Search Results (546)

The semantic segmentation of bone structures demands pixel-level classification accuracy to create reliable bone models for diagnosis. While Convolutional Neural Networks (CNNs) are commonly used for segmentation, they often struggle with complex shapes due to their focus on texture features and limited ability to incorporate positional information. As orthopedic surgery increasingly requires precise automatic diagnosis, we explored SegFormer, an enhanced Vision Transformer model that better handles spatial awareness in segmentation tasks. However, SegFormer’s effectiveness is typically limited by its need for extensive training data, which is particularly challenging in medical imaging, where obtaining labeled ground truths (GTs) is a costly and resource-intensive process. In this paper, we propose two models and their combination to enable accurate feature extraction from smaller datasets by improving SegFormer. Specifically, these include the data-efficient model, which deepens the hierarchical encoder by adding convolution layers to transformer blocks and increases feature map resolution within transformer blocks, and the FPN-based model, which enhances the decoder through a Feature Pyramid Network (FPN) and attention mechanisms. Testing our model on spine images from the Cancer Imaging Archive and our own hand and wrist dataset, ablation studies confirmed that our modifications outperform the original SegFormer, U-Net, and Mask2Former. These enhancements enable better image feature extraction and more precise object contour detection, which is particularly beneficial for medical imaging applications with limited training data. Full article

This study focuses on optical twin-core couplers, which facilitate light transmission between two closely aligned optical fibers. These couplers operate based on the principle of coupling, allowing signals in one core to interact with those in the other. The Kerr effect, which describes how a material’s refractive index changes in response to the intensity of light, induces the nonlinear behavior essential for generating solitons—self-sustaining wave packets that preserve their shape and speed. In our research, we employ fractional derivatives to investigate how fractional-order variations influence wave propagation and soliton dynamics. By utilizing the modified extended mapping method (MEMM), we derive solitary wave solutions for the equations governing the behavior of optical twin-core couplers under Kerr nonlinearity. This methodology produces novel fractional traveling wave solutions, including dark, bright, singular, and combined bright–dark solitons, as well as hyperbolic, Jacobi elliptic function (JEF), periodic, and singular periodic solutions. To enhance understanding, we present physical interpretations through contour plots and include both 2D and 3D graphical representations of the results. Full article

With the advancement of service robot technology, the demand for higher boundary precision in indoor semantic segmentation has increased. Traditional methods of extracting Euclidean features using point cloud and voxel data often neglect geodesic information, reducing boundary accuracy for adjacent objects and consuming significant computational resources. This study proposes a novel network, the Euclidean–geodesic network (EGNet), which uses point cloud–voxel–mesh data to characterize detail, contour, and geodesic features, respectively. The EGNet performs feature fusion through Euclidean and geodesic branches. In the Euclidean branch, the features extracted from point cloud data compensate for the detail features lost by voxel data. In the geodesic branch, geodesic features from mesh data are extracted using inter-domain fusion and aggregation modules. These geodesic features are then combined with contextual features from the Euclidean branch, and the simplified trajectory map of the grid is used for up-sampling to produce the final semantic segmentation results. The Scannet and Matterport datasets were used to demonstrate the effectiveness of the EGNet through visual comparisons with other models. The results demonstrate the effectiveness of integrating Euclidean and geodesic features for improved semantic segmentation. This approach can inspire further research combining these feature types for enhanced segmentation accuracy. Full article

Image segmentation is a crucial task in artificial intelligence fields such as computer vision and medical imaging. While convolutional neural networks (CNNs) have achieved notable success by learning representative features from large datasets, they often lack geometric priors and global object information, limiting their accuracy in complex scenarios. Variational methods like active contours provide geometric priors and theoretical interpretability but require manual initialization and are sensitive to hyper-parameters. To overcome these challenges, we propose a novel segmentation approach, named PolarVoting, which combines the minimal path encoding rich geometric features and CNNs which can provide efficient initialization. The introduced model involves two main steps: firstly, we leverage the PolarMask model to extract multiple source points for initialization, and secondly, we construct a voting score map which implicitly contains the segmentation mask via a modified circular geometric voting (CGV) scheme. This map embeds global geometric information for finding accurate segmentation. By integrating neural network representation with geometric priors, the PolarVoting model enhances segmentation accuracy and robustness. Extensive experiments on various datasets demonstrate that the proposed PolarVoting method outperforms both PolarMask and traditional single-source CGV models. It excels in challenging imaging scenarios characterized by intensity inhomogeneity, noise, and complex backgrounds, accurately delineating object boundaries and advancing the state of image segmentation. Full article

