Explore the vast range of titles available.

Landslide spatial prediction studies predominantly focus on estimating the likelihood of landslide occurrence by considering a set of geo-environmental factors. Nevertheless, most of these studies fail to account for the spatial dependency between landslide occurrences across different terrain units. This study explored how the understanding of spatial dependency can enhance predictions of landslide occurrence across the entire Three Gorges Reservoir area, China. Specifically, we develop spatial binomial generalized additive models (GAMs) that incorporate Duchon spline (DS) and Markov random field (MRF) functions to represent continuous and discrete spatial dependencies, respectively. To test the validity of our proposed models, we compare them against the common GAM model as well as two popular machine learning models: the support vector machine (SVM) and the random forest (RF). The experimental results reveal that our models achieve superior predictive performance, with scores ranging from 0.929 to 0.938, compared to the benchmark methods, which scored between 0.912 and 0.914, based on a tenfold cross-validation procedure. These results demonstrate that incorporating spatial dependency significantly enhances the performance of landslide susceptibility prediction. Moreover, the comparison between GAM-DS and GAM-MRF models prove that the continuous smoothing of spatial dependency offers a more detailed and precise representation of landslide susceptibility. We believe that our approach, which integrates spatial dependency, will lay the foundation for the landslide community to assess landslide susceptibility from a spatially informed perspective.

Traditional feature-based classification methods require objects to have the explicit, independent, and identifiable set of features, while most geo-referenced objects do not have the explicit features required by classifiers. Therefore, developing classificatory features under geospatial context is a prerequisite for effective spatial classification. Considering the spatial dependency, objects are correlated with each other, and for the object of interest its features (e.g., the distribution of neighboring objects) exist in a wide range of neighboring areas. However, the uncertainty of neighborhood size makes the dimensionality of potential feature set particularly high for spatial classification. Therefore, we propose a new model to automatically select a subset of spatially explicit features through continuous decision making by multiple agents in reinforcement learning (RL). A novel reward mechanism is developed to feed the knowledge of the downstream classification task back to the loop of feature selection. Through extensive experiments with facility points-of-interest datasets, we demonstrate that the subset of classificatory features selected by our RL model can help significantly improve the accuracy of spatial classification. Moreover, our feature selection has potential explainability for the spatial classification rules as it can determine the neighboring areas which have an impact on the classification result.

Japan is becoming depopulated, with declining fertility rates and massive urban agglomeration due to emigrations from rural areas, which results in rural–urban disparities. As demographic and social divisions between rural and urban areas increase, maintenance of infrastructure and social facilities becomes much more difficult. For social and demographic sustainability, accurate predictions of long-term population distributions are needed. This study improves the Cohort Component Analysis (CCA) into two aspects of “dependent structure” in the model system. The migration sub-model is expanded to include related structures between available job opportunities and the available workforce in each region, which are specified using the spatial autoregressive model. The advantage of the improved CCA to provides rational future projections by considering the longitudinal changes in the spatial distribution of the workforce. The simulation of the proposed model gives an alternative long-term impact of population distribution in Japan, which is compared with the conventional CCA. The results show that the future Japanese populations will become more concentrated in urban areas, with a lower fertility rate. Furthermore, the manufacturing employees will be attracted to metropolitan areas or to regions with industrial zones, and that the number of retailers will undergo changes over time, even in urbanized areas.

Predicting human motion based on historical motion sequences is a fundamental problem in computer vision, which is at the core of many applications. Existing approaches primarily focus on encoding spatial dependencies among human joints while ignoring the temporal cues and the complex relationships across non-consecutive frames. These limitations hinder the model’s ability to generate accurate predictions over longer time horizons and in scenarios with complex motion patterns. To address the above problems, we proposed a novel multi-level spatial and temporal learning model, which consists of a Cross Spatial Dependencies Encoding Module (CSM) and a Dynamic Temporal Connection Encoding Module (DTM). Specifically, the CSM is designed to capture complementary local and global spatial dependent information at both the joint level and the joint pair level. We further present DTM to encode diverse temporal evolution contexts and compress motion features to a deep level, enabling the model to capture both short-term and long-term dependencies efficiently. Extensive experiments conducted on the Human 3.6M and CMU Mocap datasets demonstrate that our model achieves state-of-the-art performance in both short-term and long-term predictions, outperforming existing methods by up to 20.3% in accuracy. Furthermore, ablation studies confirm the significant contributions of the CSM and DTM in enhancing prediction accuracy.

