Search Results (1,284)

This study proposes a scenario-based conceptual model for transforming underground parking structures into sustainable interior green spaces, directly addressing two core research dimensions: energy efficiency and user experience. The originality of the research lies in repositioning subterranean spaces—often overlooked in urban planning—as climate-responsive, multi-functional public environments. Using a site-specific case in downtown Rize, Türkiye, three design scenarios—passive green walls, active modular systems, and experimental micro-farming—were comparatively analyzed. These scenarios were assessed through AI-assisted simulations and climate-based performance evaluations in terms of environmental benefits, thermal regulation, carbon reduction, and experiential quality. Underground space leads to green design interventions, which in turn generate environmental, energy, and social benefits. The results demonstrate that passive systems provide cost-effective improvements, active modular systems achieve balanced performance, and experimental micro-farming yields the highest ecological and social benefits. The study uniquely contributes to urban sustainable design by integrating climate-adaptive strategies, biophilic design principles, and AI-supported visualization into the transformation of underground structures. This research not only advances academic discourse but also provides policy-relevant insights for local governments, developers, and communities in the context of urban renewal. Full article

Although GPUs are widely used in Computer-Generated Holography (CGH), their specific application to concrete problems such as occlusion or speckle filtering through temporal multiplexing is not yet standardized and has not been fully explored. This work aims to optimize the software architecture by taking the GPU architecture into account in a novel way for these particular tasks. We present an optimized algorithm for CGH computation that provides a joint solution to the problems of speckle noise and occlusion. The workflow includes the generation and illumination of a 3D scene, the calculation of the CGH including color, occlusion, and temporal speckle-noise filtering, followed by scene reconstruction through both simulation and experimental methods. The research focuses on implementing a temporal multiplexing technique that simultaneously performs speckle denoising and occlusion culling for point clouds, evaluating two types of occlusion that differ in whether the occlusion effect dominates over the depth effect in a scene stored in a CGH, while leveraging the parallel processing capabilities of GPUs to achieve a more immersive and high-quality visual experience. To this end, the total computational cost associated with generating color and occlusion CGHs is evaluated, quantifying the relative contribution of each factor. The results indicate that, under strict occlusion conditions, temporal multiplexing filtering does not significantly impact the overall computational cost of CGH calculation. Full article

Objective: Medical Visual Question Answering (Med-VQA), a key technology that integrates computer vision and natural language processing to assist in clinical diagnosis, possesses significant potential for enhancing diagnostic efficiency and accuracy. However, its development is constrained by two major bottlenecks: weak few-shot generalization capability stemming from the scarcity of high-quality annotated data and the problem of catastrophic forgetting when continually learning new knowledge. Existing research has largely addressed these two challenges in isolation, lacking a unified framework. Methods: To bridge this gap, this paper proposes a novel Evolvable Clinical-Semantic Alignment (ECSA) framework, designed to synergistically solve these two challenges within a single architecture. ECSA is built upon powerful pre-trained vision (BiomedCLIP) and language (Flan-T5) models, with two innovative modules at its core. First, we design a Clinical-Semantic Disambiguation Module (CSDM), which employs a novel debiased hard negative mining strategy for contrastive learning. This enables the precise discrimination of “hard negatives” that are visually similar but clinically distinct, thereby significantly enhancing the model’s representation ability in few-shot and long-tail scenarios. Second, we introduce a Prompt-based Knowledge Consolidation Module (PKC), which acts as a rehearsal-free non-parametric knowledge store. It consolidates historical knowledge by dynamically accumulating and retrieving task-specific “soft prompts,” thus effectively circumventing catastrophic forgetting without relying on past data. Results: Extensive experimental results on four public benchmark datasets, VQA-RAD, SLAKE, PathVQA, and VQA-Med-2019, demonstrate ECSA’s state-of-the-art or highly competitive performance. Specifically, ECSA achieves excellent overall accuracies of 80.15% on VQA-RAD and 85.10% on SLAKE, while also showing strong generalization with 64.57% on PathVQA and 82.23% on VQA-Med-2019. More critically, in continual learning scenarios, the framework achieves a low forgetting rate of just 13.50%, showcasing its significant advantages in knowledge retention. Conclusions: These findings validate the framework’s substantial potential for building robust and evolvable clinical decision support systems. Full article

