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

Confocal Fluorescence Microscopy01:16

Confocal Fluorescence Microscopy

13.3K
Confocal microscopy is an advanced microscopic technique. The prime advantage of the confocal microscope over other microscopy techniques is its ability to block the out-of-focus light from the illuminated samples using pinholes. It is widely used with fluorescence optics to obtain high-resolution, sharp contrast images. Unlike optical microscopes, confocal microscopes use a focused beam of light laser to scan the entire sample surface at different z-planes. These microscopes are, therefore,...
13.3K

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

Runtime-Robust Edge Inference System with Masking-Based Partial Update on Dynamic Reconfigurable FPGA.

Sensors (Basel, Switzerland)·2025
Same author

RFDNet: Robust Frequency-Based Denoising Network for 3D Ultrasound Vascular Imaging Using a Row-Column Addressed Array.

Ultrasonic imaging·2025
Same author

DAUS-Net: Toward Ultrasound Scanner-Agnostic Domain Generalized Robust and Accurate Segmentation.

Ultrasonic imaging·2025
Same author

Ultrasound localization microscopy lite (ULM lite): ultrasound localization microscopy with resource-efficient signal processing scheme.

Ultrasonics·2025
Same author

Contrast agent-free 3D ultrasound deep-depth vascular imaging with a 2D row column addressed Array: In vivo human clinical feasibility study.

Ultrasonics·2025
Same author

Unsupervised speckle noise reduction technique for clinical ultrasound imaging.

Ultrasonography (Seoul, Korea)·2024

相关实验视频

Updated: Jul 11, 2025

Combining Eye-tracking Data with an Analysis of Video Content from Free-viewing a Video of a Walk in an Urban Park Environment
08:25

Combining Eye-tracking Data with an Analysis of Video Content from Free-viewing a Video of a Walk in an Urban Park Environment

Published on: May 7, 2019

9.0K

在嵌入式处理器中使用语义兴趣区域生成与轻量级LiDAR集群进行高效的对象检测.

Dongkyu Jung1, Taewon Chong2,3, Daejin Park1

  • 1School of Electronic and Electrical Engineering, Kyungpook National University, Daegu 41566, Republic of Korea.

Sensors (Basel, Switzerland)
|November 14, 2023
PubMed
概括

本研究引入了一种使用轻量级系统进行3D物体检测的新方法. 它通过利用3D数据特征来增强2D卷积神经网络 (CNN) 对象检测,提高准确性和减少处理时间.

关键词:
这是一个激光雷达传感器.卷积神经网络 (CNN) 是一个神经网络.对象检测检测对象检测对象检测一个点云,一个点云.语义检测 语义检测 语义检测

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

564
Automatic Identification of Dendritic Branches and their Orientation
06:08

Automatic Identification of Dendritic Branches and their Orientation

Published on: September 17, 2021

2.0K

相关实验视频

Last Updated: Jul 11, 2025

Combining Eye-tracking Data with an Analysis of Video Content from Free-viewing a Video of a Walk in an Urban Park Environment
08:25

Combining Eye-tracking Data with an Analysis of Video Content from Free-viewing a Video of a Walk in an Urban Park Environment

Published on: May 7, 2019

9.0K
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

564
Automatic Identification of Dendritic Branches and their Orientation
06:08

Automatic Identification of Dendritic Branches and their Orientation

Published on: September 17, 2021

2.0K

科学领域:

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

背景情况:

  • 卷积神经网络 (CNN) 在3D数据中被广泛用于对象检测.
  • 基于3D数据的算法提供了稳定性,但需要大量的计算资源,限制了它们在嵌入式系统中的使用.
  • 现有的2D CNN与照明变化作斗争,并需要复杂的处理3D数据.

研究的目的:

  • 在轻量级嵌入式系统上开发一个计算高效的3D对象检测方法.
  • 通过整合3D数据特征来提高2D CNN对象检测的准确性.
  • 在不同的环境条件下克服传统的3D和2D物体检测方法的局限性.

主要方法:

  • 预处理LiDAR点云数据并使用聚类来进行对象分离.
  • 使用机器学习分类器从3D数据中提取语义检测的物理特征.
  • 从检测到的3D对象生成2D图像区域,以执行2D CNN对象检测,绕过边界框跟踪.

主要成果:

  • 拟议的模型在嵌入式板上使用YOLO v5实现了81.84%的准确性,超过了典型模型的1.92%.
  • 在非最佳照明条件下表现出卓越的性能,在高亮度 (+40%) 中精度为47.41%,在低亮度 (-40%) 中精度为54.12%.
  • 在具有挑战性的照明中,与一般模型相比,实现了显著的精度改进 (分别为8.97%和13.58%).

结论:

  • 拟议的方法有效地利用3D数据特征来提高2D CNN对象检测在资源有限的环境中的准确性.
  • 这种方法为纯粹基于3D数据的检测算法提供了一个更轻便的系统替代方案.
  • 该技术在不同的照明条件下显示出强度,并减少了执行时间,使其适用于现实世界的嵌入式应用程序.