Search Results (3,185)

Dense infrared small target detection (ISTD) in long-range remote sensing is critical for multi-target surveillance, yet existing benchmarks mostly contain only sparsely distributed targets and rarely reflect dense scenes. To address this limitation, we construct a new dense satellite ISTD dataset, IR-SatDense, by compositing small targets onto real satellite infrared backgrounds and partitioning it into subsets using the Average Minimum Inter-Target Distance (AMID) to explicitly control target density. By visualizing multi-stage backbone features, we observe that in dense scenes the deepest stage naturally forms compact, high-response target clusters in the semantic feature maps, while low- and middle-level features remain heavily cluttered. This motivates us to treat high-level semantic density as a global prior to guide low-level feature enhancement. Therefore, we propose Semantic Density-Guided ResNet (SDG-ResNet), a plug-in backbone that attaches a lightweight semantic density head to the deepest stage and injects the predicted density map into intermediate layers through Semantic Density-Guided Refine (SDGR) blocks with residual spatial gating. Integrated into representative transformer-based detectors, including Deformable DETR, DETA, and DINO, SDG-ResNet consistently improves the probability of detection (PD) at comparable false alarm (FA) levels on IR-SatDense while maintaining competitive performance on the sparse dataset IRSTD-1K. Full article

Video prediction is a fundamental task in computer vision with broad applications in intelligent robotics, autonomous driving, and related fields. However, existing methods often struggle to simultaneously model long-term temporal dependencies, preserve local details, and alleviate error accumulation during autoregressive prediction. To address these issues, this paper proposes a two-stage video prediction framework composed of a HybridResSwin Autoencoder (HRS-AE) and an Enhanced FAR Transformer (EFAR). In the first stage, HRS-AE learns compact and discriminative latent representations from input video frames while preserving essential spatial structures and fine-grained details. In the second stage, EFAR performs autoregressive temporal prediction in the latent space, and the predicted latent representations are then decoded to reconstruct future video frames. Experiments on the KTH, BAIR, and Moving MNIST datasets show that the proposed method achieves competitive performance under the adopted evaluation protocol. Specifically, the proposed framework achieves a PSNR of 30.27 dB and an LPIPS of 0.0722 on KTH, a PSNR of 20.95 dB on BAIR, and an SSIM of 0.961 with an MSE of 22.9 on Moving MNIST. In addition, ablation studies further indicate that the proposed components contribute to latent representation learning and long-horizon prediction stability. These results suggest that the proposed framework provides a promising approach for video prediction with favorable reconstruction quality, perceptual consistency, and temporal coherence. Full article

This study investigates the forecasting of the XBANK banking index traded on Borsa Istanbul by integrating financial and textual data within a deep learning framework. Unlike the majority of existing studies that focus on stable market environments, this paper explicitly examines a period of heightened political uncertainty, namely the cancellation and re-run of the 2019 Istanbul local elections. This setting provides a unique opportunity to analyze how political events and news-driven information flows influence financial market dynamics. The empirical analysis is based on a comprehensive dataset that includes daily price indicators (opening, closing, high, and low values), technical indicators, selected macroeconomic variables, and Turkish-language news headlines. Textual data are processed using topic modeling techniques to extract latent information embedded in financial news, allowing for the incorporation of qualitative signals into the forecasting framework. From a methodological perspective, this study employs a feedforward deep neural network model designed to capture nonlinear relationships across heterogeneous and contemporaneous features. Feature selection is conducted using the Boruta algorithm, while hyperparameters are optimized via grid search. The model structure reflects a deliberate design choice aimed at capturing short-term, news-driven shocks and cross-feature interactions, which are particularly relevant during periods of political uncertainty. The results indicate that incorporating textual information significantly improves forecasting performance and that news topics related to political decisions, central bank policies, and geopolitical developments have a measurable impact on the XBANK index. Furthermore, the findings suggest that the political uncertainty surrounding the 2019 local elections led to increased market sensitivity and volatility, highlighting the role of information shocks in emerging financial markets. Overall, this study contributes to the literature by combining financial and textual data in an emerging market context, utilizing Turkish-language news sources, and providing empirical evidence on the impact of political uncertainty on the BIST bank index. Full article

