Search Results (395)

Accurate cutting of salmon parts and surface defect detection are the key steps to enhance the added value of its processing. At present, mainstream manual inspection methods have low accuracy and efficiency, making it difficult to meet the demands of industrialized production. A machine vision inspection method based on a two-stage fusion network is proposed in this paper, aiming to achieve accurate cutting of salmon parts and efficient recognition of defects. The fish body image is collected by building a visual inspection system, and the dataset is constructed by preprocessing and data enhancement. For the part cutting, the improved U-Net model that introduces the CBAM attention mechanism is used to strengthen the extraction ability of the fish body texture features. For defect detection, the two-stage fusion architecture is designed to quickly locate the defective region by adding the YOLOv5 of the P2 small target detection layer first, and then the cropped region is fed into the improved U-Net for accurate cutting. The experimental results demonstrate that the improved U-Net achieves a mean average precision (mAP) of 96.87% and a mean intersection over union (mIoU) of 94.33% in part cutting, representing improvements of 2.44% and 1.06%, respectively, over the base model. In defect detection, the fusion model attains an mAP of 94.28% with a processing speed of 7.30 fps, outperforming the single U-Net by 28.02% in accuracy and 236.4% in efficiency. This method provides a high-precision, high-efficiency solution for intelligent salmon processing, offering significant value for advancing automation in the aquatic product processing industry. Full article

The mammalian skull, which surrounds and protects the brain, is one of the most morphologically diverse and functionally important structures in the vertebrate body. As one of the most ecologically diverse mammals, the developmental dynamics of morphological and structural changes and functional diversity in the skull of bats need to be revealed. Here, we focused on the developmental characteristics of the Vespertilio sinensis skull, and used statistical analysis, spatial morphology visualization, and comparative analysis of the Stretch Factors (SF) of the masticatory muscles to better understand the connection between the morphology of the skull and the development of the body size during the developmental process of V. sinensis, the changes in the three-dimensional (3D) spatial morphology and structure, and the correlations between opening capacity and the transformation of feeding habits. This study not only provides a new perspective for understanding the morphological adaptive mechanism of ecological niche expansion that accompanies the transition of mammalian skulls from juvenile to adult feeding but also provides a crucial scientific basis for an in-depth understanding of the growth and developmental mechanism of bats’ skull and even vertebrates as a whole, which is potentially useful for the development of ecological conservation and evolutionary biology. Full article

State Departments of Transportation (DOTs) face challenges with traditional bridge inspections that are time-consuming, inconsistent, and paper-based. This study focused on an existing research gap regarding automated methods that streamline the bridge inspection process, prioritize maintenance effectively, and allocate resources efficiently. Thus, this paper introduces a digitalized bridge inspection framework by integrating Building Information Modeling (BIM) and Business Intelligence (BI) to enable near-real-time monitoring and digital documentation. This study adopts a Design Science Research (DSR) methodology, a recognized paradigm for developing and evaluating the innovative SmartBridge to address pressing bridge inspection problems. The method involved designing an Autodesk Revit-based plugin for data synchronization, element-specific comments, and interactive dashboards, demonstrated through an illustrative 3D bridge model. An illustrative example of the digitalized bridge inspection with the proposed framework is provided. The results show that SmartBridge streamlines data collection, reduces manual documentation, and enhances decision-making compared to conventional methods. This paper contributes to this body of knowledge by combining BIM and BI for digital visualization and predictive analytics in bridge inspections. The proposed framework has high potential for hybridizing digital technologies into bridge infrastructure engineering and management to assist transportation agencies in establishing a safer and efficient bridge inspection approach. Full article

Background/Objectives: Understanding how muscle fatigue contributes to musculoskeletal injuries is critical in sports science. Although joint biomechanics during landing under fatigue has been studied before, limited research has focused on the layup phase under fatigue. This study examined the effects of fatigue on ankle, knee, and hip-joint biomechanics during layup and landing. We hypothesized that fatigue would increase peak vertical ground reaction force (GRF), peak knee extension angle, and peak joint moments. Methods: Fourteen healthy male participants performed 3-step layups and drop landings using their dominant leg on force plates. The fatigue protocol consisted of squat jumps, step-ups, and repeated countermovement jumps (CMJs), with fatigue defined as three consecutive CMJs below 80% of the participant’s pre-established maximum jump height. After a fatigue protocol, they repeated the tasks. Kinematic data were collected using an eight-camera Vicon system (100 Hz), and GRF data were recorded with two AMTI force plates (1000 Hz). Thirty-six reflective markers were placed on lower-limb anatomical landmarks, and data were processed using Visual 3D. Paired t-tests ($\alpha$ = 0.05) were conducted using SPSS (V26.0) to compare pre- and post-fatigue outcomes. Results: Significant increases were found in peak GRF during landing (pre: 3.41 $\pm$ 0.81 BW [Body Weight], post: 3.95 $\pm$ 1.05 BW, p = 0.036), and peak negative hip joint work during landing (pre: $-$0.34 $\pm$ 0.18 J/kg, post: $-$0.66 $\pm$ 0.43 J/kg, p = 0.025). Conclusions: These findings indicate that fatigue may alter landing mechanics, reflected in increased ground reaction forces and negative hip joint work. These preliminary findings should be interpreted cautiously, and future studies with larger samples and additional neuromuscular measures under sport-specific conditions are needed to improve ecological validity. Full article

