Search Results (388)

This paper presents a pilot educational workflow that couples 3D remote sensing with heritage-driven pedagogy by engaging architecture master’s students in the documentation and digital archiving of Transylvanian cultural sites. Using terrestrial and mobile 3D scanning, students documented multiple typologies—wooden churches (Târgușor, Tioltiur), historical ensembles (Mociu, Coplean), industrial sites (1 Mai–Luduș, Vânătorilor–Luduș), and an urban street segment (Potaissa)—to generate dense point clouds that served as the basis for geometric reconstruction, semantic interpretation, and condition assessment. The study describes how the characteristics of different construction systems (timber, brick, stone, mixed structures) relate to point-cloud quality, survey coverage, and subsequent CAD/BIM drafting, with attention to the qualitative reading of minor deformations in wooden churches and of degradation patterns in masonry and industrial buildings. We also consider how artefacts in the data (noise, occlusions, registration errors) affect scene understanding and the interpretation of derived observations relevant to condition assessment and, prospectively, to monitoring. For the Tioltiur dual-sensor case, the TLS and SLAM datasets were compared through an internal CloudCompare registration check (final RMS 0.1121 on 50,000 points, fixed scale 1.0 and theoretical overlap 100%), surface-density displays (r = 0.005 for the Z+F dataset and for the GeoSLAM dataset), fitted-wall-plane readings (dip values around 89 deg. and 85 deg.) and a longitudinal section documenting roof/vault deformation. Beyond technical performance, the paper examines the self-reported formative impact on students’ digital skills and their understanding of cultural values, arguing that participation in 3D data acquisition, processing, and interpretation positions them as co-creators of a living digital archive. Pre- and post-workshop questionnaires (n = 13 each) are analysed descriptively—counts, percentages and medians with interquartile ranges—because the two instruments are unmatched and carry no shared identifier, so no paired test is applied; post-workshop self-ratings of technical competence, heritage understanding, archival awareness and collaboration were consistently high (medians 4–5), with uneven access to VR the main gap. By connecting point-cloud-based documentation workflows with heritage education, the project outlines a transferable, monitoring-ready baseline model in which 3D remote sensing supports both careful documentation and the transmission of regional identity and cultural meaning in architectural training. As an exploratory pilot with a small, self-reported sample, the study reports descriptive and qualitative findings rather than validated metric or statistical results. Full article

This paper presents RICO-3D (Roadway Infrastructure in Context), a new large-scale Mobile Laser Scanning (MLS) dataset for semantic segmentation of French urban roadways, together with GA-Attention, a geometry-aware attention U-Net designed for this task. RICO-3D was acquired with a Leica Pegasus TRK300 mobile mapping system across Marseille, Rennes, and Opoul-Périllos (France), and provides per-point geometry, RGB, intensity, GPS time, scan angle rank, and semantic labels for 6 classes: vegetation, road, pole, building, cable, and vehicle. The dataset contains 780,981,961 labeled points and captures realistic MLS challenges, including severe class imbalance, sparse thin structures, occlusions, and varying seasonal and weather conditions. GA-Attention combines enriched geometric descriptors, attentive local aggregation, saliency-guided downsampling, attention-gated skip fusion, and curriculum-based training within a point-based encoder-decoder framework. On RICO-3D, the proposed method achieves 83.36% overall accuracy and the best IoU for road (91.35%), pole (49.91%), and cable (56.08%), with an inference time of 8.17 s. On Toronto-3D, it reaches 82.18% overall accuracy and 56.50% mIoU. These results show the relevance of RICO-3D for infrastructure-oriented MLS segmentation and the effectiveness of GA-Attention for thin and under-represented roadway infrastructure classes. To support reproducible research, the RICO-3D dataset, source code, trained models, and evaluation scripts will be publicly available once the Orange’s legal and data-governance validation process has been completed. Full article

