Unleashing the full potential of hyperspectral imaging: Decoupled image and frequency-domain spatial-spectral framework
文献类型:期刊论文
作者 | He, Shuang1,2,3; Tian, Jia1,4; Hao, Lina5; Zhang, Sen1,2,3; Tian, Qingjiu1,2 |
刊名 | EXPERT SYSTEMS WITH APPLICATIONS |
出版日期 | 2024-06-01 |
卷号 | 243页码:20 |
ISSN号 | 0957-4174 |
关键词 | Frequency domain Feature extraction Cross-domain fusion Hyperspectral image classification |
DOI | 10.1016/j.eswa.2023.122870 |
通讯作者 | Tian, Jia(tianjia@buaa.edu.cn) |
英文摘要 | Hyperspectral image classification (HSIC) is a rapidly developing field that utilizes deep learning methods. However, the reliance on convolutional neural networks (CNNs) for spectral-spatial feature extraction presents certain limitations. Specifically, the use of the fixed-position convolutional kernels in CNNs hinders their ability to capture fully the spectral information around the spatial central pixel, thereby overlooking critical differences between features. To address this issue, a decoupled image- and frequency-domain spectral-spatial framework for HSIC was developed in this study. This method incorporates image- and frequency-domain-based multiscale learnable convolutional attention to refine the differentiating features of the different feature distributions. Additionally, a novel frequency-domain information enhancement module was designed to extract the structural shape and texture details under the semantic constraints of the frequency phase, complementing the image domain to improve the extracted feature maps. Furthermore, a simple and efficient hierarchical feature representation module was introduced to extract both local and global information effectively from the fused features. The experimental results obtained using three open datasets and a practical hyperspectral image of the Gaofen-5 satellite demonstrate that the proposed method outperforms other state-of-the-art HSIC methods. |
WOS关键词 | CLASSIFICATION ; CLASSIFIERS ; NETWORKS |
资助项目 | Open Fund of State Key Laboratory of Urban and Regional Ecology[SKLURE2023-2-6] ; National Natural Science Foundation of China[42101321] ; Open Fund of State Key Laboratory of Remote Sensing Science[OFSLRSS202119] |
WOS研究方向 | Computer Science ; Engineering ; Operations Research & Management Science |
语种 | 英语 |
出版者 | PERGAMON-ELSEVIER SCIENCE LTD |
WOS记录号 | WOS:001138306400001 |
资助机构 | Open Fund of State Key Laboratory of Urban and Regional Ecology ; National Natural Science Foundation of China ; Open Fund of State Key Laboratory of Remote Sensing Science |
源URL | [http://ir.igsnrr.ac.cn/handle/311030/202039] |
专题 | 中国科学院地理科学与资源研究所 |
通讯作者 | Tian, Jia |
作者单位 | 1.Nanjing Univ, Int Inst Earth Syst Sci, Nanjing 210023, Peoples R China 2.Nanjing Univ, Jiangsu Prov Key Lab Geog Informat Sci & Technol, Nanjing 210023, Peoples R China 3.Nanjing Univ, Collaborat Innovat Ctr South Sea Studies, Nanjing 210093, Peoples R China 4.Beihang Univ, Sch Instrumentat & Optoelect Engn, Beijing 100191, Peoples R China 5.Chinese Acad Sci, Inst Geog Sci & Nat Resources Res, State Key Lab Resources & Environm Informat Syst, Beijing 100101, Peoples R China |
推荐引用方式 GB/T 7714 | He, Shuang,Tian, Jia,Hao, Lina,et al. Unleashing the full potential of hyperspectral imaging: Decoupled image and frequency-domain spatial-spectral framework[J]. EXPERT SYSTEMS WITH APPLICATIONS,2024,243:20. |
APA | He, Shuang,Tian, Jia,Hao, Lina,Zhang, Sen,&Tian, Qingjiu.(2024).Unleashing the full potential of hyperspectral imaging: Decoupled image and frequency-domain spatial-spectral framework.EXPERT SYSTEMS WITH APPLICATIONS,243,20. |
MLA | He, Shuang,et al."Unleashing the full potential of hyperspectral imaging: Decoupled image and frequency-domain spatial-spectral framework".EXPERT SYSTEMS WITH APPLICATIONS 243(2024):20. |
入库方式: OAI收割
来源:地理科学与资源研究所
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。