Search Results (201)

Abstract: Chlorophyll prediction facilitates the comprehension of red tide characteristics and enables early warning. In practice, it is formulated as a multivariate time series forecasting problem aimed at forecasting future chlorophyll concentrations by considering both exogenous factors and chlorophyll. However, the multi-step prediction of chlorophyll concentration poses a formidable challenge due to the intricate interaction between factors and the long temporal dependence between input sequences. In this work, we propose a Multi-attention Recurrent Neural Network (MaRNN) for the multi-step prediction of chlorophyll concentration. The MaRNN comprises an encoder incorporating two-stage spatial attention and a decoder employing temporal attention. The encoder first learns the significance of exogenous factors for prediction in the first phase, and subsequently captures the spatial correlation between the exogenous sequence and chlorophyll sequence in the second phase. The decoder further excavates input sequences that exhibit a strong correlation with the task through temporal attention module, thereby enhancing the prediction accuracy of the model. Experiments conducted on two real-world datasets reveal that MaRNN not only surpasses state-of-the-art methods in performance, but also offers interpretability for chlorophyll prediction. Full article

Human action recognition (HAR) is a critical area in computer vision with wide-ranging applications, including video surveillance, healthcare monitoring, and abnormal behavior detection. Current HAR methods predominantly rely on full-body data, which can limit their effectiveness in real-world scenarios where occlusion is common. In such situations, the face often remains visible, providing valuable cues for action recognition. This paper introduces Face in Action (FIA), a novel two-stream method that leverages facial action cues for robust action recognition under conditions of significant occlusion. FIA consists of an RGB stream and a landmark stream. The RGB stream processes facial image sequences using a fine-spatio-multitemporal (FSM) 3D convolution module, which employs smaller spatial receptive fields to capture detailed local facial movements and larger temporal receptive fields to model broader temporal dynamics. The landmark stream processes facial landmark sequences using a normalized temporal attention (NTA) module within an NTA-GCN block, enhancing the detection of key facial frames and improving overall recognition accuracy. We validate the effectiveness of FIA using the NTU RGB+D and NTU RGB+D 120 datasets, focusing on action categories related to medical conditions. Our experiments demonstrate that FIA significantly outperforms existing methods in scenarios with extensive occlusion, highlighting its potential for practical applications in surveillance and healthcare settings. Full article

The increasing congestion in major global maritime routes poses significant threats to international maritime safety, exacerbated by the proliferation of large, high-speed vessels. To improve the detection of abnormal ship behavior, this research employed automatic identification system (AIS) data for ship trajectory forecasting. Traditional methods primarily focus on spatial and temporal correlations but often lack accuracy and reliability. In this study, ship path predictions were enhanced using the WTG model, which combines wavelet transform, temporal convolutional networks (TCN), and gated recurrent units (GRU). Initially, wavelet decomposition was applied to deconstruct the input trajectory time series. The TCN and GRU modules then extracted features from both the time series and the decomposed data. The predicted elements were reassembled using a multi-head attention mechanism and a pooling layer to produce the final predictions. Comparative experiments demonstrated that the WTG model surpasses other models in the accuracy of ship trajectory prediction. The model proposed in this study proves to be reliable for forecasting ship paths, which is crucial for marine traffic management and ensuring safe navigation. Full article

Nature-based solutions (NbSs) of biodiversity conservation and ecosystem restoration have been paid increasing attention as an essential approach to slow down climate change. However, to what degree an NbS approach will contribute to the combined effects of human intervention and climate change has not been well studied. From a habitat quality perspective, we set four NbS scenarios to analyze whether the NbS—mangrove restoration in particular—will be enough for climate change in Xiamen Bay of Fujian Province, China. The habitat quality module of the InVEST model (InVEST-HQ) and the Sea Level Affecting Marshes Model (SLAMM) were used to simulate the spatial-temporal changes in habitat types and habitat quality. Results show that (1) rising sea levels will cause coastal squeeze effects, impacting habitat quality due to erosion and inundation in the study area; (2) mangrove restoration is an effective way to mitigate climate change effects and to increase habitat quality; and (3) further analysis of the effectiveness of mangrove restoration shows that the consideration of mangrove fragmentation effects and sea-use impacts are necessary. The findings in this study will enrich the international discussion of NbSs to climate change in coastal areas. Full article