SA-ConvLSTM is a recently proposed spatiotemporal model for total electron content (TEC) prediction, which effectively catches long-term temporal evolution and global-scale spatial correlations in TEC. However, its reliance on standard convolution limits spatial feature extraction to fixed regular regions, reducing the flexibility for irregular TEC variations. To address this limitation, we enhance SA-ConvLSTM by incorporating deformable convolution, proposing SA-DConvLSTM. This achieves adaptive spatial feature extraction through learnable offsets in convolutional kernels. Building on this improvement, we design ED-SA-DConvLSTM, a TEC spatiotemporal prediction model based on an encoder–decoder architecture with SA-DConvLSTM as its fundamental block. Firstly, the effectiveness of the model improvement was verified through an ablation experiment. Subsequently, a comprehensive quantitative comparison was conducted between ED-SA-DConvLSTM and baseline models (C1PG, ConvLSTM, and ConvGRU) in the region of 12.5° S–87.5° N and 25° E–180° E. The experimental results showed that the ED-SA-DConvLSTM exhibited superior performance compared to C1PG, ConvGRU, and ConvLSTM, with prediction accuracy improvements of 10.27%, 7.65%, and 7.16% during high solar activity and 11.46%, 4.75%, and 4.06% during low solar activity, respectively. To further evaluate model performance under extreme conditions, we tested the ED-SA-DConvLSTM during four geomagnetic storms. The results showed that the proportion of its superiority over the baseline models exceeded 58%.

The ionospheric total electron content (TEC) has complex spatiotemporal variations, making its spatiotemporal prediction challenging. Capturing long-range spatial dependencies is of great significance for improving the spatiotemporal prediction accuracy of TEC. Existing work based on Convolutional Long Short-Term Memory (ConvLSTM) primarily relies on convolutional operations for spatial feature extraction, which are effective at capturing local spatial correlations, but struggle to model long-range dependencies, limiting their predictive performance. Self-Attention Convolutional Long Short-Term Memory (SA-ConvLSTM) can selectively store and focus on long-range spatial dependencies, but it requires the input length and output length to be the same due to its “n vs. n” structure, limiting its application. To solve this problem, this paper proposes an encoder-decoder SA-ConvLSTM, abbreviated as ED-SA-ConvLSTM. It can effectively capture long-range spatial dependencies using SA-ConvLSTM and achieve unequal input-output lengths through encoder–decoder structure. To verify its performance, the proposed ED-SA-ConvLSTM was compared with C1PG, ConvLSTM, and PredRNN from multiple perspectives in the area of 12.5° S–87.5° N, 25° E–180° E, including overall quantitative comparison, comparison across different months, comparison at different latitude regions, visual comparisons, and comparison under extreme situations. The results have shown that, in the vast majority of cases, the proposed ED-SA-ConvLSTM outperforms the comparative models.

Some recent work on conditional formulation of multivariate Gaussian Markov random fields is presented. The focus is on model constructions by compatible conditionals and coregionalization. Special attention is given to multivariate generalizations of univariate models. Beginning with univariate model constructions, a survey of key approaches to formulating multivariate extensions is presented. Two challenges in the formulation and implementation of multivariate models are highlighted: (1) entanglement of spatial and non-spatial components, and (2) enforcement for positivity condition. Managing the two challenges by decomposition, separation, and constrained parameterization is discussed. Also highlighted is the challenge of flexible modeling of (conditional) cross-spatial dependencies and, in particular, asymmetric cross-spatial dependencies. Interpretation of asymmetric cross-spatial dependencies is also discussed. A coregionalization framework which connects and unifies the various lines of model development is presented. The framework enables a systematic development of a broad range of models via linear and spatially varying coregionalization, respectively, with extensions to locally adaptive models. Formulation of multivariate models over variable-specific lattices is discussed. Selected models are illustrated with examples of Bayesian multivariate and spatiotemporal disease mapping. Potential applications of coregionalization models in imaging analysis, covariance modeling, dimension reduction, and latent variable analysis are briefly mentioned.

