With the development of hyperspectral sensors, the availability of hyperspectral images (HSIs) has increased significantly, prompting advancements in deep learning-based hyperspectral image classification (HSIC) methods. Recently, graph convolutional networks (GCNs) have been proposed to process graph-structured data in non-Euclidean domains, and have been used for HSIC. The superpixel segmentation should be implemented first in the GCN-based methods, however, it is difficult to manually select the optimal superpixel segmentation sizes to obtain the useful information for classification. To solve this problem, we constructed a HSIC model based on a multiscale feature search-based graph convolutional network (MFSGCN) in this study. Firstly, pixel-level features of HSIs are extracted sequentially using 3D asymmetric decomposition convolution and 2D convolution. Then, superpixel-level features at different scales are extracted using multilayer GCNs. Finally, the neural architecture search (NAS) method is used to automatically assign different weights to different scales of superpixel features. Thus, a more discriminative feature map is obtained for classification. Compared with other GCN-based networks, the MFSGCN network can automatically capture features and obtain higher classification accuracy. The proposed MFSGCN model was implemented on three commonly used HSI datasets and compared to some state-of-the-art methods. The results confirm that MFSGCN effectively improves accuracy.

Typhoons, extratropical cyclones, and cold fronts cause strong winds leading to storm surges and waves in the Bohai Sea. A wave–flow coupled numerical model is established for storm events observed in 2022 caused by three weather systems, to investigate how storm waves are modulated by wave–tide–surge interaction (WTSI). Wave response is basically controlled by water level change in coastal areas, where bottom friction or breaking dominates the energy dissipation, and determined by the current field in deep water by altering whitecapping. Wave height increases/decreases are induced by positive/negative water level or obtuse/acute wave–current interaction angle, leading to six types of field patterns for significant wave height (Hs) responses. For the three storm events, Hs basically changed within ±5% in central deep water, while the maximum increase/decrease reached 160%/−60% in the coastal area of Laizhou Bay/Liaodong Bay. Based on maximum Hs and its occurrence time, WTSI modulation is manifested as the superposition effect of wave–tide and wave–surge interactions in both space and time scales, and occurrence time depends more on tide than surge for all three storms. The enhancement/abatement of WTSI modulation happens for consistent/opposite changing trends of wave–tide and wave–surge interaction, with the ultimate result showing the side with a higher effect.