Extreme precipitation events often present significant risks to human life and property, making their accurate prediction an essential focus of current research. Recent studies have primarily concentrated on exploring the formation mechanisms of extreme precipitation. Existing prediction methods do not adequately account for the combined terrain and atmospheric effects, resulting in shortcomings in extreme precipitation forecasting accuracy. Additionally, the satellite data resolution used in prior studies fails to precisely capture nuanced details of abrupt changes in extreme precipitation. To address these shortcomings, this study introduces an innovative approach for accurately predicting extreme precipitation: the multimodal attention ConvLSTM-GAN for extreme rainfall nowcasting (ER-MACG). This model employs high-resolution Fengyun-4A(FY4A) satellite precipitation products, as well as terrain and atmospheric datasets as inputs. The ER-MACG model enhances the ConvLSTM-GAN framework by optimizing the generator structure with an attention module to improve the focus on critical areas and time steps. This model can alleviate the problem of information loss in the spatial–temporal convolutional long short-term memory network (ConvLSTM) and, compared with the standard ConvLSTM-GAN model, can better handle the detailed changes in time and space in extreme precipitation events to achieve more refined predictions. The main findings include the following: (a) The ER-MACG model demonstrated significantly greater predictive accuracy and overall performance than other existing approaches. (b) The exclusive consideration of DEM and LPW data did not significantly enhance the ability to predict extreme precipitation events in Zhejiang Province. (c) The ER-MACG model significantly improved in identifying and predicting extreme precipitation events of different intensity levels. Full article

Accurate prediction of ship traffic flow is essential for developing intelligent maritime transportation systems. To address the complexity of ship traffic flow data in the port and the challenges of capturing its dynamic spatiotemporal dependencies, a dynamic spatiotemporal model called Temporal convolutional network-bidirectional Gated recurrent unit-Pearson correlation coefficient-Graph Attention Network (TG-PGAT) is proposed for predicting traffic flow in port waters. This model extracts spatial features of traffic flow by combining the adjacency matrix and spatial dynamic coefficient correlation matrix within the Graph Attention Network (GAT) and captures temporal features through the concatenation of the Temporal Convolutional Network (TCN) and Bidirectional Gated Recurrent Unit (BiGRU). The proposed TG-PGAT model demonstrates higher prediction accuracy and stability than other classic traffic flow prediction methods. The experimental results from multiple angles, such as ablation experiments and robustness tests, further validate the critical role and strong noise resistance of different modules in the TG-PGAT model. The experimental results of visualization demonstrate that this model not only exhibits significant predictive advantages in densely trafficked areas of the port but also outperforms other models in surrounding areas with sparse traffic flow data. Full article

Accurate temperature prediction is essential for economic production and human society’s daily life. However, most current methods only focus on time-series temperature modeling and prediction, ignoring the complex interplay of meteorological variables in the spatial domain. In this paper, a novel temperature prediction model (D3AT-LSTM) is proposed by combining the three-dimensional convolutional neural network (3DCNN) and the attention-based gated cyclic network. Firstly, the historical meteorological series of eight surrounding pixels are combined to construct a multi-dimensional feature tensor that integrates variables from the temporal domain as the input data. Convolutional units are used to model and analyze the spatiotemporal patterns of the local sequence in CNN modules by combining them with parallel attention mechanisms. The fully connected layer finally makes the final temperature prediction. This method is subsequently compared with both classical and state-of-art prediction models such as ARIMA (AR), long short-term memory network (LSTM), and Transformer using three indices: the root mean square error (RMSE), the mean absolute error (MAE), and the coefficient of determination ($R^2)$. The results indicate that the D3AT-LSTM model can achieve good prediction accuracy compared to AR, LSTMs, and Transformer. Full article

Accurately estimating the state of charge of a lithium-ion battery plays an important role in managing the health of a battery and estimating its charging state. Traditional state-of-charge estimation methods encounter difficulties in processing the diverse temporal data sequences and predicting adaptive results. To address these problems, we propose a spatial transformer network (STN) for multi-temperature state-of-charge estimation of lithium-ion batteries. The proposed STN consists of a convolutional neural network with a temporal–spatial module and a long short-term memory transformer network, which together are able to efficiently capture the spatiotemporal features. To train the STN under multi-temperature conditions, denoising augmentation and attention prediction are proposed to enhance the model’s generalizability within a unified framework. Experimental results show that the proposed method reduces the mean absolute error and root mean square error by 41% and 43%, respectively, compared with existing methods; in the semi-supervised setting, the respective reductions are 23% and 38%, indicating that effective extraction of the spatiotemporal features along with denoising augmentation is beneficial for estimating the state of charge and can promote the development of battery management systems using semi-supervised learning methods. Full article