Autoimmune pancreatitis (AIP) represents a rare multifaceted disorder group with currently three types that were recently supplemented by a newly described AIP type causally related to sunlight exposure, not previously reported in any of the AIP publications. The internet search disclosed that the new AIP type was different from three existing AIP types: the AIP-1 or AIP-2 types, both featured by idiopathy, and the AIP-3 type, triggered by immune checkpoint inhibitors. Accordingly, it seemed appropriate to classify the novel AIP as the AIP-4 type, with typical features such as a clear culprit of sunlight exposure, ascertained by a positive result of an unintentional re-exposure and considered a diagnostic gold standard; coexisting secondary sclerosing cholangitis without progression to vanishing bile duct syndrome; and ultimately unavoidable pancreatic atrophy with clinical exocrine insufficiency despite long-term treatment with immunosuppressive drugs. Thus, the recent description of a new AIP type, now classified as the AIP-4 type, is strongly associated with significant sunlight exposure and calls for a reclassification of AIP types that includes AIP-4, whereby additional efforts are essential to identify the causative factors of the AIP-1 and AIP-2 types, including drugs commonly used in the respective cohorts, which also comprise older patients with comorbidities. Full article

Semantic segmentation of high-resolution remote-sensing images is critical for urban planning, land-cover mapping, and ecological monitoring. However, existing methods face limitations in handling complex land-cover types, multi-scale objects, and modeling long-range dependencies. To address these challenges, we propose RFA2Net, an enhanced semantic segmentation model based on the DeepLabv3+ framework. The key innovations include the integration of the RFCSA-Conv module into the ResNet101 backbone to enhance feature representation and long-range dependency modeling, the design of the RFA-DASPP structure built upon the Dense ASPP framework with the novel RFCA-DConv dilated convolution module to reduce information loss during multi-scale feature fusion and enhance the model’s ability to perceive long-range directional structures, and the introduction of a Dual-Branch Fusion Network to improve segmentation accuracy for small-scale objects. Experimental results on the ISPRS Potsdam and LoveDA datasets demonstrate that RFA2Net outperforms several CNN and Transformer-based models, achieving 78.94% and 59.46% mean intersection over union (mIoU) on the ISPRS Potsdam and LoveDA datasets, respectively, with improvements of 3.19% and 3.08% over the original DeepLabv3+. Ablation studies and comparative experiments further confirm the model’s effectiveness, robustness, and practical applicability in high-resolution remote-sensing image segmentation, with particular relevance to environmental monitoring and sustainable energy applications. Full article

Hainan is the dominant production area of the red-fleshed pitaya (Hylocereus undatus) cv. ‘Jindu No.1’ in China, and cutting propagation is the main method for its large-scale seedling cultivation. Plant growth regulators (PGRs) are the key factors regulating the rooting of cuttings. Existing studies mostly focus on the concentration optimization of a single agent, lack systematic broad-spectrum screening of commonly used PGRs in agriculture, and have the problem of disconnection between laboratory results and field production. To screen an efficient root-promoting PGR scheme suitable for large-scale seedling cultivation in Hainan production areas, this study established a three-level experimental system of “broad-spectrum primary screening→gradient re-screening→soil culture scenario verification”, used 14 kinds of PGRs commonly used in agricultural production as materials, and carried out a systematic evaluation combined with principal component analysis (PCA). 1-naphthaleneacetic acid (NAA), indole-3-acetic acid (IAA) and potassium indole-3-butyrate (K-IBA) were identified as high-efficiency agents in the primary screening, with a rooting rate of 100%, and the core root morphological indexes were significantly better than those of the water control (p < 0.05). Two independent experiments verified the stability of the “total growth–thickness” binary regulation mechanism of the pitaya root system. In the re-screening test, 400 mg·L⁻¹ NAA had the best comprehensive performance, synergistically improving the total root growth and root thickness, and 125 mg·L⁻¹ K-IBA had the most significant effect in promoting the longitudinal extension of roots, with the average root length increased by 760.0% compared with the control. Soil culture tests confirmed that the two optimal schemes had stable and reliable application effects in field substrate cultivation. The results of this study can provide technical support for the large-scale seedling cultivation of ‘Jindu No.1’ pitaya, and the established three-level screening system also provides a methodological reference for PGR screening in cutting propagation of similar tropical crops. Full article