Accurate forecasting of city-level large-scale traffic flow is crucial for efficient traffic management and effective transport planning. However, previously proposed traffic flow prediction methods model dynamic spatial correlations across entire traffic networks, leading to high computational complexity, elevated memory usage, and model overfitting. Therefore, a novel grid partition-based dynamic spatial–temporal graph convolutional network was developed in this study to capture correlations within a large-scale traffic network. It includes the following: a dynamic graph convolution module to divide the traffic network into grid regions and thereby effectively capture the local spatial dependencies inherent in large-scale traffic topologies, an attention-based dynamic graph convolutional network to capture the local spatial correlations within each region, a global spatial dependency aggregation module to model inter-regional correlation weights using sequence similarity methods and comprehensively reflect the overall state of the traffic network, and multi-scale gated convolutions to capture both long- and short-term temporal correlations across varying time ranges. The performance of the proposed model was compared with that of different baseline models using two large-scale real-world datasets; the proposed model significantly outperformed the baseline models, demonstrating its potential effectiveness in managing large-scale traffic networks.

Remote sensing image change detection (CD) is done to identify desired significant changes between bitemporal images. Given two co-registered images taken at different times, the illumination variations and misregistration errors overwhelm the real object changes. Exploring the relationships among different spatial–temporal pixels may improve the performances of CD methods. In our work, we propose a novel Siamese-based spatial–temporal attention neural network. In contrast to previous methods that separately encode the bitemporal images without referring to any useful spatial–temporal dependency, we design a CD self-attention mechanism to model the spatial–temporal relationships. We integrate a new CD self-attention module in the procedure of feature extraction. Our self-attention module calculates the attention weights between any two pixels at different times and positions and uses them to generate more discriminative features. Considering that the object may have different scales, we partition the image into multi-scale subregions and introduce the self-attention in each subregion. In this way, we could capture spatial–temporal dependencies at various scales, thereby generating better representations to accommodate objects of various sizes. We also introduce a CD dataset LEVIR-CD, which is two orders of magnitude larger than other public datasets of this field. LEVIR-CD consists of a large set of bitemporal Google Earth images, with 637 image pairs (1024 × 1024) and over 31 k independently labeled change instances. Our proposed attention module improves the F1-score of our baseline model from 83.9 to 87.3 with acceptable computational overhead. Experimental results on a public remote sensing image CD dataset show our method outperforms several other state-of-the-art methods.

Traffic accident prediction is a crucial component of an intelligent traffic system, which is important to maintain citizen safety and decrease economic losses. Current methods for traffic accident prediction based on deep learning fail to consider the driving mechanisms of traffic accidents, so a novel traffic accident prediction method based on multi-view spatial-temporal learning is proposed, which represents the driving mechanism of traffic accidents from multiple views. Firstly, for the urban regions divided by grids, a new augmentation was designed to augment the spatial semantic information of regions through learnable semantic embedding, then deformable convolutional networks with non-fixed convolution kernels are used to learn dynamic spatial dependencies between regions and gated recurrent units are used to learn temporal dependencies, which can capture dynamic spatial-temporal evolution patterns of traffic accidents. Secondly, long short-term memory is employed to learn the traffic flow breakdown from the flow difference of adjacent time steps in each region to recognize the traffic accident precursor in the risk environment. Thirdly, accident patterns in different regions are learned from historical traffic flow to determine whether the flow is the dominant factor and capture the spatial heterogeneity of traffic accidents. Finally, the above features are fused for accident prediction at the regional level. Experiments are conducted on two real datasets, and the experimental results show that the proposed method outperforms eight benchmark methods.