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

What Are Outliers?01:12

What Are Outliers?

3.6K
Outliers are observed data points that are far from the least squares line. They have unusual values and need to be examined carefully. Though an outlier may result from erroneous data, at other times, it may hold valuable information about the population under study and should be included in the data. Hence, it is crucial to examine what causes a data point to be an outlier.
The z score is used to find outliers or unusual values. It should be noted that any values beyond -2 and +2 are...
3.6K

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

Short-term exposure to CO and NO<sub>2</sub> and Kawasaki disease onset in a cold region of China: a distributed lag non-linear time-series study.

Frontiers in public health·2026
Same author

Measured corneal astigmatism vs. predicted refractive astigmatism in determining astigmatism correction in cataract patients with low astigmatism.

Graefe's archive for clinical and experimental ophthalmology = Albrecht von Graefes Archiv fur klinische und experimentelle Ophthalmologie·2026
Same author

Pressure-induced softening of locust bean gum hydrogels: A counterintuitive alternative to freeze-thaw stiffening.

Proceedings of the National Academy of Sciences of the United States of America·2026
Same author

Can switching between different types of myopia control spectacles enhance effectiveness? Findings from a real-world study.

The British journal of ophthalmology·2026
Same author

Causal Mediation Analysis for Effect Heterogeneity.

Observational studies·2026
Same author

Effect of pupillary dilation on ocular biometry and intraocular lens power calculation: prospective cohort study.

Journal of cataract and refractive surgery·2026

相关实验视频

Updated: May 24, 2025

Applying Hyperspectral Reflectance Imaging to Investigate the Palettes and the Techniques of Painters
07:05

Applying Hyperspectral Reflectance Imaging to Investigate the Palettes and the Techniques of Painters

Published on: June 18, 2021

2.3K

超光谱异常检测与自我监督的异常之前的检测.

Yidan Liu1, Kai Jiang2, Weiying Xie2

  • 1College of Electrical and Information Engineering, Hunan University, Changsha, 410082, China.

Neural networks : the official journal of the International Neural Network Society
|February 28, 2025
PubMed
概括

这项研究引入了一种新的自我监督异常前 (SAP) 超谱异常检测 (HAD). 通过学习异常特征,SAP增强了超谱图像中的目标识别,比传统方法提高了准确性.

关键词:
异常检测检测异常检测深度先验的先验.超光谱图像 (HSI) 是一种超光谱图像.低级别的代表 (LRR)自主监督学习 (SSL)

更多相关视频

Excitation-Scanning Hyperspectral Imaging Microscopy to Efficiently Discriminate Fluorescence Signals
00:07

Excitation-Scanning Hyperspectral Imaging Microscopy to Efficiently Discriminate Fluorescence Signals

Published on: August 22, 2019

7.9K
Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

451

相关实验视频

Last Updated: May 24, 2025

Applying Hyperspectral Reflectance Imaging to Investigate the Palettes and the Techniques of Painters
07:05

Applying Hyperspectral Reflectance Imaging to Investigate the Palettes and the Techniques of Painters

Published on: June 18, 2021

2.3K
Excitation-Scanning Hyperspectral Imaging Microscopy to Efficiently Discriminate Fluorescence Signals
00:07

Excitation-Scanning Hyperspectral Imaging Microscopy to Efficiently Discriminate Fluorescence Signals

Published on: August 22, 2019

7.9K
Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

451

科学领域:

  • 遥感 遥感 遥感 遥感
  • 计算机视觉 计算机视觉
  • 信号处理 信号处理

背景情况:

  • 超光谱异常检测 (HAD) 对于地球观测和军事应用至关重要.
  • 现有的方法通常依赖于手工制作的稀疏先验的低级表示 (LRR),可以忽略空间结构和手工稀疏性设置.
  • 这限制了超光谱图像中异常检测的准确性和解释性.

研究的目的:

  • 开发一种更准确,更易于解释的超谱异常检测方法.
  • 为了克服基于LRR的HAD中手工制作的先驱的局限性.
  • 引入一种自我监督的方法来学习异常特征.

主要方法:

  • 一个自我监督的网络,自我监督的异常前 (SAP),被建议重新定义在LRR模型中的异常优化标准.
  • 一个新的借口任务涉及分类,以区分原始和伪异常的高光谱图像 (HSI) 用于学习.
  • 采用双净化策略,使用丰富的背景词典来完善背景表示.

主要成果:

  • 与现有的先进的HAD技术相比,所提出的SAP方法显示出更高的性能.
  • 对各种高频谱数据集的实验验验证了SAP方法的有效性和可解释性.
  • 该方法成功地识别和定位目标,没有事先的信息,即使在复杂的背景.

结论:

  • 自主监督异常前 (SAP) 在超谱异常检测方面取得了重大进展.
  • 这种方法提供了一个更强大的,数据驱动的替代手工制作的先验.
  • 在超光谱成像中,SAP提高了异常检测的准确性和解释性.