Emerging digital technologies are reshaping interior design practice, increasing the need for educational approaches that intentionally support students’ perceived workforce readiness. This study examined the implementation of a mobile 3D scanning application within a university interior design course to examine its association with students’ technology learning self-efficacy and perceived workforce readiness. Using a one-group pre–post survey design, data were collected from interior design students who participated in a classroom-based 3D scanning activity using Polycam and completed surveys before and after the instructional experience. A total of 29 pre-test and 21 post-test responses were collected. Survey instruments assessed students’ perceived importance, familiarity, and experience with 3D scanning, along with technology adoption characteristics. Post-intervention measures also examined technology learning self-efficacy and perceived workforce readiness domains, including career development, self-regulated learning, problem solving, and technological competence. Welch’s independent-samples t-tests and Mann–Whitney U tests indicated that the post-test group reported significantly higher perceptions of familiarity with and experience using 3D scanning, as well as several technology-adoption perceptions, than the pre-test group, including ease of use, work style compatibility, efficiency gains, result observability, and trialability. One-sample Wilcoxon signed-rank tests showed that post-intervention composite scores for technology learning self-efficacy and perceived workforce readiness domains were significantly higher than the neutral scale midpoint. Qualitative thematic analysis of open-ended responses further supported these findings, identifying themes of ease of use, engagement, enhanced visualization, creativity, and professional relevance. Overall, the findings suggest that participation in the classroom-based mobile 3D scanning activity was associated with positive post-intervention perceptions of technology learning self-efficacy and perceived workforce readiness. Full article

The complexity of micrometer-scale pore-throat structures in sandstone reservoirs strongly controls the occurrence state and mobilization degree of residual oil after water-flooding. To clarify the differences in residual oil occurrence between pure oil-zone and transition-zone reservoirs and their microscopic controlling mechanisms, sandstone samples were collected from the SII oil layer group, which belongs to the Upper Cretaceous Yaojia Formation, in the Nanqi area of the Daqing Oilfield, northern Songliao Basin, NE China, and were investigated. Mercury intrusion capillary pressure (MICP), two-dimensional nuclear magnetic resonance (2D NMR), laser scanning confocal microscopy (LSCM), micro-computed tomography (micro-CT), X-ray diffraction (XRD), wettability measurement and fractal analysis were integrated to systematically characterize the pore-throat architecture, mineral composition, seepage capacity, and residual oil occurrence of the two reservoir types. The results show that the pore-throat radius distributions are mainly unimodal. In the pure oil-zone samples, the pore-throat distribution is highly consistent with the corresponding permeability contribution curve, whereas evident deviations occur in some transition-zone samples. Large and medium pore throats exert the most significant control on seepage capacity, and the difference in fractal characteristics is mainly reflected by D1, the fractal dimension of large pore throats. The transition-zone reservoirs generally exhibit moderate to strong water-wet characteristics. Owing to the development of fine pore throats and strong capillary forces, water is prone to retention within pore-throat spaces, resulting in pronounced water-blocking and Jamin effects. After water-flooding, the pure oil-zone reservoirs exhibit lower residual oil saturation, with residual oil occurring mainly in a bound state; in contrast, the transition-zone reservoirs show higher residual oil saturation and relatively high proportions of free and semi-bound residual oil. Mineral composition further modifies pore-throat complexity and residual oil occurrence. D1 is negatively correlated with feldspar content, indicating that increased feldspar content helps improve the pore-throat structure, but positively correlated with clay mineral content, suggesting that clay minerals enhance structural complexity. In the transition-zone reservoirs, kaolinite and illite–smectite mixed-layer minerals are relatively well developed. Their velocity-sensitive and water-sensitive effects readily induce pore-throat blockage and increased flow resistance, which are important causes of residual oil enrichment and difficult oil mobilization in the transition zone. Full article

