Sign in to use this feature.

Years

Between: -

Subjects

remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline

Journals

remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline

Article Types

Countries / Regions

remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline

Search Results (675)

Search Parameters:
Keywords = dual-baseline

Order results
Result details
Results per page
Select all
Export citation of selected articles as:
18 pages, 6413 KiB  
Article
A Recognition Method for Marigold Picking Points Based on the Lightweight SCS-YOLO-Seg Model
by Baojian Ma, Zhenghao Wu, Yun Ge, Bangbang Chen, He Zhang, Hao Xia and Dongyun Wang
Sensors 2025, 25(15), 4820; https://doi.org/10.3390/s25154820 - 5 Aug 2025
Abstract
Accurate identification of picking points remains a critical challenge for automated marigold harvesting, primarily due to complex backgrounds and significant pose variations of the flowers. To overcome this challenge, this study proposes SCS-YOLO-Seg, a novel method based on a lightweight segmentation model. The [...] Read more.
Accurate identification of picking points remains a critical challenge for automated marigold harvesting, primarily due to complex backgrounds and significant pose variations of the flowers. To overcome this challenge, this study proposes SCS-YOLO-Seg, a novel method based on a lightweight segmentation model. The approach enhances the baseline YOLOv8n-seg architecture by replacing its backbone with StarNet and introducing C2f-Star, a novel lightweight feature extraction module. These modifications achieve substantial model compression, significantly reducing the model size, parameter count, and computational complexity (GFLOPs). Segmentation efficiency is further optimized through a dual-path collaborative architecture (Seg-Marigold head). Following mask extraction, picking points are determined by intersecting the optimized elliptical mask fitting results with the stem skeleton. Experimental results demonstrate that SCS-YOLO-Seg effectively balances model compression with segmentation performance. Compared to YOLOv8n-seg, it maintains high accuracy while significantly reducing resource requirements, achieving a picking point identification accuracy of 93.36% with an average inference time of 28.66 ms per image. This work provides a robust and efficient solution for vision systems in automated marigold harvesting. Full article
(This article belongs to the Section Smart Agriculture)
Show Figures