Drug-resistant epilepsy is frequent, persistent, and brings a heavy economic burden to patients and their families. Traditional epilepsy detection methods ignore the causal relationship of seizures and focus on a single time or spatial dimension, and the effect varies greatly in different patients. Therefore, it is necessary to research accurate automatic detection technology of epilepsy in different patients. We propose a causal-spatio-temporal graph attention network (CSTGAT), which uses transfer entropy (TE) to construct a causal graph between multiple channels, combining graph attention network (GAT) and bi-directional long short-term memory (BiLSTM) to capture temporal dynamic correlation and spatial topological structure information. The accuracy, specificity, and sensitivity of the SWEZ dataset were 97.24%, 97.92%, and 98.11%. The accuracy of the private dataset reached 98.55%. The effectiveness of each module was proven through ablation experiments and the impact of different network construction methods was compared. The experimental results indicate that the causal relationship network constructed by TE could accurately capture the information flow of epileptic seizures, and GAT and BiLSTM could capture spatiotemporal dynamic correlations. This model accurately captures causal relationships and spatiotemporal correlations on two datasets, and it overcomes the variability of epileptic seizures in different patients, which may contribute to clinical surgical planning. Full article

Meteorological prediction is crucial for various sectors, including agriculture, navigation, daily life, disaster prevention, and scientific research. However, traditional numerical weather prediction (NWP) models are constrained by their high computational resource requirements, while the accuracy of deep learning models remains suboptimal. In response to these challenges, we propose a novel deep learning-based model, the Spatiotemporal Fusion Model (STFM), designed to enhance the accuracy of meteorological predictions. Our model leverages Fifth-Generation ECMWF Reanalysis (ERA5) data and introduces two key components: a spatiotemporal encoder module and a spatiotemporal fusion module. The spatiotemporal encoder integrates the strengths of convolutional neural networks (CNNs) and recurrent neural networks (RNNs), effectively capturing both spatial and temporal dependencies. Meanwhile, the spatiotemporal fusion module employs a dual attention mechanism, decomposing spatial attention into global static attention and channel dynamic attention. This approach ensures comprehensive extraction of spatial features from meteorological data. The combination of these modules significantly improves prediction performance. Experimental results demonstrate that STFM excels in extracting spatiotemporal features from reanalysis data, yielding predictions that closely align with observed values. In comparative studies, STFM outperformed other models, achieving a 7% improvement in ground and high-altitude temperature predictions, a 5% enhancement in the prediction of the u/v components of 10 m wind speed, and an increase in the accuracy of potential height and relative humidity predictions by 3% and 1%, respectively. This enhanced performance highlights STFM’s potential to advance the accuracy and reliability of meteorological forecasting. Full article

Autonomous driving pays considerable attention to pedestrian trajectory prediction as a crucial task. Constructing effective pedestrian trajectory prediction models depends heavily on utilizing the motion characteristics of pedestrians, along with their interactions among themselves and between themselves and their environment. However, traditional trajectory prediction models often fall short of capturing complex real-world scenarios. To address these challenges, this paper proposes an enhanced pedestrian trajectory prediction model, M²Tames, which incorporates comprehensive motion, interaction, and semantic context factors. M²Tames provides an interaction module (IM), which consists of an improved multi-head mask temporal attention mechanism (M²Tea) and an Interaction Inference Module (I²). M²Tea thoroughly characterizes the historical trajectories and potential interactions, while I² determines the precise interaction types. Then, IM adaptively aggregates useful neighbor features to generate a more accurate interactive feature map and feeds it into the final layer of the U-Net encoder to fuse with the encoder’s output. Furthermore, by adopting the U-Net architecture, M²Tames can learn and interpret scene semantic information, enhancing its understanding of the spatial relationships between pedestrians and their surroundings. These innovations improve the accuracy and adaptability of the model for predicting pedestrian trajectories. Finally, M²Tames is evaluated on the ETH/UCY and SDD datasets for short- and long-term settings, respectively. The results demonstrate that M²Tames outperforms the state-of-the-art model MSRL by 2.49% (ADE) and 8.77% (FDE) in the short-term setting and surpasses the optimum Y-Net by 6.89% (ADE) and 1.12% (FDE) in the long-term prediction. Excellent performance is also shown on the ETH/UCY datasets. Full article

Graph Convolutional Networks (GCNs) have emerged as a game-changer in skeleton-based action recognition. However, most previous works are resource-heavy, with large FLoating-number OPerations (FLOPs) limiting the model’s potential. A recent work involving shift operators to GCN (Shift-GCN) has successfully introduced a lightweight GCN, but there is still a performance gap compared to previous results. Inspired by Shift-GCN, we propose an innovative and novel model named attention-guided and topology-enhanced shift graph convolutional network (AT-Shift-GCN), which continues the lightweight benchmark and provides a more powerful performance. We employ a topological transfer operation to aggregate the information flow of different channels and extract spatial information. In addition, to extract temporal information across scales, we apply attention to interacting with shift convolution kernels of different lengths. Furthermore, we integrate an ultralight spatiotemporal attention module to fuse spatiotemporal details and provide robust neighborhood representation. In summary, AT-Shift-GCN is a breakthrough in skeleton-based action recognition that provides a lightweight model with enhanced performance on three datasets. Full article

