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

SD-IDD: Selective Distillation for Incremental Defect Detection.

Sensors (Basel, Switzerland)·2026
Same author

Hand-Object Pose Estimation Based on Anchor Regression from a Single Egocentric Depth Image.

Sensors (Basel, Switzerland)·2025
Same author

An enhanced microstate clustering algorithm based on canopy, K-means, and genetic simulated annealing.

Biomedical physics & engineering express·2025
Same author

Filter bank temporally delayed CCA for uncalibrated SSVEP-BCI.

Medical & biological engineering & computing·2024
Same author

MBA-DNet: A mask block attention-based foreign matter detection network for tobacco packages.

The Review of scientific instruments·2024
Same author

Effects of GLP-1 receptor agonists on asprosin levels in normal weight or overweight/obesity patients with type 2 diabetes mellitus.

Medicine·2022
Same journal

Compressed multi-scale entropy and its application in mechanical fault diagnosis.

The Review of scientific instruments·2026
Same journal

Bidirectional drive and multi-resolution adjustment across frequency bands in inertial impact piezoelectric motors via multimodal resonant vibration.

The Review of scientific instruments·2026
Same journal

A magnetic field sensor based on flaky Terfenol-D material and dual fiber grating.

The Review of scientific instruments·2026
Same journal

A novel E-field eight-way cavity combiner for high-power S-band applications.

The Review of scientific instruments·2026
Same journal

Constant radius blade spring suspended bench for vibration isolation.

The Review of scientific instruments·2026
Same journal

Qualification of infrared optical fibers and emitters for a spectrometer for in situ planetary exploration: Results from the TRIS (TRansmission and Illumination System) project.

The Review of scientific instruments·2026
查看所有相关文章

相关实验视频

Updated: Jun 23, 2025

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

517

一个阶段的多级高效网络用于水下目标检测.

Huaqiang Zhang1, Chenggang Dai1, Chengjun Chen1

  • 1School of Mechanical and Automotive Engineering, Qingdao University of Technology, Qingdao 266520, Shandong, China.

The Review of scientific instruments
|June 18, 2024
PubMed
概括
此摘要是机器生成的。

这项研究引入了改进的YOLOv5s方法用于水下目标检测,提高了小型和密集物体的精度. 这种新的方法显著提高了复杂的水下环境中的检测精度.

更多相关视频

Long-term Behavioral Tracking of Freely Swimming Weakly Electric Fish
10:56

Long-term Behavioral Tracking of Freely Swimming Weakly Electric Fish

Published on: March 6, 2014

12.5K
Early Detection of Cyanobacterial Blooms and Associated Cyanotoxins using Fast Detection Strategy
07:13

Early Detection of Cyanobacterial Blooms and Associated Cyanotoxins using Fast Detection Strategy

Published on: February 25, 2021

3.8K

相关实验视频

Last Updated: Jun 23, 2025

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

517
Long-term Behavioral Tracking of Freely Swimming Weakly Electric Fish
10:56

Long-term Behavioral Tracking of Freely Swimming Weakly Electric Fish

Published on: March 6, 2014

12.5K
Early Detection of Cyanobacterial Blooms and Associated Cyanotoxins using Fast Detection Strategy
07:13

Early Detection of Cyanobacterial Blooms and Associated Cyanotoxins using Fast Detection Strategy

Published on: February 25, 2021

3.8K

科学领域:

  • 计算机视觉 计算机视觉
  • 机器人技术 机器人技术 机器人技术
  • 机器学习 机器学习

背景情况:

  • 由于环境的复杂性,水下目标检测具有挑战性,导致小或密集的目标的精度较低.
  • 现有的方法在复杂的水下场景中难以准确,需要检测算法的进步.

研究的目的:

  • 提出基于YOLOv5s的新型水下目标检测方法,以提高精度和稳定性.
  • 改进特征提取,表示和融合,以便更准确地识别水下物体.

主要方法:

  • 采用了高效的特征提取网络和具有可变形卷积的新型注意力机制.
  • 在YOLOv5s部引入了自适应空间融合操作,以实现有效的多层特征融合.
  • 使用自适应融合特征金字塔网络来整合全球语义信息并减少特征语义差距.

主要成果:

  • 拟议的方法在2020年中国水下机器人专业比赛数据集上获得了86.97%的mAP50,比YOLOv5s有3.07%的改进.
  • 该方法在PASCAL VOC2007数据集上显示了76.0%的检测精度,超过了几种现有方法.
  • 实验结果证实了水下目标检测的提高精度和稳定性.

结论:

  • 基于YOLOv5s的新方法显著提高了水下目标检测的准确性,特别是对于小型和密集的物体.
  • 集成先进的特征提取,注意力机制和自适应融合可以提高复杂环境中的检测性能.
  • 拟议的方法为水下目标检测提供了一个强大的解决方案,在基准数据集上表现优于标准方法.