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

Synthesis of Mechanically Isomeric Rotaxanes from Cross-Chain Bridging Cryptands and a Secondary Ammonium Ion.

Organic letters·2026
Same author

R1 Prognostic Significance of T-Wave Amplitude Variability for Adverse Cardiovascular Outcomes: A Systematic Review and Meta-Analysis.

Journal of arrhythmia·2026
Same author

Durvalumab After Chemoradiotherapy for Locally Advanced NSCLC: A Real-World Analysis Using a Nationwide Claims Database in Japan.

Clinical lung cancer·2026
Same author

Radiotherapy utilization in the last 30 days before death among patients with malignant neoplasms in Japan: a claims database study.

International journal of clinical oncology·2026
Same author

Influence of Age on the Effectiveness of Lee Silverman Voice Treatment<sup>®</sup> BIG in Patients with Parkinson's Disease: A Retrospective Exploratory Observational Study.

Geriatrics (Basel, Switzerland)·2026
Same author

Comparison of three arm-positioning techniques for minimizing motion artifacts in breast magnetic resonance imaging: a prospective volunteer study.

Breast cancer (Tokyo, Japan)·2026

相关实验视频

Updated: Jul 11, 2025

Measuring the Structure, Composition, and Change of Underwater Environments with Large-area Imaging
09:19

Measuring the Structure, Composition, and Change of Underwater Environments with Large-area Imaging

Published on: April 18, 2025

527

L-DIG:一种基于GAN的方法,用于在雪地驾驶条件下的LiDAR点云处理.

Yuxiao Zhang1, Ming Ding1,2, Hanting Yang1

  • 1Graduate School of Informatics, Nagoya University, Furo-cho, Chikusa-Ward, Nagoya 464-8601, Japan.

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

激光雷达深度图像GAN (L-DIG) 有效地消除自动驾驶点云的雪声,并合成雪效应. 这种新型的生成对抗性网络模型可以在恶劣的天气条件下提高感知系统的可靠性.

关键词:
循环GANAN是一个循环.激光雷达 (LiDAR) 是一个点云处理器.产生的雪效应的产生.降雪噪声去除 降雪噪声去除

更多相关视频

Author Spotlight: Innovative Ice Cream Melting Behavior Analysis Through a Computer Vision System
08:02

Author Spotlight: Innovative Ice Cream Melting Behavior Analysis Through a Computer Vision System

Published on: October 4, 2024

2.4K
Laser-Induced Fluorescence Emission L.I.F.E. as Novel Non-Invasive Tool for In-Situ Measurements of Biomarkers in Cryospheric Habitats
13:38

Laser-Induced Fluorescence Emission L.I.F.E. as Novel Non-Invasive Tool for In-Situ Measurements of Biomarkers in Cryospheric Habitats

Published on: October 26, 2019

8.0K

相关实验视频

Last Updated: Jul 11, 2025

Measuring the Structure, Composition, and Change of Underwater Environments with Large-area Imaging
09:19

Measuring the Structure, Composition, and Change of Underwater Environments with Large-area Imaging

Published on: April 18, 2025

527
Author Spotlight: Innovative Ice Cream Melting Behavior Analysis Through a Computer Vision System
08:02

Author Spotlight: Innovative Ice Cream Melting Behavior Analysis Through a Computer Vision System

Published on: October 4, 2024

2.4K
Laser-Induced Fluorescence Emission L.I.F.E. as Novel Non-Invasive Tool for In-Situ Measurements of Biomarkers in Cryospheric Habitats
13:38

Laser-Induced Fluorescence Emission L.I.F.E. as Novel Non-Invasive Tool for In-Situ Measurements of Biomarkers in Cryospheric Habitats

Published on: October 26, 2019

8.0K

科学领域:

  • 计算机视觉 计算机视觉
  • 自主驾驶系统 自主驾驶系统
  • 传感器数据处理 传感器数据处理

背景情况:

  • 降雪在驾驶场景中显著降低了LiDAR点云数据质量.
  • 现有的除雪方法,主要是异常波器,其有效性有限.
  • 自动驾驶系统需要强大的感知能力,即使在雪地条件下.

研究的目的:

  • 引入一种新的生成对抗网络 (GAN) 模型,L-DIG (LiDAR深度图像GAN),用于LiDAR点云除雪和雪合成.
  • 为了提高自动驾驶系统在雪环境中的感知能力.
  • 开发一种能够在LiDAR数据中消除和产生雪效应的模型.

主要方法:

  • 从未配对的数据集中利用点云的深度图像表示用于训练.
  • 实现深度图像的定制损失功能,以保持尺寸和结构的一致性.
  • 开发了一种双分辨器架构,其中包括一个像素注意力分辨器,用于在自我车辆附近捕获雪,以及一个下采样卷积分辨器,用于雪.
  • 采用3D聚类算法对各种雪状况进行自适应评估.

主要成果:

  • L-DIG模型在捕捉LiDAR点云中的雪和物体特征方面表现出卓越的性能.
  • 实验结果显示,对损坏的数据有明显的除雪效应.
  • 该模型成功地将现实的雪效应合成到清晰的LiDAR数据上.
  • 双歧视者方法有效地处理了各种各样的雪状况,从分散点到密集集群.

结论:

  • 拟议的L-DIG模型为减轻自动驾驶LiDAR数据中的雪干扰提供了有效的解决方案.
  • 能够同时去除和合成雪的能力为研究和开发提供了多功能工具.
  • 该模型的性能表明,在冬季条件下,自动驾驶汽车在强大的感知方面取得了显著的进步.