Massive amount of water level data has been collected by using Internet of Things (IoT) techniques in the Yangtze River and other rivers. In this paper, utilizing these data to construct deep neural network models for water level prediction is focused. To achieve higher accuracy, both the factors of time and locations of data collection sensors are considered to perform prediction. And the network structures of gated recurrent unit (GRU) and convolutional neural network (CNN) are combined to build a CNN-GRU model in which the GRU part learns the changing trend of water level, and the CNN part learns the spatial correlation among water level data observed from adjacent water stations. The CNN-GRU model that using data from multiple locations to predict the water level of the middle location has higher accuracy than the model only based on GRU and other state-of-the-art methods including autoregressive integrated moving average model (ARIMA), wavelet-based artificial neural network (WANN) and long-short term memory model (LSTM), because of its ability to decrease the affections of abnormal value and data randomness of a single water station to some extent. The results are verified on an experiment dataset that including 30-year observed data of water level at several collection stations in the Yangtze River. For forecasting the 8-o’clock water levels of future 5 days, accuracy of the CNN-GRU model is better than that of ARIMA, WANN and LSTM models with three evaluation factors including Nash-Sutcliffe efficiency coefficient (NSE), average relative error (MRE) and root mean square error (RMSE).

With accumulation of data and development of artificial intelligence, human activity recognition attracts lots of attention from researchers. Many classic machine learning algorithms, such as artificial neural network, feed forward neural network, K-nearest neighbors, and support vector machine, achieve good performance for detecting human activity. However, these algorithms have their own limitations and their prediction accuracy still has space to improve. In this study, we focus on K-nearest neighbors (KNN) and solve its limitations. Firstly, kernel method is employed in model KNN, which transforms the input features to be the high-dimensional features. The proposed model KNN with kernel (K-KNN) improves the accuracy of classification. Secondly, a novel reduced kernel method is proposed and used in model K-KNN, which is named as Reduced Kernel KNN (RK-KNN). It reduces the processing time and enhances the classification performance. Moreover, this study proposes an approach of defining number of K neighbors, which reduces the parameter dependency problem. Based on the experimental works, the proposed RK-KNN obtains the best performance in benchmarks and human activity datasets compared with other models. It has super classification ability in human activity recognition. The accuracy of human activity data is 91.60 for HAPT and 92.67 for Smartphone, respectively. Averagely, compared with the conventional KNN, the proposed model RK-KNN increases the accuracy by 1.82 and decreases standard deviation by 0.27. The small gap of processing time between KNN and RK-KNN in all datasets is only 1.26 seconds.

With the exponential growth of the Internet population, scientists and researchers face the large-scale data for processing. However, the traditional algorithms, due to their complex computation, are not suitable for the large-scale data, although they play a vital role in dealing with large-scale data for classification and regression. One of these variants, which is called Reduced Kernel Extreme Learning Machine (Reduced-KELM), is widely used in the classification task and attracts attention from researchers due to its superior performance. However, it still has limitations, such as instability of prediction because of the random selection and the redundant training samples and features because of large-scaled input data. This study proposes a novel model called Reformed Reduced Kernel Extreme Learning Machine with RELIEF-F (R-RKELM) for human activity recognition. RELIEF-F is applied to discard the attributes of samples with the negative values in the weights. A new sample selection approach, which is used to further reduce training samples and to replace the random selection part of Reduced-KELM, solves the unstable classification problem in the conventional Reduced-KELM and computation complexity problem. According to experimental results and statistical analysis, our proposed model obtains the best classification performances for human activity data sets than those of the baseline model, with an accuracy of 92.87 % for HAPT, 92.81 % for HARUS, and 86.92 % for Smartphone, respectively.

Accurate water-level prediction is a critical component for ensuring safe maritime navigation, optimizing port operations, and mitigating coastal flooding risks. However, the complex, non-linear spatiotemporal dynamics of water systems pose significant challenges for current forecasting models. The proposed framework introduces three key innovations. First, a dual-weight graph construction mechanism integrates geographical proximity with Dynamic Time Warping (DTW)-derived temporal similarity to better represent hydrodynamic connectivity in coastal and estuarine environments. Second, a state-aware weighted loss function is designed to enhance predictive accuracy during critical hydrological events, such as storm surges and extreme tides, by prioritizing the reduction in errors in these high-risk periods. Third, the WS-STGTN architecture combines graph attention with temporal self-attention to capture long-range dependencies in both space and time. Extensive experiments are conducted using water-level data from five stations in the tidal-influenced lower Yangtze River, a vital artery for shipping and a region susceptible to coastal hydrological extremes. The results demonstrate that the model consistently surpasses a range of baseline methods. Notably, the WS-STGTN achieves an average reduction in Mean Squared Error (MSE) of 27.6% compared to the standard Transformer model, along with the highest coefficient of determination ($R^2\approx 0.96$) across all datasets, indicating its stronger explanatory power for observed water-level variability. This work provides a powerful tool that can be directly applied to improve coastal risk management, marine navigation safety, and the operational planning of port and coastal engineering projects.

Ship trajectory prediction is crucial for intelligent navigation applications, especially in the field of simulation of the flow of traffic on ships. Traditional models face limitations due to environmental uncertainty and the impact of collaborative ship behavior on trajectories. Long-term predictions suffer from accumulated errors, leading to reduced accuracy. A new multi-ship trajectory prediction model called ShipVAE is proposed based on variational autoencoder principles. ShipVAE, an extension of the pedestrian model SocialVAE, improves network structure, collision avoidance loss, and adds navigation mark information to enhance trajectory consistency with actual ship navigation. Experiments in the #49 red buoy areas of Fujiangsha water region on an AIS-based ship dataset reveal a 15% reduction in ADE and a 13% reduction in FDE compared to the baseline SocialVAE model.

The development of China’s digital waterways has led to the extensive deployment of cameras along inland waterways. However, the limited processing and utilization of digital resources hinder the ability to provide waterway services. To address this issue, this paper introduces a novel waterway perception approach based on an intelligent navigation marker system. By integrating multiple sensors into navigation markers, the fusion of camera video data and automatic identification system (AIS) data is achieved. The proposed method of an enhanced one-stage object detection algorithm improves detection accuracy for small vessels in complex inland waterway environments, while an object-tracking algorithm ensures the stable monitoring of vessel trajectories. To mitigate AIS data latency, a trajectory prediction algorithm is employed through region-based matching methods for the precise alignment of AIS data with pixel coordinates detected in video feeds. Furthermore, an augmented reality (AR)-based traffic situational awareness framework is developed to dynamically visualize key information. Experimental results demonstrate that the proposed model significantly outperforms mainstream algorithms. It achieves exceptional robustness in detecting small targets and managing complex backgrounds, with data fusion accuracy ranging from 84.29% to 94.32% across multiple tests, thereby substantially enhancing the spatiotemporal alignment between AIS and video data.