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

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

Long-term cardiovascular risk in survivors of hematologic malignancies: a meta-analysis.

BMC cancer·2026
Same author

Remote medical system driven by medical big models: Dynamic defense model for network security threats.

PloS one·2026
Same author

Predicting poor health-related quality of life among Chinese adolescents using explainable machine learning: the role of school adjustment, family context, and lifestyle factors.

Health and quality of life outcomes·2026
Same author

Direct Observation of Interfacial Exchange Coupling in a Magnetic Tunnel Junction through Spin-Polarized Quasiparticle Interference.

Nano letters·2026
Same author

Adverse childhood experiences and mental health symptoms among aviation shift workers in China: A national cohort study.

Journal of affective disorders·2026
Same author

Isoquinoline alkaloids in <i>Coptis chinensis</i> to treat Alzheimer's disease through promoting growth of <i>Bifidobacterium breve</i> inhibiting abnormal autophagy using a novel AI high-content intelligent imaging system.

Chinese herbal medicines·2026

相关实验视频

Updated: Jun 16, 2025

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

11.1K

一种基于图像重建和异常检测的无缺陷数据的通用缺陷检查方法.

Minjie Du1, Siqi Gu1, Zihan Qin1

  • 1School of Cyber Science and Engineering, Southeast University, Nanjing, 211189, Jiangsu, China.

Neural networks : the official journal of the International Neural Network Society
|June 13, 2025
PubMed
概括

本研究引入了使用层次图像重建的新缺陷检测框架. 它实现了高精度和速度,不需要特定的缺陷数据,改善了工业检查.

关键词:
异常检测检测异常检测层次结构重建的重建.跳过连接的连接.

更多相关视频

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns
13:44

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns

Published on: August 30, 2013

42.8K
Comprehensive Characterization of Extended Defects in Semiconductor Materials by a Scanning Electron Microscope
11:14

Comprehensive Characterization of Extended Defects in Semiconductor Materials by a Scanning Electron Microscope

Published on: May 28, 2016

13.8K

相关实验视频

Last Updated: Jun 16, 2025

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

11.1K
Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns
13:44

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns

Published on: August 30, 2013

42.8K
Comprehensive Characterization of Extended Defects in Semiconductor Materials by a Scanning Electron Microscope
11:14

Comprehensive Characterization of Extended Defects in Semiconductor Materials by a Scanning Electron Microscope

Published on: May 28, 2016

13.8K

科学领域:

  • 计算机视觉 计算机视觉
  • 机器学习 机器学习
  • 工业自动化 工业自动化

背景情况:

  • 传统的监督检测缺陷的方法需要广泛的缺陷特定的培训数据.
  • 这种限制阻碍了对各种产品类型和现实世界的工业场景的概括.

研究的目的:

  • 开发一个新的,无监督的工业缺陷检查框架.
  • 为了使准确和有效的异常检测,而不需要事先了解缺陷类型.

主要方法:

  • 该框架使用层次图像重建模块.
  • 一个自我注意力机制被纳入了增强特征学习.
  • 异常检测是基于重建错误进行的.

主要成果:

  • 在MVTec AD 2D数据集上获得了97.83%的平均精度.
  • 在准确度方面,超越U-Net11.1%和U-Transformer12.9%.
  • 达到了24.1 FPS的模型推断速度,比U-变压器模型快48.1%.

结论:

  • 拟议的框架为实时工业缺陷检查提供了一个强大的解决方案.
  • 在检测准确度和推断速度方面都表现出卓越的性能.
  • 突出了无监督学习对于多功能异常检测的潜力.