The term “image captioning” refers to the process of converting an image into text through computer vision and natural language processing algorithms. Image captioning is still considered an open-ended topic despite the fact that visual data, most of which pertains to images, is readily available in today’s world. This is despite the fact that recent developments in computer vision, such as Vision Transformers (ViT) and language models using BERT and GPT, have opened up new possibilities for the field. The purpose of this review paper is to provide an overview of the present status of the field, with a specific emphasis on the use of facial expression recognition and object detection for the purpose of image captioning, particularly in the context of fault-aware systems and Prognostics and Health Management (PHM) applications within Industry 4.0 environments. However, to the best of our knowledge, no review study has focused on the significance of facial expressions in relation to image captioning, especially in industrial settings where operator facial expressions can provide valuable insights for fault detection and system health monitoring. This is something that has been overlooked in the existing body of research on image captioning, which is the primary reason why this study was conducted. During this paper, we will talk about the most important approaches and procedures that have been utilized for this task, including fault-aware methodologies that leverage visual data for PHM in smart manufacturing contexts, and we will highlight the advantages and disadvantages of each strategy. The purpose of this review is to present a comprehensive assessment of the current state of the field and to recommend topics for future research that will lead to machine-translated captions that are more detailed and accurate, particularly for Industry 4.0 applications where visual monitoring plays a crucial role in system diagnostics and maintenance. Full article

In recent years, most conventional emotion recognition approaches have concentrated primarily on facial cues, often overlooking complementary sources of information such as body posture and contextual background. This limitation reduces their effectiveness in complex, real-world environments. In this work, we present a multi-branch emotion recognition framework that separately processes facial, bodily, and contextual information using three dedicated neural networks. To better capture contextual cues, we intentionally mask the face and body of the main subject within the scene, prompting the model to explore alternative visual elements that may convey emotional states. To further enhance the quality of the extracted features, we integrate both channel and spatial attention mechanisms into the network architecture. Evaluated on the challenging NCAER-S dataset, our model achieves an accuracy of 56.42%, surpassing the state-of-the-art GLAMOUR-Net. These results highlight the effectiveness of combining multi-cue representation and attention-guided feature extraction for robust emotion recognition in unconstrained settings. The findings also highlight the importance of accurate emotion recognition for human–computer interaction, where affect detection enables systems to adapt to users and deliver more effective experiences. Full article

Background/Objectives: With the rapid development of smart healthcare, medical records have shifted from a disease-centered to a patient-centered approach. However, traditional formats, such as narratives and tables, often make it challenging for physicians to quickly grasp a patient’s condition within a limited timeframe, potentially leading to diagnostic errors and a decline in the quality of care. Recently, advances in information visualization and 3D technology have led many medical institutions to employ charts and graphs or use 3D simulations of organs to support clinical practice and education. However, few have integrated 3D models into medical records for use during physician consultations. Methods: This study presents the development and evaluation of a novel web-based 3D EMR system that integrates real-time ICD-10 diagnostic code mapping with interactive 3D human body models, enabling physicians to visualize patient-specific anatomical and diagnostic information in a dynamic and context-aware manner. Results: We employed the System Usability Scale (SUS) to evaluate the system’s usability, conducting a satisfaction survey. Results from the survey indicate that participants rated the system highly in terms of ease of use, satisfaction, and efficiency, with an average SUS score of 70.42, reflecting usability between moderate and good. Comparative evaluations and future expansion plans are also discussed. Conclusions: These findings demonstrate that integrating a 3D human model into the medical record retrieval process significantly improves visualization and interactivity, meeting the needs of healthcare professionals and enhancing both their efficiency and patient satisfaction. Full article

