

Improving Object Detection in High-Altitude Infrared Thermal Images Using Magnitude-Based Pruning and Non-Maximum Suppression.

Journal of imaging·2025
Same author

Accelerated Development of a COVID-19 Lateral Flow Test in an Academic Setting: Lessons Learned.

Accounts of chemical research·2024

相关实验视频

Updated: Jul 21, 2025

Quantitative Visualization and Detection of Skin Cancer Using Dynamic Thermal Imaging
06:08

Quantitative Visualization and Detection of Skin Cancer Using Dynamic Thermal Imaging

Published on: May 5, 2011

16.9K

使用自动编码器改善工业热图像中的视觉缺陷检测和定位.

Sasha Behrouzi1, Marcel Dix2, Fatemeh Karampanah1

  • 1Applied Data Science and Analytics, SRH University, 69123 Heidelberg, Germany.

Journal of imaging
|July 28, 2023
PubMed
概括
此摘要是机器生成的。

结合异常分数可以改善热图中的缺陷检测. 这种方法通过使用平均平方误差 (MSE),结构相似度指数 (SSIM) 和内核密度估计 (KDE) 来提高分类准确性,特别是在受污染的培训数据中.

关键词:
检测异常检测异常检测自动编码器自动编码器深度学习是一种深度学习.工业形象 工业形象 工业形象新奇发现检测新奇的检测.热图片 热图片 热图片

更多相关视频

Subsurface Defect Localization by Structured Heating Using Laser Projected Photothermal Thermography
11:34

Subsurface Defect Localization by Structured Heating Using Laser Projected Photothermal Thermography

Published on: May 15, 2017

11.2K
Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches
09:47

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches

Published on: December 15, 2023

1.1K

相关实验视频

Last Updated: Jul 21, 2025

Quantitative Visualization and Detection of Skin Cancer Using Dynamic Thermal Imaging
06:08

Quantitative Visualization and Detection of Skin Cancer Using Dynamic Thermal Imaging

Published on: May 5, 2011

16.9K
Subsurface Defect Localization by Structured Heating Using Laser Projected Photothermal Thermography
11:34

Subsurface Defect Localization by Structured Heating Using Laser Projected Photothermal Thermography

Published on: May 15, 2017

11.2K
Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches
09:47

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches

Published on: December 15, 2023

1.1K

科学领域:

  • 工业工程 工业工程 工业工程
  • 计算机视觉 计算机视觉
  • 机器学习 机器学习

背景情况:

  • 在热图像中可靠的异常检测对于工业缺陷检测至关重要.
  • 挑战来自图像序列的顺过渡,导致受污染的训练数据.
  • 基于自动编码器的方法与轻微受污染的数据集和值确定作斗争.

研究的目的:

  • 为了提高热图像数据集中的异常检测可靠性.
  • 解决工业缺陷检测中受污染的培训数据所带来的挑战.
  • 加强对健康和有缺陷的工业产品进行分类的门确定.

主要方法:

  • 使用了在健康的热图像上训练的自编码器模型,具有平均平方误差 (MSE) 和结构相似度指数 (SSIM) 损失函数.
  • 应用了三个异常分数来进行分类:MSE,SSIM和内核密度估计 (KDE).
  • 引入了MSE+和SSIM+方法,包括基于SSIM的定量异常定位参数.

主要成果:

  • 在热图数据集上实现的平均精度:MSE (95.33%),SSIM (88.37%) 和KDE (92.81%).
  • 证明结合异常分数可以提高分类准确性.
  • 展示了KDE的性能改进,特别是在受污染的健康训练数据方面.

结论:

  • 结合异常分数可以有效地分离健康和缺陷数据,克服值确定方面的挑战.
  • 拟议的方法提高了工业热成像中的异常检测可靠性.
  • 当训练数据被污染时,核密度估计 (KDE) 证明是有益的.