期刊
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
卷 59, 期 12, 页码 10425-10437出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TGRS.2021.3049282
关键词
Feature extraction; Hyperspectral imaging; Data mining; Training; Task analysis; Correlation; Image color analysis; 3-D fully convolution network (3D-FCN); attention; convolutional spatial propagation network (CSPN); deep learning (DL); hyperspectral image (HSI) classification
类别
资金
- National Natural Science Foundation of China [61871460]
- Shaanxi Provincial Key Research and Development Program [2020KW-003]
- Innovation Foundation for Doctor Dissertation of Northwestern Polytechnical University [CX202011]
This study proposes a novel neural network model FCSPN for HSI classification, overcoming the limitations of patch-level classification and convolutions with local receptive fields in traditional CNN models. Experimental results show that the proposed model achieves state-of-the-art performance in HSI classification.
In recent years, deep convolutional neural networks (CNNs) have demonstrated impressive ability to represent hyperspectral images (HSIs) and achieved encouraging results in HSI classification. However, the existing CNN-based models operate at the patch level, in which a pixel is separately classified into classes using a patch of images around it. This patch-level classification will lead to a large number of repeated calculations, and it is hard to identify the appropriate patch size that is beneficial to classification accuracy. In addition, the conventional CNN models operate convolutions with local receptive fields, which cause the failure of contextual spatial information modeling. To overcome these aforementioned limitations, we propose a novel end-to-end, pixel-to-pixel, fully convolutional spatial propagation network (FCSPN) for HSI classification. Our FCSPN consists of a 3-D fully convolution network (3D-FCN) and a convolutional spatial propagation network (CSPN). Specifically, the 3D-FCN is first introduced for reliable preliminary classification, in which a novel dual separable residual (DSR) unit is proposed to effectively capture spectral and spatial information simultaneously with fewer parameters. Moreover, the channel-wise attention mechanism is adapted in the 3D-FCN to grasp the most informative channels from redundant channel information. Finally, the CSPN is introduced to capture the spatial correlations of HSIs via learning a local linear spatial propagation, which allows maintaining the HSI spatial consistency and further refining the classification results. Experimental results on three HSI benchmark data sets demonstrate that the proposed FCSPN achieves state-of-the-art performance on HSI classification.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据