Deep-learning-based SAR ship classification has become a research hotspot in the military and civilian fields and achieved remarkable performance. However, the volume of available SAR ship classification data is relatively small, meaning that previous deep-learning-based methods have usually struggled with overfitting problems. Moreover, due to the limitation of the SAR imaging mechanism, the large intraclass diversity and small interclass similarity further degrade the classification performance. To address these issues, we propose a label smoothing auxiliary classifier generative adversarial network with triplet loss (LST-ACGAN) for SAR ship classification. In our method, an ACGAN is introduced to generate SAR ship samples with category labels. To address the model collapse problem in the ACGAN, the smooth category labels are assigned to generated samples. Moreover, triplet loss is integrated into the ACGAN for discriminative feature learning to enhance the margin of different classes. Extensive experiments on the OpenSARShip dataset demonstrate the superior performance of our method compared to the previous methods.

High-dimensional meteorological data offer a comprehensive overview of meteorological conditions. Nevertheless, predicting regional high-dimensional meteorological data poses challenges due to the vast scale and rapid changes. Apart from slow conventional numerical weather prediction methods, recently developed deep learning methods often fail to fully integrate spatial information of the high-dimensional data and require a significant amount of computational resources. This paper presents the spatiotemporal analysis fitting prediction algorithm (SA-Fit), an approximation algorithm for regional high-dimensional meteorological data prediction. SA-Fit proposes two key designs to achieve efficient prediction of the high-dimensional data. SA-Fit introduces a lightweight Transformer-based spatiotemporal analysis network to encode spatiotemporal information, which can integrate the interaction information between different coordinates in the data. Furthermore, SA-Fit introduces explicit functions with a lasso penalty to fit variations in high-dimensional meteorological data, achieving the prediction of a large amount of data with minimal prediction values. We performed experiments using the ERA5 dataset from the Shanghai and Xi’an regions. The experimental results show that SA-Fit is comparable to other advanced deep learning prediction methods in overall prediction performance. SA-Fit shortens training time and significantly reduces model parameters while using the Transformer structure to ensure prediction accuracy.

Recently, Unsupervised Domain Adaptation (UDA) has attracted increasing attention to address the domain shift problem in the semantic segmentation task. Although previous UDA methods have achieved promising performance, they still suffer from the distribution gaps between source and target domains, especially the resolution discrepany in the remote sensing images. To address this problem, this paper designs a novel end-to-end semantic segmentation network, namely Super-Resolution Domain Adaptation Network (SRDA-Net). SRDA-Net can simultaneously achieve the super-resolution task and the domain adaptation task, thus satisfying the requirement of semantic segmentation for remote sensing images which usually involve various resolution images. The proposed SRDA-Net includes three parts: a Super-Resolution and Segmentation (SRS) model which focuses on recovering high-resolution image and predicting segmentation map, a Pixel-level Domain Classifier (PDC) for determining which domain the pixel belongs to, and an Output-space Domain Classifier (ODC) for distinguishing which domain the pixel contribution is from. By jointly optimizing SRS with two classifiers, the proposed method can not only eliminates the resolution difference between source and target domains, but also improve the performance of the semantic segmentation task. Experimental results on two remote sensing datasets with different resolutions demonstrate that SRDA-Net performs favorably against some state-of-the-art methods in terms of accuracy and visual quality. Code and models are available at https://github.com/tangzhenjie/SRDA-Net.