Lightweight real-time detectors must balance long-range contextual modeling, cross-scale relational reasoning, and deployment efficiency. We present MambaACE-YOLO, an integration-oriented framework that separates computation into intra-scale encoding, cross-scale relational reasoning, sparse multi-stage feature distribution, and multi-scale detection decoding. Its primary contribution is the coordinated integration of prior frequency-decoupled and state-space ideas with detection-specific interfaces, target-scale leave-one-source aggregation, and physically prunable feature distribution, rather than a new frequency-transform, state-space, or hypergraph mechanism class. Building on prior frequency-decoupled hybrid visual Mamba research, its detection-oriented D-MobileMamba backbone applies bidirectional state-space scanning only to low-frequency components, while directional high-frequency correction and multi-kernel depthwise convolutions preserve boundaries and local shape. Compact Partial-Channel HyperACE (CP-HyperACE) models cross-scale higher-order relations in a semantic subspace and uses target-scale leave-one-source aggregation. Selective Additive FullPAD (SA-FullPAD) projects each cross-scale increment once and selects injection paths through static, physically prunable gates. On MS COCO 2017 val, the unpruned MambaACE-YOLO-N achieves 42.5 AP with 2.6 M parameters, and MambaACE-YOLO-S achieves 48.8 AP with 9.0 M parameters. Under a common documented RTX 5090 TensorRT FP16 setting, the unpruned Nano model records 1.19 ms network-forward latency at 42.5 AP, whereas its physically pruned counterpart retains 42.4 AP and records 1.02 ms. Accuracy and latency values are single-run or single-record point estimates without reported variance, and the 42.5-AP result uses the 600-epoch schedule without a matched 600-epoch YOLOv13-N control. We distinguish published cross-paper results from same-framework, same-device measurements and assess the individual design choices through controlled ablations and physical-pruning experiments. Full article

Accurate estimation of Alzheimer’s disease (AD) severity from structural magnetic resonance imaging (MRI) remains difficult, as disease-associated anatomical alterations are often subtle and publicly available datasets are typically too small to support robust deep learning model training. This feasibility study sought to determine how much Alzheimer’s disease spectrum-related information could be extracted from a small structural MRI cohort using a deliberately lightweight two-dimensional convolutional neural network (2D CNN), and whether transfer learning improves model performance. This study was intended as a methodological proof of concept rather than the development of a clinically deployable diagnostic tool. Structural scans and Clinical Dementia Rating (CDR) labels from the OASIS-1 dataset were filtered to 214 subjects: 124 cognitively normal (CN), 65 with mild cognitive impairment (MCI; CDR = 0.5), and 25 with AD-level impairment (CDR ≥ 1). A compact 2D CNN trained from scratch and a transfer learning model (frozen ImageNet MobileNetV2 features) were evaluated on four binary tasks (CN vs. AD, MCI vs. AD, CN vs. MCI, and CN vs. any impairment) under identical pre-processing and subject-level repeated 5-fold cross-validation (10 repeats), with the decision threshold tuned only on an inner split. Discrimination was summarized by ROC-AUC with 95% confidence intervals (CIs), permutation tests against chance, and per-task sensitivity and specificity. The from-scratch CNN recovered only a broad normal-versus-impaired signal (CN vs. any impairment AUC 0.59) and was at chance on adjacent-stage tasks (MCI vs. AD 0.41; CN vs. MCI 0.51). Transfer learning improved every task: CN vs. AD AUC 0.745 (95% CI 0.730–0.763), CN vs. any impairment 0.642, CN vs. MCI 0.601, and MCI vs. AD 0.599. On an independent OASIS-2 cohort, the transfer learning CN vs. AD model retained AUC 0.748. In this small-data regime, transfer learning recovers substantially more Alzheimer-spectrum signals than a from-scratch CNN, but performance remains modest because it is bounded by CDR-based, non-biomarker-confirmed labels, suggesting the model separates CDR-defined cognitive-status groups rather than detecting AD pathology. Full article