Coronary artery stenosis detection remains a challenging task due to the complex vascular structure, poor quality of imaging pictures, poor vessel contouring caused by breathing artifacts and stenotic lesions that often appear in a small region of the image. In order to improve the accuracy and efficiency of detection, a new deep-learning technique based on a coronary artery stenosis detection framework (DCA-YOLOv8) is proposed in this paper. The framework consists of a histogram equalization and canny edge detection preprocessing (HEC) enhancement module, a double coordinate attention (DCA) feature extraction module and an output module that combines a newly designed loss function, named adaptive inner-CIoU (AICI). This new framework is called DCA-YOLOv8. The experimental results show that the DCA-YOLOv8 framework performs better than existing object detection algorithms in coronary artery stenosis detection, with precision, recall, F1-score and mean average precision (mAP) at 96.62%, 95.06%, 95.83% and 97.6%, respectively. In addition, the framework performs better in the classification task, with accuracy at 93.2%, precision at 92.94%, recall at 93.5% and F1-score at 93.22%. Despite the limitations of data volume and labeled data, the proposed framework is valuable in applications for assisting the cardiac team in making decisions by using coronary angiography results. Full article

RGB-D (depth) Salient Object Detection (SOD) seeks to identify and segment the most visually compelling objects within a given scene. Depth data, known for their strong discriminative capability in spatial localization, provide an advantage in achieving accurate RGB-D SOD. However, recent research in this field has encountered significant challenges due to the poor visual qualities and disturbing cues in raw depth maps. This issue results in indistinct or ambiguous depth features, which consequently weaken the performance of RGB-D SOD. To address this problem, we propose a novel pseudo depth feature generation-based RGB-D SOD Network, named PDFNet, which can generate some new and more distinctive pseudo depth features as an extra supplement source to enhance the raw depth features. Specifically, we first introduce an RGB-guided pseudo depth feature generation subnet to synthesize more distinctive pseudo depth features for raw depth feature enhancement, since the discriminative power of depth features plays a pivotal role in providing effective contour and spatial cues. Then, we propose a cross-modal fusion mamba (CFM) to effectively merge RGB features, raw depth features, and generated pseudo depth features. We adopt a channel selection strategy within the CFM module to align the pseudo depth features with raw depth features, thereby enhancing the depth features. We test the proposed PDFNet on six commonly used RGB-D SOD benchmark datasets. Extensive experimental results validate that the proposed approach achieves superior performance. For example, compared to the previous cutting-edge method, AirSOD, our method improves the F-measure by 2%, 1.7%, 1.1%, and 2.2% on the STERE, DUTLF-D, NLPR, and NJU2K datasets, respectively. Full article

