Search Results (1,825)

Grid structured visual data such as forms, tables, and game boards require models that pair pixel level perception with symbolic consistency under global constraints. Recent Pixel Language Models (PLMs) map images to token sequences with promising flexibility, yet we find they generalize poorly when observable evidence becomes sparse or corrupted. We present GridMNIST-Sudoku, a benchmark that renders large numbers of Sudoku instances with style diverse handwritten digits and provides parameterized stress tracks for two tasks: Completion (predict missing cells) and Correction (detect and repair incorrect cells) across difficulty levels ranging from 1 to 90 altered positions in a 9 × 9 grid. Attention diagnostics on PLMs trained with conventional one dimensional positional encodings reveal weak structure awareness outside the natural Sudoku sparsity band. Motivated by these findings, we propose a lightweight Row-Column-Box (RCB) positional prior that injects grid aligned coordinates and combine it with simple sparsity and corruption augmentations. Trained only on the natural distribution, the resulting model substantially improves out of distribution accuracy across wide sparsity and corruption ranges while maintaining strong in distribution performance. Full article

With the vigorous development of deep learning technology, lane detection tasks have achieved phased results. However, existing lane detection models do not consider the unique geometric and visual features of lanes when dealing with some challenging scenarios, resulting in many difficulties and limitations. To this end, we propose a lane detection network based on full-dimensional convolutional Transformer (OTVLD-Net) to improve the adaptability of the model under extreme road conditions and better handle complex lane topology. In order to extract richer contextual features, we designed ODVT-Net, which uses full-dimensional dynamic convolution combined with improved feature flip fusion layer and non-local network layer, and aggregates lane symmetry features by utilizing the horizontal symmetry of lanes. A feature weight generation mechanism based on Transformer is designed, and a cross-attention mechanism between feature maps and lane requests is added in the decoding stage to enable the network to aggregate global feature information. At the same time, a vanishing point detection module is introduced, and a joint weighted loss function is designed to be trained in coordination with the lane detection task to improve the generalization ability of the lane detection model. Experimental results on the OpenLane and CurveLanes datasets show that the detection effect of the OTVLD-Net model has reached the current advanced level. In particular, the accuracy on the OpenLane dataset is 6.4% higher than the F1 score of the second-ranked model, and the average performance in different challenging scenarios is also improved by 8.9%. At the same time, when ResNet-18 is used as the template feature extraction network, the model achieves a speed of 103FPS and a computing power of 14.2 GFlops, achieving good performance while ensuring real-time performance. Full article

This paper presents a non-intrusive method for estimating the distance between a camera and a human subject using a monocular vision system and statistically derived interpupillary distance (IPD) values. The proposed approach eliminates the need for individual calibration by utilizing average IPD values based on biological sex, enabling accurate, scalable distance estimation for diverse users. The algorithm, implemented in Python 3.12.11 using the MediaPipe Face Mesh framework, extracts pupil coordinates from facial images and calculates IPD in pixels. A sixth-degree polynomial calibration function, derived from controlled experiments using a uniaxial displacement system, maps pixel-based IPD to real-world distances across three intervals (20–80 cm, 80–160 cm, and 160–240 cm). Additionally, a geometric correction is applied to compensate for in-plane facial rotation. Experimental validation with 26 participants (15 males, 11 females) demonstrates the method’s robustness and accuracy, as confirmed by relative error analysis against ground truth measurements obtained with a Bosch GLM120C laser distance meter. Males exhibited lower relative errors across the intervals (3.87%, 4.75%, and 5.53%), while females recorded higher mean relative errors (6.0%, 6.7%, and 7.27%). The results confirm the feasibility of the proposed method for real-time applications in human–computer interaction, augmented reality, and camera-based proximity sensing. Full article

Artificial reefs serve as a crucial measure for preventing habitat degradation, enhancing primary productivity in marine areas, and restoring and increasing fishery resources, making them an essential component of marine ranching development. Accurate identification and detection of artificial reefs are vital for ecological conservation and fishery resource management. To achieve precise segmentation of artificial reefs in multibeam sonar images, this study proposes an improved YOLOv11-based model, YOLO-AR. Specifically, the DCCA (Dynamic Convolution Coordinate Attention) module is introduced into the backbone network to reduce the model’s sensitivity to complex seafloor environments. Additionally, a small-object detection layer is added to the neck network, along with the ultra-lightweight dynamic upsampling operator DySample (Dynamic Sampling), which enhances the model’s ability to segment small artificial reefs. Furthermore, some standard convolution layers in the backbone are replaced with ADown (Advanced Downsampling) to reduce the model’s complexity. Experimental results demonstrate that YOLO-AR achieves an mAP@0.5 of 0.912, an intersection-over-union (IOU) of 0.832, and an F1 score of 0.908. Meanwhile, the parameters and model size of YOLO-AR are 2.67 million and 5.58 MB. Compared to other advanced segmentation models, YOLO-AR maintains a more lightweight structure while delivering a superior segmentation performance. In real-world multibeam sonar images, YOLO-AR can accurately segment artificial reefs, making it highly effective for practical applications. Full article