Figure 1

9 pages, 247 KiB  
Article
Hysterectomy for Benign Gynecologic Disease: A Comparative Study of Articulating Laparoscopic Instruments and Robot-Assisted Surgery in Korea and Taiwan
by Jun-Hyeong Seo, Young Eun Chung, Seongyun Lim, Chel Hun Choi, Tyan-Shin Yang, Yen-Ling Lai, Jung Chen, Kazuyoshi Kato, Yi-Liang Lee, Yu-Li Chen and Yoo-Young Lee
Medicina 2025, 61(8), 1418; https://doi.org/10.3390/medicina61081418 - 5 Aug 2025
Abstract
Background and Objectives: Hysterectomy is a common non-obstetric procedure. Minimally invasive techniques, such as laparoscopy and robot-assisted surgery, have replaced open surgery for benign gynecologic conditions. Robotic surgery offers reduced blood loss and shorter hospital stays but is limited by high costs. [...] Read more.
Background and Objectives: Hysterectomy is a common non-obstetric procedure. Minimally invasive techniques, such as laparoscopy and robot-assisted surgery, have replaced open surgery for benign gynecologic conditions. Robotic surgery offers reduced blood loss and shorter hospital stays but is limited by high costs. Articulating laparoscopic instruments aim to replicate robotic dexterity cost-effectively. However, comparative data on these two approaches in hysterectomy are limited. Materials and Methods: This multicenter study analyzed the outcomes of hysterectomies for benign gynecological diseases using articulating laparoscopic instruments (prospectively recruited) and robot-assisted surgery (retrospectively reviewed). The surgeries were performed by minimally invasive gynecological surgeons in South Korea, Japan, and Taiwan. The baseline characteristics, operative details, and outcomes, including operative time, blood loss, complications, and hospital stay, were compared. Statistical significance was set at p < 0.05. Results: A total of 151 patients were analyzed, including 67 in the articulating laparoscopy group and 84 in the robot-assisted group. The operating times were comparable (114.9 vs. 119.9 min, p = 0.22). The articulating group primarily underwent dual-port surgery (79.1%), whereas the robot-assisted group required four or more ports in 71.4% of the cases (p < 0.001). Postoperative complications occurred in both groups, without a significant difference (9.0% vs. 3.6%, p = 0.17). No severe complications or significant differences in the 30-day readmission rates were observed. Conclusions: Articulating laparoscopic instruments provide outcomes comparable to robot-assisted surgery in hysterectomy while reducing the number of ports required. Further studies are needed to explore the learning curve and long-term impact on surgical outcomes. Full article
(This article belongs to the Special Issue Recent Advances in Gynecological Surgery)
21 pages, 21837 KiB  
Article
Decoding China’s Transport Decarbonization Pathways: An Interpretable Spatio-Temporal Neural Network Approach with Scenario-Driven Policy Implications
by Yanming Sun, Kaixin Liu and Qingli Li
Sustainability 2025, 17(15), 7102; https://doi.org/10.3390/su17157102 - 5 Aug 2025
Abstract
The transportation sector, as a major source of carbon emissions, plays a crucial role in the realization of dual carbon goals worldwide. In this study, an improved least absolute shrinkage and selection operator (LASSO) is used to identify six key factors affecting transportation [...] Read more.
The transportation sector, as a major source of carbon emissions, plays a crucial role in the realization of dual carbon goals worldwide. In this study, an improved least absolute shrinkage and selection operator (LASSO) is used to identify six key factors affecting transportation carbon emissions (TCEs) in China. Aiming at the spatio-temporal characteristics of transportation carbon emissions, a CNN-BiLSTM neural network model is constructed for the first time for prediction, and an improved whale optimization algorithm (EWOA) is introduced for hyperparameter optimization, finding that the prediction model combining spatio-temporal characteristics has a more significant prediction accuracy, and scenario forecasting was carried out using the prediction model. Research indicates that over the past three decades, TCEs have demonstrated a rapid growth trend. Under the baseline, green, low-carbon, and high-carbon scenarios, peak carbon emissions are expected in 2035, 2031, 2030, and 2040. The adoption of a low-carbon scenario represents the most advantageous pathway for the sustainable progression of China’s transportation sector. Consequently, it is imperative for China to accelerate the formulation and implementation of low-carbon policies, promote the application of clean energy and facilitate the green transformation of the transportation sector. These efforts will contribute to the early realization of dual-carbon goals with a positive impact on global sustainable development. Full article
Show Figures

Figure 1

23 pages, 3055 KiB  
Article
RDPNet: A Multi-Scale Residual Dilated Pyramid Network with Entropy-Based Feature Fusion for Epileptic EEG Classification
by Tongle Xie, Wei Zhao, Yanyouyou Liu and Shixiao Xiao
Entropy 2025, 27(8), 830; https://doi.org/10.3390/e27080830 (registering DOI) - 5 Aug 2025
Abstract
Epilepsy is a prevalent neurological disorder affecting approximately 50 million individuals worldwide. Electroencephalogram (EEG) signals play a vital role in the diagnosis and analysis of epileptic seizures. However, traditional machine learning techniques often rely on handcrafted features, limiting their robustness and generalizability across [...] Read more.
Epilepsy is a prevalent neurological disorder affecting approximately 50 million individuals worldwide. Electroencephalogram (EEG) signals play a vital role in the diagnosis and analysis of epileptic seizures. However, traditional machine learning techniques often rely on handcrafted features, limiting their robustness and generalizability across diverse EEG acquisition settings, seizure types, and patients. To address these limitations, we propose RDPNet, a multi-scale residual dilated pyramid network with entropy-guided feature fusion for automated epileptic EEG classification. RDPNet combines residual convolution modules to extract local features and a dilated convolutional pyramid to capture long-range temporal dependencies. A dual-pathway fusion strategy integrates pooled and entropy-based features from both shallow and deep branches, enabling robust representation of spatial saliency and statistical complexity. We evaluate RDPNet on two benchmark datasets: the University of Bonn and TUSZ. On the Bonn dataset, RDPNet achieves 99.56–100% accuracy in binary classification, 99.29–99.79% in ternary tasks, and 95.10% in five-class classification. On the clinically realistic TUSZ dataset, it reaches a weighted F1-score of 95.72% across seven seizure types. Compared with several baselines, RDPNet consistently outperforms existing approaches, demonstrating superior robustness, generalizability, and clinical potential for epileptic EEG analysis. Full article
(This article belongs to the Special Issue Complexity, Entropy and the Physics of Information II)
Show Figures