Large-scale field video archiving under real-world operational conditions presents four compounding challenges: heterogeneous edge devices with inconsistent hardware capabilities, low-light and high-variance illumination environments, intermittent satellite and mobile network connectivity, and arbitrary camera mounting orientations. Existing QR-based video ingestion pipelines fail to address these challenges jointly, resulting in high decode failure rates and unreliable cloud archiving in distributed field deployments. This paper presents a framework that directly addresses all four challenges through a unified edge-to-cloud pipeline. Rather than introducing new computer-vision primitives, the contribution lies in the robust system architecture that integrates and orchestrates established techniques, and in its engineering validation at scale. The pipeline combines four engineered components: (i) a visibility-aware adaptive transcoding strategy; (ii) a priority-weighted non-uniform temporal sampling scheme; (iii) a 23-angle rotation ensemble decoder; and (iv) a checkpoint-resumable block-staged cloud synchronization mechanism with deferred reclassification. The framework is validated on 132 independent edge deployments across 849 operational regions, processing 87,873 production videos totaling 1.05 TB over a 10-day observation window (27 May–5 June 2026). The evaluation demonstrates 92.1% QR recall at 26.8% of full-scan CPU cost, 98.0% upload reliability, and 33.5% deferred recovery of initially unresolvable codes (p < 0.001, Cohen’s d > 0.97). To the best of our knowledge, this constitutes the largest reported real-world evaluation of an edge QR-based video ingestion framework and the first to characterize recall variance across more than 100 independent field deployments. Full article

With advances in computer vision and modern surveying technologies, intelligent inspection systems and automatic recognition methods are increasingly used in highway tunnel maintenance. However, existing mobile inspection methods still struggle to balance high-speed operation, fine-crack recognition, and comprehensive assessment of multiple defects. This study proposes an automatic recognition and quantitative assessment method for multiple visible defects in highway tunnels based on a vehicle-mounted multisensor inspection system. The system integrates high-resolution imaging, infrared illumination, 3D laser scanning, mileage positioning, and high-speed data storage, enabling continuous full-section data acquisition at speeds up to 80 km/h. A structural-feature-constrained mileage correction strategy is developed to reduce accumulated localization errors. For crack analysis, a multilevel framework combining two-stage CNN screening, cascaded segmentation, crack trajectory tracking, and subpixel edge extraction is established for crack recognition and 0.1 mm-level width measurement. Water leakage and spalling are extracted through visible–infrared image fusion and adaptive boundary refinement, while cross-sectional deformation is calculated using 3D tunnel axis reconstruction, point-cloud filtering, and cross-section fitting. Field tests and controlled experiments demonstrate that the system can rapidly identify, locate, and quantify multiple tunnel defects, providing a practical reference for intelligent tunnel inspection and maintenance. Full article

Efficient bridge scanning and documentation are crucial for creating reliable digital 3D models. However, scanning workflows often rely on implicit practitioner experience rather than standardized protocols. This paper presents practical insights derived from a Multiple Case Study (MCS) of ten heterogeneous, real-world bridges in Germany. The study evaluates Terrestrial Laser Scanning (TLS), Mobile Laser Scanning (MLS) and Unmanned Aerial Systems (UAS) photogrammetry. The findings isolate distinct performance trade-offs. TLS offers high accuracy but suffers from shadowing occlusions. Conversely, UAS provides operational flexibility but introduces geometric vulnerabilities, including photogrammetric reconstruction noise on fine structures and SLAM trajectory drift on vibrating spans. To unify these insights, a generalized, BPMN-compliant process model mapping the complete data acquisition lifecycle under legal and spatial constraints is defined. This research provides an actionable, practical guide to optimize data quality and efficiency in structural engineering workflows. Full article

Modern consumer technologies have become more intuitive and user-friendly. On the other hand, as automation levels rise in most factories and warehouses, it is becoming increasingly difficult to configure certain processes without a specialist’s involvement. Typically, these facilities employ staff without an engineering background in robotics, who can only perform their direct duties but are unable to create maps and routes for mobile robots. This article describes an alternative approach to creating a multi-purpose navigation map using mixed reality glasses, demonstrating the capabilities of Meta’s Meta Quest 3 glasses as a versatile mixed reality device for 3D room scanning and navigation mapping. In this paper, we demonstrate how a mixed reality approach can be applied to additional mapping evaluation for rooms of varying complexity and shape, including problematic areas for many similar solutions when the room contains mirrors and glass walls. Although environmental scanning in the Meta Quest 3 glasses is a standard integrated function of these glasses, it is used only for the collision avoidance system; this device can be successfully implemented in the Internet of Things system, and the collected data can be used for 3D scanning of the environment, creating navigation maps after additional data processing. This method expands the implementation of automated moving systems, including simplified map creation for mobile robots, in complex environmental and movement tasks. Full article