A nondestructive approach for accurate crop yield prediction at the field scale is vital for precision agriculture. Considerable progress has been made in the use of the spectral index (SI) derived from unmanned aerial vehicle (UAV) hyperspectral images to predict crop yields before harvest. However, few studies have explored the most sensitive wavelengths and SIs for crop yield prediction, especially for different nitrogen fertilization levels and soil types. This study aimed to investigate the appropriate wavelengths and their combinations to explore the ability of new SIs derived from UAV hyperspectral images to predict yields during the growing season of spring maize. In this study, the hyperspectral canopy reflectance measurement method, a field-based high-throughput method, was evaluated in three field experiments (Wang-Jia-Qiao (WJQ), San-Ke-Shu (SKS), and Fu-Jia-Jie (FJJ)) since 2009 with different soil types (alluvial soil, black soil, and aeolian sandy soil) and various nitrogen (N) fertilization levels (0, 168, 240, 270, and 312 kg/ha) in Lishu County, Northeast China. The measurements of canopy spectral reflectance and maize yield were conducted at critical growth stages of spring maize, including the jointing, silking, and maturity stages, in 2019 and 2020. The best wavelengths and new SIs, including the difference spectral index, ratio spectral index, and normalized difference spectral index forms, were obtained from the contour maps constructed by the coefficient of determination (R²) from the linear regression models between the yield and all possible SIs screened from the 450 to 950 nm wavelengths. The new SIs and eight selected published SIs were subsequently used to predict maize yield via linear regression models. The results showed that (1) the most sensitive wavelengths were 640–714 nm at WJQ, 450–650 nm and 750–950 nm at SKS, and 450–700 nm and 750–950 nm at FJJ; (2) the new SIs established here were different across the three experimental fields, and their performance in maize yield prediction was generally better than that of the published SIs; and (3) the new SIs presented different responses to various N fertilization levels. This study demonstrates the potential of exploring new spectral characteristics from remote sensing technology for predicting the field-scale crop yield in spring maize cropping systems before harvest. Full article

Effective detection of the contours of cloud masks and estimation of their distribution can be of practical help in studying weather changes and natural disasters. Existing deep learning methods are unable to extract the edges of clouds and backgrounds in a refined manner when detecting cloud masks (shadows) due to their unpredictable patterns, and they are also unable to accurately identify small targets such as thin and broken clouds. For these problems, we propose MDU-Net, a multiscale dual up-sampling segmentation network based on an encoder–decoder–decoder. The model uses an improved residual module to capture the multi-scale features of clouds more effectively. MDU-Net first extracts the feature maps using four residual modules at different scales, and then sends them to the context information full flow module for the first up-sampling. This operation refines the edges of clouds and shadows, enhancing the detection performance. Subsequently, the second up-sampling module concatenates feature map channels to fuse contextual spatial information, which effectively reduces the false detection rate of unpredictable targets hidden in cloud shadows. On a self-made cloud and cloud shadow dataset based on the Landsat8 satellite, MDU-Net achieves scores of 95.61% in PA and 84.97% in MIOU, outperforming other models in both metrics and result images. Additionally, we conduct experiments to test the model’s generalization capability on the landcover.ai dataset to show that it also achieves excellent performance in the visualization results. Full article

In recent years, due to the significant advancements in hardware sensors and software technologies, 3D environmental point cloud modeling has gradually been applied in the automation industry, autonomous vehicles, and construction engineering. With the high-precision measurements of 3D LiDAR, its point clouds can clearly reflect the geometric structure and features of the environment, thus enabling the creation of high-density 3D environmental point cloud models. However, due to the enormous quantity of high-density 3D point clouds, storing and processing these 3D data requires a considerable amount of memory and computing time. In light of this, this paper proposes a real-time 3D point cloud environmental contour modeling technique. The study uses the point cloud distribution from the 3D LiDAR body frame point cloud to establish structured edge features, thereby creating a 3D environmental contour point cloud map. Additionally, unstable objects such as vehicles will appear during the mapping process; these specific objects will be regarded as not part of the stable environmental model in this study. To address this issue, the study will further remove these objects from the 3D point cloud through image recognition and LiDAR heterogeneous matching, resulting in a higher quality 3D environmental contour point cloud map. This 3D environmental contour point cloud not only retains the recognizability of the environmental structure but also solves the problems of massive data storage and processing. Moreover, the method proposed in this study can achieve real-time realization without requiring the 3D point cloud to be organized in a structured order, making it applicable to unorganized 3D point cloud LiDAR sensors. Finally, the feasibility of the proposed method in practical applications is also verified through actual experimental data. Full article