This study provides a comprehensive review of ultrasonic welding research in molecular dynamics simulations, encompassing the latest advancements by scholars worldwide. Compared to traditional welding methods, ultrasonic welding offers advantages such as faster processing speed, higher mechanical strength, and environmentally friendly characteristics. However, its process parameters are subject to multiple influencing factors. Molecular dynamics simulations enable the detailed visualization of material interactions and structural changes at atomic/molecular levels during ultrasonic welding. These simulations not only predict how different process parameters affect weld quality but also facilitate the rapid identification of viable solutions, thereby reducing experimental iterations and lowering R&D costs. This review delves into the core theoretical issues pertaining to ultrasonic welding, providing robust support for practical applications. Additionally, specific optimization strategies are proposed to enhance welding performance and efficiency, promoting sustainable development in related industries. Future research could focus on exploring ultrasonic welding mechanisms under complex structures and multi-component systems. Full article

Emergency evacuation signage in metro stations plays a critical role in guiding occupants to evacuate quickly and safely. However, variations in placement height and other display attributes can affect the perceptual efficiency of signage. This study takes a metro station in Changsha, China, as an example and constructs two virtual disaster scenarios—fire and flood. An eye-tracking experiment was designed and conducted, yielding 164 valid experimental samples (89 fire, 75 flood). We compared the visual effectiveness of signage at three heights: low (0–0.8 m), medium (0.8–2 m), and high (>2 m). The results indicate that (1) low-position signage exhibits superior immediacy and should be prioritized for emergency response; (2) medium-position signage strikes a balance between perceived importance and immediacy, serving effectively as central nodes for both routine and emergency purposes; (3) high-position signage presents significant advantages in perceived importance and is suitable for conveying comprehensive, multi-level evacuation information. This research provides empirical evidence for optimizing the spatial layout of emergency evacuation signage in metro stations, offering valuable guidance for enhancing emergency evacuation capabilities in subway environments. Full article

In recent years, SLAM, visual odometry, and structure-from-motion approaches have widely addressed the problems of 3D reconstruction and ego-motion estimation. Of the many input modalities that can be used to solve these ill-posed problems, the pure visual alternative using a single monocular RGB camera has attracted the attention of multiple researchers due to its low cost and widespread availability in handheld devices. One of the best proposals currently available is the Direct Sparse Odometry (DSO) system, which has demonstrated the ability to accurately recover trajectories and depth maps using monocular sequences as the only source of information. Given the impressive advances in single-image depth estimation using neural networks, this work proposes an extension of the DSO system, named DeepDSO. DeepDSO effectively integrates the state-of-the-art NeW CRF neural network as a depth estimation module, providing depth prior information for each candidate point. This reduces the point search interval over the epipolar line. This integration improves the DSO algorithm’s depth point initialization and allows each proposed point to converge faster to its true depth. Experimentation carried out in the TUM-Mono dataset demonstrated that adding the neural network depth estimation module to the DSO pipeline significantly reduced rotation, translation, scale, start-segment alignment, end-segment alignment, and RMSE errors. Full article

