Search Results (783)

Privacy-sensitive electricity scene classification requires robust models under data localization constraints, making federated learning (FL) a suitable framework. Existing FL frameworks face two critical challenges in multi-label electricity scene classification: (1) Label correlations and their strengths significantly impact classification performance. (2) Electricity scene data and labels show distributional inconsistencies across regions. However, current FL frameworks lack explicit modeling of label correlation strengths, and locally trained regional models naturally capture these differences, leading to regional differences in their model parameters. In this scenario, the server’s standard single-stage aggregation often over-averages the global model’s parameters, reducing its discriminative ability. To address these issues, we propose FMMAN, a federated multi-stage attention neural network for multi-label electricity scene classification. The main contributions of this FMMAN lie in label correlation learning and the stepwise model aggregation. It splits the client–server interaction into multiple stages: (1) Clients train models locally to encode features and label correlation strengths after receiving the server’s initial model. (2) The server clusters these locally trained models into K groups to ensure that models within a group have more consistent parameters and generates K prototype models via intra-group aggregation to reduce over-averaging. The K models are then distributed back to the clients. (3) Clients refine their models using the K prototypes with contrastive group-specific consistency regularization to further mitigate over-averaging, and sends the refined model back to the server. (4) Finally, the server aggregates the models into a global model. Experiments on multi-label benchmarks verify that FMMAN outperforms baseline methods. Full article

To address the significant degradation of image visibility and contrast in turbid media, this paper proposes an enhanced image dehazing algorithm. Unlike traditional polarimetric dehazing methods that exclusively attribute polarization information to airlight, our approach integrates object radiance polarization and airlight polarization for haze removal. First, sky regions are localized through multi-scale fusion of polarization and intensity segmentation maps. Second, region-specific transmittance estimation is performed by differentiating haze-occluded regions from haze-free regions. Finally, target radiance is solved using boundary constraints derived from non-haze regions. Compared with other dehazing algorithms, the method proposed in this paper demonstrates greater adaptability across diverse scenarios. It achieves higher-quality restoration of targets with results that more closely resemble natural appearances, avoiding noticeable distortion. Not only does it deliver excellent dehazing performance for land fog scenes, but it also effectively handles maritime fog environments. Full article

With the growing demand for ecological monitoring and geological exploration, high-quality satellite remote-sensing imagery has become indispensable for accurate information extraction and automated analysis. However, haze reduces image contrast and sharpness, significantly impairing quality. Existing dehazing methods, primarily designed for natural images, struggle with remote-sensing images due to their complex imaging conditions and scale diversity. Given this, we propose a novel Multi-Scale Contextual Attention Generative Adversarial Network (MCA-GAN), specifically designed for satellite image dehazing. Our method integrates multi-scale feature extraction with global contextual guidance to enhance the network’s comprehension of complex remote-sensing scenes and its sensitivity to fine details. MCA-GAN incorporates two self-designed key modules: (1) a Multi-Scale Feature Aggregation Block, which employs multi-directional global pooling and multi-scale convolutional branches to bolster the model’s ability to capture land-cover details across varying spatial scales; (2) a Dynamic Contextual Attention Block, which uses a gated mechanism to fuse three-dimensional attention weights with contextual cues, thereby preserving global structural and chromatic consistency while retaining intricate local textures. Extensive qualitative and quantitative experiments on public benchmarks demonstrate that MCA-GAN outperforms other existing methods in both visual fidelity and objective metrics, offering a robust and practical solution for remote-sensing image dehazing. Full article

Previous studies observed that emotional scenes, presented as distractors, capture attention and interfere with an ongoing task. This behavioral interference has been shown to be elicited by the semantic rather than by the perceptual properties of a scene, as it resisted the application of low-pass spatial frequency filters. Some studies observed that the visual system can adapt to perceptual conditions; however, little is known concerning whether attentional capture by emotional stimuli can also be modulated by the sequential repetition of viewing conditions or of emotional content. In the present study, we asked participants to perform a parity task while viewing irrelevant natural scenes, which could be either emotional or neutral. These scenes could be either blurred (low-pass filter) or perceptually intact, and the order of presentation was balanced to study the effects of sequential repetition of perceptual conditions. The results indicate that affective modulation was most pronounced when the same viewing condition (either intact or blurred) was repeated, with faster responses when perceptual conditions were repeated for neutral distractors, but to a lesser extent for emotional ones. These data suggest that emotional interference in an attentional task can be modulated by serial sensitization in the processing of spatial frequencies. Full article