Machine learning has significantly advanced traffic surveillance and management, with YOLO (You Only Look Once) being a prominent Convolutional Neural Network (CNN) algorithm for vehicle detection. This study utilizes YOLO version 7 (YOLOv7) combined with the Kalman-based SORT (Simple Online and Real-time Tracking) algorithm as one of the models used in our experiments for real-time vehicle identification. We developed the “ISTraffic” dataset. We have also included an overview of existing datasets in the domain of vehicle detection, highlighting their shortcomings: existing vehicle detection datasets often have incomplete annotations and limited diversity, but our “ISTraffic” dataset addresses these issues with detailed and extensive annotations for higher accuracy and robustness. The ISTraffic dataset is meticulously annotated, ensuring high-quality labels for every visible object, including those that are truncated, obscured, or extremely small. With 36,841 annotated examples and an average of 32.7 annotations per image, it offers extensive coverage and dense annotations, making it highly valuable for various object detection and tracking applications. The detailed annotations enhance detection capabilities, enabling the development of more accurate and reliable models for complex environments. This comprehensive dataset is versatile, suitable for applications ranging from autonomous driving to surveillance, and has significantly improved object detection performance, resulting in higher accuracy and robustness in challenging scenarios. Using this dataset, our study achieved significant results with the YOLOv7 model. The model demonstrated high accuracy in detecting various vehicle types, even under challenging conditions. The results highlight the effectiveness of the dataset in training robust vehicle detection models and underscore its potential for future research and development in this field. Our comparative analysis evaluated YOLOv7 against its variants, YOLOv7x and YOLOv7-tiny, using both the “ISTraffic” dataset and the COCO (Common Objects in Context) benchmark. YOLOv7x outperformed others with a [email protected] of 0.87, precision of 0.89, and recall of 0.84, showing a 35% performance improvement over COCO. Performance varied under different conditions, with daytime yielding higher accuracy compared to night-time and rainy weather, where vehicle headlights affected object contours. Despite effective vehicle detection and counting, tracking high-speed vehicles remains a challenge. Additionally, the algorithm’s deep learning estimates of emissions (CO, NO, NO₂, NO_x, PM_2.5, and PM₁₀) were 7.7% to 10.1% lower than ground-truth. Full article

To address the challenge of rapid geometric model development in the digital twin industry, this paper presents a comprehensive pipeline for constructing 3D models from images using monocular vision imaging principles. Firstly, a structure-from-motion (SFM) algorithm generates a 3D point cloud from photographs. The feature detection methods scale-invariant feature transform (SIFT), speeded-up robust features (SURF), and KAZE are compared across six datasets, with SIFT proving the most effective (matching rate higher than 0.12). Using K-nearest-neighbor matching and random sample consensus (RANSAC), refined feature point matching and 3D spatial representation are achieved via antipodal geometry. Then, the Poisson surface reconstruction algorithm converts the point cloud into a mesh model. Additionally, texture images are enhanced by leveraging a visual geometry group (VGG) network-based deep learning approach. Content images from a dataset provide geometric contours via higher-level VGG layers, while textures from style images are extracted using the lower-level layers. These are fused to create texture-transferred images, where the image quality assessment (IQA) metrics SSIM and PSNR are used to evaluate texture-enhanced images. Finally, texture mapping integrates the enhanced textures with the mesh model, improving the scene representation with enhanced texture. The method presented in this paper surpassed a LiDAR-based reconstruction approach by $20\%$ in terms of point cloud density and number of model facets, while the hardware cost was only $1\%$ of that associated with LiDAR. Full article

During coal mining, sudden inrushes of water from the floor pose significant risks, seriously affecting mine safety. This study utilizes the 3602 working face of the Chenmanzhuang coal mine as a case study, and the original influencing factors were downscaled using principal component analysis (PCA) to obtain four key evaluation factors: water inflow, aquiclude thickness, water pressure, and exposed limestone thickness. The rough set theory (RST) was applied to determine the weights of the four main influencing factors as 0.2, 0.24, 0.36, and 0.2; furthermore, 19 groups of comprehensive values were calculated using the weighting method, and a water inrush risk assessment was conducted for several blocks within the working face. The results are presented as a contour map, highlighting various risk levels and identifying the water inrush danger zone on the coal seam floor. The study concludes that water inrush poses a threat in the western part of the working face, while the eastern area remains relatively safe. The accuracy and reliability of the model are demonstrated, providing a solid basis and guidance for predicting water inrush. Full article

