Hyperspectral image (HSI) and multispectral image (MSI) fusion aims to generate high-resolution HSI by leveraging the high spectral fidelity of HSI and the fine spatial details of MSI. However, most existing methods rely on static fusion strategies that assume global consistency in modality contributions, ignoring the inherent regional variability in real-world remote sensing scenes. To address this limitation, we propose an adaptive expert learning framework (AELF) that dynamically models the modal dominance of different regions and adaptively adjusts fusion strategies accordingly. A core component of AELF is the modality-guided complementary module (MGCM), which establishes bidirectional cross-attention pathways between HSI and MSI. It enables each modality to adaptively discover complementary cues across multiple scales while suppressing irrelevant information, providing enhanced feature representation for subsequent fine-grained fusion. Building upon this, we designed the attribute-aware mixture of fusion experts (AMoFE) module, which decomposes the fused features into spectral, spatial, and edge subspaces. Each component is modeled by a specialized expert network, with a soft routing mechanism dynamically adjusting expert contributions based on contextual cues. Extensive experiments on benchmark datasets and a real-world dataset demonstrate that AELF achieves state-of-the-art performance in terms of spectral fidelity and spatial sharpness. Furthermore, our results confirm that the improved data quality brought by the proposed method effectively enhances the overall performance of downstream tasks. The code will be available at https://github.com/Hewq77/AELF.

Different tree species have different carbon storage and growth rates. Therefore, accurate segmentation and identification of individual trees can provide more detailed carbon storage data, which is the basis for accurately estimating forest carbon storage. However, individual tree segmentation and recognition in dense forest areas face challenges such as crown overlap, complex terrain, and species diversity. To address these challenges and improve recognition accuracy, this paper proposes a fuzzy boundary-aware network (FBAN) for hyperspectral individual tree segmentation and recognition in dense forests. The proposed FBAN inclues a boundary-aware module (BAM) that explores channel boundaries between trees and non-trees, spatial boundaries of trees, and spectral boundaries between different trees by intergrating channel attention, spaital attention, and spectral attention. This enhances the separability of individual trees, especially those of the same species that are contiguous in dense forest areas. Additionally, an adaptive crown-aware module (ACAM) is constructed to adapt diverse-size crown features by coupling Transformer layers with dialted convolution layers. Experimental results on different hyperspectral datasets show that the proposed FBAN network outperforms existing methods in dense forest areas and different tree canopy areas, e.g., the AP on the SZU-South dataset is 4.5 points higher than that of Mask2former. It not only improves the accuracy of individual tree segmentation and recognition but also exhibits high generalization and robustness.

Hyperspectral image super-resolution aims to reconstruct high-quality spatial-spectral cubes from RGB images. However, the limited spectral coverage of RGB inputs hinders the simultaneous modeling of spatial structures and spectral relationships. To address this limitation, we propose a differential low-rank adaptive spatial-spectral transformer (DiLAST). Initially, a differential operator is employed to enhance RGB features in a bottom-up manner, explicitly amplifying subtle inter-channel differences. The enhanced features are then fed into a U-shaped backbone, which integrates three complementary modules for joint spatial-spectral modeling. Specifically, a center spatial-spectral attention (CSSA) module employs cross-attention mechanisms to capture local-to-global dependencies across both spatial and spectral domains; an adaptive cross-scale fusion (ACF) module utilizes learnable gating weights to establish dynamic interaction pathways between shallow high-frequency details and deep semantic representations; and a low-rank spectral calibration (LRSC) module exploits low-rank matrix priors to reveal low-dimensional manifold structures among spectral bands, thereby enhancing spectral consistency. By leveraging the synergistic effects of spatial non-locality, global spectral correlation, and low-rank properties, the proposed DiLAST achieves PSNR improvements of 33.82 dB, 36.03 dB, and 37.01 dB on benchmark datasets. Moreover, the accuracy and practical applicability of the reconstructed spectra have been effectively validated in remote sensing scenarios and object tracking tasks. The code is accessible at https://github.com/renqi1998/DiLAST.

With its strong capability in modeling long-range dependencies, the Transformer achieves competitive performance in hyperspectral image (HSI) and multispectral image (MSI) fusion. However, existing Transformer-based methods face the trade-off between receptive field size and computational efficiency when dealing with spatially non-local features. Furthermore, the Transformer captures deep spectral relationships by modeling pairwise channel interactions. This global interaction may overlook features that contribute little to the overall context but are critical locally, thus affecting the accurate understanding of HSI content. To overcome these challenges, we propose a novel local-global collaborative network with Transformers (LGCT) specifically designed to achieve high-quality HSI reconstruction. The proposed LGCT includes two inverse feature streams to establish multiscale deep representations of the HSI and MSI features. The feature streams comprise collaborative Transformer blocks (CTBs) explicitly designed for the spectral and spatial domains. By combining global and local processing mechanisms, the proposed CTBs can efficiently emphasize potential crucial features that Transformer ignores when capturing deep spectral and spatial relationships, thus enabling efficient modeling of the spectral and spatial domains from details to the whole. Furthermore, to enhance the reusability of multiscale enhanced features from the spectral and spatial domains, a hierarchical and symmetric strategy is adopted to progressively fuse them to generate high-quality images. The results on both simulated and real datasets demonstrate the superior performance of the proposed method in terms of quantitative metrics and visual quality. The code will be released athttps://github.com/Hewq77/LGCT.

Deep neural networks play a significant role in hyperspectral image (HSI) processing, yet they can be easily fooled when trained with adversarial samples (generated by adding tiny perturbations to clean samples). These perturbations are invisible to the human eye, but can easily lead to misclassification by the deep learning model. Recent research on defense against adversarial samples in HSI classification has improved the robustness of deep networks by exploiting global contextual information. However, available methods do not distinguish between different classes of contextual information, which makes the global context unreliable and increases the success rate of attacks. To solve this problem, we propose a robust context-aware network able to defend against adversarial samples in HSI classification. The proposed model generates a global contextual representation by aggregating the features learned via dilated convolution, and then explicitly models intraclass and interclass contextual information by constructing a class context-aware learning module (including affinity loss) to further refine the global context. The module helps pixels obtain more reliable long-range dependencies and improves the overall robustness of the model against adversarial attacks. Experiments on several benchmark HSI datasets demonstrate that the proposed method is more robust and exhibits better generalization than other advanced techniques.

The Sunway TaihuLight, equipped with 10 million cores, is currently the world's third fastest supercomputer. SpMV is one of core algorithms in many high-performance computing applications. This paper implements a fine-grained design for generic parallel SpMV based on the special Sunway architecture and finds three main performance limitations, i.e., storage limitation, load imbalance, and huge overhead of irregular memory accesses. To address these problems, this paper introduces a customized and accelerative framework for SpMV (CASpMV) on the Sunway. The CASpMV customizes an auto-tuning four-way partition scheme for SpMV based on the proposed statistical model, which describes the sparse matrix structure characteristics, to make it better fit in with the computing architecture and memory hierarchy of the Sunway. Moreover, the CASpMV provides an accelerative method and customized optimizations to avoid irregular memory accesses and further improve its performance on the Sunway. Our CASpMV achieves a performance improvement that ranges from 588.05 to 2118.62 percent over the generic parallel SpMV on a CG (which corresponds to an MPI process) of the Sunway on average and has good scalability on multiple CGs. The performance comparisons of the CASpMV with state-of-the-art methods on the Sunway indicate that the sparsity and irregularity of data structures have less impact on CASpMV.