The potential for carbon fiber reinforced polymer (CFRP)-strengthened reinforced concrete (RC) beams to achieve satisfactory load-bearing performance through secondary strengthening, after reaching their ultimate capacity and experiencing CFRP failure, has been infrequently explored. As a result, the feasibility of secondary strengthening and the associated mechanical behavior remain inadequately understood. This paper presents an experimental study aimed at investigating the secondary strengthening of RC beams with CFRP subsequent to the occurrence of CFRP debonding or rupture under loads significantly exceeding the ultimate capacity of conventional RC beams. Four-point bending tests were conducted to evaluate key indicators for the secondary-strengthened beams, including load-bearing capacity, stiffness, crack development, and strain distribution. The following observations were made regarding the secondary-strengthened test beams in this study: the ultimate load-bearing capacity of the secondary-strengthened beams is approximately 35.83% to 62.99% greater than that of an ordinary RC beam, and 2.8% lower to 14.75% higher than that of a primary-strengthened beam. The number of cracks in the secondary-strengthened beams remains largely unchanged, while existing cracks from the primary-strengthened system exhibit slight widening, with negligible growth in length. Additionally, the secondary-strengthened beams demonstrate enhanced overall stiffness compared to the primary-strengthened beams. A zero-strain concrete zone is identified within the cross-section of the secondary-strengthened beams, with a height ranging from approximately 0.25h to 0.8h when flat CFRP sheets are employed for strengthening. These experimental results indicate that CFRP-strengthened RC beams may retain the potential for secondary strengthening even following the failure of the initial CFRP system. The findings provide preliminary data that could facilitate further research into the secondary strengthening of RC beams using CFRP, and they may also offer a new exploratory perspective for re-evaluating the reliability and load-carrying performance of primary CFRP strengthening. Full article

Emergency leakage repair in subway shield tunnels requires a technique to encapsulate highly reactive sodium silicate that is simple and field-deployable, yet no mature solution currently exists. The challenge lies in sodium silicate’s strong alkalinity and high osmotic pressure, both of which corrode most shell materials. This study proposes a “composite core” concept—functionally re-engineering the core rather than relying on complex shell chemistries. Using hydroxypropyl methylcellulose (HPMC) as the key material, temperature-triggered microcapsules with a nano-silica shell and sodium silicate–HPMC core were fabricated via low-temperature shear. Low temperature (10–15 °C) is critical: it suppresses side reactions and tunes viscosity to 2000–5000 cP, facilitating shear dispersion. The resulting microcapsules exhibit well-defined morphology with a dense shell. Temperature response tests reveal distinct release onset at ~30 °C (HPMC’s LCST): HPMC chain collapse generates internal stress that ruptures the shell, driving progressive sodium silicate release. Alkaline resistance tests confirm that intact microcapsules remain stable in high-pH environments (pH ≈ 13.2) for 30 min. This work validates the “composite core” concept and provides a simple, field-operable route to fabricate temperature-triggered microcapsules for emergency repair applications. Full article

Background/Objectives: Zeaxanthin and lutein, which are essential dietary xanthophylls existing abundantly in free and esterified forms, require efficient intestinal absorption due to their insufficient synthesis in humans. However, limited knowledge on intestinal uptake and transport of xanthophyll esters is available. Methods: This study investigated the cellular uptake and transport mechanism of free and esterified xanthophylls using human Caco-2 cell monolayer, with lutein, zeaxanthin and their dipalmitates as representatives. Results: The results showed that free xanthophylls were uptaken without cellular re-esterification. Esterified xanthophylls were predominantly uptaken in free forms, as evidenced by Caco-2 cells incubated with zeaxanthin and lutein dipalmitates containing 80.8% and 89.4% of zeaxanthin and lutein, along with minor amounts of monoesters and diesters, respectively. Subsequent basolateral detection of both free xanthophylls and monoesters also confirmed intact ester uptake. Additionally, time- and concentration-dependent uptake patterns were observed, with all xanthophylls showing moderate permeability. Mechanistically, SR-BI and NPC1L1 were involved in the uptake of both free and esterified xanthophylls. At the expression level, free and esterified xanthophylls differentially affected ABCG5, with significant upregulation observed only in response to free xanthophylls. Tight junction integrity remained unaffected, excluding paracellular transport. Uptake of free and esterified xanthophyll micelles also involved clathrin- and caveolae-dependent endocytosis, whereas macropinocytosis was excluded. Conclusions: These findings provide insight into the uptake behavior of free and esterified xanthophylls and the transporter- and endocytosis-related processes involved. Full article