High-quality green gardens can markedly enhance the quality of life and mental well-being of their users. However, health and lifestyle constraints make it difficult for people to enjoy urban gardens, and traditional methods struggle to offer the high-fidelity experiences they need. This study introduces a 3D scene reconstruction and rendering strategy based on implicit neural representation through the efficient and removable neural radiation fields model (NeRF-RE). Leveraging neural radiance fields (NeRF), the model incorporates a multi-resolution hash grid and proposal network to improve training efficiency and modeling accuracy, while integrating a segment-anything model to safeguard public privacy. Take the crabapple tree, extensively utilized in urban garden design across temperate regions of the Northern Hemisphere. A dataset comprising 660 images of crabapple trees exhibiting three distinct geometric forms is collected to assess the NeRF-RE model’s performance. The results demonstrated that the ‘harvest gold’ crabapple scene had the highest reconstruction accuracy, with PSNR, LPIPS and SSIM of 24.80 dB, 0.34 and 0.74, respectively. Compared to the Mip-NeRF 360 model, the NeRF-RE model not only showed an up to 21-fold increase in training efficiency for three types of crabapple trees, but also exhibited a less pronounced impact of dataset size on reconstruction accuracy. This study reconstructs real scenes with high fidelity using virtual reality technology. It not only facilitates people’s personal enjoyment of the beauty of natural gardens at home, but also makes certain contributions to the publicity and promotion of urban landscapes. Full article

Sensing transparent objects has many applications in human daily life, including robot navigation and grasping. However, this task presents significant challenges due to the unpredictable nature of scenes that extend beyond/behind transparent objects, particularly the lack of fixed visual patterns and strong background interference. This paper aims to solve the transparent object segmentation problem by leveraging the intrinsic global modeling capabilities of transformer architectures. We design a Query Parsing Module (QPM) that innovatively formulates segmentation as a dictionary lookup problem, differing fundamentally from conventional pixel-wise mechanisms, e.g., via attention-based prototype matching, and a set of learnable class prototypes as query inputs. Based on QPM, we propose a high-performance transformer-based end-to-end segmentation model, Transparent Object Segmentation through Query (TOSQ). TOSQ’s encoder is based on the Segformer’s backbone, and its decoder consists of a series of QPM modules, which progressively refine segmentation masks by the proposed QPMs. TOSQ achieves state-of-the-art performance on the Trans10K-V2 dataset (76.63% mIoU, 95.34% Acc), with particularly significant gains in challenging categories like windows (+23.59%) and glass doors (+11.22%), demonstrating its superior capability in transparent object segmentation. Full article

Due to diverse lighting conditions and photo environments, shadows are almost ubiquitous in images, especially document images captured with mobile devices. Shadows not only seriously affect the visual quality and readability of a document but also significantly hinder image processing. Although shadow removal research has achieved good results in natural scenes, specific studies on document images are lacking. To effectively remove shadows in document images, the dark illumination correction network is proposed, which mainly consists of two modules: shadow detection and illumination correction. First, a simplified shadow-corrected attention block is designed to combine spatial and channel attention, which is used to extract the features, detect the shadow mask, and correct the illumination. Then, the shadow detection block detects shadow intensity and outputs a soft shadow mask to determine the probability of each pixel belonging to shadow. Lastly, the illumination correction block corrects dark illumination with a soft shadow mask and outputs a shadow-free document image. Our experiments on five datasets show that the proposed method achieved state-of-the-art results, proving the effectiveness of illumination correction. Full article

Geo-historical information deduced from geo-iconographical resources, derived from extensive research and the selection of cartographies and historical documents, enabled the investigation of the natural and anthropic transformations of the perifluvial area of the Po River in the Emilia-Romagna region (Italy). This territory, significant in terms of its historical, cultural, and environmental contexts, for centuries has been the scene of flood events. These have characterised the morphological and dynamic variability in the riverbed and relative floodplain. The close relationship between man and river is well documented: the interference induced by anthropic activity has alternated with the sometimes-damaging effects of river dynamics. The attention given to the fluvial region of the Po River and its main tributaries, in a peculiar lowland sector near Parma, is critical for understanding spatial–temporal changes contributing to current geo-hydrological risks. A GIS project outlined the geomorphological aspects that define the considerable variations in the course of the Po River (involving width reductions of up to 66% and length changes of up to 14%) and its confluences from the 16th to the 21st century. Knowledge of anthropic modifications is essential as a tool within land-use planning and enhancing community awareness in risk-mitigation activities and strategic management. This study highlights the importance of interdisciplinary geo-historical studies that are complementary in order to decode river dynamics in damaging flood events and latent hazards in an altered river environment. Full article