To achieve efficient vineyard grape picking, a vision-based information processing framework integrating two-stage segmentation with morphological perception is proposed. In the first stage, an improved YOLOv8s-seg model is employed for coarse segmentation, incorporating two key enhancements: first, a dynamic deformation feature aggregation module (DDFAM), which facilitates the extraction of complex structural and morphological features; and second, an efficient asymmetric decoupled head (EADHead), which improves boundary awareness while reducing parameter redundancy. Compared with mainstream segmentation models, the improved model achieves superior performance, attaining the highest mAP@0.5 of 86.75%, a lightweight structure with 10.34 M parameters, and a real-time inference speed of 10.02 ms per image. In the second stage, the fine segmentation of fruit stems is performed using an improved OTSU thresholding algorithm, which is applied to a single-channel image derived from the hue component of the HSV color space, thereby enhancing robustness under complex lighting conditions. Morphological features extracted from the preprocessed fruit stem, including centroid coordinates and a skeleton constructed via medial axis transform (MAT), are further utilized to establish the spatial relationships with a picking point and cutting axis. The visualization analysis confirms the high feasibility and adaptability of the proposed framework, providing essential technical support for the automation of grape harvesting. Full article

Background/Objectives: Multiple sclerosis (MS) often leads to gait impairments, even in early stages, and can affect autonomy and quality of life. Traditional assessment methods, while widely used, have been criticized because they lack sensitivity to subtle gait changes. This scoping review aims to map the landscape of advanced gait analysis technologies—both wearable and non-wearable—and evaluate their application in detecting, characterizing, and monitoring possible gait dysfunction in individuals with MS. Methods: A systematic search was conducted across PubMed and Scopus databases for peer-reviewed studies published in the last decade. Inclusion criteria focused on original human research using technological tools for gait assessment in individuals with MS. Data from 113 eligible studies were extracted and categorized based on gait parameters, technologies used, study design, and clinical relevance. Results: Findings highlight a growing integration of advanced technologies such as inertial measurement units, 3D motion capture, pressure insoles, and smartphone-based tools. Studies primarily focused on spatiotemporal parameters, joint kinematics, gait variability, and coordination, with many reporting strong correlations to MS subtype, disability level, fatigue, fall risk, and cognitive load. Real-world and dual-task assessments emerged as key methodologies for detecting subtle motor and cognitive-motor impairments. Digital gait biomarkers, such as stride regularity, asymmetry, and dynamic stability demonstrated high potential for early detection and monitoring. Conclusions: Advanced gait analysis technologies can provide a multidimensional, sensitive, and ecologically valid approach to evaluating and detecting motor function in MS. Their clinical integration supports personalized rehabilitation, early diagnosis, and long-term disease monitoring. Future research should focus on standardizing metrics, validating digital biomarkers, and leveraging AI-driven analytics for real-time, patient-centered care. Full article

Background: Accurate route planning is a core challenge in logistics, particularly for small- and medium-sized enterprises that lack access to costly geospatial tools. This study explores whether usable distance matrices and routing outputs can be generated solely from geographic coordinates without relying on full map-based infrastructure. Methods: A dataset of over 5000 Hungarian postal locations was used to evaluate five models: Haversine-based scaling with circuity, linear regression, second- and third-degree polynomial regressions, and a trained artificial neural network. Models were tested on the full dataset, and three example routes representing short, medium, and long distances. Both statistical accuracy and route-level performance were assessed, including a practical optimization task. Results: Statistical models maintained internal consistency, but systematically overestimated longer distances. The ANN model provided significantly better accuracy across all scales and produced routes more consistent with map-based paths. A new evaluation method was introduced to directly compare routing outputs. Conclusions: Practical route planning can be achieved without GIS services. ML-based estimators offer a cost-effective alternative, with potential for further improvement using larger datasets, additional input features, and the integration of travel time prediction. This approach bridges the gap between simplified approximations and commercial routing systems. Full article

