Synthetic aperture radar (SAR) maritime target classification serves as a critical component in modern maritime surveillance. While deep learning networks, particularly convolutional neural networks (CNNs), have driven substantial progress in this domain, three key challenges constrain their performance and practical deployment: 1) In SAR maritime images, complex inshore backgrounds and speckle noise are prevalent. Targets such as ships span a wide range of scales due to different imaging resolutions and intrinsic size variability, exacerbating inter-class similarity and intra-class variability, 2) Labeled data for SAR maritime target classification are scarce, and sensor imaging modes differ markedly across platforms, and 3) Existing CNNs that fuse traditional hand-crafted features often explicitly treat hand-crafted feature extraction as a necessary component of the network and primarily focus on classification performance, overlooking the requirement to efficiently leverage their feature extraction capabilities in downstream tasks. To overcome these challenges, this article proposes a novel SAR maritime target classification network (MBLKNet) based on large kernel convolution and multi-task self-supervised learning. In MBLKNet, four improved designs for network structure are proposed to enhance classification accuracy: 1) macro design, 2) multi-branch large kernel convolution module (MBLKCM), 3) lightweight channel-interactive multi-layer perceptron (LCIMLP), and 4) micro design. In addition, a multi-resolution unlabeled SAR maritime target dataset (SL-SARShip) and a masked image modeling framework, HOGSparK, are proposed to enable the pre-training of MBLKNet under joint supervision of pixel and HOG features. Comparison results on OpenSARShip 2.0 and FUSAR-Ship with state-of-the-art networks, as well as experiments on SSDD for SAR downstream target detection and instance segmentation, demonstrate that the proposed MBLKNet achieves superior performance and strong feature extraction ability.

The outbreak of the COVID-19 pandemic has significantly reshaped population mobility, exerting a sustained impact on the patterns and dynamics of population mobility in China over the next three years. To comprehend the changes in population mobility patterns during the early stages of the COVID-19 outbreak, as well as standard epidemic prevention and control measures, we conducted an analysis using data from Baidu Huiyan’s migration scale index. This data was used to examine the characteristics of population movement in China during the Spring Festival and National Day from 2020 to 2022. We employed the Louvain algorithm and SVD decomposition to examine the spatiotemporal patterns of population movement. In addition, we calculated the response speed of urban population arrival flow to the pandemic using the Pearson correlation coefficient. Furthermore, we analyzed the factors influencing this correlation and response speed using random forest eigenvalues. The findings suggest that daily commuting and holiday travel patterns were not significantly altered by the pandemic. Over the past three years, there has been a trend in population mobility toward quicker responses to the pandemic, influenced primarily by economic, policy, medical conditions, and population density. Areas with higher population density and greater structural complexity exhibit increased sensitivity of population mobility to the severity of the pandemic. Examining population movement patterns and influencing factors against the backdrop of the COVID-19 pandemic can offer valuable insights for devising more targeted and effective prevention and control measures. Ultimately, this endeavor contributes to enhancing health-related urban resilience and sustainability.