This study examines the effects of hailstorms on photovoltaic (PV) modules, focussing on damage mechanisms, testing standards, numerical simulations, damage detection techniques, and mitigation strategies. A comprehensive review of the recent literature (2017–2025), experimental results, and case studies is complemented by advanced simulation methods such as finite element analysis (FEA) and smoothed particle hydrodynamics (SPH). The research emphasises the crucial role of protective glass thickness, cell type, number of busbars, and quality of lamination in improving hail resistance. While international standards such as IEC 61215 specify test protocols, actual hail events often exceed these conditions, leading to glass breakage, micro-cracks, and electrical faults. Numerical simulations confirm that thicker glass and optimised module designs significantly reduce damage and power loss. Detection methods, including visual inspection, thermal imaging, electroluminescence, and AI-driven imaging, enable rapid identification of both visible and hidden damage. The study also addresses the financial risks associated with hail damage and emphasises the importance of insurance and preventative measures. Recommendations include the use of certified, robust modules, protective covers, optimised installation angles, and regular inspections to mitigate the effects of hail. Future research should develop lightweight, impact-resistant materials, improve simulation modelling to better reflect real-world hail conditions, and improve AI-based damage detection in conjunction with drone inspections. This integrated approach aims to improve the durability and reliability of PV modules in hail-prone regions and support the sustainable use of solar energy amidst increasing climatic challenges. Full article

Despite the extensive cultivation of Moringa trees in tropical regions, understanding of accession-specific performance across diverse agroecological zones remains inadequate. Thus, this study evaluated the growth, adaptability, and leaf yield performance of 12 Moringa accessions (11 M. oleifera and 1 M. stenopetala) over three years in a subtropical climate (Pretoria, South Africa). Seeds were planted in seedling trays in the glasshouse at the University of Pretoria’s experimental farm. Vigorous seedlings were transplanted to the field at the Roodeplaat experimental site of the Agricultural Research Council two months after establishment, following a randomized complete block design (RCBD). Data were measured on establishment (emergence, survival), growth and yield parameters, and monitored plant health via leaf greenness, vigour, chlorosis, and pest and disease incidence. Accessions exhibited substantial variation for most traits, except for stem diameter. Moringa stenopetala showed the highest initial emergence rate but later displayed lower survival rates than most M. oleifera accessions. Survival rates, morphological features (plant height, canopy diameter, and branching), visual scores for leaf greenness and plant vigour, and leaf yield (fresh and dry) varied considerably among the accessions. Moringa oleifera A2 consistently performed well, exhibiting vigorous growth, the maximum survival rate (78%), and fresh leaf production (6206 kg ha⁻¹). Accessions A3 and A8 showed intermediate yield and longevity, indicating potential for cultivation or breeding. Conversely, M. oleifera A10 and M. stenopetala markedly underperformed in most traits, limiting their cultivation potential. Based on multi-year performance, A2 is suggested for large-scale cultivation due to its vigour, yield, and stress tolerance, while A3 and A8 hold breeding potential. The study emphasizes the critical role of genetic variation and selection in enhancing Moringa productivity under subtropical environments. Future work should focus on genetic characterization and agronomic practices optimization of superior accessions. Full article

Urban soundscapes are complex due to the interaction of different sound sources and the influence of structures on sound propagation. Moreover, the dynamic nature of sounds over time and space adds to this complexity. Virtual reality (VR) has emerged as a powerful tool to simulate acoustic and visual environments, offering users an immersive sense of presence in controlled settings. This technology facilitates more accurate and predictive assessment of urban environments. It serves as a flexible tool for exploring, analyzing, and interpreting them under repeatable conditions. This study presents a systematic literature review focusing on research that integrates VR technology for the audiovisual reconstruction of urban environments. This topic remains relatively underrepresented in the existing literature. A total of 69 peer-reviewed studies were analyzed in this systematic review. The studies were classified according to research goals, selected urban environments, VR technologies used, technical equipment, and experimental setups. In this study, the relationship between the tools used in urban VR representations is examined, and experimental setups are discussed from both technical and perceptual perspectives. This paper highlights existing challenges and opportunities in using VR to assess soundscapes and offers practical insights for future applications of VR in urban environments. Full article