Background: Machine learning-based Artificial Intelligence (AI) models have shown significant potential in the biomedical field, offering promising advances in diagnostics, personalized medicine, and patient care. However, to build these models, we have to deal with important challenges, including (1) the scarcity and low quality of available datasets in many important applications and (2) privacy concerns associated with sensitive patient data. Synthetic data (SD) generation has emerged as a promising strategy to address these challenges, yet many existing approaches struggle to simultaneously preserve privacy and accurately model tabular data, the predominant format in healthcare. Methods: We propose Kernel Density Estimation–K-Nearest Neighbors (KDE-KNN), a privacy-aware tabular data generation method, and evaluate its performance against state-of-the-art techniques. Using sepsis detection as a real-world case study, we assess both data utility and privacy protection. Results: Models trained on KDE-KNN-generated SD outperformed those trained on real data across both internal testing and external validation. In particular, a support vector machine achieved superior performance when trained on SD relative to real data. This gain is likely driven by the balanced class distribution of the synthetic dataset, underscoring KDE-KNN’s utility as an effective data balancing strategy. Consistent performance in external validation further supports the robustness and generalizability of the proposed approach. Privacy evaluation indicated a lower re-identification risk, with a mean distance to closest record of 4.971 between synthetic and real samples, compared with 2.715 among real samples. Conclusions: KDE-KNN effectively captures underlying population distributions while generating high-quality SD that preserve statistical fidelity and protect sensitive information. By balancing the trade-off between utility and privacy, the method produces representative datasets without exposing individual records. These findings position KDE-KNN as a valuable tool for data-scarce and privacy-sensitive applications, with broad potential across healthcare and other data-driven domains. Full article

The increasing electrification of urban transportation has formulated a tightly coupled energy-mobility nexus. Under extreme disaster events or grid faults, rapidly restoring power supply capacity and re-dispatching shared electric vehicle (EV) fleets are critical for enhancing system resilience. Existing co-optimization methods face the curse of dimensionality when dealing with high-dimensional discrete grid reconfigurations and continuous spatio-temporal EV queuing dynamics. While multi-agent deep reinforcement learning (MADRL) offers real-time responsiveness, it inherently struggles to satisfy strict physical constraints, frequently generating infeasible and unsafe actions. To bridge this gap, this paper proposes a heuristic-guided safe multi-agent reinforcement learning (Safe-MADRL) framework for the resilient dispatch of the energy-mobility nexus. Instead of relying solely on black-box neural networks, the framework structurally embeds physical models and heuristic solvers into the learning loop. A quantum particle swarm optimization (QPSO) algorithm acts as a heuristic action refiner to ensure that grid topology actions strictly comply with non-linear power flow and voltage constraints. Simultaneously, a mixed-integer linear programming (MILP) model coupled with a single-queue multi-server (SQMS) model serves as a safety projection layer. This layer mathematically guarantees EV battery energy continuity and accurately quantifies spatio-temporal queuing delays at charging stations. Case studies on a coupled IEEE 33-node distribution system and a regional transportation network demonstrate that the proposed Safe-MADRL framework achieves zero physical violations during training and significantly outperforms traditional mathematical optimization and pure learning-based methods in computational efficiency, system power loss reduction, and overall operational economy. Full article

Accurate cloud and cloud-shadow segmentation is a crucial step in optical remote sensing image preprocessing, playing a significant role in subsequent applications such as land-cover classification and change detection. However, the complexity of cloud/shadow shapes and noise interference (e.g., snow and ice, buildings, complex backgrounds, and atmospheric optics) make this task challenging. Although existing deep learning methods have achieved remarkable results in cloud segmentation tasks, a better balance between computational efficiency and segmentation accuracy is still needed. Traditional deep learning models have good detail and generalization capabilities due to their local feature extraction ability and spatial invariance, but they are relatively weak in processing global context information, leading to false positives and false negatives in complex scenarios. Encoders based on state space models (such as VMamba) can effectively capture global context through long-range dependency modeling, but there is still room for optimization in computational efficiency. Additionally, complex attention mechanisms (such as CBAM) can improve feature representation capability, but the large number of parameters limits the deployment efficiency of models. This paper conducts a systematic architectural exploration of the MCloudX cloud segmentation network, seeking a balance between efficiency and accuracy from three directions: backbone network modernization, encoder efficiency optimization, and attention mechanism lightweighting. Through comprehensive ablation experiments on SPARCS and L8-Biome datasets, we systematically evaluate the independent and synergistic effects of each component and validate them on Biome_3 and SPARCS datasets. Experimental results show that the proposed optimization configuration (ResNet50+LocalMamba+ECA-Net) significantly improves computational efficiency while maintaining comparable accuracy to the baseline. We name this optimization configuration LECloud, providing valuable empirical references for future research on efficient remote sensing segmentation architectures. Full article

