Search research articles

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 Journal Methods Collections JoVE Encyclopedia of Experiments 存档

教育

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual 教师资源中心教师网站

使用条款与条件

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

Alternol inhibits GAPDH activity and disrupts glycolytic flux preferentially in cancer cells.

Frontiers in pharmacology·2026

Same author

ASSR-Net: Anisotropic Structure-Aware and Spectrally Recalibrated Network for Hyperspectral Image Fusion.

IEEE transactions on neural networks and learning systems·2026

Same author

Alternol-Induced Oxidative Modification of SQSTM1/p62 Is Associated with Nrf2 Signaling and Autophagy-Related Responses in Prostate Cancer Cells.

Antioxidants (Basel, Switzerland)·2026

Same author

Oxygen Minimum Zone and Organic Carbon Structure Benthic Prokaryotic Communities and Metabolism in Warm Deep-Red Sea Sediments.

Microorganisms·2026

Same author

Deep Error-Aware Iterative Optimization Network for Broadband Mosaiced Hyperspectral Imaging.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026

Same author

Directional Concerted Proton-Electron Transfer in COFs for Efficient Photocatalytic H<sub>2</sub>O<sub>2</sub> Production.

Advanced materials (Deerfield Beach, Fla.)·2026

Search research articles

相关实验视频

Updated: Sep 15, 2025

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

从视觉学习基础模型用于跨域遥感图像分割

Wang Liu, Puhong Duan, Zhuojun Xie

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society

|July 16, 2025

概括

本研究介绍了LFMDA,这是一种用于遥感图像细分的新方法. 它通过使用视觉基础模型 (VFMs) 来改进域调整,以创建更准确,更适应的细分模型.

科学领域:

遥感遥感遥感遥感
计算机视觉计算机视觉
机器学习机器学习

背景情况:

跨域图像细分对于遥感至关重要.
现有的平均教师模型在领域差异和类重叠方面扎,限制了绩效.
在遥感图像细分方面,需要强大的域调整方法.

研究的目的:

引入LFMDA,一种用于遥感中跨域语义细分的新领域适应方法.
利用视觉基础模型 (VFMs) 来提高不同领域的细分性能.
通过提高特征不变性和可区分性来解决当前方法的局限性.

主要方法:

提出一个原型的对比知识蒸 (PCD) 损失,以从一个域普用VFM教师的知识蒸.
实施本地区域同质化 (LRH) 战略,使用分段任何模型 (SAM) 来生成高质量的伪标签.
开发一个强大的域适应框架 (LFMDA) 进行远程传感图像细分.

主要成果:

在跨领域遥感图像细分方面,LFMDA显著优于现有的方法.
该方法通过产生域不变和类别歧视性特征来实现最先进的 (SOTA) 性能.
PCD损失和LRH策略有效地提高了跨域的细分精度和稳定性.

更多相关视频

From Voxels to Knowledge: A Practical Guide to the Segmentation of Complex Electron Microscopy 3D-Data

From Voxels to Knowledge: A Practical Guide to the Segmentation of Complex Electron Microscopy 3D-Data

Published on: August 13, 2014

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

相关实验视频

Last Updated: Sep 15, 2025

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

From Voxels to Knowledge: A Practical Guide to the Segmentation of Complex Electron Microscopy 3D-Data

From Voxels to Knowledge: A Practical Guide to the Segmentation of Complex Electron Microscopy 3D-Data

Published on: August 13, 2014

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

结论:

LFMDA代表了域适应式遥感图像细分的重大进步.
利用新的蒸和伪标签策略的VFM提供了一种强大的方法来克服领域转移的挑战.
拟议的方法为遥感领域的跨领域细分任务设定了新的基准.