RipenessGAN is a novel Generative Adversarial Network (GAN) designed to generate synthetic images across different ripeness stages of jujubes (green fruit, white ripe fruit, semi-red fruit, and fully red fruit), aiming to provide balanced training data for diverse applications beyond classification accuracy. This study addresses the problem of data imbalance by augmenting each ripeness stage using our proposed Growth Day Embedding mechanism, thereby enhancing the performance of downstream classification models. The core innovation of RipenessGAN lies in its ability to capture continuous temporal transitions among discrete ripeness classes by incorporating fine-grained growth day information (0–56 days) in addition to traditional class labels. The experimental results show that RipenessGAN produces synthetic data with higher visual quality and greater diversity compared to CycleGAN. Furthermore, the classification models trained on the enriched dataset exhibit more consistent and accurate performance. We also conducted comprehensive comparisons of RipenessGAN against CycleGAN and class-conditional diffusion models (DDPM) under strictly controlled and fair experimental settings, carefully matching model architectures, computational resources, training conditions, and evaluation metrics. The results indicate that although diffusion models yield highly realistic images and CycleGAN ensures stable cycle-consistent generation, RipenessGAN provides superior practical benefits in training efficiency, temporal controllability, and adaptability for agricultural applications. This research demonstrates the potential of RipenessGAN to mitigate data imbalance in agriculture and highlights its scalability to other crops. Full article

The integration of Large Language Models (LLMs), particularly Visual Language Models (VLMs), into robotics promises more intuitive human–robot interactions; however, challenges remain in efficiently translating high-level commands into precise physical actions. This paper presents a novel architecture for vision-based object manipulation that leverages a VLM’s reasoning capabilities while incorporating symmetry principles to enhance operational efficiency. Implemented on a Yahboom DOFBOT educational robot with a Jetson Nano platform, our system introduces a prompt-based framework that uniquely embeds symmetry-related cues to streamline feature extraction and object detection from visual data. This methodology, which utilizes few-shot learning, enables the VLM to generate more accurate and contextually relevant commands for manipulation tasks by efficiently interpreting the symmetric and asymmetric features of objects. The experimental results in controlled scenarios demonstrate that our symmetry-informed approach significantly improves the robot’s interaction efficiency and decision-making accuracy compared to generic prompting strategies. This work contributes a robust method for integrating fundamental vision principles into modern generative AI workflows for robotics. Furthermore, its implementation on an accessible educational platform shows its potential to simplify complex robotics concepts for engineering education and research. Full article

Wayfinding with minimal effort is essential for reducing cognitive load and emotional stress in unfamiliar environments. This exploratory quasi-experimental study investigated wayfinding challenges in a university building housing three spatially dispersed counseling centers and three academic departments that share the building entrances, lobby, and hallways. Using mobile eye tracking with concurrent think-aloud protocols and schematic mapping, we examined visual attention patterns during predefined navigation tasks performed by 24 first-time visitors. Findings revealed frequent fixations on non-informative structural features, while existing wayfinding cues were often overlooked. High rates of null gazes indicated unsuccessful visual searching. Thematic analysis of verbal data identified eight key issues, including spatial confusion, aesthetic monotony, and inadequate signage. Participants frequently described the environment as disorienting and emotionally taxing, comparing it to institutional settings such as hospitals. In response, we developed wayfinding design proposals informed by our research findings, stakeholder needs, and contextual priorities. We used an experiential digital twin that prioritized perceptual fidelity to analyze the current wayfinding challenges, develop experimental protocols, and discuss design options and costs. This study offers a transferable methodological framework for identifying wayfinding challenges through convergent analysis of gaze patterns and verbal protocols, demonstrating how empirical findings can inform targeted wayfinding design interventions. Full article

Molecular docking has emerged as a pivotal computational approach in agri-food research, offering a rapid and targeted means to discover bioactive molecules for crop protection and food safety. Its ability to predict and visualize interactions between natural or synthetic compounds and specific biological targets provides valuable opportunities to address urgent agricultural challenges, including climate change and the rise in resistant crop pathogens. By enabling the in silico screening of diverse chemical entities, this technique facilitates the identification of molecules with antimicrobial and antifungal properties, specifically designed to interact with critical enzymatic pathways in plant pathogens. Recent advancements, such as the integration of molecular dynamics simulations and artificial intelligence-enhanced scoring functions, have significantly improved docking accuracy by addressing limitations like protein flexibility and solvent effects. These technological improvements have accelerated the discovery of eco-friendly biopesticides and multifunctional nutraceutical agents. Promising developments include nanoparticle-based delivery systems that enhance the stability and efficacy of bioactive molecules. Despite its potential, molecular docking still faces challenges related to incomplete protein structures, variability in scoring algorithms, and limited experimental validation in agricultural contexts. This work highlights these limitations while outlining current trends and future prospects to guide its effective application in agri-food biotechnology. Full article