Figure 1

32 pages, 2102 KiB  
Article
D* Lite and Transformer-Enhanced SAC: A Hybrid Reinforcement Learning Framework for COLREGs-Compliant Autonomous Navigation in Dynamic Maritime Environments
by Tianqing Chen, Yamei Lan, Yichen Li, Jiesen Zhang and Yijie Yin
J. Mar. Sci. Eng. 2025, 13(8), 1498; https://doi.org/10.3390/jmse13081498 - 4 Aug 2025
Abstract
Autonomous navigation in dynamic, multi-vessel maritime environments presents a formidable challenge, demanding strict adherence to the International Regulations for Preventing Collisions at Sea (COLREGs). Conventional approaches often struggle with the dual imperatives of global path optimality and local reactive safety, and they frequently [...] Read more.
Autonomous navigation in dynamic, multi-vessel maritime environments presents a formidable challenge, demanding strict adherence to the International Regulations for Preventing Collisions at Sea (COLREGs). Conventional approaches often struggle with the dual imperatives of global path optimality and local reactive safety, and they frequently rely on simplistic state representations that fail to capture complex spatio-temporal interactions among vessels. We introduce a novel hybrid reinforcement learning framework, D* Lite + Transformer-Enhanced Soft Actor-Critic (TE-SAC), to overcome these limitations. This hierarchical framework synergizes the strengths of global and local planning. An enhanced D* Lite algorithm generates efficient, long-horizon reference paths at the global level. At the local level, the TE-SAC agent performs COLREGs-compliant tactical maneuvering. The core innovation resides in TE-SAC’s synergistic state encoder, which uniquely combines a Graph Neural Network (GNN) to model the instantaneous spatial topology of vessel encounters with a Transformer encoder to capture long-range temporal dependencies and infer vessel intent. Comprehensive simulations demonstrate the framework’s superior performance, validating the strengths of both planning layers. At the local level, our TE-SAC agent exhibits remarkable tactical intelligence, achieving an exceptional 98.7% COLREGs compliance rate and reducing energy consumption by 15–20% through smoother, more decisive maneuvers. This high-quality local control, guided by the efficient global paths from the enhanced D* Lite algorithm, culminates in a 10–32 percentage point improvement in overall task success rates compared to state-of-the-art baselines. This work presents a robust, verifiable, and efficient framework. By demonstrating superior performance and compliance with rules in high-fidelity simulations, it lays a crucial foundation for advancing the practical application of intelligent autonomous navigation systems. Full article
(This article belongs to the Special Issue Motion Control and Path Planning of Marine Vehicles—3rd Edition)
Show Figures

