Recently, networks consider spectral-spatial information in multiscale inputs less, even though there are some networks that consider this factor, however these networks cannot guarantee to get optimal features, which are extracted from each scale input. Furthermore, these networks do not consider the complementary and related information among different scale features. To address these issues, a multiscale deep middle-level feature fusion network (MMFN) is proposed in this paper for hyperspectral classification. In MMFN, the network fully fuses the strong complementary and related information among different scale features to extract more discriminative features. The training of network contains two stages: the first stage obtains the optimal models corresponding to different scale inputs and extracts the middle-level features under the corresponding scale model. It can guarantee the multiscale middle-level features are optimal. The second stage fuses the optimal multiscale middle-level features in the convolutional layer, and the subsequent residual blocks can learn the complementary and related information among different scale middle-level features. Moreover, the idea of identity mapping in residual learning can help the network obtain a higher accuracy when the network is deeper. The effectiveness of our method is proved on four HSI data sets and the experimental results show that our method outperforms the other state-of-the-art methods especially with small training samples.

Deep learning has attracted extensive attention in the field of hyperspectral images (HSIs) classification. However, supervised deep learning methods heavily rely on a large amount of label information. To address this problem, in this paper, we propose a two-stage deep domain adaptation method for hyperspectral image classification, which can minimize the data shift between two domains and learn a more discriminative deep embedding space with very few labeled target samples. A deep embedding space is first learned by minimizing the distance between the source domain and the target domain based on Maximum Mean Discrepancy (MMD) criterion. The Spatial–Spectral Siamese Network is then exploited to reduce the data shift and learn a more discriminative deep embedding space by minimizing the distance between samples from different domains but the same class label and maximizes the distance between samples from different domains and class labels based on pairwise loss. For the classification task, the softmax layer is replaced with a linear support vector machine, in which learning minimizes a margin-based loss instead of the cross-entropy loss. The experimental results on two sets of hyperspectral remote sensing images show that the proposed method can outperform several state-of-the-art methods.

Hyperspectral image classification holds significant application value in precision agriculture, environmental monitoring, and other fields. However, the high cost of large-scale labeling limits the widespread application of deep learning methods. Utilizing a large amount of labeled hyperspectral image data for auxiliary training of cross-domain few-shot learning methods is an effective means to address this issue. Yet, in practical applications, spectral and spatial features vary across different scenarios, posing significant challenges to the generalizability and accuracy of cross-domain learning models. To tackle this problem, this paper proposes a Multi-Level Prototype Alignment (MLPA) method for cross-domain few-shot classification, which adjusts feature representations by implementing multi-level feature alignment strategies at various hierarchical levels of the feature extraction network. This approach achieves fine-grained alignment of the source and target domain feature distributions, effectively reducing domain shift and enhancing the model’s generalization capability on target domain data. Furthermore, by employing class prototype-based domain adversarial training, the method aligns the prototypes of the source and target domains without simply aligning the entire feature space, thus avoiding overlap in the feature distribution of different classes within the domain and mitigating negative transfer. The MLPA method effectively enhances the generalizability and discriminative power of features in the target domain, thereby improving the performance of cross-domain hyperspectral image classification. Experimental results demonstrate that MLPA outperforms other cross-domain few-shot hyperspectral image classification methods. Our source code is available at https://github.com/hejinrong/MLPA.