Neuroimaging is a promising modality for early AD detection, facilitating timely clinical intervention. This study proposes an enhanced deep learning framework that extracts critical AD biomarkers from structural MRI (sMRI) data acquired from the ADNI. Our novel CNN architecture integrates conventional convolutional layers with residual and skip connections for efficient feature extraction, achieving substantially lower computational cost than standard deep architectures such as VGG-16 (138 M), while remaining more parameter-intensive than highly compact architectures such as MobileNet and EfficientNet, which are designed explicitly for resource-constrained deployment. A comprehensive preprocessing pipeline converts 3D MRI scans into 2D slices through quality control (discarding slices with mean intensity < 5% of the maximum), bilinear resizing to 96 × 96 pixels, normalization using training-set statistics, and data augmentation. Stratified, subject-level data partitioning combined with robust statistical validation via bootstrapping demonstrates superior multiclass classification performance across AD, early and late MCI, and cognitively normal groups compared to state-of-the-art methods. Additionally, Grad-CAM-based interpretability maps were generated to highlight disease-relevant brain regions, confirming consistent activation around the hippocampus and temporal lobe. Full article

The pore-throat structure is a key factor in the exploration and development of tight sandstone reservoirs. In the present study, 14 tight sandstone samples from the Chang 8 member of the Ordos Basin were analyzed using high-pressure mercury intrusion, cast thin section analysis, scanning electron microscopy and cathodoluminescence imaging techniques. Fractal dimensions, obtained from the slopes of log(S_W) versus log(P_c) double-logarithmic plots, were applied to quantitatively characterize pore-throat structures and classify reservoirs through multifractal analysis, and discuss the diagenetic controlling factors affecting the pore-throat structure of different reservoir types. The results showed that the Chang 14 tight sandstones are characterized as two segments fractal features, which indicated that these samples have complex pore-throat structure and consist of two types of spaces: mesopore-throat spaces and micropore-throat spaces. The mesopore-throat system shows a higher fractal dimension (D1: 2.74–2.99), indicating greater heterogeneity and irregularity, while the micropore-throat system exhibits a lower dimension (D2: 2.28–2.61). D1 exhibits a negative correlation with the porosity and permeability of mesopores, while D2 shows a weak positive correlation with the properties of micropores. The total fractal dimension (D) is weakly correlated with overall reservoir properties, confirming that reservoir storage and flow capacity are primarily governed by the mesopore system rather than the micropore system. By analyzing the contribution of pore throats to sample physical properties, the results indicate that the 14 samples can be classified into two types based on 35% porosity contribution and 60% permeability contribution thresholds. Type 1, reservoirs dominated by microporous throat space (D values ranging from 2.603 to 2.644); Type 2, reservoirs dominated by mesoporous throat space (D values ranging from 2.544 to 2.598). Type 1 is characterized by primary intergranular pores, residual intergranular pores and intergranular dissolution pores, which enhance connectivity and reduce network complexity, thereby improving fluid permeability. In contrast, Type 2 consists mainly of intragranular dissolution pores, intergranular gap pores and micro-dissolution pores in clay minerals, which significantly inhibit fluid mobility. Diagenesis, including compaction, dissolution and cementation, exerts a significant control on the fractal characteristics and pore-throat structure evolution. The fractal characteristics exhibited in the pore-throat structure could provide a desirable analytical method, distinguishing from classification based on scale or size, for the evaluation and classification of tight sandstone reservoirs. Full article