Figure 1

15 pages, 682 KiB  
Article
Hypergraph-Driven High-Order Knowledge Tracing with a Dual-Gated Dynamic Mechanism
by Fanglan Ma, Changsheng Zhu and Peng Lei
Appl. Sci. 2025, 15(15), 8617; https://doi.org/10.3390/app15158617 (registering DOI) - 4 Aug 2025
Abstract
Knowledge tracing (KT), a core educational data mining task, models students’ evolving knowledge states to predict future learning. In online education systems, the exercises are numerous, but they are typically associated with only a few concepts. However, existing models rarely integrate exercise information [...] Read more.
Knowledge tracing (KT), a core educational data mining task, models students’ evolving knowledge states to predict future learning. In online education systems, the exercises are numerous, but they are typically associated with only a few concepts. However, existing models rarely integrate exercise information with high-order exercise–concept correlations, focusing solely on optimizing models’ final predictive performance. To address these limitations, we propose the Hypergraph-Driven High-Order Knowledge Tracing with a Dual-Gated Dynamic Mechanism (HGKT), a novel framework that (1) captures correlations between exercises and concepts through a two-layer hypergraph convolution; (2) integrates hypergraph-driven exercise embedding and temporal features (answer time and interval time) to characterize learning behavioral dynamics; and (3) designs a learning layer and a forgetting layer, with the dual-gating mechanism dynamically balancing their impacts on the knowledge state. Experiments on three public datasets demonstrate that the proposed HGKT model achieves superior predictive performance compared to all baselines. On the longest interaction sequence dataset, ASSISChall, HGKT improves prediction AUC by least 1.8%. On the biggest interaction records dataset, EdNet-KT1, it maintains a state-of-the-art AUC of 0.78372. Visualization analyses confirm its interpretability in tracing knowledge state evolution. These results validate HGKT’s effectiveness in modeling high-order exercise–concept correlations while ensuring practical adaptability in real-world online education platforms. Full article
Show Figures

Figure 1

28 pages, 845 KiB  
Article
Place Identity and Environmental Conservation in Heritage Tourism: Extending the Theory of Planned Behavior to Iranian Rural Heritage Villages
by Zabih-Allah Torabi, Mohammad Reza Rezvani, Colin Michael Hall, Pantea Davani and Boshra Bakhshaei
Tour. Hosp. 2025, 6(3), 150; https://doi.org/10.3390/tourhosp6030150 - 4 Aug 2025
Abstract
This study examines the determinants of environmentally responsible behavior among tourists in the heritage villages of Paveh County, Iran, through an integrated theoretical framework that synthesizes place-related psychological constructs with the Theory of Planned Behavior (TPB). Employing structural equation modeling on data collected [...] Read more.
This study examines the determinants of environmentally responsible behavior among tourists in the heritage villages of Paveh County, Iran, through an integrated theoretical framework that synthesizes place-related psychological constructs with the Theory of Planned Behavior (TPB). Employing structural equation modeling on data collected from 443 tourists across three heritage villages (July–November 2024), the investigation tested comparative theoretical models with differing explanatory capacities. The baseline TPB model confirmed significant positive effects of environmental attitudes (β = 0.388), environmental norms (β = 0.398), and perceived behavioral control (β = 0.547) on behavioral intentions, which subsequently influenced environmental behavior (β = 0.561). The extended model incorporating place-related variables demonstrated enhanced explanatory power, with the R2 values increasing from 48.2% to 52.7% for behavioral intentions and from 49.2% to 54.7% for actual behavior. Notably, place identity exhibited dual psychological functions: moderating the intention–behavior relationship (β = 0.155) and mediating between place attachment and environmental behavior (β = 0.163). These findings advance sustainable tourism theory by illuminating the complex pathways through which place-based psychological connections influence environmental behavior formation in heritage contexts, suggesting that more sophisticated theoretical frameworks are required for understanding and promoting sustainable practices in culturally significant destinations. Full article
Show Figures

Figure 1

23 pages, 22135 KiB  
Article
Road Marking Damage Degree Detection Based on Boundary Features Enhanced and Asymmetric Large Field-of-View Contextual Features
by Zheng Wang, Ryojun Ikeura, Soichiro Hayakawa and Zhiliang Zhang
J. Imaging 2025, 11(8), 259; https://doi.org/10.3390/jimaging11080259 - 4 Aug 2025
Abstract
Road markings, as critical components of transportation infrastructure, are crucial for ensuring traffic safety. Accurate quantification of their damage severity is vital for effective maintenance prioritization. However, existing methods are limited to detecting the presence of damage without assessing its extent. To address [...] Read more.
Road markings, as critical components of transportation infrastructure, are crucial for ensuring traffic safety. Accurate quantification of their damage severity is vital for effective maintenance prioritization. However, existing methods are limited to detecting the presence of damage without assessing its extent. To address this limitation, we propose a novel segmentation-based framework for estimating the degree of road marking damage. The method comprises two stages: segmentation of residual pixels from the damaged markings and segmentation of the intact markings region. This dual-segmentation strategy enables precise reconstruction and comparison for severity estimation. To enhance segmentation performance, we proposed two key modules: the Asymmetric Large Field-of-View Contextual (ALFVC) module, which captures rich multi-scale contextual features, and the supervised Boundary Feature Enhancement (BFE) module, which strengthens shape representation and boundary accuracy. The experimental results demonstrate that our method achieved an average segmentation accuracy of 89.44%, outperforming the baseline by 5.86 percentage points. Moreover, the damage quantification achieved a minimum error rate of just 0.22% on the proprietary dataset. The proposed approach was both effective and lightweight, providing valuable support for automated maintenance planning, and significantly improving the efficiency and precision of road marking management. Full article
(This article belongs to the Section Image and Video Processing)
Show Figures

