通过近递归的光谱注意力和跨层特征融合,通过高光谱图像否定
Yanhua Xiao1, Huayan Zhou1, Wenfeng Li1
1School of Information Engineering, Chenzhou Vocation Technical College, Chenzhou 424500, China.
Sensors (Basel, Switzerland)
|November 27, 2025
概括
通过近递归光谱注意网络 (QRSAN) 改进了高光谱图像消噪. 这种新的框架有效地消除噪音,同时保留关键的空间和光谱细节,以提高图像质量.
科学领域:
- 遥感 遥感 遥感 遥感
- 图像处理 图像处理
- 计算机视觉 计算机视觉
背景情况:
- 超光谱图像 (HSI) 拥有丰富的空间光谱信息.
- 噪音显著降低了HSI质量和应用可靠性.
研究的目的:
- 引入一个新的端到端拒绝超光谱图像的框架.
- 在降噪过程中保持高质量的空间和光谱信息.
主要方法:
- 提出准递归光谱注意网络 (QRSAN).
- 使用近递归注意力单位 (QRAU) 进行空间光谱依赖模型.
- 在空间特征上使用二维卷积,在光谱表示上使用频率聚合.
- 在不对称的编码器-解码器架构中实现多头光谱注意力和自适应交叉层跳过连接.
主要成果:
- 在合成和真实HSI数据集上,QRSAN展示了优越的无噪性能.
- 与现有方法相比,实现了视觉质量和客观评估指标的改进.
- 在消噪后保持高空间频谱保真度.
结论:
- QRSAN提供有效的高光谱图像消噪.
- 拟议的框架显示出强大的概括能力.
- QRSAN成功地平衡了噪声抑制与维护基本图像特征.
相关概念视频
Deconvolution
532
Deconvolution, also known as inverse filtering, is the process of extracting the impulse response from known input and output signals. This technique is vital in scenarios where the system's characteristics are unknown, and they must be inferred from the observable signals.
Deconvolution involves several mathematical techniques to derive the impulse response. One common approach is polynomial division. In this method, the input and output sequences are treated as coefficients of...
Deconvolution involves several mathematical techniques to derive the impulse response. One common approach is polynomial division. In this method, the input and output sequences are treated as coefficients of...
532
Attenuated Total Reflectance (ATR) Infrared Spectroscopy: Overview
1.1K
Attenuated total reflectance (ATR) infrared spectroscopy is a powerful analytical technique used to study the composition of materials. It is widely employed in chemistry, materials science, forensic science, and other fields where sample characterization is required. ATR has several advantages over traditional transmission IR spectroscopy, including the requirement of little to no sample preparation and the ability to analyze a wide range of samples.
The ATR process begins by directing a beam...
The ATR process begins by directing a beam...
1.1K
Upsampling
569
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...
569
Aliasing
530
Accurate signal sampling and reconstruction are crucial in various signal-processing applications. A time-domain signal's spectrum can be revealed using its Fourier transform. When this signal is sampled at a specific frequency, it results in multiple scaled replicas of the original spectrum in the frequency domain. The spacing of these replicas is determined by the sampling frequency.
If the sampling frequency is below the Nyquist rate, these replicas overlap, preventing the original...
If the sampling frequency is below the Nyquist rate, these replicas overlap, preventing the original...
530
Downsampling
578
When considering a sampled sequence with zero values between sampling instants, one can replace it by taking every N-th value of the sequence. At these integer multiples of N, the original and sampled sequences coincide. This process, known as decimation, involves extracting every N-th sample from a sequence, thereby creating a more efficient sequence.
The Fourier transform of the decimated sequence reveals a combination of scaled and shifted versions of the original spectrum. This...
The Fourier transform of the decimated sequence reveals a combination of scaled and shifted versions of the original spectrum. This...
578