The spatiotemporal forecasting of temperature is a critical issue in meteorological prediction, with significant implications for fields such as agriculture and energy. With the rapid advancement of data-driven deep learning methods, deep learning-based spatiotemporal sequence forecasting models have seen widespread application in temperature spatiotemporal forecasting. However, statistical analysis reveals that temperature evolution varies across temporal and spatial scales due to factors like terrain, leading to a lack of existing temperature prediction models that can simultaneously learn both large-scale global features and small to medium-scale local features over time. To uniformly model temperature variations across different temporal and spatial scales, we propose the Multi-Scale Large Kernel Spatiotemporal Attention Neural Network (MSLKSTNet). This model consists of three main modules: a feature encoder, a multi-scale spatiotemporal translator, and a feature decoder. The core module of this network, Multi-scale Spatiotemporal Attention (MSSTA), decomposes large kernel convolutions from multi-scale perspectives, capturing spatial feature information at different scales, and focuses on the evolution of multi-scale spatial features over time, encompassing both global smooth changes and local abrupt changes. The results demonstrate that MSLKSTNet achieves superior performance, with a 35% improvement in the MSE metric compared to SimVP. Ablation studies confirmed the significance of the MSSTA unit for spatiotemporal forecasting tasks. We apply the model to the regional ERA5-Land reanalysis temperature dataset, and the experimental results indicate that the proposed method delivers the best forecasting performance, achieving a 42% improvement in the MSE metric over the widely used ConvLSTM model for temperature prediction. This validates the effectiveness and superiority of MSLKSTNet in temperature forecasting tasks. Full article

Efficient operation of urban metro systems depends on accurate passenger flow predictions, a task complicated by intricate spatiotemporal correlations. This paper introduces a novel spatiotemporal graph neural network (STGNN) designed explicitly for predicting multistep passenger flow within metro stations. In the spatial dimension, previous research primarily focuses on local spatial dependencies, struggling to capture implicit global information. We propose a spatial modeling module that leverages a dynamic global attention network (DGAN) to capture dynamic global information from all-pair interactions, intricately fusing prior knowledge from the input graph with a graph convolutional network. In the temporal dimension, we design a temporal modeling module tailored to navigate the challenges of both long-term and recent-term temporal passenger flow patterns. This module consists of series decomposition blocks and locality-aware sparse attention (LSA) blocks to incorporate multiple local contexts and reduce computational complexities in long sequence modeling. Experiments conducted on both simulated and real-world datasets validate the exceptional predictive performance of our proposed model. Full article

I report on long-term patterns of outbreak cycling in four study systems across Canada and illustrate how forecasting in these systems is highly imprecise because of complexity in the cycling and a lack of spatial synchrony amongst sample locations. I describe how a range of bottom-up effects could be generating complexity in these otherwise periodic systems. (1) The spruce budworm in Québec exhibits aperiodic and asynchronous behavior at fast time-scales, and a slow modulation of cycle peak intensity that varies regionally. (2) The forest tent caterpillar across Canada exhibits eruptive spiking behavior that is aperiodic locally, and asynchronous amongst regions, yet aggregates to produce a pattern of periodic outbreaks. In Québec, forest tent caterpillar cycles differ in the aspen-dominated northwest versus the maple-dominated southeast, with opposing patterns of cycle intensity between the two regions. (3) In Alberta, forest tent caterpillar outbreak cycles resist synchronization across a forest landscape gradient, even at very fine spatial scales, resulting in a complex pattern of cycling that defies simple forecasting techniques. (4) In the Border Lakes region of Ontario and Minnesota, where the two insect species coexist in a mixedwood landscape of hardwood and conifers, outbreak cycle intensity in each species varies spatially and temporally in response to host forest landscape structure. Much attention has been given to the effect of top-down agents in driving synchronizable population cycles. However, foliage loss, tree death, and forest succession at stem, stand, and landscape scales affect larval and adult dispersal success, and may serve to override regulatory processes that cause otherwise top-down-driven periodic, synchronized, and predictable population oscillations to become aperiodic, asynchronous, and unpredictable. Incorporating bottom-up effects at multiple spatial and temporal scales may be the key to making significant improvements in forest insect outbreak forecasting. Full article