Figure 1

16 pages, 1928 KiB  
Review
Intensive Lipid-Lowering Therapy Following Acute Coronary Syndrome: The Earlier the Better
by Akshyaya Pradhan, Prachi Sharma, Sudesh Prajapathi, Maurizio Aracri, Ferdinando Iellamo and Marco Alfonso Perrone
J. Cardiovasc. Dev. Dis. 2025, 12(8), 300; https://doi.org/10.3390/jcdd12080300 - 4 Aug 2025
Abstract
Elevated levels of atherogenic lipoproteins are known to be associated with an increased risk of incident and recurrent cardiovascular events. Knowing that the immediate post-acute coronary syndrome (ACS) period is associated with the maximum risk of recurrent events, the gradual escalation of therapy [...] Read more.
Elevated levels of atherogenic lipoproteins are known to be associated with an increased risk of incident and recurrent cardiovascular events. Knowing that the immediate post-acute coronary syndrome (ACS) period is associated with the maximum risk of recurrent events, the gradual escalation of therapy allows the patient to remain above the targets during the most vulnerable period. In addition, the percentage of lipid-lowering levels for each class of drugs is predictable and has a ceiling. Hence, it is prudent to immediately start with a combination of lipid-lowering drugs following ACS according to the baseline lipid levels. Multiple studies with injectable lipid-lowering agents (PCSK9 inhibitors) such as EVOPACS, PACMAN MI, and HUYGENS MI have shown the feasibility of achieving LDL-C goals by day 28 and beneficial plaque modification in non-infarct-related coronary arteries. Recently, a study from India demonstrated that an upfront triple combination of oral lipid-lowering agents was able to achieve LDL-C goals in a majority of patients in the early post-ACS period. This notion is also supported by a few recent lipid-lowering guidelines advocating for an upfront dual combination of a high-intensity statin and ezetimibe following ACS. Henceforth, the goal should not only be the achievement of lipid targets but also their early achievement. However, the impact of this strategy on long-term cardiovascular outcomes is yet to be ascertained. Full article
(This article belongs to the Special Issue Effect of Lipids and Lipoproteins on Atherosclerosis)
Show Figures

Figure 1

17 pages, 1455 KiB  
Article
STID-Mixer: A Lightweight Spatio-Temporal Modeling Framework for AIS-Based Vessel Trajectory Prediction
by Leiyu Wang, Jian Zhang, Guangyin Jin and Xinyu Dong
Eng 2025, 6(8), 184; https://doi.org/10.3390/eng6080184 - 3 Aug 2025
Viewed by 56
Abstract
The Automatic Identification System (AIS) has become a key data source for ship behavior monitoring and maritime traffic management, widely used in trajectory prediction and anomaly detection. However, AIS data suffer from issues such as spatial sparsity, heterogeneous features, variable message formats, and [...] Read more.
The Automatic Identification System (AIS) has become a key data source for ship behavior monitoring and maritime traffic management, widely used in trajectory prediction and anomaly detection. However, AIS data suffer from issues such as spatial sparsity, heterogeneous features, variable message formats, and irregular sampling intervals, while vessel trajectories are characterized by strong spatial–temporal dependencies. These factors pose significant challenges for efficient and accurate modeling. To address this issue, we propose a lightweight vessel trajectory prediction framework that integrates Spatial–Temporal Identity encoding with an MLP-Mixer architecture. The framework discretizes spatial and temporal features into structured IDs and uses dual MLP modules to model temporal dependencies and feature interactions without relying on convolution or attention mechanisms. Experiments on a large-scale real-world AIS dataset demonstrate that the proposed STID-Mixer achieves superior accuracy, training efficiency, and generalization capability compared to representative baseline models. The method offers a compact and deployable solution for large-scale maritime trajectory modeling. Full article
Show Figures

