Search Results (395)

Monocular 3D reconstruction remains a persistent challenge for autonomous driving systems in Degraded Visual Environments (DVEs) with extreme glare and low illumination, such as highway tunnels, due to the lack of reliable texture cues. This paper proposes a physics-aware deep learning framework that overcomes these limitations by fusing polarization sensing with conventional intensity imaging. Unlike traditional end-to-end data-driven fusion strategies, we propose a Modality-Aligned Parameter Injectionstrategy. By remapping the weight space of the input layer, this strategy achieves a smooth transfer of the pre-trained Vision Transformer (i.e., MiDaS) to multi-modal inputs. Its core advantage lies in the seamless integration of four-channel polarization geometric information while fully preserving the pre-trained semantic representation capabilities of the backbone network, thereby avoiding the overfitting risk associated with training from scratch on small-sample data. Furthermore, we design a Reliability-Aware Gating mechanism that dynamically re-weights appearance and geometric cues based on intensity saturation and the physical validity of polarization signals as measured by the Degree of Linear Polarization (DoLP). We validate the proposed method on our self-constructed POLAR-GLV benchmark, a real-world dataset collected specifically for high dynamic range tunnel scenarios. Extensive experiments demonstrate that our method consistently outperforms intensity-only baselines, reducing geometric reconstruction error by 24.2% in high-glare tunnel exit zones and 10.0% at tunnel entrances. Crucially, compared to multi-stream fusion architectures, these performance gains come with negligible additional computational cost, making the framework highly suitable for resource-constrained onboard inference environments. Full article

For Autonomous Vehicles (AVs), recognizing traffic lights and signs is critical for safety because perception errors directly affect navigation decisions. Real-world disturbances such as glare, rain, dirt, and graffiti, as well as digital adversarial attacks, can lead to dangerous misclassifications. Current research lacks (i) temporal continuity (stable detection across consecutive frames to prevent flickering misclassifications), (ii) multi-field-of-view (FoV) sensing, and (iii) integrated defenses against both digital and natural degradation. This paper presents two principal contributions: (1) a three-layer defense framework integrating feature squeezing, inference-time temperature scaling (softmax $\tau$ = 3 without distillation training), and entropy-based anomaly detection with sequence-level temporal voting; (2) a 500 sequence dual-FoV benchmark (30k base frames, 150k with perturbations) from aiMotive, Waymo, Udacity, and Texas sources across four operational design domains. The unified defense stack achieves 79.8% mAP on a 100-sequence test set (6k base frames, 30k with perturbations), reducing attack success rate from 37.4% to 18.2% (51% reduction) and high-risk misclassifications by 32%. Cross-FoV validation and temporal voting enhance stability under lighting changes (+3.5% mAP) and occlusions (+2.7% mAP). Defense improvements (+9.5–9.6% mAP) remain consistent across native 3D (aiMotive, Waymo) and projected 2D (Udacity, Texas) annotations. Preliminary recapture experiments (n = 15 scenarios) show 2.5% synthetic–physical ASR gap (p = 0.18), though larger validation is needed. Code, models, and dataset reconstruction tools are publicly available. Full article

This article presents an analysis of the impact of atmospheric turbulence on the quality of images obtained during photogrammetric missions in Antarctica using a fixed-wing UAV operating in BVLOS mode. Image quality was evaluated primarily by the degree of blurring, which served as the main assessment criterion. In the Antarctic region, turbulence is a frequent phenomenon and can occur even under very light wind conditions, which formed the basis of this study. Autopilot log data were used to conduct a series of analyses, resulting in maps of areas where turbulence symptoms were recorded. In parallel, the quality of images captured during the mission was examined, producing a map of blurring levels assessed on a five-point scale. The study shows that UAV image blurring is mainly caused by sudden camera movements, mechanical vibrations from the propulsion system, and atmospheric turbulence that disrupts flight stability and overloads image stabilization. Additional factors such as low-light conditions, fog, haze, precipitation, glare, and moving shadows further reduce image clarity. Full article

High-speed solid particles ejected during battery thermal runaway pose severe safety threats, yet their velocity measurement is hindered by high density, microscopic size, and intense glare. This study proposes a non-intrusive velocimetry framework that integrates an enhanced single-stage object detector with a structural similarity matching algorithm. The detector incorporates specialized feature extraction modules and a high-resolution layer to identify microscopic targets in extreme environments, while the matching algorithm employs adaptive direction constraints to ensure precise trajectory tracking. Experimental validation demonstrates that the framework achieves a mean average precision of 92.7% and supports real-time processing. The method successfully quantifies a three-stage velocity evolution in battery failure events, identifying a peak particle speed exceeding 120 m/s. These findings provide critical kinematic data for optimizing battery safety structures and modeling fire propagation mechanisms. Full article