Background/Objectives: Intraoperative margin assessments of oral squamous cell carcinoma (SCC) are fundamentally limited by sampling errors and freezing artifacts inherent to standard frozen section analysis. We developed a mobile, multi-scale, wide field-of-view (FOV) swept-source optical coherence tomography/microscopy (SS-OCT/OCM) system operating in the Near-Infrared-II (NIR-II) window (1.68 μm) to provide a rapid, non-destructive, volumetric evaluation of excised oral mucosal tissues. Methods: To correlate optical images with histopathology, we engineered a custom 3D-printed tissue cassette that physically mitigates macroscopic shrinkage during scanning and subsequent tissue fixation. A three-axis motorized assembly extends the effective imaging FOV without compromising resolution, while a custom 3D multi-resolution pyramid stitching algorithm synthesizes wide-FOV mosaics. Results: The customized cassette enabled precise, one-to-one spatial correlation between optical volumes and histopathology sections. Crucially, a 3 × 3 mosaic scan acquired with a 10× objective balanced imaging resolution and acquisition time, providing sufficient structural clarity to visualize basement membrane loss—a hallmark of SCC invasion. Conclusions: This 1.68 μm, fully automatic, multiscale SS-OCT/OCM platform demonstrates the feasibility of serving as a rapid, three-dimensional imaging tool for potential future use as an adjunct to conventional frozen sections. Full article

Accurate and efficient semantic segmentation of point cloud data is critical in many application areas involving indoor scene understanding. In particular, fine-grained object categories, high data density, and class imbalance in high-resolution indoor datasets significantly limit class discrimination in 3D semantic segmentation. The multimodal data structure, high-fidelity geometry, and long-tail class distribution of the recently popular ScanNet++ dataset further exacerbate these challenges. This study proposes a novel Long-Tail Aware Cross-Modal Graph Attention Network (LT-CM-GACNet++) to address fine-grained 3D semantic segmentation under long-tail distributions. The proposed method integrates dynamic graph-based geometric feature extraction with a lightweight visual feature extractor based on MobileNetV3, enabling effective fusion of geometric and RGB-based information. The proposed Cross-Modal Graph Attention (CMGA) module facilitates adaptive information transfer between modalities, enabling more effective representation learning of both local and global contextual features. To mitigate the adverse effects of long-tail class distributions, prototype-based representation learning and a class frequency-aware loss function are jointly employed. This strategy improves the learning of rare classes while enhancing the discrimination between visually and geometrically similar categories. In the preprocessing stage, density-based sampling, normal vector estimation, and block-based fixed-size point cloud generation are applied to high-resolution mesh-derived data. The proposed model is evaluated on 50 scenes and 100 semantic classes selected from the ScanNet++ dataset. Experimental results demonstrate that the proposed method achieves significant improvements over existing approaches in terms of both overall segmentation performance and rare-class performance. In particular, notable gains are observed in mean Intersection over Union (mIoU) and rare-class mIoU metrics. These results highlight the effectiveness of cross-modal learning for high-resolution 3D scene segmentation under long-tail distributions. Full article

The COVID-19 pandemic highlighted the need for effective disinfection strategies in order to minimize human exposure and reduce the risk of contagion in indoor environments. Ultraviolet-C (UV-C) irradiation has proven to be an effective solution for inactivating a wide range of pathogens. However, traditional fixed UV-C systems suffer from limited coverage and lack operational flexibility. To address these limitations, this paper proposes an augmented reality (AR)-based teleoperation framework for an omnidirectional mobile robot equipped with a UV-C disinfection light. Unlike traditional toolchain integrations, our framework synergizes immersive spatial visualization of a reconstructed environment, operator-guided waypoint-based remote navigation, and real-time interaction with the disinfection payload in a single operational workflow. The system is implemented using a ROSMASTER X3 Plus robotic platform, which generates a three-dimensional representation of the environment through visual simultaneous localization and mapping using RTAB-Map. The result is a 3D map that is imported into the Unity game engine and deployed to a Meta Quest 3 head-mounted display, enabling immersive visualization and interaction. Communication between the AR interface and the robotic system is achieved via the ROS-TCP-Connection, allowing real-time data exchange and remote robot control. Through the AR interface, the operator can navigate the robot within the scanned environment and activate the UV-C light. Experimental validation conducted in a classroom demonstrates the feasibility of the proposed approach and shows measurable reductions in surface microbial load. These results indicate that our system-level integration of AR-assisted teleoperation with mobile UV-C robotics represents a feasible proof-of-concept for flexible, operator-guided disinfection of indoor spaces. Full article