Current cavitating water jet technology for mineral liberation predominantly relies on the micro-jet impact generated by bubble collapse. Consequently, conventional nozzle designs often overlook the shear effects on mineral particles within the internal flow path. Moreover, the cavitation cloud evolution mechanisms in nozzles operating on this innovative principle remain insufficiently explored. This study systematically evaluates the cavitation performance of an innovatively designed cavitating jet nozzle with friction-shear effects (CJN-FSE), whose optimized internal structure enhances the interlayer shear and stripping effects crucial for the liberation of layered minerals. Utilizing high-speed imaging, we visualized submerged friction-shear cavitating water jets and systematically investigated the dynamic evolution patterns of cavitation clouds under jet pressures ranging from 15 to 35 MPa. The results demonstrate that the nozzle achieves effective cavitation, with jet pressure exerting a significant influence on the morphology and evolution of the cavitation clouds. As the jet pressure increased from 15 to 35 MPa, the cloud length, width, and average shedding distance increased by 37.05%, 45.79%, and 211.25%, respectively. The mean box-counting dimension of the cloud contour rose from 1.029 to 1.074, while the shedding frequency decreased from 1360 to 640 Hz. Within the 15–25 MPa range, the clouds showed periodic evolution, with each cycle comprising four stages: inception, development, shedding, and collapse. At 30 MPa, mutual interference between adjacent clouds emerged, leading to unsteady shedding behavior. This study thereby reveals the influence of jet pressure on the dynamic evolution patterns and unsteady shedding mechanisms of the clouds. It provides a theoretical and experimental basis for subsequent research into the nozzle’s application in liberating layered minerals and proposes a new design paradigm for cavitation nozzles tailored to the mechanical properties of specific minerals. Full article

Metaheuristic algorithms are widely utilized as effective tools for solving complex optimization problems. Among them, the Harmony Search (HS) algorithm has garnered significant attention for its simple structure and excellent performance. The efficacy of the HS algorithm is heavily dependent on the configuration of its internal parameters, with the Harmony Memory Considering Rate (HMCR) and Pitch Adjusting Rate (PAR) playing pivotal roles. These parameters determine the probabilities of using the Random Generation (RG), Harmony Memory Consideration (HMC), and Pitch Adjustment (PA) operators, thereby controlling the balance between exploration and exploitation. However, a systematic empirical analysis of the interaction between these parameters and the characteristics of the problem at hand remains insufficient. Thus, this study conducts a comprehensive empirical analysis of the performance sensitivity of the HS algorithm to variations in HMCR and PAR values. The analysis is performed on a suite of 23 benchmark functions, encompassing diverse characteristics such as unimodality/multimodality and separability/non-separability, along with 5 real-world optimization problems. Through extensive experimentation, the performance for each parameter combination was evaluated on a rank-based system and visualized using heatmaps. The results experimentally demonstrate that the algorithm’s performance is most sensitive to the HMCR value across all function types, establishing that setting a high HMCR value (≥0.9) is a prerequisite for securing stable performance. Conversely, the optimal PAR value showed a direct correlation with the topographical features of the problem landscape. For unimodal problems, a low PAR value between 0.1 and 0.3 was more effective, whereas for complex multimodal problems with numerous local optima, a relatively higher PAR value between 0.3 and 0.5 proved more efficient in searching for the global optimum. This research provides a guideline into the parameter settings of the HS algorithm and contributes to enhancing its practical applicability by proposing a systematic parameter tuning strategy based on problem characteristics. Full article