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

Quantifying the Waddington landscape and biological paths for development and differentiation.

Proceedings of the National Academy of Sciences of the United States of America·2011
Same author

[Evaluation on the effects of an education program regarding the sedentary behavior among school-aged children using Transtheoretical Model].

Zhonghua liu xing bing xue za zhi = Zhonghua liuxingbingxue zazhi·2011
Same author

[The application of near-infrared diffuse reflection spectra based on the principle of linear additive in tobacco redrying formula].

Guang pu xue yu guang pu fen xi = Guang pu·2011
Same author

Genetic study of families affected with aggressive periodontitis.

Periodontology 2000·2011
Same author

[Experimental study on Qi deficiency and blood stasis induced by muti-factor stimulation in rats].

Zhongguo Zhong yao za zhi = Zhongguo zhongyao zazhi = China journal of Chinese materia medica·2011
Same author

Improved visible light photocatalytic activity of sphere-like BiOBr hollow and porous structures synthesized via a reactable ionic liquid.

Dalton transactions (Cambridge, England : 2003)·2011
Same journal

DARUMA: a gateway to fast and easy prediction of intrinsically disordered regions.

PeerJ. Computer science·2026
Same journal

Alzheimer's disease detection using a quantum deep neural network with Haralick feature extraction and simulated annealing optimization.

PeerJ. Computer science·2026
Same journal

Network anomaly detection using Deep Autoencoder and parallel Artificial Bee Colony algorithm-trained neural network.

PeerJ. Computer science·2026
Same journal

An anomaly detection model for multivariate time series with anomaly perception.

PeerJ. Computer science·2026
Same journal

Retraction: A wormhole attack detection method for tactical wireless sensor networks.

PeerJ. Computer science·2026
Same journal

Evaluation of mental disorder with prioritization of its type by utilizing the bipolar complex fuzzy decision-making approach based on Schweizer-Sklar prioritized aggregation operators.

PeerJ. Computer science·2026
查看所有相关文章

相关实验视频

Updated: Jun 5, 2025

Author Spotlight: Efficient Image Recognition Using Directional Gradient Histogram Technique and Support Vector Machines
08:27

Author Spotlight: Efficient Image Recognition Using Directional Gradient Histogram Technique and Support Vector Machines

Published on: January 5, 2024

979

基于特征金字塔的自适应原型少数镜头图像分类方法.

Linshan Shen1, Xiang Feng1, Li Xu1

  • 1College of Computer Science And Technology, Harbin Engineering University, Harbin, HeiLongJiang, China.

PeerJ. Computer science
|December 9, 2024
PubMed
概括
此摘要是机器生成的。

本研究介绍了基于特征金字塔 (APFP) 的自适应原型少数镜头图像分类方法,以改善有限数据的机器学习. APFP 增强了特征提取,并动态计算类原型,在基准数据集上实现高精度.

关键词:
有几次射击学习学习.图像的分类图像的分类.计量学学习的学习方法网络原型网络原型

更多相关视频

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

468
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

相关实验视频

Last Updated: Jun 5, 2025

Author Spotlight: Efficient Image Recognition Using Directional Gradient Histogram Technique and Support Vector Machines
08:27

Author Spotlight: Efficient Image Recognition Using Directional Gradient Histogram Technique and Support Vector Machines

Published on: January 5, 2024

979
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

468
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

科学领域:

  • 计算机科学 计算机科学
  • 人工智能的人工智能
  • 机器学习 机器学习

背景情况:

  • 短暂学习 (FSL) 旨在用最小的数据对新课程进行分类,模仿人类学习.
  • 对FSL来说,度量学习至关重要,它依赖于有效的特征提取和原型计算.
  • 现有的原型方法在少数拍摄场景中与多样化的样本分布作斗争.

研究的目的:

  • 引入一种基于特征金字塔 (APFP) 的自适应原型短拍图像分类方法.
  • 为FSL增强特征表示和解决传统原型计算中的局限性.

主要方法:

  • 开发了FResNet,这是一种基于ResNet的新型特征提取器,具有用于详细信息保留的特征金字塔.
  • 提出了自适应原型 (AP) 方法,该方法使用样本查询相似性动态计算类原型.
  • 将FResNet和AP集成到APFP框架中,用于少数镜头图像的分类.

主要成果:

  • 在MiniImageNet上,APFP在5路一拍中达到67.98%的精度,在5路5拍中达到85.32%的精度.
  • 在CUB数据集中,APFP在五向一射中达到84.02%,在五向五射中达到94.44%的准确性.
  • 与传统方法相比,表现出优异的性能,验证了APFP的有效性.

结论:

  • 拟议的APFP方法有效地解决了少数拍摄图像分类的挑战.
  • 适应原型计算和增强的功能提取是提高FSL性能的关键.
  • APFP显示了对现实世界应用的巨大潜力,这些应用需要从有限的示例中快速学习.