This article examines the potential risk of impaired visibility for drivers caused by sunlight reflecting off the surfaces of PV panels installed within the right-of-way of motorways and expressways. A literature review was conducted to describe the current state of knowledge and identify the requirements applicable to this area. The procedures for measuring the reflective properties of PV panels using a goniophotometer and a luminance camera (imaging luminance measuring device—ILMD) were evaluated. The measurement results for three PV panels with different surface structures are presented, allowing the properties of PV panels to be determined in terms of their potential impact on driver safety. A computer application was developed to determine whether the sun’s rays will reflect off a photovoltaic panel’s surface toward a vehicle’s direction of travel. The application graphically displays information on whether the sun will be reflected in the direction of a moving vehicle and whether this reflection poses a threat to driver safety. A comprehensive procedure for assessing the risk of glare caused by sunlight reflecting off the surfaces of photovoltaic panels was developed, along with detailed requirements. This study supports sustainable development by promoting renewable energy deployment in motorway corridors while simultaneously ensuring road safety. It integrates environmental (renewable energy use), social (driver safety), and technical (quantitative glare assessment methods) dimensions of sustainability. Full article

Extensive glazing is a common feature of modern buildings, intended to maximize daylight and strengthen visual connections with the outdoors. While this strategy can enhance energy performance, its effectiveness strongly depends on climate, orientation, and seasonal variations, and it often introduces challenges related to visual comfort, particularly glare. This paper proposes a refurbishment methodology that systematically integrates the view out, often neglected in current practice, into the decision-making framework, focusing on its relationship with daylight. The methodology follows a stepwise process encompassing the identification of discomfort conditions, evaluation of intervention feasibility, and design of targeted refurbishment strategies. Its main innovation lies in integrating and verifying a balance between view quality and daylight within a unified analytical framework. Validation through a university building in València confirmed that optimizing these parameters represents a significant design challenge, as enhancing one may compromise the other. The analysis also revealed limitations of current standards, such as EN 17037, whose static approach fails to capture the dynamic interactions among daylight, shading operation, and user perception. Furthermore, the proposed methodology introduces a scalable level of analytical granularity, enabling the assessment depth to be adapted to economic resources and time constraints, thereby supporting informed and sustainable decisions in building refurbishment. Full article

Digital twin, a technology that offers an opportunity to access dynamic and real-time data for efficient decision-making, has witnessed minimal utilization in smart facilities management. Additionally, combining stakeholders’ views with a digital twin provides more efficient building management. Thus, this study aimed to combine digital twin technology and stakeholders’ views to develop a best practice framework for enhancing indoor conditions of a typical university building. It analyses feedback received from building stakeholders and results from a digital twin to develop the best practice framework. This study adopted a case study approach by using a university library at Western Sydney University, Australia. It used a multi-stage approach and a series of interviews with facility management experts for the development and validation of the framework, respectively. The key findings revealed that all the monitored parameters in the digital twin system were within acceptable thresholds. However, the building occupants expressed concerns regarding excessive solar heat gain, inadequate airflow, and direct glare. It was also revealed that heat was the most disturbing environmental parameter in the library, and built-in energy efficiency measures were also not adequately maintained, contributing to the building’s energy consumption. The proposed framework provides strategic measures for improving building occupants’ comfort and energy consumption. Furthermore, the best practice framework aids facility managers in holistically considering key aspects of building services management in managing such buildings. Full article

Robust visual perception of Aids to Navigation (AtoN) is essential for Maritime Autonomous Surface Ships (MASS) operating in restricted navigational waters, where estuarine clutter, fog, glare, and dense traffic can severely degrade detection reliability. Existing maritime vision datasets largely emphasize open-sea targets or coarse AtoN categories, leaving a granularity gap for IALA-compliant fine-grained understanding in river–sea transition and port-approach channels. The River–Sea AtoN Navigation Dataset (RSAND) is introduced, a large-scale benchmark collected along the Yangtze River Deepwater Channel from inland corridors to open estuarine waters. RSAND contains 39,926 images with expert-verified bounding-box annotations and a hierarchical taxonomy that jointly captures AtoN location, shape, and functional semantics across 29 categories. To support both realistic long-tailed evaluation and standardized model comparison, two protocols are provided: RSAND-Full (29 categories) and RSAND-Balanced (10 critical categories). All quantitative benchmarking results in this paper are reported on RSAND-Balanced, while RSAND-Full is released for future large-scale long-tailed robustness studies. Benchmarking experiments on 14 state-of-the-art detectors demonstrate that YOLOv12x achieves superior performance with an mAP50-95 of 80.7%, significantly outperforming previous baselines. However, the analysis reveals persistent challenges in detecting small, distant targets and distinguishing visually similar functional markers. RSAND and the accompanying evaluation toolkit are released to facilitate reproducible research toward safer and smarter marine and coastal navigation. Full article