Background: ErisNet, a novel AI model to reduce noise in Computed Tomography images. Methods: We trained ErisNet on 23 post-mortem whole-body CT scans. We assessed the objective performance with mean square error (MSE), peak signal-to-noise ratio (PSNR), structural similarity index (SSIM) measure, visual information fidelity (VIF), edge preservation index (EPI) and noise variance (NV). We assessed the qualitative performance by six radiologists. To support the visual assessment, we placed circular regions of interest (ROI) in the vitreous body, brain, liver and spleen parenchyma and paravertebral muscle. Results: ErisNet achieved MSE 64.07 ± 46.81, PSNR 31.32 ± 3.69 dB, SSIM 0.93 ± 0.06, VIF 0.49 ± 0.09, EPI 0.97 ± 0.01 and NV 64.69 ± 46.80. The ROI analysis showed a reduction in noise: the SD of the HU decreased by 8% in the vitreous body (from 17.6 to 16.2 HU), by 18% in the brain parenchyma (from 18.85 to 15.40 HU) and by 15–19% in the liver, spleen and paravertebral muscle. The six radiologists confirmed these results by assigning high scores (scale from one to five): overall quality 4.5 ± 0.6, noise suppression/detail preservation 4.7 ± 0.5 and diagnostic confidence 4.8 ± 0.4 (p < 0.01). Conclusions: ErisNet improves the quality of CT images and shows strong potential for processing low-dose scans. Full article

Aiming at the problem of image quality degradation caused by turbid water, non-uniform illumination, and scattering effect in the surface defect detection of underwater concrete structures, firstly, the concrete surface images under different shooting distances, different sediment concentrations, and different illumination conditions were collected through laboratory experiments to simulate the concrete surface images of a reservoir dam with higher sediment concentration and deeper water depth. On this basis, an underwater image enhancement algorithm named DIVE (Dynamic Illumination and Vision Enhancement) is proposed. DIVE solves the problems of luminance unevenness and color deviation in stages through the illumination–scattering decoupling processing framework, and combines efficient computing optimization to achieve real-time processing. The lighting correction of Gaussian distributions (dynamic illumination module) was processed in stages with suspended particle scattering correction (visual enhancement module), and the bright and dark areas were balanced and color offset was corrected by local gamma correction in Lab space and dynamic decision-making of G/B channel. Through thread pool parallelization, vectorization and other technologies, the real-time requirement can be achieved at the resolution of 1920 × 1080. Tests show that DIVE significantly improves image quality in water bodies with sediment concentration up to 500 g/m³, and is suitable for complex scenes such as reservoirs, oceans, and sediment tanks. Full article

Prolonged poor sitting posture increases the risk of musculoskeletal disorders and chronic diseases. We developed a smart cushion system that integrated pressure sensing and machine learning for posture recognition. Nine FSR406 sensors were used to measure pressure distribution on the system. A calibration and normalization process improves data consistency, and a heatmap visualizes the result. Among the five machine learning models evaluated, the narrow neural network achieved the best performance, with a validation accuracy of 97.63% and a test accuracy of 91.73%. When body mass index (BMI) was included as an additional input feature, the test accuracy improved to 95.49%, indicating that BMI positively impacts recognition performance. Full article

This study aims to bridge the research gap emerging from the relationships between Visual electronic Word-of-Mouth (VeWOM) and brand factors, and their impact on consumers’ behavior by exploring the causal effects of eWOM attributes on hotel brand factor spreading through Brand Awareness (BA) and Brand Perceived Value (BV) and its consequences on Purchase Decisions (PD) in the hospitality context. Attribution Theory was extended to incorporate brand-mediated effects and crisis-specific factors. The study investigates the impact of VeWOM on consumer Purchase Decisions (PD) in terms of hotel room bookings in the British hospitality market, emphasizing the mediating role of brand-related constructs. Drawing on Attribution Theory, the research proposes a structural model to assess both direct and indirect pathways through which VeWOM influences behavioral outcomes. A stratified, non-probability sampling approach yielded 443 valid responses from hotel bookers who engaged with user-generated visual content prior to booking. The Partial Least Squares Structural Equation Model (PLS-SEM) was employed to test the hypothesized relationships. The findings reveal that VeWOM significantly influences Brand Value (BV), eWOM Credibility, and Information Quality, which in turn shape consumer purchase behavior. Crucially, Brand Value emerges as a key mediating variable, bridging VeWOM and Purchase Decisions, while VeWOM alone does not directly affect booking behavior. Moreover, Brand Awareness showed no significant mediating effect. The study underscores the indirect attribution process in visual review contexts, demonstrating that the influence of VeWOM is channeled primarily through brand perception mechanisms rather than direct persuasion. These insights extend Attribution Theory by highlighting the distinct cognitive pathways activated by visual content compared to text-based reviews. Practically, the research suggests that hoteliers should focus on enhancing Brand Value via bundled offerings and relationship-based marketing rather than relying solely on visual appeal or awareness to drive bookings. The study contributes to the growing body of VeWOM literature by clarifying its nuanced effects on decision-making in digital hospitality environments. Full article