Background/Objectives: Dental anxiety in children is a common issue that can hinder the delivery of effective dental care. Traditional approaches to managing this are often insufficient or involve pharmacological interventions. This study shows the potential of virtual reality (VR) to aid in reducing anxiety in children undergoing simple dental procedures. By immersing children in relaxing VR environments (such as beaches, forests, mountains, or underwater scenes with calm music), the objective is to assess VR’s effectiveness in calming pediatrics patients during these procedures. Methods: Children scheduled for minor dental treatments wore a wearable device that monitored pulse, perspiration, and stress levels. Each child’s baseline data was collected without the VR headset, followed by data collection during VR exposure before and during dental procedures. VR scenarios ranged from soothing nature scenes to animated cartoons, designed to foster relaxation. Results: The data collected showed a reduction in physiological indicators of stress, such as lower heart rate and reduced perspiration, when the VR headset was used. Children appeared more relaxed, with a calmer response during the procedure itself, compared to baseline levels without VR. Conclusions: This study provides preliminary evidence supporting VR as an effective tool for reducing anxiety and stress in pediatric dental patients. By offering an engaging, immersive experience, VR can serve as an alternative or complementary approach to traditional anxiety management strategies in pediatric dentistry, potentially improving patient comfort and cooperation during dental procedures. Further research could determine if VR may serve as an alternative to local anesthesia for non-intrusive pediatric dental procedures. Full article

Land surface temperature (LST) is an important parameter in the surface system with drastic variation in spatial and temporal domains. The protection of the ecological environment in mountainous areas and the monitoring of natural disasters require the support of surface temperature data with high spatiotemporal resolution. LST downscaling is an effective method to improve the spatial and temporal resolution of remote sensing LST data. However, at present, the LST downscaling research mainly focuses on plain and urban areas, while the area of rugged terrain is less studied, and the accuracy of LST in rugged terrain is lower than in plain and urban areas. In the few studies that discuss auxiliary parameters for LST downscaling in rugged terrain, only elevation is considered as an auxiliary parameter. In this study, we selected parameters that have evident correlation with LST as potential auxiliary factors and discussed the benefits of adding shortwave radiation to the LST downscaling process. We chose four scene images in the Beijing suburbs and the Loess Plateau and conducted the LST downscaling experiments. In this study, we used the Taylor expansion method for LST downscaling. We selected Landsat 8 and MODSI LST data as fine and coarse study datasets, respectively. The results show that the accuracy of LST downscaling in rugged terrain areas can be improved by using elevation and shortwave radiation as auxiliary factors, and the benefits of shortwave radiation is independent of that of elevation. Therefore, it is suggested that these two parameters be simultaneously used to achieve the best LST downscaling result over rugged terrain areas. Full article

Accurate multi-class semantic segmentation of disaster-affected areas is essential for rapid response and effective recovery planning. We present CSDNet, a context-aware segmentation model tailored to disaster scene scenarios, designed to improve segmentation of both large-scale disaster zones and small, underrepresented classes. The architecture combines a lightweight transformer module for global context modeling with depthwise separable convolutions (DWSCs) to enhance efficiency without compromising representational capacity. Additionally, we introduce a detection-guided feature fusion mechanism that integrates outputs from auxiliary detection tasks to mitigate class imbalance and improve discrimination of visually similar categories. Extensive experiments on several public datasets demonstrate that our model significantly improves segmentation of both man-made infrastructure and natural damage-related features, offering a robust and efficient solution for post-disaster analysis. Full article

Timely and accurate detection of forest fires is crucial for protecting forest ecosystems. However, traditional monitoring methods face significant challenges in effectively detecting forest fires, primarily due to the dynamic spread of flames and smoke, irregular morphologies, and the semi-transparent nature of smoke, which make it extremely difficult to extract key visual features. Additionally, deploying these detection systems to edge devices with limited computational resources remains challenging. To address these issues, this paper proposes a lightweight hybrid receptive field model (LHRF-YOLO), which leverages deep learning to overcome the shortcomings of traditional monitoring methods for fire detection on edge devices. Firstly, a hybrid receptive field extraction module is designed by integrating the 2D selective scan mechanism with a residual multi-branch structure. This significantly enhances the model’s contextual understanding of the entire image scene while maintaining low computational complexity. Second, a dynamic enhanced downsampling module is proposed, which employs feature reorganization and channel-wise dynamic weighting strategies to minimize the loss of critical details, such as fine smoke textures, while reducing image resolution. Furthermore, a scale weighted Fusion module is introduced to optimize multi-scale feature fusion through adaptive weight allocation, addressing the issues of information dilution and imbalance caused by traditional fusion methods. Finally, the Mish activation function replaces the SiLU activation function to improve the model’s ability to capture flame edges and faint smoke textures. Experimental results on the self-constructed Fire-SmokeDataset demonstrate that LHRF-YOLO achieves significant model compression while further improving accuracy compared to the baseline model YOLOv11. The parameter count is reduced to only 2.25M (a 12.8% reduction), computational complexity to 5.4 GFLOPs (a 14.3% decrease), and mAP50 is increased to 87.6%, surpassing the baseline model. Additionally, LHRF-YOLO exhibits leading generalization performance on the cross-scenario M4SFWD dataset. The proposed method balances performance and resource efficiency, providing a feasible solution for real-time and efficient fire detection on resource-constrained edge devices with significant research value. Full article