Figure 1

21 pages, 4314 KiB  
Article
Panoptic Plant Recognition in 3D Point Clouds: A Dual-Representation Learning Approach with the PP3D Dataset
by Lin Zhao, Sheng Wu, Jiahao Fu, Shilin Fang, Shan Liu and Tengping Jiang
Remote Sens. 2025, 17(15), 2673; https://doi.org/10.3390/rs17152673 - 2 Aug 2025
Viewed by 189
Abstract
The advancement of Artificial Intelligence (AI) has significantly accelerated progress across various research domains, with growing interest in plant science due to its substantial economic potential. However, the integration of AI with digital vegetation analysis remains underexplored, largely due to the absence of [...] Read more.
The advancement of Artificial Intelligence (AI) has significantly accelerated progress across various research domains, with growing interest in plant science due to its substantial economic potential. However, the integration of AI with digital vegetation analysis remains underexplored, largely due to the absence of large-scale, real-world plant datasets, which are crucial for advancing this field. To address this gap, we introduce the PP3D dataset—a meticulously labeled collection of about 500 potted plants represented as 3D point clouds, featuring fine-grained annotations for approximately 20 species. The PP3D dataset provides 3D phenotypic data for about 20 plant species spanning model organisms (e.g., Arabidopsis thaliana), potted plants (e.g., Foliage plants, Flowering plants), and horticultural plants (e.g., Solanum lycopersicum), covering most of the common important plant species. Leveraging this dataset, we propose the panoptic plant recognition task, which combines semantic segmentation (stems and leaves) with leaf instance segmentation. To tackle this challenge, we present SCNet, a novel dual-representation learning network designed specifically for plant point cloud segmentation. SCNet integrates two key branches: a cylindrical feature extraction branch for robust spatial encoding and a sequential slice feature extraction branch for detailed structural analysis. By efficiently propagating features between these representations, SCNet achieves superior flexibility and computational efficiency, establishing a new baseline for panoptic plant recognition and paving the way for future AI-driven research in plant science. Full article
Show Figures

Figure 1

20 pages, 15898 KiB  
Article
Design of a Humanoid Upper-Body Robot and Trajectory Tracking Control via ZNN with a Matrix Derivative Observer
by Hong Yin, Hongzhe Jin, Yuchen Peng, Zijian Wang, Jiaxiu Liu, Fengjia Ju and Jie Zhao
Biomimetics 2025, 10(8), 505; https://doi.org/10.3390/biomimetics10080505 - 2 Aug 2025
Viewed by 170
Abstract
Humanoid robots have attracted considerable attention for their anthropomorphic structure, extended workspace, and versatile capabilities. This paper presents a novel humanoid upper-body robotic system comprising a pair of 8-degree-of-freedom (DOF) arms, a 3-DOF head, and a 3-DOF torso—yielding a 22-DOF architecture inspired by [...] Read more.
Humanoid robots have attracted considerable attention for their anthropomorphic structure, extended workspace, and versatile capabilities. This paper presents a novel humanoid upper-body robotic system comprising a pair of 8-degree-of-freedom (DOF) arms, a 3-DOF head, and a 3-DOF torso—yielding a 22-DOF architecture inspired by human biomechanics and implemented via standardized hollow joint modules. To overcome the critical reliance of zeroing neural network (ZNN)-based trajectory tracking on the Jacobian matrix derivative, we propose an integration-enhanced matrix derivative observer (IEMDO) that incorporates nonlinear feedback and integral correction. The observer is theoretically proven to ensure asymptotic convergence and enables accurate, real-time estimation of matrix derivatives, addressing a fundamental limitation in conventional ZNN solvers. Workspace analysis reveals that the proposed design achieves an 87.7% larger total workspace and a remarkable 3.683-fold expansion in common workspace compared to conventional dual-arm baselines. Furthermore, the observer demonstrates high estimation accuracy for high-dimensional matrices and strong robustness to noise. When integrated into the ZNN controller, the IEMDO achieves high-precision trajectory tracking in both simulation and real-world experiments. The proposed framework provides a practical and theoretically grounded approach for redundant humanoid arm control. Full article
(This article belongs to the Special Issue Bio-Inspired and Biomimetic Intelligence in Robotics: 2nd Edition)
Show Figures

