The convolutional neural network (CNN) has been gradually applied to the hyperspectral images (HSIs) classification, but the lack of training samples caused by the difficulty of HSIs sample marking and ignoring of correlation between spatial and spectral information seriously restrict the HSIs classification accuracy. In an attempt to solve these problems, this paper proposes a dual-branch extraction and classification method under limited samples of hyperspectral images based on deep learning (DBECM). At first, a sample augmentation method based on local and global constraints in this model is designed to augment the limited training samples and balance the number of different class samples. Then spatial-spectral features are simultaneously extracted by the dual-branch spatial-spectral feature extraction method, which improves the utilization of HSIs data information. Finally, the extracted spatial-spectral feature fusion and classification are integrated into a unified network. The experimental results of two typical datasets show that the DBECM proposed in this paper has certain competitive advantages in classification accuracy compared with other public HSIs classification methods, especially in the Indian pines dataset. The parameters of the overall accuracy (OA), average accuracy (AA), and Kappa of the method proposed in this paper are at least 4.7%, 5.7%, and 5% higher than the existing methods.

Hyperspectral images (HSIs) are usually composed of hundreds of bands, which are highly correlated and redundant, leading to dimension disaster and high complexity of classification. In this paper, we propose an end-to-end dense spatial–spectral attention network (DSSAN) for hyperspectral image band selection to reduce the complexity of classification while ensuring the accuracy. In this network, an embeddable spatial–spectral attention module is designed, which can adaptively select the spectral bands from the raw input data. Moreover, this module is a plug-and-play complementary component and embedded in a dense convolutional network (DenseNet) for end-to-end training. The experimental results on two classic hyperspectral data sets demonstrate that the proposed method is superior to several mainstream band selection methods in classification accuracy and the selected band subset has lower redundancy, which can meet the application requirements.