Analyzing the spatial-temporal changes in tropical cyclone (TC) tracks in the east China coastal ocean (ECCO) to quantify the magnitude of poleward and landward migration of TCs is of significant importance for coastal disaster mitigation and planning due to its susceptibility to the impacts of TCs. In this study, the TCs that affected the ECCO from 1949 to 2022 are classified into three typical types of tracks using the k-means clustering method, mass moments, and track interpolation based on TC location, shape, and intensity information. Type 1 is a northwestward track, Type 2 is a northwest to northeast-turning track, and Type 3 is a northwest to northeast-turning offshore track. Type 1 tracks mainly make landfall in southern China, while Type 2 predominantly makes landfall in eastern China. Moreover, the proportion of Type 1 decreases while their landfall percentage increases over time, and the proportion of Type 2 tracks is increasing. The probability of TC effects on the eastern and northern parts of the ECCO is increasing, and the boundary where the TC center reaches after landfall is shifting landward. During the period from 1994 to 2022, there has been a significant migration in TC tracks, with the mean centroid of the TCs affecting the ECCO shifting westward by 0.66° in longitude and northward by 1.26° in latitude, which means the magnitude of the poleward shift is about twice that of the landward shift. This migration appears to have been pre-conditioned by a combined influence of a weakening westward steering flow, reduced vertical wind shear, and warmer sea surface temperature Our findings provide valuable insights into the longitudinal and latitudinal migration of TC tracks and have important implications for disaster prevention, mitigation planning, and the adjustment of crucial coastal protection zones in the ECCO and similar regions around the globe.

@en

Image-sentence matching is a challenging task in the field of language and vision, which aims at measuring the similarities between images and sentence descriptions. Most existing methods independently map the global features of images and sentences into a common space to calculate the image-sentence similarity. However, the image-sentence similarity obtained by these methods may be coarse as (1) an intermediate common space is introduced to implicitly match the heterogeneous features of images and sentences in a global level, and (2) only the inter-modality relations of images and sentences are captured while the intra-modality relations are ignored. To overcome the limitations, we propose a novel Cross-Modal Hybrid Feature Fusion (CMHF) framework for directly learning the image-sentence similarity by fusing multimodal features with inter- and intra-modality relations incorporated. It can robustly capture the high-level interactions between visual regions in images and words in sentences, where flexible attention mechanisms are utilized to generate effective attention flows within and across the modalities of images and sentences. A structured objective with ranking loss constraint is formed in CMHF to learn the image-sentence similarity based on the fused fine-grained features of different modalities bypassing the usage of intermediate common space. Extensive experiments and comprehensive analysis performed on two widely used datasets - Microsoft COCO and Flickr30K - show the effectiveness of the hybrid feature fusion framework in CMHF, in which the state-of-the-art matching performance is achieved by our proposed CMHF method.

@en

In wireless sensor network applications where each node sends a packet to a sink, the stochastic nature of the link affects the delivery rate (total number of packets delivered at the sink). Based on a simple analytical model, we study the statistical properties of the delivery rate for scenarios without link or transport layer retransmission (best-effort routing). For these best-effort scenarios, we derive bounds for the expectation and variance of the delivery rate for single-sink and multiple-sink architectures. Our analytical findings are further validated through simulations using a realistic link-layer model.@en