Figure 1

22 pages, 24173 KiB  
Article
ScaleViM-PDD: Multi-Scale EfficientViM with Physical Decoupling and Dual-Domain Fusion for Remote Sensing Image Dehazing
by Hao Zhou, Yalun Wang, Wanting Peng, Xin Guan and Tao Tao
Remote Sens. 2025, 17(15), 2664; https://doi.org/10.3390/rs17152664 - 1 Aug 2025
Viewed by 172
Abstract
Remote sensing images are often degraded by atmospheric haze, which not only reduces image quality but also complicates information extraction, particularly in high-level visual analysis tasks such as object detection and scene classification. State-space models (SSMs) have recently emerged as a powerful paradigm [...] Read more.
Remote sensing images are often degraded by atmospheric haze, which not only reduces image quality but also complicates information extraction, particularly in high-level visual analysis tasks such as object detection and scene classification. State-space models (SSMs) have recently emerged as a powerful paradigm for vision tasks, showing great promise due to their computational efficiency and robust capacity to model global dependencies. However, most existing learning-based dehazing methods lack physical interpretability, leading to weak generalization. Furthermore, they typically rely on spatial features while neglecting crucial frequency domain information, resulting in incomplete feature representation. To address these challenges, we propose ScaleViM-PDD, a novel network that enhances an SSM backbone with two key innovations: a Multi-scale EfficientViM with Physical Decoupling (ScaleViM-P) module and a Dual-Domain Fusion (DD Fusion) module. The ScaleViM-P module synergistically integrates a Physical Decoupling block within a Multi-scale EfficientViM architecture. This design enables the network to mitigate haze interference in a physically grounded manner at each representational scale while simultaneously capturing global contextual information to adaptively handle complex haze distributions. To further address detail loss, the DD Fusion module replaces conventional skip connections by incorporating a novel Frequency Domain Module (FDM) alongside channel and position attention. This allows for a more effective fusion of spatial and frequency features, significantly improving the recovery of fine-grained details, including color and texture information. Extensive experiments on nine publicly available remote sensing datasets demonstrate that ScaleViM-PDD consistently surpasses state-of-the-art baselines in both qualitative and quantitative evaluations, highlighting its strong generalization ability. Full article
Show Figures