Carbonate fracture–cavity reservoirs are significant oil and gas reservoirs globally, and their efficient development is influenced by the connectivity between fracture–cavity units within the reservoir. These reservoirs primarily consist of large caves, dissolution holes, and natural fractures, which serve as the primary storage and flow spaces. The S91 unit of the Tarim Oilfield is a karstic fracture–cavity reservoir with shallow coverage. It exhibits significant heterogeneity in the fracture–cavity reservoirs and presents complex connectivity between the fracture–cavity bodies. The integration of static and dynamic data, including geology, well logging, seismic, and production dynamics, resulted in the development of a set of static and dynamic connectivity evaluation processes designed for highly heterogeneous fracture–cavity reservoirs. Methods include using structural gradient tensors and stratigraphic continuity attributes to delineate the boundaries of caves and holes; performing RGB fusion analysis of coherence, curvature, and variance attributes to characterize large-scale fault development features; applying ant-tracking algorithms and fracture simulation techniques to identify the distribution and density characteristics of fracture zones; utilizing 3D visualization technology to describe the spatial relationship between fracture–cavity units and large-scale faults and fracture development zones; and combining dynamic data to verify interwell connectivity. This process will provide a key geological basis for optimizing well network deployment, improving water and gas injection efficiency, predicting residual oil distribution, and formulating adjustment measures, thereby improving the development efficiency of such complex reservoirs. Full article

Extra-osseous calcification refers to the pathological deposition of calcium salts in soft tissues. Its most recognized forms affect the cardiovascular system, leading to vascular and heart valve calcifications. This process is active and regulated, involving the phenotype transition of resident cells into osteo/chondrogenic lineage. Chronic kidney disease (CKD) patients frequently suffer from vascular and other soft tissue calcification. OsteoSense dyes are fluorescent imaging agents developed to visualize calcium deposits during bone formation. In addition to its application in bone physiology, it has been used to detect vascular smooth muscle cell calcification in vitro and to evaluate calcification ex vivo. Here, we investigated CKD-associated soft tissue calcification by applying OsteoSense in vivo. CKD was induced by a diet containing adenine and elevated phosphate. OsteoSense (80 nmol/kg body weight) was injected intravenously through the retro-orbital venous sinus 18 h before the measurement on an IVIS Spectrum In Vivo Imaging System. OsteoSense staining detected calcium deposition in the aorta, kidney, heart, lung, and liver in CKD mice. On the other hand, no calcification occurred in the brain, eye, or spleen. OsteoSense positivity in the calcified soft tissues in CKD mice was associated with increased mRNA levels of osteo/chondrogenic transcription factors. Our findings demonstrate that OsteoSense is a sensitive and effective tool for detecting soft tissue calcification in vivo, and may be particularly valuable for studies of CKD-related ectopic calcification. Full article

In this study, emissivity was employed as the primary metric for evaluating the radiative-cooling performance of a supersonic nozzle under flight-like conditions. A supersonic nozzle was fabricated by PBF, after which combustion (hot-fire) tests and emissivity measurements were carried out. These data enabled quantification and visualization of radiant energy in the 8–14 µm-wavelength band during combustion. The hot-fire tests revealed a clear cap-shock pattern, confirming that the nozzle flow was fully developed. Emissivity measurements showed that the additively manufactured surface—subsequently treated by the AMS 5662 heat-treatment process—followed the angular-emission trends reported in previous studies. The surface exhibited a high roughness (Ra ≈ 30 µm) and an emissivity of 0.85 in the 8–14 µm band; a temperature-dependent emissivity fitting function was accordingly derived. By coupling the combustion test results with the emissivity data, the actual temperature distribution along the nozzle surface and the corresponding radiant energy in the 8–14 µm band were quantitatively reconstructed and visualized. The maximum emissive power in this band reached 2214 W m⁻², representing at least 16.61% of the total black-body radiation at 700 K. Full article

A large number of cities have created semantic 3D city models, but these models are rarely used as input data for simulations, such as noise and flooding, in the urban planning process. Reasons for this are that many simulations require detailed land cover (LC) and elevation data that are often not included in the 3D city models, and that there is no linkage between the elevation and land cover data. In this study, we design, implement and evaluate methods to handle LC and elevation data in a 3D city model. The LC data is stored in 2.5D or 3D in the CityGML modules Transportation, Vegetation, WaterBody, CityFurniture and LandUse, and a complete 3D LC partition is created by combining data from these modules. The entire workflow is demonstrated in the paper: creating 2D LC data, extending CityGML, creating 2.5D/3D data from the 2D LC data, dividing the LC data into CityGML modules, storing it in a database (3DCityDB) and finally visualizing the data in Unreal Engine. The study is part of the 3CIM project where a national profile of CityGML for Sweden is created as an Application Domain Extension (ADE), but the result is generally applicable for CityGML implementations. Full article