Tea disease detection is of great significance to the tea industry. In order to solve the problems such as mutual occlusion of leaves, light disturbance, and small lesion area under complex background, YOLO-SSM, a tea disease detection model, was proposed in this paper. The model introduces the SSPDConv convolution module in the backbone of YOLOv8 to enhance the global information perception of the model under complex backgrounds; a new ESPPFCSPC module is proposed to replace the original spatial pyramid pool SPPF module, which optimizes the multi-scale feature expression; and the MPDIoU loss function is introduced to optimize the problem that the original CIoU is insensitive to the change of target size, and the positioning ability of small targets is improved. Finally, the map values of 89.7% and 68.5% were obtained on a self-made tea data set and a public tea disease data set, which were improved by 3.9% and 4.3%, respectively, compared with the original benchmark model, and the reasoning speed of the model was 164.3 fps. Experimental results show that the proposed YOLO-SSM algorithm has obvious advantages in accuracy and model complexity and can provide reliable theoretical support for efficient and accurate detection and identification of tea leaf diseases in natural scenes. Full article

At crime scenes, apart from the detection of blood, it may be important to determine whether a person was alive at the time of blood deposition. Based on the rapid onset of fibrinolysis after death, this pathway could be considered to identify potential biomarkers for postmortem blood. Fibrinolysis is the natural process that breaks down blood clots after healing a vascular injury. One of its products, D-dimer, could be a potential biomarker for postmortem blood. SERATEC^® (SERATEC^® GmbH, Göttingen, Germany) has developed the PMB immunochromatographic assay to simultaneously detect human hemoglobin and D-dimer. The main goals of this study were to assess the possibility of using this test to detect postmortem blood, evaluate D-dimer levels in antemortem, menstrual, and postmortem blood, and assess the ability to obtain STR profiles from postmortem blood. Except for one degraded sample, all postmortem blood samples reacted positively for the presence of D-dimer using the SERATEC^® PMB test. All antemortem blood samples from living individuals showed negative results for D-dimer detection, except for one liquid sample with a weak positive result, probably due to pre-existing health conditions. Menstrual blood samples gave variable results for D-dimer. The DIMERTEST^® Latex assay was used for semi-quantitative measurement of D-dimer concentrations, with postmortem and menstrual blood yielding higher D-dimer concentrations compared to antemortem blood. Full STR profiles were developed for all postmortem samples tested except for one degraded sample, pointing to the possibility of not only detecting postmortem blood at the crime scene but also the potential identification of the victim. Full article

Fluid effects are important in films and advertisements, where their realism and aesthetic quality directly impact the visual experience. With the rapid advancement of digital technology and computer-aided design (CAD), modern visual effects are used to simulate various water-related phenomena, such as flowing water, ocean waves, and raindrops. However, creating these realistic effects is not solely dependent on advanced software and hardware; it also requires an understanding of the technical and artistic aspects of visual effects artists. In the creation process, the artist must possess a keen aesthetic sense and innovative thinking to craft stunning visual effects to overcome technological constraints. Whether depicting the grandeur of turbulent ocean scenes or the romance of gentle rain, the artist needs to transform fluid effects into expressive visual language to enhance emotional impact, aligning with the storyline and the director’s vision. The production process of fluid effects typically involves the following critical steps. First, the visual effects artist utilizes CAD-based tools, particle systems, or fluid simulation software to model the dynamic behavior of water. This process demands a solid foundation in physics and the ability to adjust parameters flexibly according to the specific needs of the scene, ensuring that the fluid motion appears natural and smooth. Next, in the rendering stage, the simulated fluid is transformed into realistic imagery, requiring significant computational power and precise handling of lighting effects. Finally, in the compositing stage, the fluid effects are seamlessly integrated with live-action footage, making the visual effects appear as though they are parts of the actual scene. In this study, the technical details of creating fluid effects using free software such as Blender were explored. How advanced CAD tools are utilized to achieve complex water effects was also elucidated. Additionally, case studies were conducted to illustrate the creative processes involved in visual effects production to understand how to seamlessly blend technology with artistry to create unforgettable visual spectacles. Full article