Facing 2050 climate uncertainties, enhancing building resilience is critical3. This study addresses the “black-box” and interoperability gaps in traditional multi-objective optimization (MOO) by proposing an intelligent framework based on the Model Context Protocol (MCP) and Large Language Models (LLMs). Unlike stochastic algorithms, the MCP-LLM framework uses semantic reasoning to bridge building performance simulation (BPS) engines like EnergyPlus 24.2.0 and Radiance 5.4. Through a case study of an educational building in Taiwan under the IPCC RCP 8.5 scenario, results show the framework improves optimization convergence speed by 55% compared to NSGA-II. The optimized shading system reduced peak cooling loads by 18.5% and annual EUI by 12.3%, while maintaining uncomfortable glare (DGP > 0.35) below 5% of annual hours. Crucially, the system provides explainable design logic via natural language, marking a shift from automated simulation to human-machine collaboration. This framework offers a transparent decision-support tool for forward-looking climate adaptation in educational environments. Full article

Glare is a critical factor in the design of LED luminaires for street lighting, particularly in environments where pedestrians, cyclists and drivers coexist. Generally, glare assessments are performed for fixed geometries and a single observer, limiting their applicability to real urban environments. This study examines the effect of angular redistribution of the beam on glare and illuminance by introducing the relative angular parameter α into the photometric model and the UGR calculation. A generic LED luminaire is modelled using a cosine-type luminous intensity distribution raised to a power, and the emitting surface is also discretized to evaluate the luminance, solid angle and Guth position index at the patch level. This approach is applied to three distinct observer geometries—pedestrian, cyclist and driver—allowing direct comparison using a unified mathematical formulation. The results show that beam redistribution affects each observer differently, reducing glare for pedestrians while simultaneously increasing it for drivers, whereas cyclists show limited sensitivity to angular changes. Although relative illuminance and UGR show similar monotonic trends, their physical and perceptual interpretation is different. This paper presents a novel tool for the preliminary analysis of trade-offs between visual comfort and luminous efficiency in urban lighting design. Full article

The automated localization of the flange interface in LNG tanker loading and unloading imposes stringent requirements for accuracy and illumination robustness. Traditional monocular vision methods are prone to localization failure under extreme illumination conditions, such as intense glare or low light, while LiDAR, despite being unaffected by illumination, suffers from limitations like a lack of texture information. This paper proposes an illumination-robust localization method for LNG tanker flange interfaces by fusing monocular vision and LiDAR, with three scenario-specific innovations beyond generic multi-sensor fusion frameworks. First, an illumination-adaptive fusion framework is designed to dynamically adjust detection parameters via grayscale mean evaluation, addressing extreme illumination (e.g., glare, low light with water film). Second, a multi-constraint flange detection strategy is developed by integrating physical dimension constraints, K-means clustering, and weighted fitting to eliminate background interference and distinguish dual flanges. Third, a customized fusion pipeline (ROI extraction-plane fitting-3D circle center solving) is established to compensate for monocular depth errors and sparse LiDAR point cloud limitations using flange radius prior. High-precision localization is achieved via four key steps: multi-modal data preprocessing, LiDAR-camera spatial projection, fusion-based flange circle detection, and 3D circle center fitting. While basic techniques such as LiDAR-camera spatiotemporal synchronization and K-means clustering are adapted from prior works, their integration with flange-specific constraints and illumination-adaptive design forms the core novelty of this study. Comparative experiments between the proposed fusion method and the monocular vision-only localization method are conducted under four typical illumination scenarios: uniform illumination, local strong illumination, uniform low illumination, and low illumination with water film. The experimental results based on 20 samples per illumination scenario (80 valid data sets in total) show that, compared with the monocular vision method, the proposed fusion method reduces the Mean Absolute Error (MAE) of localization accuracy by 33.08%, 30.57%, and 75.91% in the X, Y, and Z dimensions, respectively, with the overall 3D MAE reduced by 61.69%. Meanwhile, the Root Mean Square Error (RMSE) in the X, Y, and Z dimensions is decreased by 33.65%, 32.71%, and 79.88%, respectively, and the overall 3D RMSE is reduced by 64.79%. The expanded sample size verifies the statistical reliability of the proposed method, which exhibits significantly superior robustness to extreme illumination conditions. Full article

