Remote Sensing Scene Classification (RSSC) is an important and challenging research topic. Transformer-based methods have shown encouraging performance in capturing global dependencies. However, recent studies have revealed that Transformers perform poorly in capturing high frequencies that mainly convey local information. To solve this problem, we propose a novel method based on High-Frequency Enhanced Vision Transformer and Multi-Layer Context Learning (HETMCL), which can effectively learn the comprehensive features of high-frequency and low-frequency information in visual data. First, Convolutional Neural Networks (CNNs) extract low-level spatial structures, and the Adjacent Layer Feature Fusion Module (AFFM) reduces semantic gaps between layers to enhance spatial context. Second, the High-Frequency Information Enhancement Vision Transformer (HFIE) includes a High-to-Low-Frequency Token Mixer (HLFTM), which captures high-frequency details. Finally, the Multi-Layer Context Alignment Attention (MCAA) integrates multi-layer features and contextual relationships. On UCM, AID, and NWPU datasets, HETMCL achieves state-of-the-art OA of 99.76%, 97.32%, and 95.02%, respectively, outperforming existing methods by up to 0.38%.

An important goal of smart cities is to ensure city safety and reduce city risks. However, because the chemical industry park is often located interior and surroundings of a city, it is easy to induce explosions in case of natural disasters, causing serious losses. To investigate the possibility of explosion damage to other storage tanks in chemical industrial parks caused by tank explosions, the evolution process of tank explosion damage was studied, and an analysis method to determine the most unfavorable process was provided. This method is based on regional grid division and shock wave overpressure calculation to obtain the probability of damage at the grid location. Based on the system fault evolution process, the space fault network model was used to calculate the explosion damage evolution process of each storage tank, and the most unfavorable situation was finally determined. The paper assumes a positive linear relationship between the damage probability of the affected storage tank and the likelihood of explosion. The paper provides a calculation method for regional grid division, tank explosion overpressure, and damage probability. A mathematical model for the evolution process of tank explosion damage was constructed, and it is believed that the damage effects of tanks are a logical superposition. The results can provide a reference for controlling the explosion process in chemical industrial parks under the most unfavorable conditions and realize intelligent analysis and prediction of regional risks.

To address the challenge of efficiently integrating multi-source heterogeneous data to improve the accuracy of public safety event prediction, this study proposes and validates a novel public safety event prediction model, GATPNet, based on multi-source heterogeneous data. The model integrates Graph Attention Networks (GAT), Spatiotemporal Transformers, and Proximal Policy Optimization (PPO) to achieve effective data fusion, spatiotemporal feature extraction, and real-time decision support. Through experiments conducted on the Los Angeles Crime Data and CrisisLexT26 datasets, this study demonstrates that GATPNet outperforms other baseline models. On the Los Angeles Crime Data dataset, GATPNet achieved an accuracy of 90%, recall of 89%, Spatiotemporal Prediction Accuracy (STPA) of 80%, and a response time of 1.9 s, showing a 5% improvement in accuracy and a 10% improvement in STPA over the best baseline method. On the CrisisLexT26 dataset, it achieved an accuracy of 89%, recall of 88%, STPA of 78%, and a response time of 2.1 s, showing a 4% improvement in accuracy and a 6% improvement in STPA over the best baseline method. Additionally, ablation experiments further indicate that each module plays a critical role in improving overall performance. Despite the model’s high computational complexity when handling large-scale heterogeneous data and the limited coverage of the datasets, GATPNet still demonstrates its broad application potential in public safety event prediction and management, offering effective technical support for social governance and emergency management.

Intracerebral hemorrhage (ICH) is a critical condition characterized by a high prevalence, substantial mortality rates, and unpredictable clinical outcomes, which results in a serious threat to human health. Improving the timeliness and accuracy of prognosis assessment is crucial to minimizing mortality and long-term disability associated with ICH. Due to the complexity of ICH, the diagnosis of ICH in clinical practice heavily relies on the professional expertise and clinical experience of physicians. Traditional prognostic methods largely depend on the specialized knowledge and subjective judgment of healthcare professionals. Meanwhile, existing artificial intelligence (AI) methodologies, which predominantly utilize features derived from computed tomography (CT) scans, fall short of capturing the multifaceted nature of ICH. Although existing methods are capable of integrating clinical information and CT images for prognosis, the effectiveness of this fusion process still requires improvement. To surmount these limitations, the present study introduces a novel AI framework, termed the ICH Network (ICH-Net), which employs a joint-attention cross-modal network to synergize clinical textual data with CT imaging features. The architecture of ICH-Net consists of three integral components: the Feature Extraction Module, which processes and abstracts salient characteristics from the clinical and imaging data, the Feature Fusion Module, which amalgamates the diverse data streams, and the Classification Module, which interprets the fused features to deliver prognostic predictions. Our evaluation, conducted through a rigorous five-fold cross-validation process, demonstrates that ICH-Net achieves a commendable accuracy of up to 87.77%, outperforming other state-of-the-art methods detailed within our research. This evidence underscores the potential of ICH-Net as a formidable tool in prognosticating ICH, promising a significant advancement in clinical decision-making and patient care.

Building polygons plays an important role in urban management. Although leveraging deep learning techniques for building polygon extraction offers advantages, the models heavily rely on a large number of training samples to achieve good generalization performance. In scenarios with small training samples, the models struggle to effectively represent diverse building structures and handle the complexity introduced by the background. A common approach to enhance feature representation is fine-tuning a pre-trained model on a large dataset specific to the task. However, the fine-tuning process tends to overfit the model to the task area samples, leading to the loss of generalization knowledge from the large dataset. To address this challenge and enable the model to inherit the generalization knowledge from the large dataset while learning the characteristics of the task area samples, this paper proposes a knowledge distillation-based framework called Building Polygon Distillation Network (BPDNet). The teacher network of BPDNet is trained on a large building polygon dataset containing diverse building samples. The student network was trained on a small number of available samples from the target area to learn the characteristics of the task area samples. The teacher network provides guidance during the training of the student network, enabling it to learn under the supervision of generalization knowledge. Moreover, to improve the extraction of buildings against the backdrop of a complex urban context, characterized by fuzziness, irregularity, and connectivity issues, BPDNet employs the Dice Loss, which focuses attention on building boundaries. The experimental results demonstrated that BPDNet effectively addresses the problem of limited generalization by integrating the generalization knowledge from the large dataset with the characteristics of the task area samples. It accurately identifies building polygons with diverse structures and alleviates boundary fuzziness and connectivity issues.