Explainable Artificial Intelligence (XAI) has become a critical requirement in medical image analysis, where transparency and interpretability are essential for clinical trust and decision support. Melanoma is recognized as one of the most deadly types of skin cancer, with its occurrence exhibiting an increasing pattern in recent times. However, detecting this cancer in its initial stages greatly increases patients’ chances of long-term survival. Various computer-based techniques have recently been proposed to diagnose skin lesions at their early stages. Even though the machine learning community has achieved a certain degree of success, there is still an unresolved research challenge regarding high error margins and the limited interpretability of automated systems. This study focuses on addressing both segmentation and classification tasks, with particular emphasis on two key concepts: (1) improving image quality to maximize distinguishability between foreground and background regions, thereby enhancing visual interpretability and segmentation accuracy and (2) eliminating redundant and cluttered feature information to generate the most discriminative and compact feature representations. The input images are initially processed using a novel metaheuristic contrast-stretching method to estimate image-specific key parameters, thereby enhancing lesion boundary clarity in a clinically interpretable manner. Following this, the improved images are fed into selected pre-trained deep models, including DenseNet-201, Inception-ResNet v2, and NASNet-Mobile. The extracted features from all pre-trained models are fused to produce resultant vectors, which are then refined using a bio-inspired feature selection method, termed entropy-controlled whale optimization, to retain only the most informative attributes. The selected discriminative feature set is subsequently classified using multiple classifiers. The results indicate that the proposed framework achieves superior performance compared to existing methods in terms of accuracy, sensitivity, specificity, and F1-score. Additionally, it facilitates a more explainable, transparent, and structured diagnostic pipeline appropriate for medical applications. Full article

Against the backdrop of deepening coordinated policy governance, the systemic synergy between digitalization and green transformation policies and their impact on corporate ESG performance has become a key issue requiring urgent exploration. Unlike existing studies that focus on the effects of individual policies, this paper adopts a policy system synergy framework to systematically investigate the impact of the coordinated implementation of big data administrative reform and low-carbon city pilot policies on corporate ESG performance. Using a sample of Chinese A-share listed companies from 2010 to 2022, this study applies a multi-period difference-in-differences (DID) method for empirical analysis. The findings show that the systemic synergy between digital and green policies significantly enhances corporate ESG performance, with this promoting effect substantially stronger than that of single pilot policies. Further causal re-identification using a double machine learning (DML) approach verifies the robustness of the baseline conclusion. Heterogeneity analysis indicates that the synergistic effect of digital and green policies is more pronounced in firms with higher levels of digital transformation, greater patient capital, and heavier tax burdens. Mechanism tests reveal that digital–green policy synergy improves ESG performance by enhancing external supervision from government, society, and the market, increasing green government subsidies, and incentivizing firms to engage in green innovation. At the same time, policy system synergy also reduces firms’ perceived uncertainty regarding economic policies and stabilizes their expectations, further enhancing ESG performance. This paper extends the research on the determinants of corporate ESG performance from the perspective of system synergy governance, providing new empirical evidence for understanding the interaction mechanisms between digital governance and green transformation policies. Full article

This work solves the problem associated with the lack of analytical models capable of describing the nonlinear vibration behavior of re-entrant hexagonal units when geometric nonlinearity and structural modifications, such as soft filling, are taken into account. The purpose of this study is to develop an analytical framework that enables prediction and control of vibration characteristics, with particular emphasis on achieving low-frequency response and enhanced energy storage and redistribution within the structure. The proposed approach is based on Lagrangian modeling of the unit cell, leading to a nonlinear equation of motion of the Liénard type that admits a first integral. By exploiting the existence of this integral, an approximate analytical expression for the oscillation period is derived using energy-based methods. The analysis is performed for two configurations: an empty unit and a unit filled with a soft material, allowing direct comparison of their dynamic responses. The analytical results are validated through comparison with numerical simulations and available experimental data. A parametric study is conducted to evaluate the influence of the mass ratio and the re-entrant angle on the oscillation period and frequency. Furthermore, the effects of filling mass, stiffness, and degree of filling are systematically investigated, revealing distinct inertia-dominated and stiffness-dominated regimes. The obtained results provide clear design guidelines for tailoring the dynamic response of re-entrant hexagonal units. It is shown that low-frequency vibration and increased capacity for energy storage can be achieved through appropriate selection of geometric parameters and filling properties, with potential applications in vibration control and structural design. Full article