Temporal traffic prediction is critical for ITS yet remains challenging in handling complex spatio-temporal dynamics of traffic systems. The continuous traffic data (e.g., traffic flow, and speed) from various channels and nodes in a traffic network are coupled with each other over the time points of each channel, spatially between traffic nodes, and jointly in both spatial and temporal dimensions. Such multi-aspect traffic data couplings reflect the conditions of a real-life traffic system and evolve over traffic movement and network dynamics. The recent studies formulate traffic prediction by high-profile graph neural networks. However, they mainly focus on hidden relations captured by neural graph mechanisms while overlooking or simplifying the above multi-aspect traffic data couplings. By modeling a traffic system as a coupled traffic network, we learn the multi-aspect traffic data couplings by a Multi-relational Synchronous Graph Attention Network (MS-GAT). Specifically, MS-GAT learns three embeddings to respectively but synchronously represent the traffic data-based channel, temporal, and spatial relations between nodes by specific graph attention designs. The embeddings are further adaptively coupled according to their respective importance to prediction. Tested on five real-world datasets, MS-GAT outperforms six SOTA graph networks-based traffic predictors. MS-GAT captures not only spatial and temporal couplings but also traffic data-based channel interactions over traffic evolution.

Aiming at the problem of low recognition rate caused by local deformations such as pose, angle, expression change and occlusion, especially small local deformations in face recognition. A face recognition algorithm based on Gabor wavelet transform and central symmetric local direction pattern (CSLDP) is proposed. Gabor is used to extract multi-dimensional features, and CSLDP is used to reduce the computational complexity. Then histogram feature information is divided into blocks and counted in series as feature vectors of face images. Finally, sparse representation classifier (SRC) is used for classification and recognition. The experimental results on ORL, YALE and AR face databases show that the proposed method can adapt to the changes of fine expressions and light, and has strong robustness.

Waterline usually plays as an important visual cue for the autonomous navigation of marine unmanned surface vehicles (USVs) in specific waters. However, the visual complexity of the inland waterline presents a significant challenge for the development of highly efficient computer vision algorithms tailored for waterline detection in a complicated inland water environment that marine USVs face. This paper attempts to find a solution to guarantee the effectiveness of waterline detection for the USVs with a general digital camera patrolling variable inland waters. To this end, a general deep-learning-based paradigm for inland marine USVs, named DeepWL, is proposed, which consists of two cooperative deep models (termed WLdetectNet and WLgenerateNet, respectively). They afford a continuous waterline image-map estimation from a single video stream captured on board. Experimental results demonstrate the effectiveness and superiority of the proposed approach via qualitative and quantitative assessment on the concerned performances. Moreover, due to its own generality, the proposed approach has the potential to be applied to the waterline detection tasks of other water areas such as coastal waters.

The step voltage carried on by personnel is researched when single-phase grounding fault of medium voltage shore power system occurs. The experiments are designed to measure resistance value from the coastline to the ground point and the resistivity of sea water. According to the data obtained from experiments, the finite element model is established though ANSYS, the potential distribution from the hull to the fault point is calculated when single-phase fault of medium voltage shore power system occurs, and the decay of surface potential in the wharf is obtained, especially drop speed of potential near the point of failure is paid close attention, the safety range of step voltage to terminal staffs is obtained.