相关实验视频

Updated: Jan 7, 2026

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

HSFAN:一个双分支混合规模特征聚合网络,用于远程传感超分辨率图像

Jiawei Yang¹, Hongliang Ren¹, Mengjie Zeng¹

¹College of Information Science and Engineering, Fujian Key Laboratory of Light Propagation and Transformation, Huaqiao University, Xiamen 361021, China.

Entropy (Basel, Switzerland)

|December 24, 2025

概括

一个新的双分支混合规模特征聚合网络 (HSFAN) 通过改善细节恢复和降低复杂性来提高远程传感图像超分辨率. 这种方法平衡了模型效率和高质量的重建,以获得更好的结果.

科学领域:

计算机视觉计算机视觉
遥感遥感遥感遥感
图像处理图像处理

背景情况:

现有的遥感图像超分辨率网络在高率区域,细节恢复和计算复杂性方面扎.
功能利用不足和过度的模型参数限制了当前的方法.

研究的目的:

提出一种新的双分支混合规模特征聚合网络 (HSFAN),用于远程传感图像超分辨率.
为了实现模型复杂性和重建质量之间的最佳平衡.
为了提高特征利用在高的地区,并改善细节恢复.

主要方法:

HSFAN采用双分支架构:一个主分支,用于全球结构建模的多尺度并行大型卷积内核 (MSPLCK) 模块,以及用于特征优先级的增强并行注意 (EPA) 模块.
一个辅助分支使用一个多尺度的大内核注意力 (MSLA) 模块,具有深度可分离的卷积,以减少计算开销和捕获本地高频信息.
在整个主分支的特征空间中保持一致性约束.

主要成果:

拟议的HSFAN在UC Merced数据集上实现了27.91dB的PSNR和0.7616的SSIM,用于X4超分辨率.
该算法超过了当前大多数主流超分辨率算法.
与现有方法相比,HSFAN保持了较低的计算成本和模型参数数量.

结论:

在HSFAN有效地解决现有的遥感图像超分辨率网络的局限性.
拟议的网络为改进远程传感图像超分辨率重建提供了一条新的技术路线.
在重建质量和计算效率之间,HSFAN提供了一个有前途的平衡.

关键词:

信息是信息的.多个尺度的多个尺度.远程传感图像远程传感图像超级分辨率的超级分辨率

更多相关视频

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

Super-resolution Imaging of the Bacterial Division Machinery

Super-resolution Imaging of the Bacterial Division Machinery

Published on: January 21, 2013

相关概念视频

Super-resolution Fluorescence Microscopy

Super-resolution Fluorescence Microscopy

Super-resolution fluorescence microscopy (SRFM) provides a better resolution than conventional fluorescence microscopy by reducing the point spread function (PSF). PSF is the light intensity distribution from a point that causes it to appear blurred. Due to PSF, each fluorescing point appears bigger than its actual size, and it is the PSF interference of nearby fluorophores that causes the blurred image. Various approaches to achieving higher resolution through SRFM have recently been...

Upsampling

Upsampling

Managing signal sampling rates is essential in digital signal processing to maintain signal integrity. A decimated signal, characterized by a reduced frequency range due to its lower sampling rate, can be upsampled by inserting zeros between each sample. This upsampling process expands the original spectrum and introduces repeated spectral replicas at intervals dictated by the new Nyquist frequency. To refine this zero-inserted sequence, it is passed through a lowpass filter with a cutoff...