PAD: Phase-Amplitude Decoupling Fusion for Multimodal Land Cover Classification
文献类型：期刊论文
| 作者 | Zheng, Huiling2,3,4; Zhong, Xian1,8; Liu, Bin7; Xiao, Yi6; Wen, Bihan5; Li, Xiaofeng2 |
| 刊名 | IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
 |
| 出版日期 | 2025 |
| 卷号 | 63页码:14 |
| 关键词 | Frequency-domain analysis Synthetic aperture radar Feature extraction Land surface Remote sensing Urban areas Transfer learning Spectral analysis Representation learning Decoding Land cover classification (LCC) multimodal segmentation remote sensing RGB-synthetic aperture radar (SAR) multimodality |
| ISSN号 | 0196-2892 |
| DOI | 10.1109/TGRS.2025.3621902 |
| 通讯作者 | Zhong, Xian(zhongx@whut.edu.cn) ; Liu, Bin(bliu@shou.edu.cn) |
| 英文摘要 | The fusion of synthetic aperture radar (SAR) and RGB imagery for land cover classification (LCC) remains challenging due to modality heterogeneity and underexploited spectral complementarity. Existing approaches often fail to decouple shared structural features from modality-complementary radiometric attributes, resulting in feature conflicts and information loss. To address this, we propose phase-amplitude decoupling (PAD), a frequency-aware framework that separates phase (modality-shared) and amplitude (modality-complementary) components in the Fourier domain. This design reinforces shared structures while preserving complementary characteristics, thereby enhancing fusion quality. Unlike previous methods that overlook the distinct physical properties encoded in frequency spectra, PAD explicitly introduces amplitude-phase decoupling for multimodal fusion. Specifically, PAD comprises two key components: 1) phase spectrum correction (PSC), which aligns cross-modal phase features via convolution-guided scaling to improve geometric consistency; and 2) amplitude spectrum fusion (ASF), which dynamically integrates high-frequency (HF) and low-frequency (LF) patterns using frequency-adaptive multilayer perceptrons (MLP), effectively exploiting SAR's morphological sensitivity and RGB's spectral richness. Extensive experiments on WHU-OPT-SAR and DDHR-SK demonstrate state-of-the-art performance. This work establishes a new paradigm for physics-aware multimodal fusion in remote sensing. The code will be available at https://github.com/RanFeng2/PAD |
| WOS关键词 | NETWORK |
| 资助项目 | National Natural Science Foundation of China[62271361] ; National Natural Science Foundation of China[42006159] ; Hubei Provincial Key Research and Development Program[2024BAB039] |
| WOS研究方向 | Geochemistry & Geophysics ; Engineering ; Remote Sensing ; Imaging Science & Photographic Technology |
| 语种 | 英语 |
| WOS记录号 | WOS:001606670800020 |
| 出版者 | IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC |
| 源URL | [http://ir.qdio.ac.cn/handle/337002/203743]  |
| 专题 | 海洋研究所_海洋环流与波动重点实验室 |
| 通讯作者 | Zhong, Xian; Liu, Bin |
| 作者单位 | 1.Wuhan Univ Technol, Sch Comp Sci & Artificial Intelligence, Hubei Key Lab Transportat Internet Things, Wuhan 430070, Peoples R China 2.Chinese Acad Sci, Key Lab Ocean Circulat & Waves, Inst Oceanol, Qingdao 266071, Peoples R China 3.Wuhan Univ Technol, Sch Comp Sci & Artificial Intelligence, Wuhan 430070, Peoples R China 4.Wuhan Univ Technol, Sanya Sci & Educ Innovat Pk, Sanya 572025, Peoples R China 5.Nanyang Technol Univ, Sch Elect & Elect Engn, Rapid Rich Object Search Lab, Singapore 639798, Singapore 6.Zhengzhou Univ, Sch Comp & Artificial Intelligence, Zhengzhou 450001, Peoples R China 7.Shanghai Ocean Univ, Coll Oceanog & Ecol Sci, Shanghai 201306, Peoples R China 8.Wuhan Univ Technol, State Key Lab Maritime Technol & Safety, Wuhan 430063, Peoples R China |
| 推荐引用方式 GB/T 7714 | Zheng, Huiling,Zhong, Xian,Liu, Bin,et al. PAD: Phase-Amplitude Decoupling Fusion for Multimodal Land Cover Classification[J]. IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING,2025,63:14. |
| APA | Zheng, Huiling,Zhong, Xian,Liu, Bin,Xiao, Yi,Wen, Bihan,&Li, Xiaofeng.(2025).PAD: Phase-Amplitude Decoupling Fusion for Multimodal Land Cover Classification.IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING,63,14. |
| MLA | Zheng, Huiling,et al."PAD: Phase-Amplitude Decoupling Fusion for Multimodal Land Cover Classification".IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING 63(2025):14. |
入库方式： OAI收割
来源：海洋研究所
浏览0
下载0
收藏0
其他版本
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。