Efficient and accurate forward modeling of electromagnetic fields is essential for advancing geophysical exploration in complex marine environments. However, realistic survey conditions characterized by low-frequency spectra, fine sedimentary strata, irregular bathymetry, and anisotropic materials pose significant challenges for conventional numerical methods. To address these issues, this work presents a parallel modeling framework that combines coordinate transformations with an adaptive high-order finite-element approach for 3D marine controlled-source electromagnetic (MCSEM) simulations. The algorithm exploits the form invariance of Maxwell’s equations to map the original boundary value problem over the physical domain to one defined over a computationally favorable domain filled with anisotropic media. The transformed model is then discretized and solved using a parallel high-order finite-element scheme enhanced with a goal-oriented adaptive mesh refinement strategy. We examine the performance of the proposed framework using both synthetic models and the realistic Marlim R3D benchmark dataset. The results demonstrate that the proposed approach can effectively reduce computational costs while maintaining high accuracy across a wide frequency range and varying water depths. These findings highlight the framework’s potential for large-scale, high-resolution CSEM exploration of offshore resources. Full article

Against the backdrop of rapid new urbanisation and the ongoing integration of urban and rural areas, the evolving spatial dynamics between public service facilities and population distribution have increasingly garnered scholarly interest. The present study employs a grid-based spatial unit and a coupling coordination model as a foundation. This model integrates POI data, Baidu heat maps, and other sources of spatial and temporal information. The objective is to explore the dynamic matching pattern of public service facilities and population distribution. The study’s findings are as follows: The population within the core urban area displays a strong propensity for agglomeration during the morning and evening peak hours, thereby forming a highly coordinated public service network characterised by high-density and piecemeal distribution of public service facilities. The population residing within the transition zone between urban and rural areas is commuting in a substantial number, and the relationship between the supply of and demand for facilities demonstrates cyclical fluctuations. Local areas are subject to time-periodic pressure on the supply of and demand for facilities. In rural areas, due to the continuous population outflow and dispersed residence, the layout of service facilities is fragmented, exhibiting the island effect. The study reveals a structural contradiction between traditional homogeneous planning and the gradient difference between urban and rural areas, providing a scientific basis for Shandong Province to promote new urbanisation and rural revitalisation strategies in an integrated manner. Full article

Background: Geographic information systems (GIS) are a promising tool for mapping vaccination coverage and identifying missed communities, yet their use in low- and middle-income countries (LMICs) remains limited. In settings without standardized addresses such as schools or outreach sites, innovative methods are needed to collect and analyse spatial data. Schools offer a unique platform for identifying under-vaccinated children missed by routine or campaign efforts. Methods: During a pilot school vaccination screening program in Zambia, GIS reference maps of health facility catchment areas were developed from hand-drawn sketch maps, catchment area shapefiles, and coordinates of prominent landmarks. These maps were iteratively refined with input from local health staff. In caregiver interviews, data collectors used the maps to identify the child’s zone of residence within the health facility catchment area. Vaccination status was extracted from paper registries used during screening. Geographic heat maps were generated in ArcGIS to visualize under-vaccination by zone. Results: Of 535 children screened across 25 zones, 29% were under-vaccinated. Under-vaccination varied by zone, with clusters of missed children identified, for example, 50% of children in Kabushi Zone 6 were under-vaccinated, compared with much lower rates elsewhere. Conclusions: Pairing school-based vaccination checks with GIS mapping offers a scalable approach to identifying missed communities in LMICs. This method enables spatial analysis without household visits, supporting targeted immunization planning where traditional data systems fall short. However, because the study was limited to children enrolled in five purposively selected schools, out-of-school children and those in other schools were not represented. This selection bias may underestimate the true extent of under-vaccination, and future evaluations should incorporate broader and more representative populations. Full article

Accurate target recognition and localization remain significant challenges for robotic fruit harvesting in unstructured orchard environments characterized by branch occlusion and leaf clutter. To address the difficulty in identifying and locating apples under such visually complex conditions, this paper proposes an improved YOLOv5-based visual recognition algorithm incorporating an efficient channel attention (ECA) module. The ECA module is strategically integrated into specific C3 layers (C3-3, C3-6, C3-9) of the YOLOv5 network architecture to enhance feature representation for occluded targets. During operation, the system simultaneously acquires apple pose information and achieves precise spatial localization through coordinate transformation matrices. Comprehensive experimental evaluations demonstrate the effectiveness of the proposed system. The custom-designed six-degree-of-freedom (6-DOF) robotic arm exhibits a wide operational range with a maximum working angle of 120°. The ECA-enhanced YOLOv5 model achieves a confidence level of 90% and an impressive in-range apple recognition rate of 98%, representing a 2.5% improvement in the mean Average Precision (mAP) compared to the baseline YOLOv5s algorithm. The end-effector positioning error is consistently controlled within 1.5 mm. The motion planning success rate reaches 92%, with the picking completed within 23 s per apple. This work provides a novel and effective vision recognition solution for future development of harvesting robots. Full article

