Spatial-Spectral Attention-Enhanced Multi-Level Wavelet-Informed Network for Hyperspectral Image Denoising.
Saved in:
| Title: | Spatial-Spectral Attention-Enhanced Multi-Level Wavelet-Informed Network for Hyperspectral Image Denoising. |
|---|---|
| Authors: | Wang, Rui1 (AUTHOR), Liu, Hong1,2 (AUTHOR), Hu, Wen-Shuai1 (AUTHOR), Huang, Shaoguang2 (AUTHOR) huangshaoguang@cug.edu.cn, Wang, Jiuping1 (AUTHOR) |
| Source: | Remote Sensing. Jun2026, Vol. 18 Issue 12, p2053. 22p. |
| Subjects: | Image denoising, Wavelet transforms, Frequency-domain analysis, Precision farming, Ecological assessment, Artificial neural networks |
| Abstract: | Highlights: What are the main findings? The proposed spatial-spectral attention-enhanced multi-level wavelet-informed network (SAMWNet) integrates a multi-level discrete wavelet transform (DWT) with spatial-spectral attention for HSI stripe noise removal. Extensive experiments on synthetic datasets (Washington DC Mall and Pavia Center) and real datasets (Indian Pines and Pavia University) demonstrate that SAMWNet outperforms representative methods, including LRMR, Restormer, and SST. Ablation studies validate the contributions of the dual-branch feature extraction module and the low-frequency hybrid enhancement (LFHE) and high-frequency enhancement (HFME) submodules. What are the implications of the main findings? By jointly exploiting frequency-domain decomposition and spatial-spectral attention, SAMWNet provides a feasible paradigm for direction-aware and frequency-adaptive HSI denoising. The preserved spatial structures and spectral consistency enable SAMWNet to serve as a practical preprocessing tool for downstream HSI applications, including crop growth monitoring in precision agriculture, ecological factor inversion in environmental assessment, and lithology identification in mineral exploration. Hyperspectral image (HSI) stripe noise removal is essential for downstream interpretation tasks. However, most existing methods exhibit incomplete joint modeling of spatial structures and inter-band spectral correlations, lack direction-aware modeling for stripe noise, and lack differentiated processing of high- and low-frequency components. To tackle these limitations, we propose a spatial-spectral attention-enhanced multi-level wavelet-informed network (SAMWNet). Its dual-branch module extracts spatial and spatial-spectral features from each band and its adjacent bands. Afterward, a discrete wavelet-informed progressive denoising (MDWPD) module conducts multi-level Haar wavelet decomposition and progressive reconstruction. Within this module, the low-frequency hybrid enhancement (LFHE) module preserves low-frequency spectral structures, while the high-frequency enhancement (HFME) module suppresses directional stripe artifacts in high-frequency subbands. We further adopt a composite loss function to balance pixel fidelity, noise estimation, structural similarity, and spectral consistency. Experimental results on simulated and real-world HSIs demonstrate that SAMWNet achieves competitive or superior performance compared with several representative HSI denoising methods. [ABSTRACT FROM AUTHOR] |
| Copyright of Remote Sensing is the property of MDPI and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.) | |
| Database: | Engineering Source |
|
Full text is not displayed to guests.
Login for full access.
|
|
| Abstract: | Highlights: What are the main findings? The proposed spatial-spectral attention-enhanced multi-level wavelet-informed network (SAMWNet) integrates a multi-level discrete wavelet transform (DWT) with spatial-spectral attention for HSI stripe noise removal. Extensive experiments on synthetic datasets (Washington DC Mall and Pavia Center) and real datasets (Indian Pines and Pavia University) demonstrate that SAMWNet outperforms representative methods, including LRMR, Restormer, and SST. Ablation studies validate the contributions of the dual-branch feature extraction module and the low-frequency hybrid enhancement (LFHE) and high-frequency enhancement (HFME) submodules. What are the implications of the main findings? By jointly exploiting frequency-domain decomposition and spatial-spectral attention, SAMWNet provides a feasible paradigm for direction-aware and frequency-adaptive HSI denoising. The preserved spatial structures and spectral consistency enable SAMWNet to serve as a practical preprocessing tool for downstream HSI applications, including crop growth monitoring in precision agriculture, ecological factor inversion in environmental assessment, and lithology identification in mineral exploration. Hyperspectral image (HSI) stripe noise removal is essential for downstream interpretation tasks. However, most existing methods exhibit incomplete joint modeling of spatial structures and inter-band spectral correlations, lack direction-aware modeling for stripe noise, and lack differentiated processing of high- and low-frequency components. To tackle these limitations, we propose a spatial-spectral attention-enhanced multi-level wavelet-informed network (SAMWNet). Its dual-branch module extracts spatial and spatial-spectral features from each band and its adjacent bands. Afterward, a discrete wavelet-informed progressive denoising (MDWPD) module conducts multi-level Haar wavelet decomposition and progressive reconstruction. Within this module, the low-frequency hybrid enhancement (LFHE) module preserves low-frequency spectral structures, while the high-frequency enhancement (HFME) module suppresses directional stripe artifacts in high-frequency subbands. We further adopt a composite loss function to balance pixel fidelity, noise estimation, structural similarity, and spectral consistency. Experimental results on simulated and real-world HSIs demonstrate that SAMWNet achieves competitive or superior performance compared with several representative HSI denoising methods. [ABSTRACT FROM AUTHOR] |
|---|---|
| ISSN: | 20724292 |
| DOI: | 10.3390/rs18122053 |