The present study explores the influence of diverse nozzle geometries on the combustion characteristics of ADN-based energetic propellants. The pressure contour maps reveal a rapid initial increase in the average pressure of ADN-based propellants across the three different nozzles. Subsequently, the pressure tapers off gradually as time elapses. Notably, during the crucial initial period of 0–5 μs, the straight nozzle exhibited the most significant pressure surge at 30.2%, substantially outperforming the divergent (6.67%) and combined nozzles (15.5%). The combustion product variation curves indicate that the contents of reactants ADN and CH₃OH underwent a steep decline, whereas the product N₂O displayed a biphasic behavior, initially rising and subsequently declining. In contrast, the CO₂ concentration remained on a steady ascent throughout the entire combustion process, which concluded within 10 μs. Our findings suggest that the straight nozzle facilitated the more expeditious generation of high-temperature and high-pressure combustion gases for ADN-based propellants, expediting reaction kinetics and enhancing combustion efficiency. This is attributed to the reduced intermittent interactions between the nozzle wall and shock waves, which are encountered in the divergent and combined nozzles. In conclusion, the superior combustion characteristics of ADN-based propellants in the straight nozzle, compared to the divergent and combined nozzles, underscore its potential in informing the design of advanced propulsion systems and guiding the development of innovative energetic propellants. Full article

The devastating effects of rain-induced attenuation on communication links operating above 10 GHz during rainy events can significantly degrade signal quality, leading to interruptions in service and reduced data throughput. Understanding the spatial and seasonal distribution of rain heights is crucial for predicting these attenuation effects and for network performance optimization. This study utilized ten years of atmospheric temperature and geopotential height data at seven pressure levels (1000, 850, 700, 500, 300, 200, and 100 hPa) obtained from the Copernicus Climate Data Store (CDS) to deduce rain heights across nine stations in South Africa. The kernel density estimation (KDE) method was applied to estimate the temporal variation of rain height. A comparison of the measured and estimated rain heights shows a correlation coefficient of 0.997 with a maximum percentage difference of 5.3%. The results show that rain height ranges from a minimum of 3.5 km during winter in Cape Town to a maximum of about 5.27 km during the summer in Polokwane. The spatial variation shows a location-dependent seasonal trend, with peak rain heights prevailing at the low-latitude stations. The seasonal variability indicates that higher rain heights dominate in the regions (Polokwane, Pretoria, Nelspruit, Mahikeng) where there is frequent occurrence of rainfall during the winter season and vice versa. Contour maps of rain heights over the four seasons (autumn, spring, winter, and summer) were also developed for South Africa. The estimated seasonal rain heights show that rain-induced attenuations were grossly underestimated by the International Telecommunication Union (ITU) recommended rain heights at most of the stations during autumn, spring, and summer but fairly overestimated during winter. Durban had a peak attenuation of 15.9 dB during the summer, while Upington recorded the smallest attenuation of about 7.7 dB during winter at a 0.01% time exceedance. Future system planning and adjustments of existing infrastructure in the study stations could be improved by integrating these localized, seasonal radio propagation data in link budget design. Full article

This article implements the Hirota bilinear (HB) transformation technique to the Landau–Ginzburg–Higgs (LGH) model to explore the nonlinear evolution behavior of the equation, which describes drift cyclotron waves in superconductivity. Utilizing the Cole–Hopf transform, the HB equation is derived, and symbolic manipulation combined with various auxiliary functions (AFs) are employed to uncover a diverse set of analytical solutions. The study reveals novel results, including multi-wave complexitons, breather waves, rogue waves, periodic lump solutions, and their interaction phenomena. Additionally, a range of traveling wave solutions, such as dark, bright, periodic waves, and kink soliton solutions, are developed using an efficient expansion technique. The nonlinear dynamics of these solutions are illustrated through 3D and contour maps, accompanied by detailed explanations of their physical characteristics. Full article