The integration of electric vehicles (EVs) into smart grids (SGs) is reshaping both energy systems and mobility infrastructures. This review presents a comprehensive and cross-disciplinary synthesis of current technologies, methodologies, and challenges associated with EV–SG interaction. Unlike prior reviews that address these aspects in isolation, this work uniquely connects three critical pillars: (i) the evolution of energy storage technologies, including lithium-ion, second-life, and hybrid systems; (ii) optimisation and predictive control techniques using artificial intelligence (AI) for real-time energy management and vehicle-to-grid (V2G) coordination; and (iii) cybersecurity risks and post-quantum solutions required to safeguard increasingly decentralised and data-intensive grid environments. The novelty of this review lies in its integrated perspective, highlighting how emerging innovations, such as federated AI models, blockchain-secured V2G transactions, digital twin simulations, and quantum-safe cryptography, are converging to overcome existing limitations in scalability, resilience, and interoperability. Furthermore, we identify underexplored research gaps, such as standardisation of bidirectional communication protocols, regulatory inertia in V2G market participation, and the lack of unified privacy-preserving data architectures. By mapping current advancements and outlining a strategic research roadmap, this article provides a forward-looking foundation for the development of secure, flexible, and grid-responsive EV ecosystems. The findings support policymakers, engineers, and researchers in advancing the technical and regulatory landscape necessary to scale EV–SG integration within sustainable smart cities. Full article

Accurate boundary detection is critical for autonomous trackless rubber-wheeled vehicles in underground coal mines, as it prevents lateral collisions with tunnel walls. Unlike open-road environments, underground tunnels suffer from poor illumination, water mist, and dust, which degrade visual imaging. To address these challenges, this paper proposes a navigable area computation for underground autonomous vehicles via ground feature and boundary processing, consisting of three core steps. First, a real-time point cloud correction process via pre-correction and dynamic update aligns ground point clouds with the LiDAR coordinate system to ensure parallelism. Second, corrected point clouds are projected onto a 2D grid map using a grid-based method, effectively mitigating the impact of ground unevenness on boundary extraction; third, an adaptive boundary completion method is designed to resolve boundary discontinuities in junctions and shunting chambers. Additionally, the method emphasizes continuous extraction of boundaries over extended periods by integrating temporal context, ensuring the continuity of boundary detection during vehicle operation. Experiments on real underground vehicle data validate that the method achieves accurate detection and consistent tracking of dual-sided boundaries across straight tunnels, curves, intersections, and shunting chambers, meeting the requirements of underground autonomous driving. This work provides a rule-based, real-time solution feasible under limited computing power, offering critical safety redundancy when deep learning methods fail in harsh underground environments. Full article

We present a deep learning-based approach to optimize the central angle between adjacent camera viewpoints for the efficient generation of natural 360-degree holographic 3D content. High-quality 360-degree digital holograms require the acquisition of densely sampled RGB–depth map pairs, a process that traditionally requires significant computational costs. Our method introduces a novel pipeline that systematically evaluates the impact of varying central angles—defined as the angular separation between equidistant viewpoints in an object-centered coordinate system—on both depth map estimation and holographic 3D image reconstruction. By systematically applying this pipeline, we determine the optimal central angle that achieves an effective balance between image quality and computational efficiency. Experimental investigations demonstrate that our approach significantly reduces computational demands while maintaining superior fidelity of the reconstructed 3D holographic images. The relationship between central angle selection and the resulting quality of 360-degree digital holographic 3D content is thoroughly analyzed, providing practical guidelines for the creation of immersive holographic video experiences. This work establishes a quantitative standard for the geometric configuration of viewpoint sampling in object-centered environments and advances the practical realization of real-time, high-quality holographic 3D content. Full article

This systematic review synthesizes evidence on economic instruments that mobilize renewable-energy investment in emerging economies, analyzing 50 peer-reviewed studies published between 2015 and 2025 under PRISMA 2020. We advance an Institutional Capacity Integration Framework that ties instrument efficacy to regulatory, market, and coordination capabilities. Green bonds have mobilized roughly USD 500 billion yet work only where robust oversight and liquid markets exist, offering limited gains for decentralized access. Direct subsidies cut renewable electricity costs by 30–50% and connect 45 million people across varied contexts, but pose fiscal–sustainability risks. Carbon pricing schemes remain rare given their administrative complexity, while multilateral climate funds show moderate effectiveness (coefficients 0.3–0.8) dependent on national coordination strength. Bibliometric mapping with Bibliometrix reveals three fragmented paradigms—market efficiency, state intervention, and international cooperation—and highlights geographic gaps: sub-Saharan Africa represents just 16% of studies despite acute financing barriers. Sixty-eight percent of articles employ descriptive designs, constraining causal inference and reflecting tensions between SDG 7 (affordable energy) and SDG 13 (climate action). Our framework rejects one-size-fits-all prescriptions, recommending phased, context-aligned pathways that progressively build capacity. Policymakers should tailor instrument mixes to institutional realities, and researchers must prioritize causal methods and underrepresented regions through focused initiatives for equitable global progress. Full article