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

Predictive coding explains asymmetric connectivity in the brain: A neural network study.

PLoS computational biology·2026
Same author

Regulation of the growth-to-ripening transition in tomato fruits by energy charge involving SlATP-PRT and SlAPRT1.

Molecular horticulture·2026
Same author

Histone acetyltransferase Kat2a regulates glioma stem cell differentiation via Hsp90aa1-dependent oxidative phosphorylation.

Biochemical pharmacology·2026
Same author

PMTNet: A Part-Centric Missing-Aware Temporal Network for Cat Behavior Recognition in Unconstrained Videos.

Animals : an open access journal from MDPI·2026
Same author

Gut microbiota and postoperative delirium: mechanistic integration of the "gut-liver-anesthesia axis" in the context of liver disease and perioperative intervention strategies.

Frontiers in medicine·2026
Same author

Design and implementation of a thermoelectric cooler-based temperature control system for the resonator tube wall of an optically pumped terahertz gas laser.

The Review of scientific instruments·2026

相关实验视频

Updated: Jan 11, 2026

Failure Analysis of Batteries Using Synchrotron-based Hard X-ray Microtomography
08:11

Failure Analysis of Batteries Using Synchrotron-based Hard X-ray Microtomography

Published on: August 26, 2015

9.3K

一个实时细分网络用于电池表面缺陷检测和检测缺陷.

Jiaxing Xie1,2,3,4, Peiwen Wu5, Jiasi Chen5

  • 1College of Electronic Engineering (College of AI), South China Agricultural University, Guangzhou, 510642, China. xjx1998@scau.edu.cn.

Scientific reports
|November 19, 2025
PubMed
概括
此摘要是机器生成的。

本研究介绍了用于电池表面缺陷检测的双重注意力金字塔细分网络 (DAPSeg). DAPSeg有效地识别微小的缺陷,并平衡高精度与实时性能.

关键词:
注意力机制注意力机制编码器 解码器功能增强 功能增强 功能增强多个尺度的特征聚变聚变.稳定的扩散 稳定的扩散表面缺陷检测检测的表面缺陷检测.

更多相关视频

Nanoscale Characterization of Liquid-Solid Interfaces by Coupling Cryo-Focused Ion Beam Milling with Scanning Electron Microscopy and Spectroscopy
11:03

Nanoscale Characterization of Liquid-Solid Interfaces by Coupling Cryo-Focused Ion Beam Milling with Scanning Electron Microscopy and Spectroscopy

Published on: July 14, 2022

3.9K
Focused Ion Beam Fabrication of LiPON-based Solid-state Lithium-ion Nanobatteries for In Situ Testing
10:58

Focused Ion Beam Fabrication of LiPON-based Solid-state Lithium-ion Nanobatteries for In Situ Testing

Published on: March 7, 2018

10.6K

相关实验视频

Last Updated: Jan 11, 2026

Failure Analysis of Batteries Using Synchrotron-based Hard X-ray Microtomography
08:11

Failure Analysis of Batteries Using Synchrotron-based Hard X-ray Microtomography

Published on: August 26, 2015

9.3K
Nanoscale Characterization of Liquid-Solid Interfaces by Coupling Cryo-Focused Ion Beam Milling with Scanning Electron Microscopy and Spectroscopy
11:03

Nanoscale Characterization of Liquid-Solid Interfaces by Coupling Cryo-Focused Ion Beam Milling with Scanning Electron Microscopy and Spectroscopy

Published on: July 14, 2022

3.9K
Focused Ion Beam Fabrication of LiPON-based Solid-state Lithium-ion Nanobatteries for In Situ Testing
10:58

Focused Ion Beam Fabrication of LiPON-based Solid-state Lithium-ion Nanobatteries for In Situ Testing

Published on: March 7, 2018

10.6K

科学领域:

  • 材料科学 材料科学 材料科学
  • 计算机视觉 计算机视觉
  • 人工智能的人工智能

背景情况:

  • 电池表面缺陷检测对于工业应用至关重要.
  • 现有的方法在不同的缺陷尺度,特别是微小的缺陷上扎.
  • 高精度和实时性能对于缺陷检测至关重要.

研究的目的:

  • 提出一个新的网络,双注意力金字塔细分网络 (DAPSeg),用于精确和实时的电池表面缺陷细分.
  • 为了应对尺度变化的挑战和同时高精度和速度的需求.

主要方法:

  • 开发了DAPSeg,采用了选择性内核模块 (SKM) 进行自适应的多尺度特征提取.
  • 采用了一种轻量级的细分头,配备蓝图可分离层 (BSL) 和双重注意力特征融合模块 (DAFFM).
  • 在LB-SD数据集上使用扩散模型进行数据增强,以减轻过度拟合.

主要成果:

  • DAPSeg获得了79.57% (LB-SD),83.53% (MT) 和89.10% (MSD) 的mIoU分数.
  • 该模型的处理速度为74.09 FPS.
  • 在平衡准确性和推断速度方面超越了最先进的模型.

结论:

  • DAPSeg为电池表面缺陷检测提供了一个强大的解决方案.
  • 该网络实现了高精度和实时处理能力.
  • 在不同的数据集中,DAPSeg表现出强大的泛化性能.