The building sector accounts for approximately 30% of global energy use. The demand for energy-efficient, high-performance buildings is increasing given the increasing awareness of the climate crisis. The building envelope greatly influences overall building energy performance. Considering the broad shift from passive to adaptive systems, smart window technologies are attracting attention. Despite their potential, few scholars have examined occupant comfort in spaces with smart windows. This gap is addressed herein by comparatively analyzing occupants’ responses to thermal and visual environments in a room with a smart window (RoomSW) and a room with a conventional window (RoomCW) in a residential building in winter. The smart window is operated via a glare-prevention tint control strategy. The results reveal that under thermal conditions comparable to those in an actual dwelling, wintertime smart window tinting for glare prevention does not decrease occupants’ thermal sensation or satisfaction. Regarding visual comfort, conditions in RoomSW and RoomCW satisfy the minimum illuminance requirement of 200 lx, but glare occurs in RoomCW with a mean New Daylight Glare Index (DGI_N) of 24.1, compared to 9.6 in RoomSW. Questionnaire results indicate greater satisfaction with the luminous environment in RoomSW relative to RoomCW, with scores of +1.4 and +0.2, respectively. Full article

Silica nanoparticles (SNPs) are pivotal in designing functional optical films, but accurately modeling their properties is hindered by the limitations of classical effective medium theories, which break down for larger particles and complex morphologies. We introduce a robust, effective medium theory that overcomes these limitations by incorporating full Mie scattering solutions, thereby accounting for size-dependent and multipolar effects. Our model is comprehensively developed for unshelled, shelled, mixed, and hollow SNPs randomly dispersed in a host medium. Its accuracy is rigorously benchmarked against 3D finite-element method simulations. This work establishes a practical and reliable framework for predicting the optical response of SNP composites, significantly facilitating the rational design of high-performance coatings, such as anti-glare layers, with minimal computational cost. Full article

The increasing adoption of highly glazed façades in contemporary office building has improved daylight penetration but has also intensified glare risk and sunlight overexposure in Mediterranean climates, with direct implications for occupant visual comfort and environmental sustainability. While daylight optimization has been widely discussed, fewer studies have examined how façade morphology systematically shapes the balance between daylight sufficiency and visual comfort in Mediterranean island contexts. This study investigates the relationship between façade configuration, daylight availability, and glare performance in office buildings in Northern Cyprus using climate-based daylight simulation. Six façade morphologies are evaluated across a range of window-to-wall ratios (WWR) using EN 17037-aligned criteria and metrics, including spatial daylight autonomy (sDA), annual sunlight exposure (ASE), and daylight glare probability (DGP). Usable daylight is not simply a function of more glass. As WWR increases, fully glazed façades in Mediterranean conditions tend to admit excessive direct sun and intensify glare, so daylight becomes less workable even when illuminance is high. Instead, hybrid and adaptive morphologies that control lighting through a combined approach of shade, diffusion, and redirection provide the most dependable performance, reducing both overexposure and glare while ensuring sufficient daylight sufficiency. The findings also indicate a distinct turning point at about 50–55% WWR, beyond which performance is mostly dependent on the façade’s ability to modulate its morphology and further glass offers minimal advantage. Based on this, the article suggests a contextual framework to encourage façade options for Mediterranean office environments that are more sustainable, aesthetically pleasing, and climate-responsive. Full article

The Daylight Glare Probability (DGP) includes the luminance of a glare source quadratically, but the solid angle only linearly. While this is in line with formulae of other glare metrics, it must be questioned for small glare sources, if the glare stimulus can no longer be distinguished from larger stimuli causing equal vertical illuminance at the eye, especially in the peripheral visual field. To account for this, the modified version Daylight Glare Metric (DGM) was previously developed. We conducted two studies to evaluate the effect of the modified DGM. First, in a laboratory study under an artificial sky with an LED sun, 35 test subjects evaluated different glare situations. Second, we performed a comprehensive simulation study for an office space, including three locations, three view directions, and 17 window systems (electrochromic glazing, fabric shades). The results from the perception study under the artificial sky provide evidence that the adapted DGM is better suited to predict glare from small, bright sources. The results from the simulation study for a realistic office setting show that, compared to the DGP, the DGM reduces glare ratings for many hours of the year, thus underscoring the practical relevance of improving the DGP formula. Full article