Figure 1

37 pages, 7429 KiB  
Article
Study on the Influence of Window Size on the Thermal Comfort of Traditional One-Seal Dwellings (Yikeyin) in Kunming Under Natural Wind
by Yaoning Yang, Junfeng Yin, Jixiang Cai, Xinping Wang and Juncheng Zeng
Buildings 2025, 15(15), 2714; https://doi.org/10.3390/buildings15152714 - 1 Aug 2025
Viewed by 153
Abstract
Under the dual challenges of global energy crisis and climate change, the building sector, as a major carbon emitter consuming 33% of global primary energy, has seen its energy efficiency optimization become a critical pathway towards achieving carbon neutrality goals. The Window-to-Wall Ratio [...] Read more.
Under the dual challenges of global energy crisis and climate change, the building sector, as a major carbon emitter consuming 33% of global primary energy, has seen its energy efficiency optimization become a critical pathway towards achieving carbon neutrality goals. The Window-to-Wall Ratio (WWR), serving as a core parameter in building envelope design, directly influences building energy consumption, with its optimized design playing a decisive role in balancing natural daylighting, ventilation efficiency, and thermal comfort. This study focuses on the traditional One-Seal dwellings (Yikeyin) in Kunming, China, establishing a dynamic wind field-thermal environment coupled analysis framework to investigate the impact mechanism of window dimensions (WWR and aspect ratio) on indoor thermal comfort under natural wind conditions in transitional climate zones. Utilizing the Grasshopper platform integrated with Ladybug, Honeybee, and Butterfly plugins, we developed parametric models incorporating Kunming’s Energy Plus Weather meteorological data. EnergyPlus and OpenFOAM were employed, respectively, for building heat-moisture balance calculations and Computational Fluid Dynamic (CFD) simulations, with particular emphasis on analyzing the effects of varying WWR (0.05–0.20) on temperature-humidity, air velocity, and ventilation efficiency during typical winter and summer weeks. Key findings include, (1) in summer, the baseline scenario with WWR = 0.1 achieves a dynamic thermal-humidity balance (20.89–24.27 °C, 65.35–74.22%) through a “air-permeable but non-ventilative” strategy, though wing rooms show humidity-heat accumulation risks; increasing WWR to 0.15–0.2 enhances ventilation efficiency (2–3 times higher air changes) but causes a 4.5% humidity surge; (2) winter conditions with WWR ≥ 0.15 reduce wing room temperatures to 17.32 °C, approaching cold thresholds, while WWR = 0.05 mitigates heat loss but exacerbates humidity accumulation; (3) a symmetrical layout structurally constrains central ventilation, maintaining main halls air changes below one Air Change per Hour (ACH). The study proposes an optimized WWR range of 0.1–0.15 combined with asymmetric window opening strategies, providing quantitative guidance for validating the scientific value of vernacular architectural wisdom in low-energy design. Full article
Show Figures

Figure 1

16 pages, 4587 KiB  
Article
FAMNet: A Lightweight Stereo Matching Network for Real-Time Depth Estimation in Autonomous Driving
by Jingyuan Zhang, Qiang Tong, Na Yan and Xiulei Liu
Symmetry 2025, 17(8), 1214; https://doi.org/10.3390/sym17081214 - 1 Aug 2025
Viewed by 212
Abstract
Accurate and efficient stereo matching is fundamental to real-time depth estimation from symmetric stereo cameras in autonomous driving systems. However, existing high-accuracy stereo matching networks typically rely on computationally expensive 3D convolutions, which limit their practicality in real-world environments. In contrast, real-time methods [...] Read more.
Accurate and efficient stereo matching is fundamental to real-time depth estimation from symmetric stereo cameras in autonomous driving systems. However, existing high-accuracy stereo matching networks typically rely on computationally expensive 3D convolutions, which limit their practicality in real-world environments. In contrast, real-time methods often sacrifice accuracy or generalization capability. To address these challenges, we propose FAMNet (Fusion Attention Multi-Scale Network), a lightweight and generalizable stereo matching framework tailored for real-time depth estimation in autonomous driving applications. FAMNet consists of two novel modules: Fusion Attention-based Cost Volume (FACV) and Multi-scale Attention Aggregation (MAA). FACV constructs a compact yet expressive cost volume by integrating multi-scale correlation, attention-guided feature fusion, and channel reweighting, thereby reducing reliance on heavy 3D convolutions. MAA further enhances disparity estimation by fusing multi-scale contextual cues through pyramid-based aggregation and dual-path attention mechanisms. Extensive experiments on the KITTI 2012 and KITTI 2015 benchmarks demonstrate that FAMNet achieves a favorable trade-off between accuracy, efficiency, and generalization. On KITTI 2015, with the incorporation of FACV and MAA, the prediction accuracy of the baseline model is improved by 37% and 38%, respectively, and a total improvement of 42% is achieved by our final model. These results highlight FAMNet’s potential for practical deployment in resource-constrained autonomous driving systems requiring real-time and reliable depth perception. Full article
Show Figures

Figure 1

Back to TopTop